Axisvm Manualx6 en
Axisvm Manualx6 en
USE R ’S M AN UAL
Copyright © 1991-2021 Inter-CAD Kft. of Hungary. All rights reserved. No part of this publication may be
Copyright
reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechan-
ical, photocopying, recording or otherwise, for any purposes.
Disclaimer The material presented in this text is for illustrative and educational purposes only and is not intended to
be exhaustive or to apply to any particular engineering problem for design. While reasonable efforts had
been made in the preparation of this text to assure its accuracy, Inter-CAD Kft. assumes no liability or
responsibility to any person or company for direct or indirect damages resulting from the use of any in-
formation contained herein.
Changes Inter-CAD Kft. reserves the right to revise and improve its product as it sees fit. This publication describes
the state of this product at the time of its publication and may not reflect the product at all times in the
future.
Version This is an International Version of the product that may not conform to corresponding standards in a
respective country and is available solely on an “as is” basis.
Limited warranty Inter-CAD Kft. makes no warranty, either expressed or implied, including but not limited to any implied
warranties of merchantability or fitness for a particular purpose, regarding these materials.
In no event shall Inter-CAD Kft. be liable to anyone for special, collateral, incidental, or consequential
damages in connection with or arising out of purchase or use of these materials. The sole and exclusive
liability to Inter-CAD KFT., regardless of the form of action, shall not exceed the purchase price of the
material described herein.
Technical support If you have questions about installing or using the AxisVM, check this User’s Manual first - you will find
and services answers to most of your questions here. If you need further assistance, please contact your software pro-
vider.
User’s Manual X6 Release1 e2 3
CONTENTS
General
Smart command line and search 2.17 Smart command line
Special window layouts in the Windows menu 3.6.6 Special window layouts
Enhanced selection filter 2.16.1 Selection
Logical parts by type of special domains 2.16.14 Parts
Customization of further graphic symbols 3.3.11.3 Graphic symbols
Editing
Moving virtual strips with the underlying elements
Copying system of structural gridlines 3.3.5 Guidelines
Mesh check selecting free edges 4.11.4 Select free edges
Elements
Parametric double wedged I, crossed I, SFB, IFB, symmetric and asymmetric HSQ 3.1.16 Cross-Section Library
profiles
Loads
Load cases can be selected from a structured dropdown list on the Loads tab 4.10 Loads
New load split functions 4.10.33 Split loads into separate load
cases
Displaying color-coded surface loads of all load cases of a load group in one view 2.16.5 Display mode
Generating wind loads from a CFD model (new CFD module) 4.10.15 Generating wind load from a CFD
(Computational Fluid Dynamics) file –
CFD module
Displaying color-coded snow and wind loads
Applying eccentric concentrated or distributed loads on beams and ribs 4.10.4 Concentrated load on beam
4.10.6 Distributed line load on beam/rib
Options to handle multiple loads 3.3.2 Model options
Default color of surface and line loads can be customized 3.3.11.3 Graphic symbols
Rules to generate custom combinations in the table of critical load group combi- 4.10.2 Load combinations
nations
Defining the direction of spectra for seismic analysis 4.10.24 Seismic loads – SE1 module
12
Analysis
Obtaining imperfect shape from buckling shape(s) (New IMP module) 6.2.1 Creating imperfect geometry from
buckling shapes – IMP module
Nonlinear analysis with reinforcement calculated from the ULS envelope of 5.1 Static analysis
nonlinear results
Results
Result display in rendered view 2.16.4 Display mode
6.1 Static
Rendered view can display distribution of normal stress within the cross-section 6.1.21 Truss/beam/rib stresses
of a beam
Rendered view can display top and bottom components of surface stresses, rein- 6.1.22 Surface element stresses
forcement amounts, crack width values simultaneously (RC1 module)
New semi-automatic critical combinations 6.1 Static
Selection filtering by any result component 2.17 Smart command line
Bicolor (positive/negative) color display mode 2.19.4 Color legend window
New color legend option to hide hatching of values out of range 2.19.4 Color legend window
New color legend option 6.1.1 Minimum and maximum values
Design
Support of Norwegian NAD for Eurocode
Cracking results both for calculated and applied reinforcement (RC1 module) 6.5.6 Nonlinear analysis of RC surfaces
Calculating required reinforcement for reinforced trapezoid steel decks. (RC1 6.5.1.4 Design of reinforced concrete cores
module) and walls – RC5 module
Speed improvements in reinforced concrete beam design calculation by running 6.5.10 Composite column design – RC2
multiple threads (RC2 module) module
Shear design of reinforced concrete walls (RC5 module) 6.5.14.4 Shear verification of reinforced
concrete walls
Reports
New table for specific amount of reinforcement calculated for each domain (RC1
module)
Utilization result component and diagrams for reinforced concrete columns, walls 6.5.8 Shear resistance and reinforcement
and masonry walls in the Table Browser (RC2, RC5, MD1 modules) calculation for plates and shells
Utilization result component and diagrams for reinforced concrete columns (RC2
module)
Predefined window layouts for checking column reinforcement (RC2 module)
Custom force values to check a reinforced concrete column are saved with the
column (RC2 module)
Custom forces for column check can be imported via clipboard (RC2 module)
Design calculation report for reinforced concrete columns (RC2 module)
Steel design calculation report includes effective cross-section parameters and
lateral supports (SD1 module)
New features of the bolted joint design (SC1 module) 6.6.4 Steel connection design - SC1 mod-
ule
User’s Manual X6 Release1 e2 13
Postprocessing Displaying the results: deformed/undeformed shape display; diagram, and iso-line/surface plots;
animation; customizable tabular reports.
After your analysis, AxisVM provides powerful visualization tools that let you quickly interpret your
results, and numerical tools to search, report, and perform further calculations using those results. The
results can be used to display the deformed or animated shape of your geometry or the isoline/surface
plots. AxisVM can linearly combine or envelope the results.
Reporting Reporting is always part of the analysis, and a graphical user interface enhances the process and
simplifies the effort. AxisVM provides direct, high quality printing of both text and graphics data to
document your model and results. In addition data and graphics can be easily exported (DXF, BMP, JPG,
WMF, EMF, RTF, HTML, TXT, DBF).
The table below shows the minimum/recommended hardware and software requirements, so you can
experience maximum productivity with AxisVM.
Recommended 8 GB RAM
configuration 50 GB of free hard disk space
17” color monitor (or larger), at least 1920x1080 pixels
A dual or multi-core processor over 2 GHz
Windows10 64-bit operating system
Mouse or other pointing devices
Windows compatible laser or inkjet printer
Minimal 4 GB RAM
configuration 10 GB of free hard disk space
15” color monitor, at least 1024x768 resolution
Mouse
Windows 7
Supported operat- Windows 10, Windows 8.1, Windows 8, Windows 7, Windows Server 2008 (32-bit or 64-bit operating
ing systems systems)
Partially supported (some add-ons or plugins may fail): Windows Vista, Windows XP/ SP3, Windows
Server 2003/SP1
Memory access To reach more memory is very important as it speeds up the analysis considerably.
64-bit and 32-bit The native 64-bit version of AxisVM X6 runs only on 64-bit operating systems. It has direct access to
versions the physical memory, so no further settings are required.
The 32-bit version of AxisVM X6 runs on both 32-bit and 64-bit operating systems. It has direct access
to the lower 2 GB of the physical memory.
14
Using a protection by hardware key the software can be installed on multiple computers but can be
launched only if the hardware key is plugged in. Plug the key only after completing the installation as
certain operating systems try to recognize the plugged device and this process may interfere with the
driver installation.
Super Pro The most common protection device is the Sentinel SuperPro hardware key available as a single or a
hardware key multi-user network license. A single license can be used on one computer only (on Windows Terminal
Server networks on one terminal only).
The operation of the hardware key requires a driver. It is automatically installed by the AxisVM Installer
but in case of a failure the installation can be done later.
After completing the AxisVM installation look for the Sentinel folder and run Sentinel Protection Istaller
from there. If you have an installation USB flash drive run Startup.exe and choose Reinstall driver.
Super Pro If you have a network license you must install the network key. In most cases AxisVM and the key are on
network different computers but to make the key available through the network the Sentinel driver must be
hardware key installed on both computers using the Sentinel Protection Installer. Network keys can be used both on
local networks (LAN) and remote networks (VPN).
Connect the key to the USB port of one of the computers. This way you select the AxisVM server.
After completing the installation a service called Sentinel Protection Server is launched to establish a
connection between the hardware key plugged into the server machine and the client machines.
Network licenses allow running as many AxisVM copies on the network as the purchased number of
licenses. Launched copies are registered into the hardware key. Closing the copy will remove the
registration making the license available on other machines. If the application becomes unresponsive
for some reason the registration is not removed immediately. The license will be available only after 3
minutes when the lack of response is detected by the server.
The hardware key manufacturer provides a tool called Sentinel License Monitor to check the currently
registered network licenses, the number of free licenses and remove the unresponsive licenses if
necessary. To run this tool start you internet browser and type in the local network IP address of the
server (i.e., the machine where the hardware key is plugged in) and append a reference to the port 6002,
e.g.: http://192.168.18.167:6002. If you run the internet browser on the machine where the hardware key
is plugged in you can type simply http://localhost:6002.
FIREWALL SETUP: Enable TCP / IP and UDP protocols on ports 6001 and 6002 on the firewall to
ensure network operation.
User’s Manual X6 Release1 e2 15
To run AxisVM on any computer on the network SuperPro Server must be running on the server. If it
stops all running AxisVM programs stop.
Sentinel HL The Sentinel HL key is a new kind of protection device that does not require driver installation. Plugging
hardware key it into the USB port the Windows operating system identifies it as a HID (Human Interface Device).
Sentinel HL keys also have single user and network models. A single license can be used on one
computer only (on Windows Terminal Server networks on one terminal only).
Sentinel HL Network keys can be used on local (LAN) or remote networks (VPN) as well. Connect the key to a
network machine to use it as a key licence server and install AxisVM copies on client machines. Sentinel Runtime
hardware key Environment must be installed both on the server and on all clients by launching InstallSentinelRuntime
located in the Sentinel LDK folder under the AxisVM folder created during program installation.
The Sentinel Runtime Environment provides a connection between the hardware key plugged into the
server machine and the client machines. If the environment is not running AxisVM cannot be launched
on client machines.
To purchase further licenses for a network key run the Admin Control Center (see below), choose Sentinel
Keys from the menu on the left and choose the appropriate network key from the list. Click on the C2V
icon on the right to save a c2v file. Send this file to your distributor with your license order. Your
distributor will send a v2c file with the additional licenses. Save it to your machine. To update the number
of licenses choose Update/Attach from the menu on the left then browse for the v2c file you saved. Click
the Open button to transfer license information to the network key. The new licenses will become
available.
FIREWALL SETUP: Enable TCP / IP and UDP protocols on port 1947 on the firewall to ensure net-
work operation.
Configuration of To ensure the proper functioning of a Sentinel HL key, the Sentinel Admin Control Center must be con-
server and client figured on server and client machines.
computers
Server
Open the web browser and type localhost:1947 to get to the Sentinel Admin Control Center. Go to the
Access from Remote Clients tab and set Anyone under Allow Access from Remote Clients.
16
Client
Open the web browser and type localhost:1947 to get to the Sentinel Admin Control Center. Go to the
Access to Remote License Manager tab and enable Allow Access to Remote Licenses, Broadcast Search
for Remote Licenses and Aggressive Search for Remote Licenses. Enter the IP address of the server to
the Remote License Search Parameters field and click on the Submit button.
If both the server and the clients are properly configured, choosing Sentinel Keys from the Options list
on the left displays a list of all Sentinel HL and SL keys installed.
User’s Manual X6 Release1 e2 17
Software key protection by an installed license replaces the hardware key plugged into a USB port.
Software key licenses are also available in a local or a network version. A local license allows running
AxisVM on one computer, where the software key is installed. A network license allows a limited number
of concurrent users on the local network.
Software keys can be transferred to another machine using a tool (see later). After transferring a soft-
ware license, the program will run only on the other machine and fails to run on the first one.
A software license is linked to the hardware elements of the computer. Therefore, any change in
the hardware (replacement of the motherboard, CPU, Ethernet card, hard drive, will invalidate
the license.
To prevent such loss, the license must be transferred temporarily to another machine before changing
the hardware, replace the hardware element then transfer the license from the other machine
Sentinel SL Sentinel SL is not a physical device attached to a port but a software key specific to the given hardware.
software key It cannot be copied directly but can be transferred to another machine using the RUS tool of the pro-
tection package (see details below).
Software key may be a reasonable solution if storage, connection or protection of the physical key raises
problems.
As the software key is stored on the hard drive / SSD of the machine, viruses, malware or flaws in the
update of the operating system may corrupt the key. In this case another software key must be pur-
chased.
Installing Sentinel A software key requires the installation of the Sentinel Run-time Environment.
HASP Run-time En- Download it from www.axisvm.eu (find it under Downloads / Software Updates / Sentinel driver) or look
vironment for InstallSentinelRuntime.exe in the Sentinel LDK folder of your AxisVM installation.
Sentinel Admin To check the status of the environment run your internet browser on the server then type http://lo-
Control Center calhost:1947. The Sentinel Admin Control Center will be displayed.
18
The Sentinel Admin Control Center provides overview of the Sentinel HL and SL keys, their usage infor-
mation and helps disconnecting clients from licenses. Choose Sentinel Keys form the menu on the left
to display the list of available keys on the network.
If more than one hardware key is plugged and you are not sure which one belongs to AxisVM select
Products from the menu on the left. The list of keys will be reorganized by protected products. Look for
the one protecting AxisVM.
Requesting a A software key requires the installation of the Sentinel Run-time Environment. See Installing Sentinel
Sentinel SL HASP Run-time Environment above for the details. After completing the installation run the RUS_Ax-
software license
isVM_En.exe application from the SentinelRuntime subfolder in the AxisVM installation folder.
To request a software license, select the option Installation of new protection key at the bottom and click
on the Collect information button.
Save the file with c2v extension generated by this application and send it attached to a license request
email to your AxisVM distributor.
Applying a Software license is a v2c file sent by your distributor. Save it to your computer.
Sentinel SL Run the RUS_AxisVM_En.exe application from the SentinelRuntime subfolder in the AxisVM installation
software license
folder. Choose the Apply License File tab and enter the file name with full path to the Update File field
or click on the […] button to browse.
Clicking on the Apply Update button activates the software license.
Check your license using the Admin Control Center tool (see above). Choose Sentinel Keys from the
menu on the left, and check if your Sentinel SL license appears in the list.
Sentinel SL A software license can be transferred to another machine. After completing the transfer AxisVM will run
License transfer only on the other machine (works just like a hardware protection key removed and plugged into the
other machine). To transfer the license run the RUS_AxisVM_En.exe application on the other machine
too, select the Transfer License tab and generate a recipient information file (*.id) and copy that file to
the first machine. Go back to the first machine, select the Transfer License tab, go to the Read the recip-
ient information file from field, click on the […] button and load the *.id file you have copied from the
other machine, fill out the Generate the license transfer file to field then click on the Generate License
Transfer File button. Copy the generated file with h2h extension to the other machine, go there, select
the Apply License File tab, go to the Update File field, browse for the h2h file you have copied and click
on Apply Update.
Configuration of the To ensure the proper functioning of a software key, the Sentinel Admin Control Center must be config-
server and client ured on server and client machines the same way described at the end of the previous chapter.
machines
2.2.3. Installation
Preparing AxisVM Most professional structural analysis software programs support only Windows. The following methods
installation will enable AxisVM use on a Mac:
on a Mac
1. Boot Camp
Apple’s Boot Camp allows for installing Windows on a Mac. This will provide the best performance for
running AxisVM on a Mac.
2. Virtual Machines
Software such as Parallels or VMWare Fusion use virtual machines to install Windows on a Mac, allowing
AxisVM to run.
3. Remote Desktop
The Remote Desktop software on a Mac computer allows for remote access of a Windows computer
running AxisVM.
Installation If you have downloaded the installer application from our website, run it. Otherwise, connect the AxisVM
installation USB flash drive. The Startup program starts automatically if the autoplay option is enabled.
If autoplay is not enabled, click the Start button, and select Run.... Open the Startup.exe program from
the USB flash drive. Select AxisVM X6 Setup and follow the instructions.
User’s Manual X6 Release1 e2 19
By default the application and the example models will be installed on drive C: in
C:\AxisVM_X6 and C:\AxisVM_X6\Examples
folders. You can specify the drive and the folders during the installation process. The setup program
creates the AxisVM program group that includes the AxisVM application icon.
The application can be installed to the usual C: \ Program Files \ AxisVM_X6 folder (C: \ Program Files
(x86) \ AxisVM_X6 under 64-bit operating systems). However, in this case the Run as administrator prop-
erty must be set for AxisVM.exe, AxisVM_x64.exe \ IDTFConverter \ DTFConverter.exe. Find these files
through Start Menu / Computer, right click on the files and choose Properties from the popup menu, go
to the Compatibility tab, find Privilege level and turn on the above option. Users without administrative
rights has to ask the administrator to set write access to the C: \ Program Files \ AxisVM_X6 folder (see
Permissions under the Security tab).
32/64-bit versions On 64-bit operating systems the user can choose to install either the 32-bit or the 64-bit version of
AxisVM. Installing the 64-bit version also copies the 32-bit version to the hard disk but no shortcut is
created on the desktop for this file. On 32-bit operating systems only the 32-bit version is installed.
It is not recommended to install AxisVM under the c:\Program Files folder as the program placed
there can be started only with administrative rights and there may be failures in running libraries
like the 3D PDF generator.
Silent installation The software can be installed in silent mode using command line. The Setup program accepts optional
command line parameters. These can be useful to system administrators, and to other programs calling
the Setup program.
Two typical examples are shown below considering 32 and 64-bit versions:
AxisVM setup is made by Inno Setup installer. The most important optional command line parameters
are listed below. For more information please see the Inno Setup help.
/SP-
Disables the This will install... Do you wish to continue? prompt at the beginning of Setup. Of course,
this will have no effect if the DisableStartupPrompt [Setup] section directive was set to yes.
/SILENT, /VERYSILENT
Instructs Setup to be silent or very silent. When Setup is silent the wizard and the background window
are not displayed but the installation progress window is. When a setup is very silent this installation
progress window is not displayed. Everything else is normal so for example error messages during in-
stallation are displayed and the start-up prompt is (if you haven't disabled it with DisableStartupPrompt
or the '/SP-' command line option explained above).
If a restart is necessary and the '/NORESTART' command isn't used (see below) and Setup is silent, it
will display a Reboot now? message box. If it's very silent it will reboot without asking.
20
/SUPPRESSMSGBOXES
Instructs Setup to suppress message boxes. Only has an effect when combined with '/SILENT' and
'/VERYSILENT'.
/LOG
Causes Setup to create a log file in the user's TEMP directory detailing file installation and [Run]
actions taken during the installation process. This can be a helpful debugging aid. For example, if you
suspect a file isn't being replaced when you believe it should be (or vice versa), the log file will tell you
if the file was really skipped, and why.
The log file is created with a unique name based on the current date. (It will not overwrite or append
to existing files.)
The information contained in the log file is technical in nature and therefore not intended to be
understandable by end users. Nor is it designed to be machine-parsable; the format of the file is subject
to change without notice.
/LOG="filename"
Same as /LOG, except it allows you to specify a fixed path/filename to use for the log file. If a file with
the specified name already exists it will be overwritten. If the file cannot be created, Setup will abort
with an error message.
/NOCANCEL
Prevents the user from cancelling during the installation process, by disabling the Cancel button and
ignoring clicks on the close button. Useful along with '/SILENT' or '/VERYSILENT'.
/NORESTART
Instructs Setup not to reboot even if it's necessary.
/LANG=language
Specifies the language to use. language specifies the internal name of the language as specified in a
[Languages] section entry.
When a valid /LANG parameter is used, the Select Language dialog will be suppressed.
/DIR="x:\dirname"
Overrides the default directory name displayed on the Select Destination Location wizard page. A
fully qualified pathname must be specified.
User’s Manual X6 Release1 e2 21
/GROUP="folder name"
Overrides the default folder name displayed on the Select Start Menu Folder wizard page. If the
[Setup] section directive DisableProgramGroupPage was set to yes, this command line parameter is
ignored.
/NOICONS
Instructs Setup to initially check the Don't create a Start Menu folder check box on the Select Start
Menu Folder wizard page.
/TYPE=type name
Overrides the default setup type.
If the specified type exists and isn't a custom type, then any /COMPONENTS parameter will be ig-
nored.
Only the specified components will be selected; the rest will be deselected.
If a component name is prefixed with a "*" character, any child components will be selected as well
(except for those that include the dontinheritcheck flag). If a component name is prefixed with a "!"
character, the component will be deselected.
This parameter does not change the state of components that include the fixed flag.
Example:
Deselect all components, then select the "help" and "plugins" components:
/COMPONENTS="help,plugins"
Example:
Deselect all components, then select a parent component and all of its children with the exception of
one:
/COMPONENTS="*parent,!parent\child"
Only the specified tasks will be selected; the rest will be deselected. Use the /MERGETASKS parameter
instead if you want to keep the default set of tasks and only select/deselect some of them.
If a task name is prefixed with a "*" character, any child tasks will be selected as well (except for those
that include the dontinheritcheck flag). If a task name is prefixed with a "!" character, the task will be
deselected.
Example:
Deselect all tasks, then select the "desktopicon" and "fileassoc" tasks:
/TASKS="desktopicon,fileassoc"
Example:
Deselect all tasks, then select a parent task and all of its children with the exception of one:
/TASKS="*parent,!parent\child"
22
Running add-ons You can use add-ons created for AxisVM by external developers – or yourself. To run these programs
the AxisVM COM server must be registered in the Windows Registry. If you installed AxisVM with ad-
ministrative rights this registration is already completed. If the registration failed you can run! Regis-
ter_AxisVM.bat (on 32-bit operating systems) or !Register_AxisVM_x64.bat (on 64 bit operating systems)
with administrative rights.
32-bit add-ons launch the 32-bit version and are compatible only with that. 64-bit add-ons can
be used only on 64-bit versions.
False virus alarms Certain antivirus products running on the PC can send a false alarm during installation. This is caused by
heuristic algorithms searching for virus-like activities. These algorithms may detect the operation of the
special protection system of AxisVM and send a false alarm. If this happens you can do the following
• If the antivirus product put AxisVM.exe into quarantine restore it
• Add AxisVM.exe to the exceptions (files not checked by the antivirus software)
• Reduce the sensitivity of the heuristic check on the control panel of the antivirus product
The VirusTotal website offers antivirus check of 47 different products.
Starting AxisVM Click the Start button, select Programs, AxisVM folder, and click the AxisVM X6 icon.
At startup a splash screen is displayed (see 3.7.5 About) then a welcome screen is shown where you can
select a previous model or start a new one.
Clearing the checkbox at the bottom turns the welcome screen off for the future. To turn it on choose
the Settings\Preferences\Data Integrity dialog and check the Show welcome screen on startup checkbox.
When setting up a new model the following dialog is displayed.
User’s Manual X6 Release1 e2 23
Running AxisVM Both AxisVM.exe (32-bit version) and AxisVM_x64.exe (64-bit version) can be started in safe mode en-
in safe mode tering axisvm.exe /SAFE or axisvm_x64.exe /SAFE in the command line.
It is recommended to start AxisVM in safe mode in the following cases: (1) graphic card or driver prob-
lems, (2) if problems are detected in multithreaded mode (3) if AxisVM hangs when trying to recover
the latest file damaged in a crash (4) if a plugin or addon module causes errors.
Launching AxisVM If a problem is encountered during startup or AxisVM cannot be launched, run AxisVM.exe (32-bit ver-
in debug mode sion) or AxisVM_x64.exe (64-bit version) with the /USERDEBUG and /NETDONGLELOG switches. In the
case of a network license, first, launch AxisVM with /USERDEBUG then with /NETDONGLELOG. In case
of a local single license, it is enough to use the /USERDEBUG switch.
These switches activate logging of the startup process, creating a log file. Send the log file to your
distributor to find how to fix the problem.
Upgrading It is recommended to install the new version to a new folder. This way the previous version will remain
available.
Converting earlier Models created in a previous versions are recognized and converted automatically. Saving files will use
models the latest format by default. Saving files in the file format of one of the previous is possible but this way
the information specific to the newer versions will be lost.
Capacity Practically, the model size is limited by the amount of free space on your hard disk.
The restrictions on the model size and on the parameters of an analysis are as follows:
Professional
Entity Maximum
Nodes Unlimited
Materials Unlimited
Elements Truss Unlimited
Beam Unlimited
Rib Unlimited
Membrane Unlimited
Plate Unlimited
Shell Unlimited
Support Unlimited
Gap Unlimited
Diaphragm Unlimited
Spring Unlimited
Rigid Unlimited
Link Unlimited
Load cases Unlimited
Load combinations Unlimited
Frequencies Unlimited
Small Business
Entity Maximum
Nodes Unlimited
Materials Unlimited
Elements Only trusses 500
Truss+Beam+Rib * 250
Rib on the edge of a surface 1500
Any combination of membrane, 2000
plate or shell
Support Unlimited
Gap Unlimited
Diaphragm Unlimited
Spring Unlimited
Rigid Unlimited
Link Unlimited
Load cases Unlimited
Load combinations Unlimited
The best way to discover AxisVM is to build a simple model, analyse it and evaluate the results.
The steps of model building are described in the AxisVM Step by step tutorial.
There are three major steps in a modeling process:
Geometry The first step is to create the geometry model of the structure (in 2D or 3D).
Geometry can be drawn by hand or can be imported from other CAD programs. It is also possible to
draw elements (columns, beams, walls, slabs) directly.
Elements If you chose to draw the geometry first you must specify material and element properties, mesh the
geometry into elements (assigning the properties and a mesh, to the wire-frame model), and define the
support conditions.
User’s Manual X6 Release1 e2 25
Loads In the third step you must apply different loads on the model.
The end result will be a finite element model of the structure.
Once the model is created it is ready for analysis.
You can find the steps of model building in detail in the AxisVM Step by step tutorial made available
from the Help menu. This tutorial helps to build and analyse simple structures.
• Beam model
• Frame model
• Plate model
• Membrane model (with parametric mesh)
• Membrane model (with domains)
• Shell model
AxisVM Step by step (for advanced users) guides the user through dynamic and seismic analysis of two
advanced models.
English language video tutorials for the example models can be found here:
AxisVM YouTube channel
It is recommended that you read the entire User’s Manual at least once while exploring AxisVM.
In Chapter 1 you can find the timely, new features of the version.
Chapter 2 contains general information about using AxisVM. In other chapters the explanation follows
the pre- and postprocessor menu structures. Please, consult this User’s Manual every time you are using
AxisVM.
This section describes the working environment of the full AxisVM graphical user interface. Please read
these instructions carefully. Your knowledge of the program increases the modeling speed and
productivity.
AxisVM screen Elements of the user interface
26
Unicode is a computing industry standard for the consistent encoding, representation and handling of
text expressed in most of the world's writing systems. AxisVM provides full Unicode support.
All windows appear according to the current Windows theme.
Graphics cursor As you move your mouse, the graphics cursor symbol tracks the movement on the screen. To select an
entity, an icon or menu item, move the cursor over it and click the left mouse button. The shape of the
cursor will change accordingly (see 4.7.1 Cursor identification), and will appear on the screen in one of
the following forms:
If you pick an entity when the cursor is in its default mode (info mode), the properties of that entity will
be displayed as a tooltip.
User’s Manual X6 Release1 e2 27
Tooltip Depending on the menu your cursor is on, you may get the properties of the following entities:
[Ctrl] + Moves the graphics cursor in the current plane with a step size enlarged/reduced by a factor set in the
Arrow keys, Settings dialog box.
[Shift]+ Moves the graphics cursor in the current plane on a line of angle n· , custom or +n·90°.
[][][][→],
[Home] [End] Moves the graphics cursor perpendicular to the current plane.
[Ctrl]+ Moves the graphics cursor perpendicular to the current plane with a step size enlarged/reduced by a
[Home], [End] factor set in the Settings dialog box.
right button Displays the Quick Menu. See 2.7 Quick Menu
28
Keyboard shortcuts can be assigned to common operations, toolbar buttons, menu items. The default settings of AxisVM are
the following. The main toolbar button shortcuts are context-sensitive, i.e., the same shortcut can perform different tasks on
different tabs (Geometry, Elements, Loads, etc.).
General commands
Open ............................................. Ctrl+O
Save ............................................. Ctrl+S
Print ............................................ Ctrl+P
Undo ............................................. Alt+BkSp
Redo ............................................. Shift+Alt+BkSp
Layer Manager .................................... F11
storeys .......................................... F7
Table Browser .................................... F12
Report Maker ..................................... F10
Drawings Library ................................. F6
Save to Drawings Library ......................... F9
Delete ........................................... Del
Set relative origin .............................. Ins
Previous load case ............................... Ctrl+PgUp
Next load case ................................... Ctrl+PgDn
Icon bar
Selection ........................................ S
Color coding ..................................... Shift+C
Translate / Copy ................................. Shift+T
Rotate ........................................... Shift+R
Mirror ........................................... Shift+M
Scale ............................................ Shift+S
Dimension lines .................................. Ctrl+Alt+D
Edit background layers ........................... Ctrl+Alt+E
Part ............................................. Shift+P
Section lines .................................... Shift+Ctrl+X
Find ............................................. F3
Display options .................................. Ctrl+Y
Model information ................................ Shift+I
View
Zoom in ........................................ Ctrl+ü
Zoom out ....................................... Shift+Ctrl+ü
Fit in window .................................. Ctrl+W
Pan ............................................ Ctrl+M
Rotate ......................................... Ctrl+R
Front view ..................................... Ctrl+1
Side view ...................................... Ctrl+3
Top view ....................................... Ctrl+2
Perspective .................................... Ctrl+4
Wireframe ...................................... Alt+F5
Hidden line removal ............................ Alt+F6
Rendered ....................................... Alt+F7
Texture ........................................ Alt+F8
Guidelines
Structural grid ................................ Shift+G
Perpendicular .................................. Alt+V
Parallel ....................................... Alt+P
Bisector ....................................... Alt+B
Dividing point ................................. Alt+M
Intersection point of two lines ................ Alt+I
Toolbar
[Geometry] ....................................... Shift+F1
[Elements] ....................................... Shift+F2
[Loads] .......................................... Shift+F3
[Mesh] ........................................... Shift+F4
[Static] ......................................... Shift+F5
[Buckling] ....................................... Shift+F6
[Vibration] ...................................... Shift+F7
[Dynamic] ........................................ Shift+F8
[R. C. Design] ................................... Shift+F9
[Steel design] ................................... Shift+F10
[Timber design] .................................. Shift+F11
Geometry
Node ........................................... N
Line ........................................... L
Polygon ........................................ P
Rectangle ...................................... R
Arc ............................................ A
User’s Manual X6 Release1 e2 29
Arc based on three points ...................... B
Horizontal division ............................ H
Divides lines .................................. D
Intersect ...................................... I
Elements
Material ....................................... Shift+Ctrl+M
Cross-section .................................. Shift+Ctrl+C
Draw objects directly .......................... F4
Draw supports directly ......................... F5
Domain ......................................... D
Hole............................................ H
Line elements .................................. L
Nodal support .................................. T
Line support ................................... U
Node to node interface element ................. I
Line to line interface element ................. J
Loads
Load cases and load groups ..................... L
Load combinations............................... C
Nodal loads .................................... N
Concentrated loads on beams .................... B
Domain point load .............................. A
Load along line elements ....................... J
Surface edge loads ............................. E
Domain line load ............................... I
Distributed surface load ....................... H
Distributed load on domain ..................... D
Derived surface load over trusses/beams/ribs ... K
Load panel ..................................... P
Snow load ...................................... O
Wind load ...................................... W
Fluid loads .................................... F
Self Weight .................................... G
Moving line load definition .................... T
Mesh
Domain meshing ................................. G
Static
Linear static analysis ......................... L
Nonlinear static analysis ...................... N
Result display parameters ...................... D
Min, Max values ................................ Ctrl+X
Animation ...................................... A
Buckling
Buckling analysis .............................. L
Result display parameters ...................... D
Min, Max values ................................ Ctrl+X
Animation ...................................... A
Vibration
Vibration analysis ............................. L
Result display parameters ...................... D
Min, Max values ................................ Ctrl+X
Animation ...................................... A
Dynamic
Dynamic analysis ............................... L
Result display parameters ...................... D
Min, Max values ................................ Ctrl+X
Animation ...................................... A
R. C. Design
Reinforcement parameters ....................... P
Actual reinforcement ........................... T
Result display parameters ...................... D
Min, Max values ................................ Ctrl+X
Animation ...................................... A
Column reinforcement ........................... C
Beam reinforcement design ...................... B
Plate punching analysis ........................ U
Pad footing design ............................. F
Strip footing design ........................... I
Steel design
Design parameters .............................. P
Result display parameters ...................... D
Min, Max values ................................ Ctrl+X
Joint design ................................... J
Joint design ................................... K
Steel cross-section optimization ............... O
Timber design
Design parameters .............................. P
Result display parameters ...................... D
30
Min, Max values ................................ Ctrl+X
Animation ...................................... A
Menu
File
Open ........................................... Ctrl+O
Save ........................................... Ctrl+S
Print .......................................... Ctrl+P
Exit ........................................... Ctrl+Q
Edit
Undo ........................................... Alt+BkSp
Redo ........................................... Shift+Alt+BkSp
Select all ..................................... Num *
Copy ........................................... Ctrl+C
Paste .......................................... Ctrl+V
Delete ......................................... Del
Table Browser .................................. F12
Report Maker ................................... F10
Drawings Library ............................... F6
Save to Drawings Library ....................... F9
Weight report .................................. F8
Piano Sisma .................................... Ctrl+Alt+P
Assemble structural members .................... Shift+A
Break apart structural members ................. Shift+B
Reverse local coordinate system ................ Ctrl+E
Settings
Symbols ........................................ Ctrl+Y
Switches ....................................... Ctrl+D
Layer Manager .................................. F11
storeys ........................................ F7
Guidelines setup ............................... Ctrl+G
Structural grid ................................ Shift+G
Keyboard shortcuts ............................. Ctrl+K
Preferences
Fonts ........................................ Shift+Alt+F
Analysis ..................................... Shift+Alt+B
View
Front view ..................................... Ctrl+1
Top view ....................................... Ctrl+2
Side view ...................................... Ctrl+3
Perspective .................................... Ctrl+4
One storey up ................................... PgUp
One storey down ................................. PgDn
Zoom in ........................................ Ctrl+ü
Zoom out ....................................... Shift+Ctrl+ü
Fit in window .................................. Ctrl+W
Pan ............................................ Ctrl+M
Rotate ......................................... Ctrl+R
Wireframe ...................................... Alt+F5
Hidden line removal ............................ Alt+F6
Rendered ....................................... Alt+F7
Texture ........................................ Alt+F8
Window
Property Editor ................................ Shift+Alt+P
Status ......................................... Shift+Alt+I
Color coding ................................... Shift+Alt+C
Coordinates .................................... Shift+Alt+K
Color legend ................................... Shift+Alt+L
Load background picture ........................ Ctrl+B
Split horizontally ............................. Shift+H
Split vertically ............................... Shift+V
Close window ................................... Ctrl+F4
Decrease label font size ....................... Ctrl+Alt+I
Increase label font size ....................... Ctrl+Alt+O
Drawings Library ............................... F6
Save to Drawings Library ....................... F9
Help
Contents ....................................... F1
Speed buttons
Auto intersection ................................ Ctrl+I
Mesh display on/off .............................. M
Show only selected elements ...................... Ctrl+F
Load display on/off .............................. Ctrl+L
Result display options
None ............................................. Ctrl+F5
Diagram .......................................... Ctrl+F6
Isolines ......................................... Ctrl+F7
Isosurfaces 2D ................................... Ctrl+F8
Section line ..................................... Ctrl+F9
User’s Manual X6 Release1 e2 31
Isosurfaces 3D ................................... Ctrl+F10
Diagram + average values ......................... Ctrl+F11
List of result components ........................ Q
Coordinates
X ................................................ X
Y ................................................ Y
Z ................................................ Z
L ................................................ Shift+Ctrl+L
R ................................................ Shift+Ctrl+R
A ................................................ Shift+Ctrl+A
H ................................................ Shift+Ctrl+H
B ................................................ Shift+Ctrl+B
Temporary workplane .............................. Shift+Ctrl+W
Lock X[m] : ...................................... Ctrl+Alt+X
Lock Y[m] : ...................................... Ctrl+Alt+Y
Lock Z[m] : ...................................... Ctrl+Alt+Z
Lock L[m] : ...................................... Ctrl+Alt+L
Lock r[m] : ...................................... Ctrl+Alt+R
Lock a[°] : ...................................... Ctrl+Alt+A
Lock h[m] : ...................................... Ctrl+Alt+H
Lock b[°] : ...................................... Ctrl+Alt+B
Relative / global coordinates .................... Shift+D
Relative / global polar coordinates .............. Shift+E
Scroll up Zoom in
Scroll down Zoom out
Wheel down + move Pan (slow)
Wheel down + ALT + move Rotate
Wheel down + CTRL + move Pan (fast)
32
right button When the cursor is over the graphics area, by pressing the right mouse button a quick menu appears
in accord with the current command in use.
Selection Geometry / Elements / Loads Results
After selecting a function usually a dialog box appears on the screen. These dialog boxes can be used
the same way as any other Windows dialog.
The dialog font can be changed by selecting the Settings / Preferences / Fonts dialog and clicking the
font sample label Dialog boxes.
You can change the position of all dialog windows. The program saves the latest position and displays
the dialog on the same position next time.
AxisVM uses tables to display numerical information on the screen allowing changes in formatting. The
tables operate in the same way independent of the content displayed. All the tables AxisVM creates are
[F12] available through the Table Browser dialog box by clicking its button or pressing [F12].
The model data to be displayed in the Table Browser can be selected from the tree structure in the left
side of the browser. If you use Table Browser while working in the pre-processor, input model data is
displayed only. While working in the post-processor, the model results are also displayed.
Only the data of the current selection (if any) or of the active (i.e., displayed) part is listed by
default.
The tree view on the left lists element / load data, result tables and libraries in a hierarchy and can also
be used as a model overview.
Sorting by columns Data entry and result tables can be sorted by most of the columns. Hovering over the header of a sort-
able column makes the header cell color change. Clicking on the header turns the cell to light blue and
the sorting order changes (ascending, descending, no sort). Table rows will be sorted accordingly.
Most data entry tables can be sorted by multiple criteria. Sorting the table of nodes by Z then by X
means that nodes with an equal Z coordinate will be sorted by their X coordinate.
User’s Manual X6 Release1 e2 33
The order of criteria depends on the order of clicking. Result tables allow sorting by one result compo-
nent column only. But there may be certain columns that can be used as the first criterion (like Cross-
section in the Beam force tables, i.e., you can sort the results first by cross-section then by a force com-
ponent).
Color coding of Tables of the Table Browser apply color coding to the displayed values (e.g., domain thickness, beam
table cells cross-section, beam length, result components, etc.). Cell background colors appear only on the screen.
It helps to find common, smaller of bigger values.
Result tables show negative values in shades of blue, positive values in shades of red. Efficiencies are
displayed in shades of green under the limit and in shades of red above it. Color coding can be disabled,
see Format / Color coding of table cells.
Add new
row Copy Print
Paste
Cross-Section Fit
Delete Columns
Editor
Library Add to report
Using the table A table can contain more rows and/or columns than can be displayed at the same time.
It can be viewed in its entirety using the scroll bars and/or using the keyboard as follows:
Arrow keys Moves the edit focus up and down, to the left and to the right, and scrolls the table along the rows or
left button columns. Clicking an editable cell moves the edit focus to that cell.
[Home] Moves the focus to the first cell of the row.
[End] Moves the focus to the last cell of the row.
[Ctrl]+[Home] Moves the focus to the first cell of the first row
[Ctrl]+[End] Moves the focus to the last cell of the last row.
[Page Up] Displays the previous page of rows.
34
File
Browse from If the table of Materials, Cross-sections, Spring characteristics or XLAM timber panels is selected the first
libraries menu item allows browsing of the built-in material library, cross-section library, spring characteristics
library or XLAM timber panels library.
[Ctrl]+ [L]
Import DBase File Imports a DBase file name.dbf into the current table. The program checks the values of the fields and
sends an error message if an incompatible value is found.
Save As DBase File Exports the current table into a Dbase file name.dbf. The field names are generated based on the names
of the columns. The fields will be of text type.
Save As HTML Exports the current table into an HTML file name.htm. This file can be imported as a table into Word or
can be opened in web browser applications. Some formatting information of the columns will be lost.
Save As TXT Exports the current table into a TXT (ASCII) file name.txt.
Save As RTF Exports the current table into an RTF file name.rtf using the current template file. You can import this
file into Microsoft Word or any other word processor which can import RTF files.
See 2.10.2 Report
New cross-section Creates a new cross-section data file name.sec. The table created will be placed together with the cross-
table sections of the same type.
You can store cross sections of any type in these tables. Type of the table determines only the position
of the table in the Cross-section Library.
Cross-section table You can modify properties (table’s name, cross.section type) of a user defined table.
properties
User’s Manual X6 Release1 e2 35
New XLAM table Creates a new XLAM timber panel layer table under LIBRARIES / XLAM timber panels.
Rename XLAM table You can change the name of a user defined table within XLAM timber panels.
Delete XLAM table You can delete a user defined table within XLAM timber panels.
New spring charac- Creates a new spring characteristics table under LIBRARIES / Spring characteristics library.
teristics table
Rename spring If a custom spring characteristics table is selected within the Spring characteristics library this menu item
characteristics table appears and the name of the table can be changed.
Delete spring If a custom spring characteristics table is selected within the Spring characteristics library this menu item
characteristics table appears and the table can be deleted.
Print Prints all the information displayed in the table to the selected printer or to a file, with the page header
[Ctrl + [P] and comment row previously set with the File/Header menu command.
Turning on Description of table columns an explanation of columns appears at the bottom of the
table.
Exit Exits the table in the same way as the Cancel button (the changes are not saved).
[Alt]+ [F4]
Edit
New row Adds a new row to the list, and allows you to fill all the editable cells with data in a fixed order from left
to right.
[Ctrl]+ [Insert]
Delete rows Deletes the selected rows. Also available in the popup menu.
[Ctrl]+ [Del]
Delete textures Available only if materials are listed. Removes texture from the se-
lected materials. Available in the popup menu.
Select table Selects the entire table. Clicking the top left cell does the same.
[Ctrl]+ [A]
Design new custom Starts the graphics Cross-Section Editor, allowing the input of a new custom cross-section.
cross-section
36
[Ctrl]+[G]
Modify custom Starts the graphics Cross-Section Editor, allowing the modification of a custom cross-section previously
cross-section created with the graphics Cross-Section Editor.
[Ctrl]+[M]
Automatic cross- If this function is on changing section parameters in the table leads to the recalculation of geometry
section shape up- and cross-section parameters.
date
Copy Copies selected cells to the Clipboard as a table. Also available in the popup menu.
[Ctrl]+ [C]
Paste Pastes table cells from the Clipboard overwriting cell values.
If any of the values is unacceptable Paste aborts.
If entire rows were cut or copied and the table allows inserting new rows you can also add clipboard
[Ctrl]+ [V]
data to the end of the table instead of overwriting the existing rows.
Set common value Sets a common value for the selected cells within a column.
Example: you can set the Z coordinate of all nodes to the same value making the model absolutely flat.
Available from the Table Browser Menu / Edit / Set common value.
Also available in the popup menu.
Go to Jumps to a specified row in the table.
[F5]
Format
Data entry tables
Turn on/off columns You can specify whether a column is visible or not, by setting the check boxes of the corresponding
columns. If some columns are turned off, information on hidden columns appear below the table.
Checking the Save as default option makes the column status the default for that type of table.
[Ctrl]+ [Alt]+ [F]
The display format is set according to the settings in the Units/Settings dialogue window (See 3.3.8
Units and Formats).
Many cells require the entry of a numeric value. When entering real numbers you can use the following
characters:
+-01234567890E
and the standard Windows decimal separator specified in Start / Settings / Control Panel / Regional
Settings / Number / Decimal symbol field.
In some cases you cannot enter a negative number so the - key is deactivated while entering these kind
of values. If an integer value is required you cannot use the decimal separator and E.
Restore default Restores the default format of the entire table (column visibility and decimals).
format
Ctrl+D
Color coding of Color coding of cell backgrounds can be enabled or disabled here.
table cells
Ctrl+Alt+C
Order of load The display order of load cases can be customized.
cases... See 4.10.1 Load cases, load groups
User’s Manual X6 Release1 e2 37
Intermediate After dividing or meshing beams or ribs with variable cross-section AxisVM builds up intermediate
sections cross-sections. This menu item is to turn on/off the display of intermediate cross-sections at the end of
the list.
Show used cross- After the Delete unused cross-sections command only the sections in bold will remain in the list.
sections in boldface
The cross-section names which are signed by bold letter will remain in the table if the Delete unused
cross-sections switch is turned on.
In case of result query new items appear on the Format menu and the Toolbar.
During
result query
Result Display You can control finding the extremes for result components and set to show results (Result) and/or just
Options the extremes (Extremes).
[Ctrl]+[R] See in detail… 6.1.5 Result tables
Results On/Off Display of results can be turned on / off.
[Ctrl]+[T]
Extremes On/Off Display of extremes can be turned on / off.
[Ctrl]+[E]
Property Filtering Property filtering helps you to select elements
to include in the table.
[Ctrl]+[Q]
Report
Current report You can set the current report. Tables will be added to this report. See 2.10 Report Maker
Add table to Adds the current table to the current report. If the selected node in the treeview has sub-nodes (e.g.,
report MODEL or Loads) all tables under that node will be added. If the current table is a result table and is set
to display extremes only all sub-tables will display extremes only. See 2.10 Report Maker.
[F9]
38
[F10]
Help
Help to Use Table Displays info about the table browser operation.
Browser
Result tables also display the extremes (minimum and maximum values) of the data if you select
this option in the Display options dialog when you enter Table Browser. Displaying both the
individual values and the extremes is the default setting.
[F10]
Report Maker is a tool to compile a full report of a project using report items (tables / drawings /
pictures created by AxisVM and user-defined text blocks). Reports are stored in the model file (*.axs)
and can be printed or saved as a Rich Text Format (RTF) file. RTF files can be processed by other
programs (e.g., Microsoft Word).
Tables exported from Table Browser are automatically updated if the model has been changed or some
of its parts were deleted.
Report Maker can handle several different reports for the same project. The structure of reports is
displayed in a tree view on the left. The properties of the selected report item are shown on the right
side of the window.
User’s Manual X6 Release1 e2 39
Drawings Library By clicking the Drawings Library tab you can browse the saved drawings and add the selected ones to
the report. Unlike the pictures in the Gallery these drawings are not graphics files, but view settings
40
stored to recreate the drawing at any time. This way drawings will be automatically updated if we change
and recalculate the model.
See in detail 3.6.10 Drawings Library, 3.6.11 Save to Drawings Library.
Gallery By clicking the Gallery tab you can browse the saved pictures (BMP, JPG, WMF, EMF) located in a folder
named Images_modelname and add the selected ones to the report. This folder is automatically created
as a subfolder of the model folder.
See in detail 2.10.5 Gallery
Settings
Click the Settings... button to change the caption, size, justification, rotation color mode or scaling of
drawings.
You can save the current drawing on screen or the result tables in design modules with the function of
Edit\ Saving drawings and design result tables in main menu.
See 3.2.12 Saving drawings and design result tables
One or more selected pictures in the Gallery can be inserted into a report by selecting menu item Gal-
lery/Add pictures to the report or clicking the arrow button above the Gallery or by drag and drop.
In printed reports Report Maker automatically builds a table of contents and inserts it to the beginning
of the report. Tables are listed according to their titles. Text blocks are listed only if they were formatted
using one of the Heading styles in the Text Editor. Pictures are listed only if they have a caption.
2.10.2. Report
New report Creates a new report. Report names can be 32 characters long.
Delete entire report Deletes the current report (i.e., the report which contains the selected item). Pictures used in the report
are not deleted from Gallery.
[Del], [Ctrl]+[Del]
Rename Gives a new name to an existing report.
Save As TXT Exports the report into a ASCII text file. Drawings or pictures are not included.
Export as RTF Saves the report as name.rtf using the current template. If you save the file to a folder different from
the model folder all picture files used in the report are copied to an automatically created subfolder
Images_modelname. It is necessary because pictures are only linked and not saved into the RTF
document. To print the RTF report on a different machine make sure that picture files are also copied
to a subfolder Images_modelname.
Character and paragraph formatting of text blocks will be exported. The only exception is the character
color. Tables will be exported as RTF tables. Table titles are formatted with Heading 3 style so it is easy
to build a table of contents automatically using Microsoft Word.
Embedded PNG: Drawings are embedded into the file. This method provides portability with a smaller
size due to lossless compression of pictures.
Link to PNG files: This option keeps the RTF file smaller as drawings are stored in external files. Drawings
appear only if pictures are present in an Images_modelname subfolder relative to the folder of the RTF
file.
42
Embedded EMF: Drawings are embedded into the file as EMF (Enhanced Windows Metafile). It improves
portability but can result in huge file size.
Label font size: Here the label font size used on exported drawings can be fine-tuned.
Gridlines: Gridlines of exported tables can also be turned on/off.
Report preview Displays a print preview dialog. You can set the zoom factor between 10% and 500% (Page Width and
Full Page is also an option). Click the buttons or use the keyboard to move backward and forward
[F3]
between pages ([Home] = first page [←] = previous page, [→] = next page, [End] = last page.
Report preview can display multiple pages. [PgUp] [PgDn] steps back and forward according to the
number of pages displayed.
Print A dialog to set printing parameters and print a report. The options are the same as the table printing
options. Turning on Description of table columns an explanation of columns appears at the bottom
[Ctrl]+[P] of each table.
2.10.3. Edit
Some of the functions in the Edit menu are also available in the popup menu after clicking right mouse
button on a report item.
Undo Undoes the effect of the previous command.
Redo Executes the command which was undone.
Report template See 2.10.3.1 Template-based reports
Insert folder Inserts a new folder into the tree, below the current item. The current folder name appears on the right
side under the folder icon.
The number of expanded levels (1-7) of the report tree can be set with the level-adjustment bar.
Insert text into Starts a built-in Text Editor to create a new text block. The formatted text will be inserted after the
report selected report item.
[Ctrl]+[T]
Page break Inserts a page break after the selected report item.
User’s Manual X6 Release1 e2 43
[Ctrl]+[Alt]+[B]
Move up/down Moves up/down the selected report item by one.
selected report item
Move to / Copy to Moves / copies the selected report item to the end of another report.
Selection filter Determines which types of report items can be selected (report, table, drawing, picture, text, page break,
folder).
Select subitems If you turn this checkbox on and select a folder all subitems will be selected automatically.
automatically
Delete Deletes the selected report item (text block, picture, table, page break). If the current selection in the
tree is a report it deletes the entire report.
[Del], [Ctrl]+[Del]
Delete all report Deletes all items from the current report but does not delete the report itself.
items
Report templates can be used to generate reports based on certain presets, filters and preferences. Gen-
erated reports consist of drawings and tables. Templates can be saved as files and reused to generate a
report for other models.
The range of included elements, model data and result components can be set by filters.
Rebuild report If the model has been extended and the report should be updated (for example the steel member design
has been completed) click on the Rebuild button.
Any report item inserted by the user will be removed.
This button is to open or close the bottom part of dialog with the list of templates.
It is a complex task to create a report template. The steps of this process are listed on the left. Clicking
on these items we can edit filter options and other settings. An edit box under the list allows entering a
name for the template. The template administrator dialog lists templates by their names.
Elements
The first step is selecting element types to include in the report and choosing the element classification.
If Structural element types is selected, elements will be classified by their finite element type. If Architec-
tural element types is selected, elements will be classified by their architectural type (determined from
the element geometry).
Element data and results can be selected separately for reporting.
Next steps will display tables and drawings based on this selection.
User’s Manual X6 Release1 e2 45
Subgroups
The second step is to set the subgroups for the reported elements. A complete sub-report will be built
for each subgroup.
If the option Selected elements is activated only elements selected before opening Report Maker will be
reported.
If the option Subgroups from logical parts is activated subgroups can be created from domains with the
same thickness, structural members with the same cross-section or storeys (only the selected storeys will
be reported).
If Report domains one by one is checked each domain will be reported separately. Internal domains (being
entirely within another domain) are reported with the outer domain even in this case.
If the Entire model is selected subgroups can be created also from user defined parts.
Model data
drawings
46
If Overview is checked a view of the model will be inserted at the beginning of the report. Click Select a
view... to choose a view from the Drawings Library.
Check Model drawings if you want to include automatically created load and result diagrams. Click Select
a view... to choose a view from the Drawings Library.
Generated diagrams will inherit all settings of the selected drawing (point of view, status of graphics
symbols, numbering, labeling etc.) with minor adjustments. If no drawing is selected (e.g., the Drawings
Library is empty) drawings will follow the current view in the active window.
If a diagram is displayed only on a certain part of the structure the view is zoomed to fit drawing to the
window. The point of view and the status of graphics symbols will remain unchanged.
Checking Adjust drawings to leave space for info windows sets the zooming to fit the drawing within the
maximum rectangular space available among the info windows.
Load diagrams
If Insert load diagrams for the following load cases is checked, select load cases to add their load diagrams
to the report. Load diagrams will be generated from the view set for Model drawings in the previous step.
Select model data tables to add to the report. You can set the visible columns for tables to fine tune the
report content.
Under Elements you will find only those elements you selected in the first step.
Under Loads you can select load cases to add their load data tables to the report.
User’s Manual X6 Release1 e2 47
AxisVM provides a huge number of results. It is important to decide which load cases, combinations,
envelopes or critical combinations should contribute to the report for displacements, internal forces,
stresses, reinforcement values, steel or timber design checks.
Result diagrams
and tables
The last step is to select from the possible result tables and control the visibility of their columns. It is
also possible to generate result diagrams for result components. Click in the Drawing column in a row of
a result component. You can choose a drawing mode for that component from the dropdown list even
if you leave the checkbox unchecked (hiding the respective column). Result diagrams will be generated
from the view set for Model drawings and will be inserted before the table.
Creating reports from filter options is an earlier method which is less configurable and its content is not
updated but can be used in simple cases. Select element and load types, load cases and result compo-
nents from the filter tree on the left to control report composition. The resulting report appears in the
tree on the right. Its individual items can be checked or unchecked. Only checked items will be included
in the generated report.
48
2.10.4. Drawings
Add drawings to the Inserts the selected drawing(s) from the Drawings Library into the selected report. Place of insertion is
report determined by the selected item of the report tree. Effect of this function is the same as that of the
button on the Drawings Library tab.
Format of drawings See 2.10.2 Report
in RTF file...
2.10.5. Gallery
Add pictures to the Inserts selected pictures into the current report.
report
Copy pictures to You can copy bitmaps (.BMP, .JPG, .PNG) and Windows Metafiles (.WMF) to the folder
Gallery Images_modelname.
Delete pictures from Deletes selected pictures from the Gallery. Files are permanently deleted.
Gallery
Delete unused Deletes pictures which are not used in the reports.
pictures
Sort by name Gallery sorts pictures by filename / by type (.BMP, .EMF, .JPG, .WMF) or by date.
/ type / date
Reverse order If checked pictures are sorted in descending order. Otherwise pictures are sorted in ascending order.
You can perform certain tasks faster using these small toolbars.
After selecting Insert text to report a formatted text can be created in a simple WordPad-like text
processor.
File
Open The main purpose of this function is to load a Rich Text file written in Text Editor. If you open an RTF
[Ctrl]+[O] file created in another word processor it may contain special commands (e.g., tables, paragraph borders,
Unicode characters) which are not supported this simple editor. As a result you may get a series of rtf
control commands instead of formatted text.
Save Saves the text into an RTF file.
[Ctrl]+[S]
Exit Quits Text Editor.
Edit
Undo / Redo Undoes / redoes the last editing action.
[Alt]+[BkSp] /
[Shift]+[Alt]+[BkSp]
Cut Cuts the selected text and places it to the Clipboard.
[Ctrl]+[X]
Copy Copies the selected text to the Clipboard.
[Ctrl]+[C]
Paste Pastes the content of the Clipboard at the current position.
[Ctrl]+[V]
Find You can search for any text in the document. You can search from the beginning or from the current
[Ctrl]+[F] position. You can search whole words only and turn on and off case sensitivity.
Find next If a match was found you can get the next match with this function.
[F3]
Select all Selects the entire text.
[Ctrl]+[A]
Character
Bold Applies bold formatting to the selected text.
[Ctrl]+[B]
Italic Applies italic formatting to the selected text.
[Ctrl]+[I]
Underline Applies underline formatting to the selected text.
[Ctrl]+[U]
Color Sets the character color of the selection.
[Ctrl]+[Alt ]+[C]
Paragraph
Left justify Justifies the selected paragraphs to the left.
[Ctrl]+[L]
Centered Justifies the selected paragraphs to the centerline.
[Ctrl]+[E]
Right justify Justifies the selected paragraphs to the right.
[Ctrl]+[R]
Bullet Places bullets before the selected paragraphs.
[Ctrl]+[Alt]+[U]
50
2.11. Stories
Selection *
Zoom
Views
Display mode
Color coding
Transformations
Workplanes
Structural grid
Guidelines
Geometry tools
Dimensioning,
labeling
Background
layer editing
Renaming,
renumbering
Parts
Sections
Virtual beam
Search
Display options
Options
Model info
Dragging and The left-side icon bar and any flyout toolbar can be dragged and docked.
docking the Icon bar
and the flyout Dragging and docking of the Icon bar
toolbars If you move the mouse over the handle of the Icon bar (on its top edge), the cursor will change its shape
(moving). You can drag the Icon bar to any position on the screen. If you drag the Icon bar out of the
working area through its top or bottom edge the Icon bar becomes horiozontal. If you drag it to the
left or right edge it becomes vertical.
If the Icon bar is horizontal you can dock it at the top or at the bottom. You can change the position
and the order of docked toolbars by dragging. In the Cross-Section Editor and in Beam and Column
Reinforcement dialogs the Icon bar cannot be docked. Closing a floating Icon bar restores its original
position docked on the left.
52
The Icon bar and the flyout toolbars can be restored to their original position by selecting
Settings / Toolbars to default position from the menu
2.16.1. Selection
Activates the selection mode and displays the selection icon’s bar.
Select
Deselect Filter
Invert Region
All
Intersected lines
Previous Annular
Parts Sectorial
Polygon
Skewed rectangular
Rectangular
Lets you select a set of entities (nodes (points), lines, finite elements and loads) for processing. When
you execute commands you can use the Selection icon to specify the entity set to which to apply the
command to. If the Parts check box (See section 2.16.14 Parts) is enabled the selection will refer only to
the active (visible) parts.
You can change the view settings or continue selection in another window pane during the selection
process. These allow you to select elements in the most convenient view. The selected entities are
displayed in magenta in the graphics area.
The selection process is considered finished when the OK button is pressed.
Select Adds the currently selected entities to the set of selected entities.
Deselect Removes the currently selected entities from the set of selected entities.
All Applies the current selection mode (add, remove, or invert) to all filtered entities.
All selection commands change the selection status of elements chosen in the selection filter (see
below).
Previous Restores the previous selection set.
Selection of parts Clicking the button and a part from the list will select elements of the chosen part.
Filter Lets you specify filtering criteria to be used during selection. Check element types to select.
Property filtering lets you apply further criteria (beam length, cross-section, material, surface thickness,
reference). All elements matching certain criteria can also be selected using the smart command line
(see 2.17 Smart command line).
User’s Manual X6 Release1 e2 53
Method Selects entities using different methods (selection shapes). Rectangular, skewed rectangular, sectorial
or ring selection shapes are available. In the followings examples of the application of various selection
shapes are provided:
When drawing a rectangular frame, the following rules apply:
Dragging the selection frame from left to right selects elements falling entirely within the frame.
Dragging the selection frame from right to left selects elements that are not entirely outside the frame
(i.e., they are inside the selection rectangle or intersect it).
Selection: Result:
Rectangular
Skewed
rectangular
Polyline
Sectorial
Annular
Intersected lines
If an entity is hidden by another entity you cannot select it by simply clicking on it. In such a
case, you have to change view to select it.
54
The selected nodes are marked with a surrounding magenta rectangle. Sometimes it is necessary to
double-select nodes. In this case these nodes are marked with an additional blue rectangle surrounding
them.
Selections can also be made, without using the Selection Icon Bar. Pressing and holding the [Shift] button
while selecting with the will add entities to the selection and pressing and holding the [Ctrl] button
while selecting with the will remove entities from the selection.
Double selections can be made by pressing and holding the [Alt] button while double clicking on the
entities with the .
During selection you can modify the view or switch to another view or perspective.
Zoom in Displays an area of the model drawing specified by two points (two
opposite corners) on the graphics area defining a rectangular zoom
region. As a result, the apparent size of the model displayed in the
graphics area increases.
Zoom out Displays the model drawing from the graphics area on the area
specified by two points (two opposite corners) defining a rectangular
zoom region. As a result, the apparent size of the model displayed in
the graphics area decreases.
Zoom to fit Scales the drawing of the model to fit the graphics area, so you can view the entire model.
Pan Moves the drawing. Press and hold the left button of the while moving the mouse, until the desired
position of the drawing is obtained on the screen.
Quick Drag:
You can use the mid mouse button to drag the model drawing at any time (without the the Pan icon).
1. Click the Pan icon.
2. Drag the model to its new position.
This cursor shape indicates that you can pan the model.
Rotate After clicking this icon you can rotate the model around the centre of the encapsulating block of the
model by dragging. During the rotation the following pet palette appears at the lower part of the screen:
Rotation methods:
Free rotation around the horizontal axis of the screen and the global Z axis.
Rotation around a selected point. Click on a point to use it as the center of rotation.
This cursor shape indicates that you can rotate the model.
Undo view Undoes / redoes the action of up to 50 view commands.
/ Redo view
2.16.3. Views
Displays the projection of the model on the X-Z plane (front view).
Displays the projection of the model on the X-Y plane (top view).
Displays the projection of the model on the Y-Z plane (side view).
Perspective toolbar
Axonometry X-Z view
Perspective
X-Y view
Rotate about the
perpendicular axis Z-Y view
Rotate (activates
the pet palette)
Observation distance
Delete active
New perspective view perspective view
Perspective view list
Sets the parameters of the perspective display. The proper view can be set by rotating the model drawing
around the three axes, and by setting the observation distance. Rotation angles can be set with a
precision of 0.1 degrees. You can assign a name to each setting that you want to save for later use. Type
a name into the combo and click on the icon on the left of the combo to save the settings. To delete a
perspective setting choose it from the dropdown list and click on the Delete icon on the right side of
the combo. Palette settings are stored.
Observation Observation distance is the distance between the viewpoint and the centre of the encapsulating block
distance of the model.
Views, Displays three projection views and the perspective view of the model, and allows you select the view
perspective that you want to display. Click the view you want to select.
56
Wireframe:
Displays a wireframe model drawing. In this mode the axis of the line elements and
the mid-plane of the surface elements are displayed.
Hidden:
Displays a wireframe model drawing with the hidden lines removed.
Rendered:
Displays a rendered model drawing. The line elements are displayed with their actual
cross-section and the surface elements with their actual thickness. The elements
colors are displayed corresponding to colors assigned to their materials. Rendered
view is smoother and shows the details of thin-walled cross-sections.
User’s Manual X6 Release1 e2 57
Opaque Transparent
Branches of the tree view on the left and the horizontal list above the texture thumbnails show the
material types (brick, concrete, metal, stone, timber, other). The last type (custom) is for the user-defined
textures. Textures of the current type are displayed as thumbnails. The selected texture appears in the
preview window with a thick black frame.
Popup menu After clicking the texture with the right mouse button, a popup menu ap-
pears with the following functions:
Removing the texture from the material
Defining or deleting a custom texture
Rotation settings
3D object contour Turning on display of 3D object contours (see the speed button below) a semi-transparent solid model
settings will also be displayed. Transparency of elements can be configured in View / 3D object contour settings.
Colors By architectural type The solid model will be displayed using the colors assigned to
walls, slabs, beams, etc.
See 3.3.11 Preferences / Graphical symbols
By material The colors assigned to the material will be used.
Use color coding of rendered view Colors will be determined by the current color coding set for
rendered view.
Transparency Move trackbars to set the desired transparency. Click Use defaults to return to the default values. Check
Auto refresh to see the changes immeadiately on the model view.
Results displayed on the rendered model. If a result component is displayed with the Solid model
display mode (see 6.1 Static), results will color the rendered model where elements are displayed with
their actual physical dimensions. For settings, see 3.3.11.10 Display
Color coding helps to get an overview of element properties. Different color coding can be set for the
rendered and wireframe display modes.
Type of color coding can be chosen from a dropdown list.
The program automatically associates different colors to different properties, but colors can be changed.
User’s Manual X6 Release1 e2 61
Surface Surface
supports loads
Setting colors
Default
Restores the default values (default element type color, default material color).
Color gradient
Enter the start and end color of the gradient by
dragging arc end points on the hue / saturation
circle to the desired position. The program picks
up the necessary number of colors between the
end points.
Use the trackbar on the right to set brightness for
the selected end point.
Shorter arc connects colors with the smallest pos-
sible hue changes. Longer arc goes around the hue
circle in the other (longer) way.
Random colors
Program selects random colors but ensures that colors are not too close.
Rendered and wireframe colors are handled separately but can be synchronized. Clicking with the
right mouse button on the color list a popup menu appears. Select Apply colors assigned to rendered
view (wireframe view) to import the color set from the other display mode.
2.16.6.1. Translate
Move: moves the selected entities by the distance dX, dY, dZ. Lines running into the moved nodes remain
connected.
Detach: moves the selected entities by the distance dX, dY, dZ. Lines running into the moved nodes are
detached.
None: No nodes will be connected.
Double selected: Holding the [Alt] key pressed you can double select nodes. These nodes will be
connected.
All: All nodes to be copied will be connected.
Switches
Copy options Copy elements: You can specify the finite elements assigned to the geometric entities to be copied as
well.
Copy loads: You can specify the loads assigned to the geometric entities to be copied as well.
Copy nodal masses: You can specify the nodal masses to the geometric entities to be copied as well.
Copy dimension lines: The dimension lines will be copied only if the nodes to which they are assigned
are selected.
With guidelines All rulers will also be moved (useful when moving the entire model).
With DXF layer With this option checked the transformations will be performed on the objects of the DXF layer as well.
If individual layer elements are selected the transformation will be applied only to the selected elements.
If nothing is selected the entire layer is transformed.
Visible layers only With this option checked only the visible layers will be transformed.
If you have repetitive parts in your model, you should first create these (including the definition
of finite elements, support conditions, loads, and dimension lines), and then make copies of
them.
You can use any existing point when you have to specify the translation vector.
Selected loads can be copied or moved to another load case if load case is changed to the target
load case during the operation.
2.16.6.2. Rotate
Rotation options Incremental: makes N copies of the selected entities by the cursor angle.
Distribute: makes N copies of the selected entities by cursor angle/N increments.
Spread by angle: makes copies of the selected entities spread by a given angle specified in the dialog.
The number of copies depends on how many copies will fit into the cursor angle.
Consecutive: makes N consecutive copies of the selected entities at different cursor angles.
Move: moves the selected entities by the cursor angle. Lines running into the moved nodes remain
connected.
Detach: moves the selected entities by the cursor angle. Lines running into the moved nodes are
detached.
Nodes to connect See 2.16.6.1 Translate
In perspective view, the centerpoint, start point and end point can be specified only using existing
points or other identified 3D locations (i.e., a point on a line). In perspective view, cursor angle is
determined by the global X and Y coordinates only.
User’s Manual X6 Release1 e2 65
2.16.6.3. Mirror
In perspective view, the mirroring is possible only across a plane parallel to the global
Z axis.
2.16.6.4. Scale
Scale Makes multiple copies of, or moves the selected geometric entities, by scaling from a center. You must
specify the scaling center, a point of reference and its new position after scaling (coordinate ratios will
determine the scaling factors).
2.16.7. Workplanes
Workplanes (user coordinate systems) makes it easier to draw on oblique planes. Consider a hole for a
skylight on an oblique plane of a roof. The plane of the roof can act as a workplane so drawing can be
performed in two dimensions. In case of workplanes altitudinal coordinate means the distance along
the axis normal to the workplane.
Clicking the workplane speed button the workplane can be selected from a list. Workplanes are also
available from the main menu by selecting View \ Workplanes or from the popup menu by selecting
Workplanes.
Display options A workplane can be displayed in the global coordinate system or in its local system. After checking Hide
elements not in the workplane only those elements are displayed that are in the workplane. After checking
Show elements out of workplane grayed elements out of the workplane appears grayed.
Changing If you select a workplane from the tree, its parameters are displayed. Editing them and clicking the OK
workplane button or selecting another workplane will change the parameters of the selected workplane.
parameters
Project to If this option is activated all editing will be projected to the workplane. To work in the coordinate system
workplane of the workplane but out of the workplane (e.g., setting z to 3 meters) turn off this option.
Two buttons fly out from the Icon bar button. The first one is to define a structural grid, and the second
one is to define individual gridlines.
To define a structural grid set an X0, Y0, or Z0 origin then enter the ΔX, ΔY or ΔZ relative spacing values.
For example, with X0 = 0 entering 4*3.5; 2*5; 7.5 into the ΔX field gridlines will appear at the following
X positions: 3.50; 7.00; 10.50; 14.00; 19.00; 24.00; 31.50.
The structural grid can be rotated by a custom angle.
The length of lines is determined by the minimum and maximum coordinate positions of gridlines in
the other direction, so the shape of the grid is always a rectangle.
User’s Manual X6 Release1 e2 67
Refresh all
Update structural grids for all views.
New
structural grid
Grid plane For grids parallel to a global plane the X0, Y0, or Z0 distance between the
grid plane and the global plane can be set
If the model has storeys different structural grids can be assigned to each
storey. The grid can be associated to all storeys by selecting On all storeys
from the dropdown list. If the grid is associated to a specific storey (e.g.,
Story grid
storey 1) and Display only if the storey is active is checked the grid remains
hidden until storey 1 is activated.
e[m] It is recommended to set a nonzero Grid line extension so that the labels fall
outside the rectangle of gridlines
Automatic If the option Automatic dimensioning is checked dimension lines are also created with the structural
dimensioning gridlines. These dimension lines are associative, e.g., follow changes of the structural grid.
It is also possible to place smart dimension lines manually between intersection points of structural
gridlines (see 2.16.11.1 Orthogonal dimension lines). It means that activating smart mode and placing a
dimension line between A1 and G1 will generate a group of dimension lines: A1-B1, B1-C1, C1-D1, D1-
E1, E1-F1, F1-G1.
Modify Name, labels, positions, color of a grid can be modified. If Create structural grid is checked the entire
structural grid grid will be recreated with the new parameters. In this case all custom gridlines associated to the grid
(see below) will be removed.
Custom gridlines can also be defined by clicking on the start point and the end point. Properties of the
gridline (label position, extension, prefix, labels, colour) can be set on a pet palette.
Custom gridlines Custom grid lines must be associated to a structural grid and can be turned on / off with that grid.
Recreating the grid with new parameters deletes all associated custom gridlines.
2.16.9. Guidelines
Helps in editing the geometry of the model. Guidelines can be defined in the global coordinate system.
This way an arbitrary grid can be created, intersections can be determined and distances can be set. The
cursor identifies the guidelines.
See 4.7 Editing tools
Places a vertical and a horizontal guideline at the current position of the cursor.
Places a pair of orthogonal oblique guidelines at the current position of the cursor.
In perspective view all the guidelines are displayed but only oblique guidelines can be placed. You can
change the position of a guideline with the mouse by dragging it to a new position. You can remove
(delete) a guideline by dragging it off the graphics area.
Guidelines can be entered numerically by coordinates. Clicking with the mouse on a guideline or
selecting Settings / Guidelines Setup command from the main menu, the following dialog is displayed:
guideline
b
a
a: is the angle of the guideline’s projection on the X-Y plane and the X axis.
b: is the angle of the guideline and its projection on the X-Y plane.
70
Refresh all If checked changes will be applied to all views otherwise only in the active view.
The icons of Geometry tools allow you to lock the direction of drawing a line.
Perpendicular Parallel
baseline baseline
Begin to draw a line. Click the Perpendicular or Parallel icon then click an existing line or click two points
to define the direction. The cursor will move perpendicular or parallel to this baseline.
Begin to draw a line. Click the Perpendicular to a plane icon then click
the domain defining the plane. The cursor will move perpendicular to
Perpendicular to a plane the plane. The plane can also be defined by clicking three points.
These icons can be conveniently used while editing the geometry of the model or defining section
planes.
Begin to draw a line then click start point and end point of another line.
Midpoint will determine the direction.
Line towards a midpoint
Begin to draw a line then click the two legs of an angle. Bisector will
determine the direction of the line.
Bisector
Begin to draw a node or a line then click the icon, click the two lines or
their start and end point. A node or line point is created at the point of
Point of intersection intersection. Any of the lines (or both) can be an arc. In this case there
may be more than one point of intersection. If so, calculated points are
marked with small circles. The required point has to be selected by click-
ing.
Begin to draw a node or a line then click the icon and click the two
nodes. Specify the division by ratio or by distance in the popup dialog.
Dividing point A node or line point is created.
The action for Point of intersection and Dividing point can be set here.
Two options are available: creating a node or moving the relative origin
Point constraint operation to the position calculated.
User’s Manual X6 Release1 e2 71
This group of functions lets you assign associative orthogonal and aligned dimension lines or strings of
dimension lines to the three dimensional model, as well as angle, arc length, arc radius, level and
elevation marks, labels for result values. Click on the Dimensions icon to display the Dimension Toolbar.
That will allow you to select the proper dimension tool. Click on the left-bottom icon of the Dimension
Toolbar to set the parameters of the selected tool.
You can change the position of dimension lines or labels at any time by dragging them to their new
position. If the dimension lines were associated with the model their position and dimension will be
continuously updated as you modify the geometry of the model.
Associative orthogonal dimension lines or strings of dimension lines, parallel to the global X, Y, or Z
axes can be assigned to the model by following the next steps:
1. Click on dimension line start point and on the end point. If these points are connected by a line
you can just click on the line.
2. Move the mouse. The position of the dimension line depends on the direction in which you
moved the mouse. There is one exception: when the segment is not parallel to any global plane
and the editing is in the perspective view. In this case you have to select the direction dX, dY, or
dZ from the toolbar.
3. Click the left mouse button to set the final position of the dimension line.
To insert a string of dimension lines, click on the points in the corresponding order or on the lines if
any. Steps 2 and 3 are the same as for the individual dimension lines. A string of dimension lines can be
selected at once if you click on one of them while depressing the Shift key. It allows you to move it as
a group. To change the position of a group segment individually select it using the selection rectangle
and drag it to its new position. As a result this dimension line will be removed from the group (it can be
moved individually).
If the dimension line is assigned to the points of a model, it will always behave in an associative way
(e.g., will move with the model when the model is changed or resized or moved).
72
Tick mark Lets you set the tick marks of the dimension lines. You can select from nine predefined symbols.
Color Lets you set the color of dimension lines individually. You can get the color from the active layer. The
dimension lines, marks, and texts are placed on the Dimensions layer by default but you can change it
any time.
Sizes Lets you set the drawing parameters of the dimension line.
Dimension style/ Lets you to set the type and thickness of a dimension or extension line. You can choose a predefined
Extension style value or get it from the active layer. You can turn on/off the display of extension lines.
Label orientation Lets you set the orientation of the text labels of the dimension lines (Always horizontal, Always vertical,
Auto horizontal/vertical, or Aligned to dimension line) inside or outside the dimension line.
Use defaults Lets you restore the default setting.
Apply font to all Apply the same font to every dimension line.
symbols
Save as default Lets you save the current setting as default setting.
setting
Apply to all Applies the current setting to all existing orthogonal or aligned dimension lines to ensure a uniform
dimension lines look.
Layers Lets you select/define/set layers where the dimension lines will be placed. If there are no layers defined
when you start defining dimension lines, a Dimension layer will be automatically created.
See 3.3.3 Layer Manager
User’s Manual X6 Release1 e2 73
Text Parameters
Allows to you to define the settings of the text on the dimension lines.
Measured value Allows you to place the measured value on the dimension line, using the current prefix and suffix
settings. By clicking the Units and formats button the number format can be set in the Dimensions
section of the Settings / Units and formats dialog box.
Display unit of Display of the unit of measured value.
measurement
Units and Formats... To change the current font parameters, click the button below the Units and formats... button.
Prefix Sets the prefix used with the text on the dimension lines. You can choose from the following options:
Auto (dX, dY, dZ, dL = [depending on the direction])
Auto (DX, DY, DZ, DL = [depending on the direction])
User defined (this option will require you to enter the prefix).
Suffix Sets the suffix used with the text on the dimension lines.
The steps are the same as the steps of creating an orthogonal dimension line.
See 2.16.11.2 Aligned dimension lines
The plane of the parallel dimension line is determined automatically. There is one exception: when the
segment is not parallel to any global plane and the editing is in the perspective view. In this case you
have to select the direction X, Y, or Z from the toolbar. The plane of the section line will be defined by
the segment and the selected global axis.
74
Associative angle dimensions, as the symbol of the angle between two segments, can be assigned to
the model in the following steps:
1. Click on start point and on the end point of the first segment. If the points are connected by a
line you can just click on the line.
2. Click on start point and on the end point of the second segment. If the points are connected by
a line you can just click on the line.
3. Move the mouse. The position and radius of the angle
dimension will be determined by the mouse movement.
Based on the position of the mouse, the angle,
supplementary angle or complementary angle
dimension can be entered.
4. Click the left mouse button to set the angle dimension
in its final position.
By clicking the Text parameters / Units and formats… button the angle number format can be set in the
Dimensions section of the Settings / Units and Formats dialog box.
User’s Manual X6 Release1 e2 75
Level Selects the level mark symbol, and sets its size and format.
Elevation Selects the elevation mark symbol, and sets its size and format.
1. Enter the text in the Text box parameters window, or in case of a single line text enter it directly
into the edit field of the Toolbar.
2. Click on the point to which you want to assign the text box.
3. Move the mouse to the desired position and click to set the text box in its final position.
User’s Manual X6 Release1 e2 77
Color Sets the color of the text, frame, and extension line. You can get the color from the layer.
Text box These switches set the drawing parameters of the text box, frame, and extension line, the transparency
and alignment of the text, and the d distance of the extension line from the reference point (to which
the text box is assigned to).
Font Sets the text font, style and size.
You can reload and change default settings, apply text box or font parameters to all existing text boxes
Active Links Active links can be placed in text boxes to attach any external information to the model.
If the text contains a file reference or a link to a web page clicking the text box launches the application
associated to the file or URL instead of opening the above dialog. To change the text, select text box
first (e.g., Shift+click) then click into the box.
File reference A file reference is made of the -> characters and a file name. E. g.:
->C:\MyModel\Reports\Details.doc
If no full path is specified AxisVM starts from the folder of the model, so if our model is in C:\MyModel
we can enter: -> \Reports\Details.doc
Clicking the text box starts the application associated to the file type. This way we can attach pictures,
movies, sounds, Excel tables or other documents to any part of the model.
URL Supported protocols and link formats are:
http://..., ftp://..., https://..., file://..., www. ...
After clicking the textbox, the default web browser launches and opens the web site or file.
If the text contains more than one URL, the first one is used.
78
Object info text box Element or load properties appear in the text box depending on the current tab (Geometry, Element or
Loads). Information text box parameters can be set in a dialog:
Result text box When displaying results the cursor determines the value of the current result component on nodes,
mid-side nodes, surface centers, or intermediate points of beams or ribs and shows it as a tooltip. The
text of the tooltip is automatically entered in a text box.
The steps of result labeling are similar to creating a text box.
The result text box is visible only when the selected result component is the same as the one that was
selected when the result text box was created. For example an My result text box is displayed only when
the My component is selected as the current result component.
Below the button of Use defaults three checkboxes helps to customize the text box:
Layer Manager Lets you create new layers or modify existing ones.
This function is also available from the menu as Settings / Layer Manager.
See 3.3.3 Layer Manager
[F11]
Automatic labeling of isolines can be activated in the color legend window. See 2.19.4 Color legend
window
This editor allows making changes in the imported DXF and PDF layers and adding new shapes.
Background layers contain only geometry information and play no role in the structure.
Background layer
editor
Layer Manager
Opens the Layer Manager. See 3.3.3 Layer Manager
Selecting a layer Select a layer for editing from the dropdown tree.
To create a new layer, open the Layer Manager and create a new layer and click OK. Then you can select
the new layer.
Another way to select a layer is to click on this button beside the dropdown tree then click on a shape.
The layer associated to that shape will be selected.
User’s Manual X6 Release1 e2 81
Selection
Click on this button to activate shape selection.
Click the outline of the shape or drag a frame around shapes and click
on the outline of a selected shape then set properties in the Edit shape
properties dialog.
Using special selection functions choose the next button of the toolbar
(Special selection modes).
Right clicking on a layer object displays a popup menu. Isolate layer for
editing locks all the other layers and allows editing of the containing
layer. Delete layer allows deleting of the containing layer.
Special selection modes
Clicking on this button displays a palette for selection options. See 2.16.1 Selection
Click OK if the selection is finished and click on any selected shape to set shape properties.
Pick up properties
Click on this button and to activate pick up. Clicking on a shape picks up all properties of that shape (i.
e. all subsequent drawing functions will use these properties).
Convert points of the selected shapes to AxisVM nodes
Delete shapes To delete shapes first select them then press the Delete key of the keyboard.
Pen color Pen color is used to draw the outline of shapes and also to fill the interior of filled shapes.
There are three ways to set the current pen color.
Line style Two dropdown lists on the right show the available line styles (top) and line weights (bottom). Select the
Line weight desired values.
These settings don’t have any effect on filled shapes as they don’t have an outline.
2.16.13. Renaming/renumbering
The list on the left shows the number of selected nodes and elements. Choose what you want to re-
name / renumber.
Start at Enter the starting number. Selected elements will be renumbered in an order determined from their
position. Renumbering can have effect on elements not selected as two nodes or elements cannot have
the same number.
Name In name strings element number is represented as an underscore (_). For example: if starting number is
1, and the Name field contains T_, the names of the selected elements will be T1, T2, T3 … If only one
element is selected it is not necessary to include _ in the Name. Otherwise, it must be included as ele-
ments must have different names.
If the Name field is empty, the name will be the number itself.
Restore original If Restore original numbers is checked clicking the OK button restores the original numbers of the se-
numbers lected elements and clears their names. Element type must be selected from the list on the left.
To turn on/off the display of numbers / names of elements open the Display options dialog
(see 2.16.18 Display options) or use the speed button (see 2.18 Speed Buttons)
2.16.14. Parts
Parts can also be activated without opening this dialog box by simply clicking the Parts speed button
(at the bottom of the screen).
Depth of the tree expansion can be set by clicking on the numbers on the right-hand side of the window.
User’s Manual X6 Release1 e2 83
Delete Lets you delete the selected user-defined part from the list. This command will not affect the model.
If section segment result tables are selected only section segments within the active parts are
listed.
Creating Creating folders offer a way of sorting user-defined model parts. Parts can be moved and rearranged by
new folders dragging them to a new position. [Ctrl] and [Shift] allows multiple selection in the usual way.
Turning folders on/off turns on/off the parts within the folder.
Logical parts This dialog is to set criteria for creating logical parts.
Architectural objects are defined by their geometry. Vertical beams, ribs and trusses are considered to
be columns, horizontal ones are considered to be beams. Domains in horizontal plane are slabs, domains
in planes perpendicular to horizontal planes are considered to be walls.
If there are storeys defined logical parts can be created by storeys.
Parts created from timber and steel optimization groups make it easy to change cross-section of an
optimization group. Parts can be created from Domains by type (parametric ribbed domains, trapezoidal
steel decks, XLAM domains, etc.) or by eccentricity groups see 4.9.6 Domain.
If the model contains structural grids, it is also possible to define logical parts based on structural grid-
lines. These parts contain elements within the plane obtained by sweeping the gridline perpendicular to
the plane of the grid. Line elements and domains are included only if all their nodes fall within the plane.
Plane tolerance can be customized for each gridline. Multiple gridlines can be selected (Ctrl+click or
Shift+click on the gridline name) and a common value can be set.
84
When working on parts, only the data of the active parts will appear in the tables by default.
Auto Refresh
If it is on turning on or off parts will immediately cause a redraw. If it is off the screen is updated only
after clicking the OK button.
Refresh all
If it is on parts will be turned or on off in all window panes in multi-window mode.
If it is off part settings will be updated only in the active panel.
Show non-visible parts grayed
If it is on the entire modell wireframe is also displayed in gray to help identification of model parts.
Save as default If this option is checked the current state of Logical parts, Refresh all and Show non-visible parts grayed
will be saved as the default setting when closing the dialog.
2.16.15. Sections
Lets you create section lines, planes and segments through any surface model, that can be used to
process the results (displacements, internal forces, etc).
If a truss, rib or beam is within an active section plane and the result component has values on these
elements a diagram is displayed on these line elements too.
Section segment results can be listed in the Table Browser.
See 6.1.5.1 Section segment result tables
User’s Manual X6 Release1 e2 85
If section segment result tables are selected only section segments within the active parts are
listed.
Creating a section
segment group
86
Section segment groups can be created to make it easier to turn on/off several section segments to-
gether. Click New section segment group, enter a name for the group (name) and define any number of
section segments. End definition by pressing [Esc]. Section segments will be numbered (xx) and get into
the name folder as name_xx. Extension of an existing section segment group is also possible selecting
the folder or a member of that before starting this function.
Creating Creating folders offer a way of sorting sections. Segments can be moved and rearranged by dragging
new folders them to a new position within its own type group. [Ctrl] and [Shift] allows multiple selection in the usual
way.
Turning folders on/off turns on/off the segments within the folder.
New section To define the segment enter two points of a domain or on domains in the same plane.
segment Setting the radio buttons you can control how the internal forces diagram will be displayed. The average
and resultant integrated values are calculated for subsegments defined between adjacent changings of
the sign of the diagram.
If Envelope or Critical is selected the envelope or critical values are calculated for each point (as
it is shown in the diagram) and then integrated. Therefore the integrated value is unreal in the
sense that it is computed from not simultaneous values.
If combined seismic result (X, Y, Z, +, -) is selected, the combined values according to the CQC or
SRSS method are calculated for each point (as it is shown in the diagram) and then integrated.
Therefore, the integrated value is unreal in the sense that it is computed from not simultaneous
values.
Left or right segment width can also be specified. In this case the diagram values are computed by
averaging along the segment width. For the more expressive visualization of the segment width the
diagram can also be displayed on both side of the strip.
Diagrams are usually displayed perpendicular to the element plane but checking the option Draw dia-
gram in the plane of the elements rotates the diagram into the plane. In the Display parameters dialog
this parameter can be turned on/off for all section segments.
If the option Draw diagram in the plane of the elements is selected, the option Diagram with
segment width cannot be used.
The section segment divides the domain into two parts. For
the calculation one of them is removed. The removed part is
selected such a way that the local axis perpendicular to the
section segment (axis y denoted by yellow in the picture)
points outward from the remaining part. The arrows are dis-
played on the side of the removed part. They show the forces
and moment acting on the remaining part by the removed
one. The direction of the section segment is shown by the ar-
rows along the section plane contour. The section segment is
directed from the first to the second point picked at its defi-
nition. The picture on the right shows the positive direction of
resultant components.
The resultant integrated over the segment is displayed only for section segments parallel with
the local x or y axis.
If Envelope or Critical is selected the resultant components are displayed in Min or Max mode
only. The values of the three resultant components are not simultaneous.
If Envelope or Critical is selected the resultant value is integrated for each envelope or critical
load combination over the segment and then the minimum or maximum is shown. Therefore the
displayed result belongs to only one load combination and it is independent from the displayed
diagram that contains non-simultaneous values.
If combined seismic result (X, Y, Z, +, -) is selected the resultant value is integrated for each modal
load case over the segment and then the combined result according to the CQC or SRSS method
is shown. Therefore, the result is independent from the displayed diagram that contains non-
simultaneous values.
New section plane Click New section plane and assign a name to the section. This type of section is based on a plane. Click
or enter two points to set the section plane. Then click OK in the Selection Icon Bar to save. In perspective
view you have to click or enter three points to set the section plane. Section planes are displayed as
rectangles of dotted lines. You can enable/disable the display of section plane rectangles.
Section planes are useful when you want to display results only along a certain line through the entire
structure.
88
New section line Click New section line and assign a name to the section. You then have to select surface edges or beam
elements that define the section line. Then click OK in the Selection Icon Bar to save. Section lines can
be discontinuous.
The tracelines of the section lines are not correlated with the directions of the result components
displayed.
Virtual strips are like virtual beams but sections do not run until the domain edge (they are limited by a
fixed strip width on the left and on the right side, ΔL and ΔR).
Virtual beam sections are created from both the meshed and unmeshed domains. Virtual strips
require a mesh.
User’s Manual X6 Release1 e2 89
The definition of virtual beams and strips is much quicker on unmeshed domains, since then the
program calculates the intersections with all the finite elements only after mesh generation, not at
the moment of definition.
Axis of reduction of Centre line: The axis of reduction is the centre line of the virtual beam. The local system
virtual beams of the virtual beam can be set independently from the axis of reduction by setting the
local x and z direction under References.
Straight line: The program slices the domains along the local x direction, perpendicular
to the x direction. The cross-section at a certain point consists of the intersection of
the domains and the section plane. The first and last non-empty cross-section defines
the start point and end point of the virtual beam. The axis of reduction is a straight
line between the center of gravity of the first and last section. The axis of reduction
does not necessarily run parallel with the local x direction but depends on it.
Point and reference vector: The axis of reduction will be a line going through a given
point and parallel with a selected reference vector.
Two points: The axis of reduction goes through two selected points.
Axis of reduction of Centre line: The axis of reduction coincides with the centre line of the virtual strip,
virtual strips which is the collection of centres of gravity of the sections perpendicular to the line
defined by the two end points of the strip.
90
Excentric: The creation of segments is identical, but the internal forces are reduced to
a line segment, excentric to the one connecting the end points of the strip. The ec-
centricity is governed by the transition of the end points of the strip.
References References are set automatically. By default, the local x direction of the virtual beam is parallel with the
longest side of the bounding box of the selected domains in the global space. The local y and z direc-
tions are set the same way as the automatic references for beam elements. The references can be cus-
tomized. (See 4.9.22 References).
Cross-section A custom cross-section can be defined for virtual beams and strips. This cross-section can be read
through the COM interface and can be used in external design programs.
Creating Creating folders offer a way of sorting virtual beams. Virtual beams can be rearranged by dragging them
new folders to a new position within its own type group. Turning folders on/off turns on/off all virtual beams within
the folder. [Ctrl] and [Shift] allows multiple selection in the usual way.
Virtual beams This checkbox controls the display of virtual beams in the model.
The same virtual strip with (left) or without (right) displaying strip width.
Speed button A dedicated speed button at the bottom right allows turning on/off virtual
beams and strips.
User’s Manual X6 Release1 e2 91
2.16.17. Find
Graphics Mesh
Symbols Enables display of the inner mesh lines.
When disabled the generated mesh lines are not displayed.
Node
Enables display of the nodes (small black rectangles).
Surface center
Enables display of the center point (selection point) of the surface elements.
Color codes: plate = red, membrane = blue, shell = green.
Center of circle
Enables display of centers of circles as a small cross.
Domain
Enables display of the domain’s contour.
The color of the domain is the same as of the surface type.
Color codes: plate = red, membrane = blue, shell = green.
Nodal support
Enables display of the nodal supports.
Nodal supports appear as thick axes.
Color codes: axial displacement=yellow, axial rotation= orange.
Line support
Enables display of line supports.
The line supports appear as a thick edge.
Color codes: axial displacement=yellow, axial rotation= orange.
Surface support
Enables display of the surface supports.
Surface supports appear as a light brown hatch .
Footing
Footings designed on the R. C. Design tab appear with their calculated or specified shape and
size.
Dimension lines
Enables display of footing dimension lines.
Links
Enables display of link elements.
Node-to-node link elements are displayed as solid green lines with an arrowhead showing the
location of the link.
Line-to-line link elements are displayed as solid green lines with an arrowhead showing the
location of the link and dashed green lines at the line end points.
Rigids
Enables display of rigid bodies. They appear as thick black lines.
Diaphragm
Enables display of diaphragms as dashed thick purple lines.
Reference
Enables display of the references.
Red vector, crosshairs or triangle.
Cross-section shape
Enables display of the shape of the cross-section of the truss/beam/rib elements.
The user-defined cross-sections will be displayed as rectangles that circumscribe the shape
of the cross-sections.
End releases
Enables display of the end release and edge hinges.
End release:
Blue circle: hinge / roller
Blue circle + cross semi-rigid hinge
Red circle: spherical hinge
Solid blue circle: plastic hinge
Edge hinges
Circles on the edges.
Structural members
Enables display of the structural elements.
User’s Manual X6 Release1 e2 93
An orange line along the member and the number of the member.
Reinforcement param.
Enables display of two intersected straight lines at surface
centers where reinforcement parameters are assigned. The
direction of these lines indicates the defined reinforcement
directions. (See 6.5.1 Surface reinforcement parameters and
reinforcement calculation - RC1 module).
Reinforcement domain
Enables display of mesh independent reinforcement do-
mains as dashed brown outlines.
Top and bottom x and y reinforcements are also displayed.
Two vertices of the polygon are connected to the center
by brown lines.
Mass
Enables display of the symbol of the concentrated masses.
Double red circle.
ARBO-CRET elements
Aschwanden ARBO-CRET elements placed into the model.
A schematic drawing of the element is displayed.
AIRDECK elements
AIRDECK elements placed into the model.
Void formers are displayed as circles in wireframe mode and balls in rendered view.
Thickness reference points
Reference points entered during definition of tapered domains
Orange crosses, thickness value labels, dashed lines connecting the reference points.
Reinforcement of cores and walls
For detailed description, please see 6.5.14 Design of reinforced concrete cores and walls – RC5
module
Stress-strain analysis parameters
Structural elements with stress-strain parameters are shown in the main window with zigzag
lines.
Local Enables display of axes of the elements in the local coordinate system.
systems
Loads Display of load symbols can be set separately for each load
type (concentrated, distributed along a line, distributed on
surface, temperature, self weight, moving load, miscella-
neous (length changing, tension / compression).
To display of surface loads distribution to beams (see the
diagram on the right) check Load distribution. To display the
derived beam loads, check Derived beam load.
94
Load panel Load panel outlines are thick dashed turquoise lines. If load panels are attached to domains small rec-
tangles appear along the outline showing the linkage.
Abutting wall or Load panel edges with abutting wall or parapet can be selected when defining snow loads. These edges
parapet (for snow are displayed with a light brown rectangle along the edge.
loads)
Save as default Saves the current symbol display option settings as default for new models.
Labels
Numbering
Displaying the number of nodes, elements, ma- For meshed line elements checking Use finite ele-
terials, cross-sections, references. ment numbers displays the number of finite ele-
ments instead.
User’s Manual X6 Release1 e2 95
Use finite element Turning on this switch replaces structural member numbers on diagrams with finite element numbers.
numbers Tables also display results on a finite element basis and not for structural elements.
See 3.2.14 Assemble structural members, 3.2.15 Break apart structural members
Labels on lines seen Checking/unchecking Labels on lines seen from axis direction turns on/off labels on lines seen from the
from axis direction direction of their axis (seen as points).
Transparent labels If this option is checked all labels are transparent. Otherwise, labels are drawn on a rectangle filled with
the background color.
Prevent labels from If this option is checked the program prevents labels from overlapping by moving and hiding less im-
overlapping portant labels. In the latter case zooming in may increase the number of visible labels. If large number
of labels are to be displayed this optimization may be a time-consuming process.
Story center of Enables display of center of gravity of each storey. AxisVM converts loads of load cases
gravity used to calculate the vibration shapes for seismic analysis into masses then calculates the
center of gravity for each storey. The centers are displayed as black +s in black circles
with a label Gmi where i is the level number.
Story shear center Enables display of shear center of each storey. AxisVM calculates storey shear centers by finding wall
sections and using the same method as for thin-walled cross-sections.
The centers are displayed as red +s, with a label Si, where i is the level number.
Properties Enables display of the name and values of materials properties, cross-sections, element lengths or
thicknesses, load values, masses, etc.
If the Units option check-box is enabled, the labels will include the units as well.
Write values to Controls how the result labels are placed over the drawing.
Actual reinforce- Enables labeling for top and bottom reinforcement in x and y directions independently and sets the
ment labeling mode.
Turning on According to the displayed result component makes the current reinforcement component
the only displayed component.
Switches
Information Coordinates See 2.19.1 Coordinate window
Windows Info See 2.19.2 Info window
Color coding See 2.19.3 Color coding
Color legend See 2.19.4 Color legend window
Display The display of the actual parts and guidelines can be turned on and off.
Parts Enables/disables the display of user-defined and logical parts.
Guidelines Enables/disables the display of guidelines.
Structural grid Enables/disables the display of the structural grid.
DXF/PDF layers Enables/disables the display of background layers.
Architectural layers Enables/disables the display of imported architectural objects.
AxisVM layers Enables/disables the display of layers defined within AxisVM.
Allows the selection of the options for the settings of the grid, cursor, editing, drawing parameters and
design code.
96
Cursor step Allows to choose coordinates of an invisible dot mesh (not the grid).
You can set the cursor step parameters as follows:
Mouse snap - Restricts the movement of the mouse cursor to an invisible grid specified by the cursor
step values below.
ΔX, ΔY, ΔZ - Restricts the cursor movement to regular intervals. Each time you press a cursor movement
key the cursor moves in the corresponding direction (X, Y or Z) one step
(X, Y or Z respectively).
Ctrl x - Sets the value of a factor that increases or decreases the cursor step size if you press the [Ctrl]
key when you move the cursor. This allows you to achieve adequate positioning accuracy.
The cursor step is ignored if you position the cursor on a line not parallel to global coordinate
axes. In such a case, the cursor will move along the line.
If the editing tolerance is greater than the cursor step, the mouse will follow an invisible grid
specified by the editing tolerance.
When using with constraints, the cursor step is applied in the constrained direction with the DX value.
See 4.7.4 Constrained cursor movements
If the grid step and the cursor step is set to the same value, nodes will be placed snapped to the
grid.
2.16.19.2. Editing
Auto Sets commands that are applied automatically if the corresponding check-box is enabled.
Intersect
Sets the line intersection handling. At intersection points of lines a node will be generated and lines
will be bisected. If surfaces are intersected by lines, they will be split, and the resulting elements will
have the same material and cross-sectional properties as the original.
Part management:
Any entity drawn or modified after the check-box is enabled will be associated with all of the active
parts.
Refresh
Sets the display refresh mode to automatic.
Editing tolerance If two nodes are closer than the value set as the editing tolerance, they will be merged in the case of a
mesh check. This value is also used when comparing surface thickness or beam length.
Cursor identification The element under the cursor is identified if it is within an adjustable
cursor identification distance. The unit for cursor identification distance is
pixels.
If more than one element is within this range the closest one will be identified.
See 4.7.1 Cursor identification
Plane tolerance Nodes of domains and surfaces must be in plane. If a node of a domain or surface deviates from this
plane more than the given value the element will be deleted. Plane tolerance can be specified in two
ways:
Relative [‰] per thousand of the biggest extension of the element polygon
Absolute [m] a given value
2.16.19.3. Drawing
Load symbol Sets the display size of the load symbols. This factor
display factors is applied when the checkbox in Symbols / Graphics
Symbols / Load is enabled. These values do not af-
fect load values.
Force
Sets the display size of the symbol of concent-
rated force loads.
Moment
Sets the display size of the symbol of concent-
rated moment loads.
Line / surface load
Sets the display size of the symbol of line /
surface loads.
Section segment resultant value
This factor determines the size of the arrow rep-
resenting the resultant value over a section seg-
ment.
Contour line angle Sets the display of the inner mesh lines (between adjacent surface elements). The common edge of two
or more surface elements is displayed if the angle enclosed by the normal to the planes of the elements
is larger than the value set here.
98
Edge is displayed
Zoom factor Sets the scale of magnification/reduction of the zoom commands associated to the [ + ] and
[ - ] keys.
Display of axial Until Version 12 normal force and torsion moment diagrams were always drawn in the local x-z plane.
force and torsion From Version 13 it can be set to the local x-y plane as well.
moment diagram
Options:
• All commands in the main menu, on the icon bar, coordinate window, etc. that can have a keyboard
shortcut assigned
• All the symbols, labels and switches configurable on the Display options dialog (2.16.18 Display opti-
ons)
• Load cases (on the Loads tab or on result or design tabs)
• Result components load situations (on tabs displaying results)
• Settings available at Settings / Preferences (3.3.11 Preferences)
• Model options available at Settings / Model options (2.16.19 Model options)
• Elements can be found by their number or name
• Elements can be selected with or without filter criteria
Pressing Enter will select Node 12 and move the mouse pointer
there. In short, you can find Node 12 by typing no 12[Enter].
As the smart command line finds the term inside expressions, nodal support 5 can be found also by
typing sup 5[Enter], taking advantage of the fact, that Nodal support appears at the top of the list after
narrowing down the matches.
Selection It is also possible to select or deselect certain elements from the smart command line. If parts are acti-
vated, these commands operate on the visible parts only.
Simple selection If the first character is *, the list displays the selectable elements
only and the command will select the chosen item.
Typing *- at the beginning will deselect the chosen item.
To select everything type *[Enter], to deselect everything type *-
[Enter]
To select all nodes type *no[Enter]
Nodes with a given degree of freedom can also be selected. Se-
lection of other elements like beams, trusses, ribs, domains work
the same way.
User’s Manual X6 Release1 e2 101
Selection by Type * and start typing the name of a material or cross-section (or type mat or cr) to select elements
a property made of a certain material or having a certain cross-section. Typing *Bottom rib, *Rib in midplane,
*Top rib the appropriate ribs can be selected. It is enough to type the first characters of these expressions
and choose from the list. Similarly, typing *Variable cross-section selects these line elements, while
typing *Variable thickness selects these domains. Domains of a special type can also be selected by
typing *Ribbed, *Hollow core, *Composite ribbed slab, *Trapezoidal steel deck, *XLAM, * Custom
stiffness matrix. Architectural type can also be used (*Columns, *Beams, *Walls, *Slabs).
Selection by Elements can also be selected by a parameter value. An example: Select line elements longer than 5 m.
a parameter After typing *le[Enter] the command line changes to *Length (m) and the list is replaced by a help mes-
sage
Type the filter condition then press Enter.
Examples: >0, >=10, <>20, =30, <=40, <50
Continue typing to get *Length (m) > 5[Enter]. The appropriate line elements will be selected.
The condition can include one of the >, >=, <>, =, <= and < symbols and a numerical value.
Supported parameters are: *Length, *Beam eccentricity ey, *Beam eccentricity ez, *Rib eccentricity
ey, *Rib eccentricity ez, (line elements), *Area, *Thickness, *Eccentricity (domains).
Selection by If the active tab is Static or any design tab to the right, elements can be selected by a result component
result value value. The appropriate elements matching the condition in every cross-section or finite element point
will be selected.
An example: Select elements where eZ < 0.
Type *ez[Enter]. The command line changes to *eZ [mm]. Con-
tinue typing to get eZ [mm] < 0[Enter]. All nodes, midside nodes,
surface centers, beams and ribs where the global Z component of
the displacement is negative will be selected.
If the active view already displays the eZ component the command
is even shorter. Simply type *<0[Enter]. If no result component is
entered, the current result component is used to check the values.
Another example: Select failed design members on the Steel design
tab.
Type *u[Enter]. The command line changes to *Utilization [].
Continue typing to get *Utilization[] >1[Enter]. All steel design
members where utilization exceeds 1 will be selected.
Selection by To select elements based on multiple criteria click the right mouse button over the magnifier icon and
multiple criteria choose Selection based on multiple criteria from the popup menu. Entering multiple criteria is more com-
plex so in simple cases Selection based on one criterion is recommended.
An example: Select all columns where normal force is below –5 kN.
Choose Selection based on multiple criteria from the popup menu. Type *co[Enter]. In the other mode
this command would select all columns. But now the command line changes to *Columns | waiting for
further conditions. Pressing another [Enter] would select all columns.
Continue typing *Columns | nx< to get *Columns | Nx[kN] <,
then continue *Columns | Nx[kN] < -5[Enter].
Now a second criterion is added: *Columns | Nx < -5.00 kN |
This is the last one. Pressing [Enter] selects all columns where the normal force is below –5 kN.
Interpretation of
multiple criteria
Criteria relating to different properties (like those of normal force, material and cross-section) are always
connected with an implied AND operator, i.e., all criteria must be met. Criteria for result component
values are also connected with an implied AND operator. Selecting elements with two result criteria in
an OR relation can be completed only in two consecutive selections:
Multiple criteria for materials or cross-sections are connected with an implied OR operator within their
category (i.e., elements with any of the entered cross-sections will be selected) but is in AND relation with
other types of conditions. E. g.
Selects all IPE 80 or IPE 120 columns matching the other criteria.
102
The quick switches toolbar allows you to change the display settings without entering the Display option
/ Symbols or Model options dialog. The icons are located in the bottom right corner of the graphics area.
Auto intersection
Mouse snap
Stories
Workplanes
Virtual beams/strips
Display mesh
Display symbols
Numbering
Some of these settings are available also from Display and Options icons.
The information windows are situated in the graphics area. You can move these windows on the screen
by clicking title bar, holding down the left mouse button, and dragging it to a new location on the
screen.
Shows information about the display of the results such as: active part(s), current
perspective setting, type of analysis, current design code, current load case or load
combination, solution errors, current result component.
For the explanation of E(U), E(P), E(W), E(EQ) parameters see 5. Analysis and
5.1 Static
In case of geometric nonlinear calculation, a warning message appears if a stability
problem occurred in the current increment. See 5.1 Static analysis
In case of nonlinear calculation taking into account the reinforcement short forms
of the reinforcement warning messages also appear. See 6.5.5 Nonlinear analysis
of reinforced concrete beam and column elements and 6.5.6 Nonlinear analysis of
RC surfaces.. The warning message appears if the problem occurred in the current
or in any previous increment.
If more than one part is activated a list of active parts is displayed provided that the number of parts
does not exceed a limit. This limit can be set by right clicking the info window and clicking the Settings
menu item.
This info window appears after activating a color coding (see 2.16.5 Color coding) unless
it is turned off in the Window menu (see 3.6.2 Information Windows).
The type of the coding is displayed within the black header line.
Values
Limits Setting criteria for the interval limits:
Min/max of model
Sets the lower and upper limit values to the minimum and maximum values of the entire model.
Intermediate values are interpolated.
Min/max of parts
Sets the lower and upper limit values to the minimum and maximum values of the active parts.
Intermediate values are interpolated.
Round If the interval limits are interpolated between the minimum and maximum values (no custom values or
calculated values step value has been introduced) the interpolated values can be rounded.
By step value
Color values are determined by the given step . When entering a new level value the other levels
will be recalculated using the step. Switching from other crieria the array starts from the lowest value
and using the latest step value.
You can save the settings of the scale using the Save As button. To review saved settings click the ...
button.
Hatching for out-of- Hatching for out-of-range values can be set to Opaque or Transparent.
range values
Standard interval limit settings are also available directly from the color legend window popup menu.
To activate popup menu click right mouse button on the window.
User’s Manual X6 Release1 e2 105
Calculate
When displaying reinforcement values click Custom and Calculate to
get the amount of reinforcement from rebar diameters and distances
for the selected list item.
Color gradient A color is assigned to the scheme according to its specific rebar area.
by rebar area
Color coding by re- A color is assigned to the scheme according to the rebar diameter and spacing. Hue depends on the
bar diameter diameter and brightness depends on the spacing.
and spacing If reinforcement domains overlap the color of the overlapping area can be determined by one of the
following methods:
Grayscale by rebar area
A gray color is assigned to the area according to the total specific rebar area.
Color coding by average of overlapping reinforcements
An average spacing and diameter is calculated as
1 𝑑𝑖
𝑠= ; 𝑑=𝑠∙∑
1 𝑠𝑖
∑𝑖 ( ) 𝑖
𝑠𝑖
3.1. File
Creates a new untitled model. Use this command to start a new modeling session. If you have not saved
the current model, a prompt appears asking if you want to save it first. Refer to the Save and Save as
commands for more information on how to save your current model.
You must specify a name for the new model. You can select the appropriate Standard and system of
units. You can enter specific information in the Heading section, that will appear on each printed page.
A new model uses the default program settings.
If a page header logo was selected for printing drawings, tables and reports, the header also appears
here. Click on the logo to change it or try Settings / Preferences / Report / Company logo / Settings
108
AxisVM can preserve previous revisions of the model. This feature is activated by default. The Previous
revisions dialog lists previous revisions in reverse chronological order (the topmost item is the latest one).
The File size column shows the size of the model file (and the result file if it exists).
Revisions can be opened and locked. Comments can be entered and edited by clicking on a cell in the
Comment column.
There are two ways to open an earlier revision: opens it As a new model or Saving, then overwriting the
current model. This latter option saves the current state of the model as a revision then the earlier revision
overwrites the model.
Save a previous revision. Saves the selected revision under a new name. It is possible to preserve earlier
revisions.
Lock revisions. Locks the selected revisions. These revisions are not deleted even if the number of saved
revisions has reached the maximum defined in Settings.
Unlock revisions. Unlocks the selected revisions. If a revision is unlocked, it will be deleted if necessary.
Previous revisions are stored in a \Revisions\model name subfolder created below the folder of
the model file. Copying or moving the file does not copy or move these subfolders. In order to
copy or move the revisions the above subfolder must be copied or moved to the new location.
3.1.3. Open
Loads an existing model into AxisVM. If you have not saved the current model, a prompt appears asking
[Ctrl]+ [O] if you want to save it first. Refer to the Save and Save as commands for more information on how to
save your current model. Selecting this command will bring up the Open dialog box.
If the folder name appearing in the dialog box is what you want, simply enter the file name in the edit
box or select it from the list box. If the directory is not what you want, select the drive and directory
names along with the file name.
AxisVM saves your model data in file names appearing as Modelname.AXS (input data), and
Modelname.AXE (the results). Both file contains the same identifier unique for each save which
makes it possible to check if AXS and AXE files belong to the same version of the model.
The open dialog looks different on Windows XP and Vista / Windows 7 / Windows 8 operating systems.
3.1.4. Save
Saves the model under the name displayed at the top of the AxisVM screen. If you have not saved the
[Ctrl]+ [S] model yet, the Save as dialog box automatically appears prompting you to enter a name. Use the Save
as command if you are changing an existing model, but want to keep the original version.
If you select Create backup copy in the Settings / Preferences / Data integrity / Auto save a backup file of
your previous model will be created.
If Settings / Preferences / Data integrity / Compress AXS is checked the model file will be saved in a
compressed format. The average size of the compressed file is about 10% of the original. Compression
is more efficient on large files.
3.1.5. Save as
Names and saves the model. Use this menu command to name and save a model if you have not saved
the model yet, or if you are changing an existing model, but want to keep the original version.
Selecting this menu command will bring up the Save As dialog box.
Converting models Models created with previous AxisVM versions (if applicable) will be converted into the current version
file format when you open them for the first time.
The File / Save As / File Format command lets you save the model in earlier formats.
The save dialog looks different on Windows XP and Vista / Windows 7 / Windows 8 operating systems.
3.1.6. Export
Export functions create different types of files to transfer data to other engineering applications.
DXF file Saves the geometry of the model to a DXF file format for use in other CAD programs. The geometry is
saved with actual dimensions, in a Modelname.DXF file.
Selecting this menu command will bring up the Export DXF dialog box, which lets you specify the units
of measurement in the exported file. The results set in the current view (in the diagram, contour, surface
2D, etc.. view mode) can also be exported.
110
Reinforcement design files are isoline diagrams of required surface reinforcement exported with labels
and a color legend. Each direction is on a separate layer. To export this type of DXF file, go the R. C.
Design tab and select a result component related to required reinforcement.
It is recommended to use the AutoCAD 2004 format to avoid data loss. Earlier DXF formats support 256
palette colors and ASCII characters only.
For further DXF file export options and their description, see section 3.1.6.7 Export to other applications.
IFC 2x, 2x2, 2x3 file Export of this file type requires an IFC module.
Exports an IFC file describing the model with achitectural objects (walls, slabs, columns, beams). IFC files
can be imported in ArchiCAD, AutoDesk ADT, Revit, Nemetscheck Allplan, Tekla-Xsteel and other
architectural programs.
Checking Export reinforcement on the save dialog the and clicking on
Preferences displays a dialog to set how the actual reinforcement will be
exported.
Since the release of 13R1, users have the possibility to export reinforcement to .ifc files.
The built-in algorithm of AxisVM helps to export not only the field reinforcement based on user given
intensities, but shaped bars too, like hairpins, hooks etc.
• Closing reinforcement on free plate
edges: free plate edges are closed by
hairpins, whose lengths are calculated
based on the lap length of bars, which is in
turn calculated either by the standard set in
the program, or the user given value of l0.
This option can be turned on or off.
• Reinforcing concave corners: concave
corners can be strenghtened by simple
bars, whose lengths are calculated based
on the lap length of bars, which is in turn
calculated either by the standard set in the
program, or the user given value of l0. This
option can be turned on or off.
• Convert short rebars to stirrups: to avoid
useless duplicating of rebars, the algorithm
can be given a multiplicator value. Those
simple bars, whose lengths are greater
than the minimal lap length multiplied by
this value, will be converted to a closed
stirrup. This option can be turned on or off.
• Closing of rebars at joining plates: along
the edges of plates that do not share the
same plane, the algorithm can connect the
reinforcements of the individual plates
either by hairpins or by hooks. The actual
geometry of these shaped bars depend on the width and angle of the joining plates, and
sometimes, the algorithm revises the user's choice. The sizes of the shaped bars are calculated
with respect to the minimal lap length.
User’s Manual X6 Release1 e2 111
Export model data If the Architectural model option is selected only architectural objects will be exported. If Static model
is selected, finite element meshes, loads, load cases, load groups and load combinations will be included
in the IFC file. Dynamic, influence line or moving loads will be excluded.
Started from Revit2017 the popup window shown below will appear at the initialization phase of
every add-in or third party executable.
In this case, please press "Load Once" if you wish to load the add-in, and use its abilities, or
"Always Load" if you wish to suppress this dialog in the future.
Exported data For the time being, only geomatrical data are exported to Revit, which are:
• linear elements with their materials and sections
• regions and their materials
• supports
Usage Export process must always be started from AxisVM, the way mentioned below:
File menu -> Export -> Export to revit (*.are)
By pressing the Settings button you can choose between several options:
112
• Entire model – all exportable data will be written into the file
• Displayed parts – all parts visible on screen will be exported
• Selected elements – only elementes selected before will be exported
Having saved the .are extension file you can import it into Revit like on the attached picture:
In as much as there is an opened project in Revit, then both Export and Import tabsheets will be enabled,
otherwise only Import tabsheet can be used.
To import .are extension files you have to press the button located left to the Filename label.
After this the tree on the left is refreshed by the content of the file, and a preview of the model to be
imported is visible on the bottom right corner.
User’s Manual X6 Release1 e2 113
Elements that are actually imported can be filtered by checking the nodes of the left tree.
Import method If you have an open project at the moment of import, you can select from several options:
• Overwrite – all elements will be deleted from the project (except for the ones pinned before),
and new elements will be imported into this purged project
• Append – new elements are add the project preserving older items
• Create new project – opened projects remain unchanged, and a brand new one will be created,
into which all elements will be imported
On the other hand, if you have no opened project, then a new one will be created.
It is important to note, that if the project is read-only, then only the third option is enabled.
Mapping materials As AxisVM material records can not be directly mapped to Revit materials, user interaction is needed to
help on this mapping. It can be done on the form seen below:
114
You can only continue the import process only if all mappings are done on this form. In order to map
AxisVM material to Revit material, you have to select a material as material type on the left tree, and
another one from the rigth tree. Then if you press the button between the two trees, the selected AxisVM
material will be mapped onto the selected Revit material. Selection on the left tree will then automatically
jump on the next unprocessed node. Unprocessed nodes are drawn red, everything else is drawn black.
You do not have to do this mapping one by one. Instead, if you select a node with several children, then
all the node’s children will be assigned the selected Revit material.
If you select the last node of the right tree, a new Revit material will be inserted into the database with
properties inherited from the selected AxisVM material.
Having accomplished all mappings, OK button will be enabled, and you can continue the import process.
Importing ASF file • To import an asf file go to Create/Interfaces/Import FEA files menu item in Allplan menu.
into AllPlan • Select the folder with asf files from the directory structure in the top-left window. You can see the
asf files in the bottom-left window. Select one or more asf files from this list and select the place
of these files by clicking in the row of the list on the right side (Destination Files).
• Click on the Transfer -> button, and the selected files will appear in the list.
• Click on Import button, and the selected files will be imported into Allplan program.
• Close the window with OK
Displaying imported • Click on Create/Engineering/Bar Reinforcement/FEA reinforcement Color Image menu item in
reinforcement in the Allplan menu
AllPlan • Click on the desired asf file
• Choose layer (top/bottom) and type (compression/tension) of the imported reinforcement
values
• Click on OK
This Excel-based file format supports exchange of structural model data with other applications support-
ing this protocol. The file consists of standardized spreadsheets with the geometry, structural and load
properties of the model, thus can be viewed and/or edited with Excel. SAF protocol is supported by
ArchiCAD, AllPlan, SCIA and RFEM.
116
• trusses, beams, columns, ribs, supports, regions and connection elements (hinges, joints)
• cross-sections
• support stiffnesses
• load, load combinations
Usage Select File menu / Export / SAF export (*.xlsx), select the location and filename for the export and press
Save.
Bocad file Saves the geometry of the model into a file format that is recognized by the Bocad software. The file
includes the coordinates of i and j-end nodes, the cross-sectional properties and the reference point of
truss and beam elements.
StatikPlan file For StatikPlan AxisVM exports a DXF file including the contour of the reinforced concrete plate, the
calculated reinforcements as isolines and the result legends on different layers.
PianoCA file Generates a *.pia interface file for PianoCA. It includes the data, supports, loads and the calculated
results of the selected beam elements.
CADWork file Creates a DXF file to use in CADWork reinforcement detailing software.
Only selected domains will be exported.
As CADWork works in 2D, selected domains must be in the same plane. Each domain in the DXF file is
transformed to a local X-Y coordinate system, Z coordinate represents the calculated amount of rein-
forcement.
SDNF 2.0, 3.0 file Saves the model in SDNF (Steel Detailing Neutral Format) file readable by steel detailing products (Ad-
vance Steel, SDS/2, Tekla Structures, PDMS).
Glaser -isb cad- Exports reinforcement domains to the Glaser -isb cad-program.
*.fem file
3.1.7. Import
Import functions read and process different types of files to transfer data from other engineering
applications and provide a link to BIM systems (Building Information Modeling).
Place Lets you specify the plane of the DXF layer (X-Y, X-Z, or Y-Z).
The Place button allows to graphically position the imported DXF drawing in your model space.
Visible layers only With this option AxisVM imports only the layers set visible in the DXF file.
118
Import Hatching is represented by individual lines in a DXF file so in most cases it is not recommended to
hatch pattern lines import them. If you need the hatching check this option.
This function is available only on Geometry, Elements, Loads and Mesh tabs!
Static model From IFC version 2x3 it is possible to export details of the static model (nodes, topology, supports, loads,
load combinations). The Static model option is available only if the file contains this information. If it
describes architectural objects (columns, beams, walls, slabs, roofs) only the static model can be created
automatically in AxisVM after importing the file.
Architectural model This option can overwrite or update existing architectural model information in the AxisVM model. Ax-
objects isVM can read columns, beams, walls, slabs, roofs.
See 4.9.23 Creating model framework from an architectural model
Opening aligned to Import as opening
domain edge
Moving the struc- If checked, AxisVM calculates the centre of gravity (COG) of the imported model, translates the whole
ture to the origin model, so that X and Y coordinates of the COG get to (0, 0) of the coordinate system. It can be useful
when the imported model has large coordinate values.
If Also in Z direction checkbox is checked, a vertical translation will be also performed, moving the Z
coordinate of the COG to zero too.
Automatic detection This function lists new, modified and deleted objects when reimporting IFC files. Each change can be
of changes approved or ignored. Selecting a change in the tree shows the position of selected element within the
current model.
User’s Manual X6 Release1 e2 119
Filtering objects
Filtering by Model history – The user can choose to import new elements or modified elements
This type of filter works properly only on files previously exported from AxisVM and modified in another
program.
Filtering by Properties – The user can choose elements with a load bearing capacity and/or by fire rating.
This type of filter works properly only if this information is included in the file.
Filtering by on Layers – if the IFC file contains layer information for the entities the user can select which
layers to import and which layers to ignore.
When exporting a model from ADT (Architectural Desktop) turn off the automatic intersection
of walls before creating the IFC file.
Import from Revit AxisVM can import data through a direct connection between AxisVM and Autodesk Revit 2015 (or
newer). It is based on the Revit API which lets other programs to query and modify its database.
At the end of AxisVM setup process the installer creates a folder called "RevitImport" under the AxisVM
folder, and copies all the necessary files.
The .addin extension file is created in folder
c:\Users\<UserName>\AppData\Roaming\Autodesk\Revit\Addins\yyyy\, where "yyyy" represents an in-
teger value greater or equal to 2015. Both this file and the whole content of "RevitImport" folder are
necessary for the proper communication between the two programs.
There are two methods of communication
120
Started from Revit2017 the popup window shown below will appear at the initialization phase of
every add-in or third party executable.
In this case, please press "Load Once" if you wish to load the add-in, and use its abilities, or
"Always Load" if you wish to suppress this dialog in the future.
Usage The scheme of dataflow is
Project The project info declared in Revit is listed in the upper-left list, while the image to the right shows a
screenshot of the project as could be seen in Revit.
Import method If the AxisVM model is not empty, two options are available to overwrite the model or update it.
Structural model If this option is activated AxisVM changes the raw geometry data to build a proper structural framework.
Colliding beams are connected with a rigid body, neighbouring domains are joined, beams running on
a domain are converted to ribs. Some of these calculations require considerable resources and may slow
down the import process.
Arc resolution Certain geometry calculation requires arcs to be converted into polylines. The conversion criterion can
be set here.
Editing tolerance Tolerance used when calculating geometry.
Joining objects To improve the efficiency of collision detection bounding boxes of objects are inflated a bit. This toler-
ance value can be set here. Active only if the checkbox on panel Structural model is checked.
Import through Use this option if you process the data extracted from Revit yourself and you run Revit and AxisVM on
COM interface the same machine. After a successful export through the AxisVM COM interface, AxisVM launches and
you can continue editing the model there. The above geometry parameters must be set in the Revit
export dialog.
Moving the struc- If checked, AxisVM calculates the centre of gravity (COG) of the imported model, and translates the
ture to the origin whole model so that X and Y coordinates of the COG get to (0, 0) of the coordinate system. It can be
useful when the imported model has large coordinate values.
If Also in Z direction checkbox is checked, a vertical translation will be also performed, moving the Z
coordinate of the COG to zero too.
Section recognition Sometimes the cross-section information read from the Revit database is limited so identification of
sections requires user interaction. If so, a dialog is displayed.
User’s Manual X6 Release1 e2 123
From AxisVM li- If the automatic recognition has failed, you can select the proper section from the AxisVM section library.
brary
Mapping Revit pa- Storage of section parameters in Revit is not consistent so a unique parameter matching may be needed.
rameters to AxisVM You can select a property from the list and move it up or down into the proper position to associate
parameters parameters in AxisVM and Revit.
The add-in learns the parameter matching you defined so next time it will identify the section
properties without user interaction.
Skip this section Skipping the section means that all elements having this section will be imported as a general object
with its boundary representation but without a structural model.
Setting materials Extracting proper material information from Revit poses similar problems. If the add-in fails, it will call
your help to identify the materials that could not be recognized, and you can associate AxisVM material
names to element names.
You can leave some elements unassigned. At the import phase, AxisVM will assign a default material to
them.
If the installer If the installer fails due to any reason and Revit does not find the dll file of the add-in you can install it
fails... manually by running !REGISTER_Revit.BAT from the folder where AxisVM is installed.
124
It is important to know, that the communication between the two programs is based on the Revit
API interface. The structural model of the elements is queried from the Revit database (if Revit
does not provide this information, AxisVM can import only the wireframe of the element). This
structural model still requires many minor adjustments to make it usable. AxisVM tries to join
objects (if you check this option) but sometimes you have to edit the model to get a proper struc-
tural framework.
Comparison based Since version X5, AxisVM and Revit can share a workflow between each other. This is actually a compar-
data exchange ison based export-import.
With this, engineers working on the same project can import new or modified entities without effecting
unchanged elements.
The workflow process can be started both from AxisVM and Revit.
When starting from Revit one must leave the checkbox on the bottom of the main window checked.
The reason is that in order to inject the necessary parameters into the project file, the application needs
the permission of the user and to save the file afterwards. If it can not happen, the exported file will not
contain workflow parameters. AxisVM will still be able to read the produced file, however comparion
based import will not be possible.
When starting the workflow from AxisVM there is no need for such acts, necessary parameters will
automatically be embedded in the file.
One can use comparison based import only if there is a modell previously opened in either application.
While in AxisVM, you can see the lines shown below when importing a file previously saved with workflow
parameters:
For the comparison to actually happen you have to check the checkbox shown below:
If after this we find them equal, the imported item is skipped. However, if they are found to be different,
arriving item is processed, and the former one is deleted.
When -for any reason- we cannot perform a succesful comparison, users can select to keep the existing
item or to import the new one.
These mentioned user preferences can be adjusted prior to the actual import process on the form pasted
below:
Revit form
AxisVM form
Versions before Tekla Structures 2019 support one-way communication only (3.1.9.2 One-way connection
for Tekla Structures versions before 2019)
For Tekla Structures 2019 or later it is recommended to use the IT module to establish a two-way con-
nection between AxisVM and Tekla (3.1.9.1 Two-way connection for Tekla Structures versions 2019 and
later)
126
PDF *.pdf PDF module is required to import drawings from PDF files as a background layer or AxisVM lines. draw-
ings from PDF files as a background layer or AxisVM lines.
Only lines, curves and text objects are processed, images and other elements are ignored.
The final size of the imported drawing can be set interactively on the screen (Find proper scale interac-
tively) or can be adjusted by specifying a Scale.
Importing text and line width is optional. Note: certain pdf creating engines export text as a series of
graphical objects making it impossible to distinguish between text and graphics.
You can set the Default font to use when displaying text with an unavailable font. After clicking on the
Place button, a page number must be entered for multipage documents. Only one page can be imported
at a time.
For other settings and commands see the above part describing importing DXF files.
This Excel-based file format supports exchange of structural model data with other applications
supporting this protocol. The file consists of standardized spreadsheets for geometry, structural and load
properties of the model, hence can be viewed and/or edited with Excel. SAF protocol is supported by
ArchiCAD, AllPlan, SCIA and RFEM.
Entities imported from the incoming file are
• trusses, beams, columns, ribs, supports, regions and connection elements (hinges, joints)
• cross-sections
• support stiffnesses
• load, load combinations
Usage Select File / Import / SAF import file (*.xlsx) and select the file to be read and press Open.
Alternatively, you can drag and drop the imported file on the main window of the application.
User’s Manual X6 Release1 e2 127
Import The import process differs if a model is already opened in AxisVM or not.
• If not, then AxisVM simply parses the file and imports entities.
• In case of an opened model, the following dialog is displayed
o The first option clears the existing model and imports the incoming file
o The second option adds all incoming entities to the existing model
o The third option compares the existing and arriving model, and imports only those elements
which are either
▪ new
▪ or modified either in their material or in their geometry
If the last option is selected the users can choose to
• compare only the geometry of the two models
• or include the comparison of materials.
Geometry comparison consist of comparing the axes and cross-sections of linear elements, the poly-
gons and thickness of regions.
To handle the situation when the imported data is incomplete, or any unexpected issue happens, users
can choose to either
• ignore the arriving element
• or replacing the existing item anyway
The user can control what happens if certain elements are missing from the arriving file. The two op-
tions
• Keep the original elements unchanged
• Remove them
For the time being comparison-based file exchange supports only linear elements, domains and
supports.
If there are different properties assigned to the same merged elements, the properties of the current
model will be retained. Load groups and combinations if any, are appended to the existing ones as new
groups and combinations, and the load cases as new cases. If no load groups or combinations are
defined in the imported model, the load cases will be appended to the existing ones as new cases.
If the same load case exists in both models, loads will be merged if the Merge load cases with the same
name checkbox is checked.
If both models contains loads that are limited to one occurrence (e.g., thermal) in the same load case,
the load in the current model will be retained.
The Section Lines/Planes Parts with the same name are merged, otherwise they are appended.
When importing an AxisVM file a dialog is displayed.
Use the Place button to graphically position the imported model in your model’s space.
Stereo Lithography Reads the triangular mesh describing the surface of a model from a file in STL format (binary or text).
*.stl file Multiple nodes and degenerated triangles are filtered out. Import can be transferred to a background
layer as well.
Bocad interface Opens a data file created by Bocad steel construction software (*.sc1) and imports beam cross-sections
*.sc1 file and geometry.
Glaser -isb cad- Imports *.geo files exported by Glaser -isb cad- describing beam or surface models.
*.geo file
SDNF file Imports a file exported in Steel Detailing Neutral Format used in data exchange between steel detailing
(Steel Detailing programs.
Neutral Format)
DSTV format In its current version AxisVM is capable of importing stp files implementing "PSS_2000_04" file scheme
*.stp file based on "Standard Description for Product Interface Steel Construction 4/2000". This filetype is de-
signed for storing steel structures' data, and can describe linear structures as well as planar regions.
Even statical calculations can be imported through this interface, though with limited capabilities.
If static calculations and an architectural model are both embedded in the file, the user can choose
importing either the calculation or the architectural model. When trying to import data onto an existing
model, the user can select between overwriting or just supplementing it.
„Parametric design is based on the idea, that specific numbers, i.e., parameters control the geometry.
To fully exploit its potentials, these parameters are linked to each other, and these links form a system.
Thus, it is possible that by modifying only one parameter, the whole geometry changes. This idea is
regularly used in structural design. Imagine that by modifying only the span of a truss, the height and
the truss layout are automatically updated both in the 3D model and in the shop drawings. To build a
system capable of doing this requires time, and therefore it is essential to think it over whether it is
worth the time to create a parametric model. However, once we have built such a system of parameters,
links, and a dynamically changing model, we have a highly flexible tool.”
AxisVM Export The most commonly used parametric design software in structural design are Grasshopper and Dy-
plugins namo. These are often referred to as graphical algorithm editors that can be used to create parametric
models easily. By using the AxisVM Export plugins, the parametric models created in Grasshopper/Dy-
namo can be exported to AxisVM. First, the AxisVM Export plugin should be installed in Grasshop-
per/Dynamo. The AxisVM Export plugin communicates with AxisVM through its API (COM inter-
face). The export requires that AxisVM and a version of Grasshopper/Dynamo are all installed on the
same computer. After starting the export, AxisVM automatically launches, showing the exported model.
One advantage of this interface is that it is easy to create parametric models and even templates in
Grasshopper/Dynamo. For example, the parameters of a truss may be the span, height, layout, and
cross-section sizes. The other advantage is that in in the Grasshopper/Dynamo there are a lot of geo-
metric functions, that enables creating complex spatial models.
User’s Manual X6 Release1 e2 129
Currently, models with straight lines can be exported to AxisVM. There are two available exporting op-
tions:
• export lines;
• export elements, for which the type of the element (truss, beam, rib), the material properties,
and the cross-section have to be defined.
*.gh, *.3dm files Grasshopper is the parametric design tool of the Rhinoceros CAD software. Grasshopper can be started
from within Rhinoceros. To export a file with *.gh extension, the AxisVM Export plugin shall be installed
in Grasshopper in Rhinoceros version 6. After installation, the modeling and exporting functions (com-
ponents) are available in the Grasshopper menu, in the AxisVM tab. Rhinoceros files (*.3dm) shall be
loaded into Grasshopper before exporting to AxisVM.
After exporting, the AxisVM model is independent of the parametric model. Starting a new export (e.g.,
with modified geometric parameters) launches a new AxisVM model in a separate window.
*.dyn, *.rvt files Dynamo is a parametric design tool of the Autodesk Revit software. Dynamo can either be started from
within Revit or as an independent software (Dynamo Studio, Dynamo Sandbox). To export a file with
*.dyn extension, the AxisVM Export plugin shall be installed in Dynamo. After installation, the modeling
and exporting functions (nodes) are available on the left-hand side menu of Dynamo, at the AxisVM
drop-down menu. Revit files (*.rvt) shall be loaded into Dynamo before exporting to AxisVM.
In Dynamo, before starting the export, first, a new model shall be created using the AxModel node. Self-
weight, nodal load, and nodal supports can be added to the parametric structural model. After starting
the export, a one-directional link is created between the parametric model and the AxisVM model. This
means, that modifying a (geometrical or structural) parameter in Dynamo automatically updates the
structural model in AxisVM. In AxisVM, further structural elements can be added, but these do not in-
fluence the Dynamo parametric model. In Dynamo, it is possible to start an AxisVM linear static analysis.
The link between the models is destroyed by closing any of the models.
The simplest way to download the AxisVM Export plugin is from within the Dynamo software: go to
Packages in the Dynamo menu, then choose Search for a Package... and type AxisVM Export. For further
details, example files, and installation guide see:
https://github.com/AxisVM/DynamoToAxisVM/wiki
AxisVM is able to communicate with Tekla Structures software different ways in case of
• version 2019 and later,
• version before 2019.
130
AxisVM is able to communicate with Tekla Structures software different ways in case of
• version 2019 and later,
• version before 2019.
3.1.9.1. Two-way connection for Tekla Structures versions 2019 and later
Setup If the user chooses to install the Tekla plugin, the AxisVM Installer copies all necessary files including the
AxisVM Import &Export plugin for Tekla. No other user activity is required. If TI module is included in the
AxisVM configuration, the two-way connection is functional after the AxisVM installation.
The AxisVM Import &Export plugin for Tekla Structures can be installed separately. Download the separate
free installer from the AxisVM website.
AxisVM with a TI module can export the model to Tekla Structures by creating a file with ATE extension.
This ATE file can be read to Tekla using a dedicated Tekla plugin building the model using Tekla Open
API commands.
Exported elements AxisVM can export the following elements: trusses, beams, ribs, domains, supports and actual reinforce-
ment. The AxisVM Import & Export plugin for Tekla Structures allows filtering of elements. Exported data
includes element geometry, material and cross-section of line elements, material of domains and stiffness
values for nodal supports. Loads and load combinations are not yet exported.
Export to ATE file The export can be launched from AxisVM via the main menu: File / Export / Export to Tekla (*.ate).
Click the Settings button to set the available options.
Export actual rein- Since X5 Release 4, users can export reinforcement to .ate files.
forcement
The algorithm of AxisVM helps to export not
only the applied field reinforcement but shaped
bars too, like hairpins, hooks etc.
Calculated raw bar lengths are always rounded
up to the nearest value of this array: 0.4, 0.5, 0.6,
0.8, 1.0, 1.2, 1.5, 2.0, 2.4, 3.0, 4.0, 6.0, 12.0 m.
• Closing reinforcement on free plate edges: free plate edges are closed by hairpins, whose lengths are
calculated based on the lap length of bars, which is calculated according to the design code, or can
be entered manually as l0. This option can be turned on or off.
• Reinforcing concave corners: concave corners can be strengthened by simple bars, whose lengths are
calculated based on the lap length of bars, which is calculated according to the design code, or can
be entered manually as l0. This option can be turned on or off.
• Convert short rebars to stirrups: to avoid useless duplicating of rebars, this algorithm uses a multipli-
cator value. Simple bars having a length greater than the minimal lap length multiplied by this value,
will be converted to a closed stirrup. This option can be turned on or off.
• Closing of rebars at joining plates: along the edges of plates that do not share the same plane, the
algorithm can connect the reinforcements of the individual plates either by hairpins or by hooks. The
actual geometry of these shaped bars depends on the width and angle of the joining plates. The
algorithm may revise the user's choice. The sizes of the shaped bars are calculated with respect to the
minimal lap length.
• Reinforcement in embedded domains: the method of closing or continuing rebars of embedded do-
mains can be chosen. If the inner domain is small enough, the outer domain's bars may as well be
driven through its body.
• Slab edges with more than two connecting domains: there is no automatic solution in such cases, so
the user can choose whether to ignore those edges or close them one by one.
• Optimization of reinforcement schemes: Optimization of rebar positions to distribute reinforcement
evenly if reinforcement domains overlap. May be time-consuming.
• Stirrups of column and beam reinforcement: stirrups of the applied column and beam reinforcement
are exported. If stirrup spacing was not defined the default can be exported optionally.
132
Import ATE file to The file with .ate extension can be imported to Tekla
Tekla Structures Structures via a dedicated plugin. Make sure that the
Analysis & Design models dialog box is closed during the
import process because it slows it down significantly.
Search for “AxisVM” in the Applications & components
catalogue in Tekla Structures and double click on the Ax-
isVM Import & Export to open it and switch to its Import
tab. The following dialog will appear
Import settings Press the button left to the Filename label. A preview of the model to be imported will be displayed in
the bottom right corner.
Elements to import Imported elements can be filtered by checking the nodes.
Workflow options Active only if Model handling is set to Update. Select the analysis model to compare the imported model
to.
Editing tolerance Tolerance used when calculating geometry.
Model handling Overwrite model All the existing objects will be deleted before importing the model.
Update Choose this method to compare the existing model to the imported one. Changed
elements will be updated. After clicking on the OK button further options will be
displayed.
Support parameters Nodal supports geometry in Tekla will be created from
default parameters. Click the Support parameters button
to adjust these values.
User’s Manual X6 Release1 e2 133
Select an AxisVM material from the left tree, select its matching material in Tekla on the right then click
on the button between the two views to map the imported material to the Tekla material. The selection
in the left tree jumps to the next unmatched material. The name of the matched material turns red, un-
matched materials are black.
If a material category is selected on the left (and not a material), all materials of this type will be
matched to the same material in Tekla.
It is also possible to create a new Tekla material based on the properties of an AxisVM material by se-
lecting Create new material from the right tree.
If all imported materials are matched the OK button becomes enabled and the import of the model can
be continued.
The AxisVM Import & Export plugin in Tekla Structures can export the model data to a file with TAE ex-
tension. This file can be imported to AxisVM.
Exported data The plugin exports the following elements: line elements with their material and cross-section, domains
with their material, nodal supports, loads and load combinations. Only elements present in the chosen
analysis model in Tekla will be exported. Supports must be defined as separate point supports. Line ele-
ments with a custom cross-section are not converted to static model and exported only as a custom
shape.
134
Setting analysis You can access the part analysis properties int the menu below:
properties of parts
For beams and columns set these parameters on the Position tab:
User’s Manual X6 Release1 e2 135
For contour plates, slabs and panels set these parameters on the Position tab:
Creating the Analy- After setting the analysis properties of every part you can create the analysis model.
sis Model
1. Open the Analysis & Design models dialog box.
2. Press New... to create a new analysis model with the parameters below:
Analysis Model
Properties
136
3. Select the parts and loads you would like to add to the analysis model and click on the Add selected
objects button. To remove elements from the analysis model select them and click on Remove selected
objects.
5. Check if there are any warnings indicated by Tekla Structures and try to resolve
them.
If you see the message that Exporting data succeeded, the operation was successful.
The resulting file with .tae extension can be either dropped onto the user interface of AxisVM or imported
via File / Import.
Importing TAE file to When importing the file into AxisVM a dialog will pop up:
AxisVM
User’s Manual X6 Release1 e2 137
Import method If the AxisVM model is not empty, two options are available to overwrite the model or update it.
Structural model If this option is activated AxisVM changes the raw geometry data to build a proper structural framework.
Colliding beams are connected with a rigid body, neighbouring domains are joined, beams running on a
domain are converted to ribs. Some of these calculations require considerable resources and may slow
down the import process.
Arc resolution Certain geometry calculations require arcs to be converted into polylines. The conversion criterion can be
set here.
Editing tolerance Tolerance used when calculating geometry.
Joining objects To improve the efficiency of collision detection bounding boxes of objects are inflated a bit. This tolerance
value can be set here. Active only if the checkbox on panel Structural model is checked.
Default material Users can select the material assigned to entities for which Tekla supplied no information.
AxisVM and Tekla Structures can share a workflow between each other. This is a comparison based ex-
port-import. With this, engineers working on the same project can import new or modified entities with-
out affecting unchanged elements.
The workflow process can be started both from AxisVM and Tekla Structures.
For the comparison to happen you have to choose Update on the form and check the checkbox shown
below:
AxisVM: Tekla:
138
Importing to AxisVM When importing a Tekla model to AxisVM, three basic scenarios are handled:
1. Element to be imported cannot be found in the analysis model, so it can be simply processed
and imported.
2. One or more items of the actual model cannot be found in the file to be imported. In this case,
users can select to either delete these elements or to ignore and leave them intact.
3. The item to be imported can be found in the actual model. In this case, we perform a compari-
son based on
a) the material of the elements
b) the statical model of the elements
c) the profile or thickness of the elements.
If after this we find them equal, the imported item is skipped. However, if they are found to be different,
arriving item is processed, and the former one is deleted. When - for any reason - we cannot perform a
successful comparison, the user must decide whether to keep the existing item or import the new one.
These preferences can be adjusted prior to the actual import process on the form below:
AxisVM dialog
Importing to Tekla When importing an AxisVM model to Tekla Structures we perform a comparison based on:
Structures
a) the material of the elements
b) the statical model of the elements
c) the profile and thickness of the elements
d) spring characteristics
After the comparison you can review the comparison results and customize the actions to perform.
User’s Manual X6 Release1 e2 139
New elements This list shows the elements created in AxisVM since the last export. These elements will always get cre-
ated after the import.
Unchanged ele- No changes were done to these elements. We are leaving them as they are.
ments
Modified elements The elements which got modified in AxisVM are listed here. Every change is accepted by default. Uncheck
the changes you do not wish to perform.
Selecting a node will highlight the related elements.
Deleted elements These elements got deleted in AxisVM since the last export. They are getting deleted after the import
process.
Missing elements These elements can’t be found in the imported file but present in the Tekla model. You can select to
either delete these elements or to ignore them and leave them intact.
Ambiguous ele- If an element has incomplete information or we can’t compare it for some reason, then it will appear here.
ments You can decide if you want these elements to overwrite the current ones or you want to keep the old
elements.
The connection between the two software is made through a COM server enabled to run AxisVM. To
make the connection work first the COM server must be registered within the operating system (in the
Registry) then Tekla Structures must be notified that a compatible server is available.
AxisVM setup automatically performs these registering operations, however if Tekla Structures is not
installed the second registration cannot be completed. Therefore, after installing Tekla Structures the
registration has to be started again by running two batch files from the AxisVM program folder:
!REGISTER_AXISVM_X64.BAT
!REGISTER_TEKLA_X64.BAT
Connection After a successful registration the model built in Tekla Structures can be transferred to
AxisVM in the following way: click Analysis & Design models... in the Analysis menu then click the Prop-
erties button to set AxisVM AD Engine as the Analysis engine.
140
User’s Manual X6 Release1 e2 141
If AxisVM AD Engine does not appear in the dropdown list the registration was not successful and has
to be repeated.
Getting back to the Analysis & Design models dialog click Run to start the transfer of the model. The
process status is displayed in dialog. If the transfer is completed successfully click the OK button to see
the model in AxisVM.
142
Loads and load cases specified in Tekla Structures are also converted.
User’s Manual X6 Release1 e2 143
Lets you specify a header text (two lines), which contains the name of the project and designer. It will
appear on the top of every printed page. An additional comment line can be added.
If a page header logo was selected for printing drawings, tables and reports, the logo also appears here.
Click on the logo to change it or try Settings / Preferences / Report / Company logo / Settings…
3.1.12. Print
Lets you print the model according to the current display settings. Allows the setup of the printer, and
[Ctrl]+ [P] of the page.
Current printer Printer setup
Output
and printer setup
Printing drawing
Send To
Lets you send the output directly to the printer or to a graphics file (DXF, BMP, JPG, PNG or Windows
Metafile [WMF/EMF]).
144
Printer
Lets you select and setup the printer.
If Print to file is selected, the data sent to the printer will be stored in the Name.prn file, where Name
is a file name to be entered.
You can set the number of copies required.
Click on the printer name or on the Printer… button to invoke the standard Windows Printer Setup
dialog where you can change printer and printer settings in detail.
Scale
Lets you set the scale of the drawing to print. In case of perspective or rendered view or if the
output is sent to a Windows Metafile the scale cannot be set.
Margins (Printer/DXF)
Lets you set the size and the units of the page margins. You can also drag margin lines within the
preview area by their corner and midside handles.
Bitmap size (BMP, JPG)
Lets you set the bitmap size in pixels, inch, mm or cm and bitmap resolution in dpi (dots per inch).
Preview
Lets you view the printed image prior printing. If you select Printer as a target the graphics cursor
turns to a hand whenever it enters the preview area. By pressing the left mouse button and moving
the mouse you can specify an additional panning which will affect the printed output only.
Page header
Lets you set the date and remark that will appear on each page, and the starting number for the
page numbering. If the Page numbers checkbox is turned off a blank space will appear after Page
allowing handwritten page numbers. Setup… displays the Page header dialog (see 3.1.10 Page
header) where you can change the company logo.
Orientation
Lets you set the orientation of the page.
Color options
Lets you select printing in grayscale, color, or black and white.
If your printer cannot print in color you may get different results in the first two cases.
If you select Grayscale the output will be converted to grayscale using an internal grayscale palette
of AxisVM. If you select Color the conversion to grayscale will be performed by the Windows printer
driver. Try both to find which works better for you. When black and white printing is selected, all
entities are printed in black.
Paper size
Lets you set the size of the paper.
Change fonts
Lets you select fonts to be used in printing and set the font size.
Pen widths
Sets the size of the pens for printing.
Thick lines are used for drawing supports and rigid elements.
Medium lines are used for isolines and section line.
Thin lines are used for elements and geometry and other
entities.
Windows to Print
Lets you print either the active window or all windows displayed.
Printing to file When Print to File is selected the printing is redirected to a file, name.prn that you can print anytime
later.
User’s Manual X6 Release1 e2 145
If the file name.prn already exists, you can add your printing to it, or overwrite it.
Printing table When printing from the table browser, you can set the pages (all / even / odd) of all / current / selected
pages you want to print.
Example: Entering 1, 3, 7-10, 20-18 in the Selected field the 1st, 3rd, 7th, 8th, 9th, 10th, 20th, 19th, and 18th
page will be printed in this order.
You can print the prn file you created from the following window.
You can print more than one prn file at a time. You can set the printing order with the up/down arrows
in the right of the file list box, or dragging the file names to a new position with the mouse.
146
The File/Model Library command lets you preview, get information and manage your model files.
As in Open and Save as dialog windows the standard file access dialog box items are displayed, but in
the list box you can select multiple files.
The AxisVM model files are marked with the symbol. If a model has a result file the symbol has a blue
right-bottom corner, .
New
Creates a new sub-folder in the current folder with the name you enter.
Copy
Copies the selected files to a different folder. You can specify whether to copy the result files or not.
Rename/Move
Renames the selected files in the current folders or moves them into a different folder.
Delete
Deletes the selected files from the current folders. You can specify to delete only the result files or
all.
Open
Opens the selected file for editing.
AxisVM files are marked with . If a result file is available, the bottom right corner of the icon is blue.
Preview
Shows the model wireframe in front, side, top view or in perspective depending on the model
dimensions. Model information is also displayed in a list.
Close
Quits the Model Library.
User’s Manual X6 Release1 e2 147
AxisVM provides a preloaded material library (that contains the most frequently used structural
materials) and allows you to create material property sets that you can use over and over again in many
different models. You must assign different names to each material property set.
The material library window can also be opened using the Table Browser icon and by selecting
Libraries/Material Library. See 4.9.9 Line elements, 4.9.23 Creating model framework from an architectural
model
See the detailed description of the Table Browser in section 2.9.
Properties of This table contains the properties of materials often used in civil engineering to the MSz, Eurocode, DIN-
materials 1045, DIN-1045-1, NEN, SIA-162, a STAS and Italian codes. You can add, modify, or delete existing
material data. In case of entering a new material with an existing name it will be added as
materialname_number. These materials can be used in any model.
Changes in the material library does not reflect in models using the modified material.
When entering a new material, the following dialog is displayed:
Define new material Definig new material or clicking to a non-editable column (eg. national design code, type)
a dialog appears, in which all material properties, calculation and design parameters can be defined or
[Ctrl+Ins] changed. The fields containing the basic properties independent of the design code can be edited in the
Change table.
material properties When a material with a name identical to one existing is entered an index is attached to the name
(name_index) to differentiate from the existing one.
If no texture was assigned to the material click the sample rectangle to select one from the library. See
2.16.4 Display mode
148
Material Properties For each material the following properties are stored:
𝜈 if 𝐸𝑖 ≥ 𝐸𝑗 𝐸𝑖 𝐸𝑗
𝐸𝑧 = max{𝐸𝑥 , 𝐸𝑦 } ; 𝜈𝑖𝑗 = { 𝐸 ; 𝐺𝑖𝑗 =
𝑖
𝜈 if 𝐸𝑖 < 𝐸𝑗 𝐸𝑖 + 𝐸𝑗 + 2𝜈𝑖𝑗 𝐸𝑗
𝐸𝑗
where 𝑖𝑗 = {𝑥𝑦, 𝑥𝑧, 𝑦𝑧}
Nonlinear properties Material model used in the software is valid only in case of small strains.
Material type
Material model / yield criterion
Steel Concrete, masonry Other
Plastic / Bresler-Pister — ✓ ✓
E [kN/cm2] Young’s modulus of elasticity for nonlinear analysis (rise of the initial section of the -
diagram)
ET [kN/cm2] Young’s modulus of elasticity for nonlinear analysis (rise of the tangential section of
the - diagram)
σyT [kN/cm2] Tensile yield stress
σyC [kN/cm2] Compressive yield stress
By function
Stress-strain
diagram editor
Most of the toolbar functions are the same as in the time history diagram editor. See 4.10.30 Dynamic
loads (for time-history analysis) – DYN module
Symmetrical If this option is activated defining the positive part of the function also defines the negative part.
function
Biaxial compression σyB [kN/cm2] Biaxial compressive yield stress. 𝜎𝑦𝐵 = 𝐶𝑦𝐵 ∙ 𝜎𝑦𝐶
CyB [] Biaxial compression factor
Yield surface visuali- The diagram shows the section cut of the yield surface by the plane of the principal stresses S1-S2. This
zation boundary curve belongs to plane stress state.
Plastic behaviour
Increasing load moves the point representing the state of the material along the - curve, decreasing
load moves it parallel with the initial section of the curve.
Plastic / Bresler-Pister
This material model is recommended for concrete and masonry materials, because it can model the
low yield strength for tension and the high yield strength for uniaxial and biaxial compression simulta-
neously. [36] https://en.wikipedia.org/wiki/Bresler_Pister_yield_criterion
𝜎𝑦𝑇 = 5 MPa , 𝜎𝑦𝐶 = 50 MPa 𝜎𝑦𝑇 = 5 MPa , 𝜎𝑦𝐶 = 50 MPa 𝜎𝑦𝑇 = 5 MPa , 𝜎𝑦𝐶 = 50 MPa
𝐶𝑦𝐵 = 1.0 𝐶𝑦𝐵 = 1.1 𝐶𝑦𝐵 = 1.2
𝜎𝑦𝐵 = 𝐶𝑦𝐵 ∙ 𝜎𝑦𝐶 = 50 MPa 𝜎𝑦𝐵 = 𝐶𝑦𝐵 ∙ 𝜎𝑦𝐶 = 55 MPa 𝜎𝑦𝐵 = 𝐶𝑦𝐵 ∙ 𝜎𝑦𝐶 = 60 MPa
Stiffness Initial stiffness The materially linear initial stiffness matrix is used for the evaluation of the global
stiffness matrix. The calculation is slower (more iterations are necessary) but the
convergence is more stable.
Tangent stiffness The materially nonlinear tangent stiffness matrix is used for the evaluation of the
global stiffness matrix. The calculation is faster (less iterations are necessary) but if
the increments are not small enough convergence problems may occur.
User’s Manual X6 Release1 e2 151
B 1
A C(b)
A B
B A
A O O
O A O 2
C(b)
B
C(a)
C(a)
In case of kinematic hardening the size of the yield surface does not change but its centre moves from
the origin of the stress space to the 𝑂′ location. In such a case the relative stresses 𝜎𝑟,𝑖 = 𝜎𝑖 − 𝜎𝑂′,𝑖 with
respect to the current origin are substituted into the equation of the effective stress calculated on the
basis of the yield criterion. See 6.1.21 Truss/beam/rib stresses and 6.1.22 Surface element stresses In
case of isotropic hardening 𝜎𝑂′,𝑖 = 0 and 𝜎𝑟,𝑖 = 𝜎𝑖 .
Design Parameters Design parameters depend on the material type and the design code.
AxisVM provides preloaded cross-section libraries, that contain the most frequently used steel shapes
and concrete cross-sections, and allow you to create standard cross-section property sets that you can
use over and over again in many different models. The libraries includes products of manufacturers
worldwide.
For the description of the Table Browser see 2.9 Table Browser.
The Undo function does not work when libraries are modified.
User’s Manual X6 Release1 e2 153
Create a new library You can create a custom cross-section library by the File / New Cross-Section Table command in the
Table Browser. You have to specify a library name, a library file name and a cross-section type to
categorize the library in the Table Browser. A Geographical region and a Manufacturer can also be
entered. These fields can be used to filter cross-sections when importing from the library. See 4.9.2.1
Browse cross-section libraries
Standard and custom cross-section library files (*.sec) are stored in the folder where the application is
stored.
Hot rolled parallel flange channel (U Hot rolled tapered flange channel (U shape)
shape)
Concrete cross- The concrete cross-sections are listed starting from the size 20x20 to size 80x80 cm in steps of 2 and 5
sections cm.
Cross-section Editor allows you to edit thin / thick walled (solid) cross-sections or create certain
parametric composite sections of two or three different materials.
You can use parametric circular, rectangular, ring and polygonal shapes, or any shape listed in the cross-
section libraries to build complex cross-sections. The shapes used to build a new thin-walled or thin-
walled cross-section are referred to as components, and have to be of the same material.
You can translate, rotate, mirror, copy or move the selected components at any time during the editing.
When a component is placed to its location graphically, the principal axes and the cross-sectional
properties of the composite cross-section are computed.
You can use keyboard commands the same way as in main editing windows.
The OK button exits and closes the cross-section editor window, and saves your current cross-section
into the cross-section table of your model with a name you specify.
Cross-section editor is on the toolbar of the Cross-section Library and can also be launched from the line
element dialog. See 4.9.9 Line elements
The editor can be used when creating a native model from an architectural model through the IFC
interface. See 4.9.23 Creating model framework from an architectural model
Editor Keys See 2.5 Using the cursor, the keyboard, the mouse
From Cross-section Loads a cross-section from the Cross-section Library. Only thick or thin-walled cross-sections are
Library available depending on the cross-section editor tab position. See 4.9.2.1 Browse cross-section libraries
From DXF file Contour of thick walled cross-sections can also be imported from a DXF file.
Stress-points You can specify the points you want to calculate stresses for. The default stress-point is the center of
gravity. You can specify up to 9 stress-points for each cross-section.
When applying a move command the stress-points can also be moved.
In case of thin-walled sections the shear flow is calculated only along the centerlines. The projected
points, from where the shear stress values are assigned to the stress-points, are also displayed. If a stress-
point is located at a corner or at a joint of several centerline segments, the highest value is taken into
account.
Stress calculations are performed at the specified stress-points only. If you don’t specify any
stress-points, stress will be calculated in the center of gravity only. It means that no bending stress
will appear.
Icon bar Editor functions and settings can be found on the Icon bar on the left. The behaviour of the Icon bar is
the same as that of the main Icon bar. See 2.16 Icon bar.
The only difference is that this Icon bar can be moved above the menus at the top or at the bottom but
it is not dockable.
Geometry trans-
formations
All standard geometry transformations (moving, rotating, mirroring, scaling) can be used. All operations
are performed in the y-z plane of the cross-section editor.
Dimensioning
Dimensioning tools can be selected from the upper toolbar of the palette. These are orthogonal and
aligned dimension lines, angle, radius and arc length dimensions.
The bottom toolbar contains the buttons for fine tuning dimension line properties and choosing the
smart dimension line option which places multiple dimension lines between the start point and the end
point.
The bottom toolbar contains the buttons for fine tuning dimension line properties and choosing the
smart dimension line option.
User’s Manual X6 Release1 e2 157
Definition of a The Editor can be used to create custom thin-walled or thick-walled (solid) or composite cross-sections.
cross-section The components of a complex but non-composite cross-section will have the same material. Composite
cross-sections are parametric and created from two or three materials.
Thin-walled
cross-sections
Cross-section properties
Standard shapes can also be defined parametrically. In this case the following parameters has to be
defined in the dialog:
Manufacturing There are three options (rolled, welded, cold formed.)
process
158
Dimensions Values depending on the type of the cross-section (height, width, thickness, corner/fillet radius, diameter
etc.).
Rotation Lets you define a rotation by angle . The default value is 0.
Wedged I shape
Double wedged I
shape
Pipe
Definition of a pipe by its parameters d (outside diameter), and v (thickness). The centerline is considered
as the contour of a closed domain, which is displayed with a dashed line.
User’s Manual X6 Release1 e2 159
Other cross-sections Definition of cross-sections by height, width, thickness and in the case of rolled or bended cross-sections
by the corner/fillet radius.
Double shapes The base cross-section can be defined parametrically (width, height, web and flange thickness) or taken
from the Cross-section Library.
Special parameters for double shapes:
distance: a
orientation : facing or back-to-back (in case of 2U)
Cross shapes Definition of a cross shape by its parameters: h
(height), b (width), tw (web thickness) and tf
(flange thickness).
2I cross shapes
1. left side
2. center line
3. right side
R: Rounding (corner and fillet) radii
Delete Using the [Del] key you can invoke the Selection Icon Bar, and select the components you want to delete.
When deleting a component the stress-points will also be delete.
Stress-point Deletes the selected stress-points.
Thick-walled cross-
sections
Parametric shapes:
Rectangle Rectangle with a width of b and a height of h.
Rounded rectangle Rounded rectangle with a width of b, height of h and rounding radii r1, r2.
Hollow rectangle Hollow rectangle with a width of b, height of h and v1, v2 wall thickness.
Haunched I shape Haunched, I shape defined by the bw web thickness, bt, bb top and bottom width,
h total height and ht, hth, hb, hbh parameters.
T shape with T shape with haunched web defined by bw, bb, bt width, a h total height and ht,
haunched web flange thickness.
T shape with T shape with haunched web defined by bw, web width bt total width, h total
haunched flange height and ht , hth parameters
Hollow circle Hollow circle defined by the D outer diameter and the v wall thickness.
Symmetric trape- Symmetric trapezoid with bt, bb top and bottom width and a height of h.
zoid
Polygonal Definition of a polygonal shape by drawing a a complex polygon. Press the Esc key, click the right mouse
button or close the polygon to finish editing. During editing the following pet palette appears:
Line
Tangent
Tangential arc
Insert a vertex Insertion of a new vertex on the contour of the cross-section. Shape of the cross-section can be changed
by dragging a vertex by the mouse.
Contour If the Contour button is down the cross-section can be defined. If the Hole button is down a hole can be
specified.
Hole You can specify a hole in rectangular, circular, and closed polygonal shape components. The hole can be
rectangular, circular, and closed polygonal.
Composite
cross-sections
Materials can be imported from the Material Library, see 3.1.15 Material Library.
The tube section can be entered parametrically or selected from the pipe cross-sections of Cross-section
Library, see 3.1.16 Cross-Section Library. Inner profile can be selected from the Cross-section Library.
Concrete filled Definition of a concrete-filled box shape requires the following parameters: the
box shape material, the b width, h height, tf and tw thickness values of the box and the
material of the infill. An optional inner profile can also be defined.
Circular concrete Definition of a circular concrete-encased profile requires the following parame-
encased profile ters: the material and the d diameter of the round shape, an inner profile and its
material.
Concrete encased Definition of a concrete encased crossed I shape requires the following param-
crossed I shape eters: material of the crossed I shape and the infill, the h height, b flange width,
and two wall thickness values: tf flange thickness and tw web thickness.
Delete Using the [Del] key you can invoke the selection window, and select the components you want to delete.
When deleting a component, the stress-points will also be deleted.
Polygon Deletes the selected components.
Stress-point Deletes the selected stress-points.
You can not delete the default stress-point (from the center of gravity).
Options Lets you set the grid size, cursor step, and the zoom factors.
Compute properties Following cross-section properties are calculated:
AxisVM calculates Ax, Iy, Iz, Iyz by integration, Ay, Az, Ix, I, y, z, yz, 1, 2, A1, A2
by performing a finite element analysis of the cross-section.
User’s Manual X6 Release1 e2 165
In case of a cross-section consisting of two or more independent parts, Ay, Az, y, z, yz, 1, 2, A1, A2
are not determinded.
Ax Axial (cross-sectional) area
Ay Shear area in local y direction
Az Shear area in local z direction
Ix Torsional inertia
Iy Flexural inertia about local y axis
Iz Flexural inertia about local z axis
Iyz Centrifugal inertia
I1(*) Principal inertia about local 1st axis
I2(*) Principal inertia about local 2nd axis
α Angle between local 1st axis and the local y axis.
I Warping modulus (used for the design of steel shapes)
y shear factor in local y direction
z shear factor in local y direction
yz shear factor for local yz cross
1 shear factor for local 1st direction
2 shear factor for local 2nd direction
A1(*) Shear area associated with shear forces in local 1st direction
A2(*) Shear area associated with shear forces in local 2nd direction
W1, el,t(*) Elastic cross-section modulus, top = I1 / e2_max (see diagram below)
W1, el,b(*) Elastic cross-section modulus, bottom = I1 / e2_min
W2,el,t(*) Elastic cross-section modulus, top = I2 / e1_max
W2,el,b(*) Elastic cross-section modulus, bottom = I2 / e1_min
W1,pl(*) Plastic cross-section modulus
W2,pl(*) Plastic cross-section modulus
i1 Radius of inertia about local 1st axis
i2 Radius of inertia about local 2nd axis
yG Position of the center of gravity of the cross-section in local y direction relative to the lower-
left corner of the circumscribed rectangle
zG Position of the center of gravity of the cross-section in local z direction relative to the lower-
left corner of the circumscribed rectangle
ys, zs Position of the shear center in local y and z directions relative to the center of gravity
Po Outer circumference (cross-section contour)
Pi Inner circumference (holes)
(*) If first and second principal axes are the local y and z axes values with (*) appears with
indices y and z.
Principal inertia
𝐼𝑥 + 𝐼𝑦 𝐼𝑥 − 𝐼𝑦 2
𝐼1 = + √( 2
) + 𝐼𝑥𝑦
2 2
𝐼𝑦 + 𝐼𝑧 𝐼𝑦 − 𝐼𝑧 2
𝐼2 = + √( 2
) + 𝐼𝑦𝑧
2 2
2𝑛𝑥𝑦
tan(2𝛼𝑛 ) =
𝑛𝑥 − 𝑛𝑦
−90° < 𝛼 ≤ +90°, relative to the cross-section’s local y
axis.
166
Calculation of elas- 𝐼1 𝐼1
𝑊1,𝑒𝑙,𝑡𝑜𝑝 = 𝑊1,𝑒𝑙,𝑏𝑜𝑡𝑡𝑜𝑚 =
tic cross-section 𝑒2,𝑚𝑎𝑥 𝑒2,𝑚𝑖𝑛
modulus
𝐼2 𝐼2
𝑊2,𝑒𝑙,𝑡𝑜𝑝 = 𝑊2,𝑒𝑙,𝑏𝑜𝑡𝑡𝑜𝑚 =
𝑒1,𝑚𝑎𝑥 𝑒1,𝑚𝑖𝑛
Shear deformations For beam elements the shear deformations are not taken into account even if the cross-section was
entered with nonzero for the shear area.
The shear areas are used by the rib elements, 𝐴𝑦 = 0 and 𝐴𝑧 = 0.
In the steel design module, the shear areas are calculated according to the corresponding design code,
instead of using the values entered here.
Compute properties Properties of a composite cross-section are calculated as the modulus-weighted [16, Chapter 1.2] sum
of composite cross- of the properties of its parts. The reference moduli 𝐸𝑟𝑒𝑓 and 𝐺𝑟𝑒𝑓 belong to the material of the outer
sections part.
1 1 𝐺𝑟𝑒𝑓 𝐴𝑥
𝐴𝑥 = ∑ 𝐸𝑖 𝐴𝑥,𝑖 𝐼𝑥 = ∑ 𝐺𝑖 𝐼𝑥,𝑖 𝜌𝑦 =
𝐸𝑟𝑒𝑓 𝐺𝑟𝑒𝑓 𝐺𝐴
∑ 𝑖 𝑥,𝑖
𝜌𝑦,𝑖
1 1
𝐴𝑦 = ∑ 𝐺𝑖 𝐴𝑦,𝑖 𝐼𝑦 = ∑ 𝐸𝑖 𝐼𝑦,𝑖 𝐺𝑟𝑒𝑓 𝐴𝑥
𝐺𝑟𝑒𝑓 𝐸𝑟𝑒𝑓 𝜌𝑧 =
𝐺𝐴
1 1 ∑ 𝑖 𝑥,𝑖
𝐴𝑧 = ∑ 𝐺𝑖 𝐴𝑧,𝑖 𝐼𝑧 = ∑ 𝐸𝑖 𝐼𝑧,𝑖 𝜌𝑧,𝑖
𝐺𝑟𝑒𝑓 𝐸𝑟𝑒𝑓
𝐺𝑟𝑒𝑓 𝐴𝑥
1 𝜌𝑦𝑧 =
𝐺𝐴
𝐼𝑦𝑧 = ∑ 𝐸𝑖 𝐼𝑦𝑧,𝑖 ∑ 𝑖 𝑥,𝑖
∑ 𝑦𝐺,𝑖 𝐸𝑖 𝐴𝑖 𝐸𝑟𝑒𝑓 𝜌𝑦𝑧,𝑖
𝑦𝐺 =
𝐸𝑟𝑒𝑓 𝐴𝑥
∑ 𝑧𝐺,𝑖 𝐸𝑖 𝐴𝑖
𝑧𝐺 =
𝐸𝑟𝑒𝑓 𝐴𝑥
The modulus-weighted computation method is based on the assumption that there is no shear
connection between the parts. However, it has no effect for the parameters Ax, Iy, Iz, Iyz, yG, zG,
because of the coincidence of the center of gravity of the parts. But the values of Ay, Az, Ix, I,
yS, zS, ρy, ρz, ρyz would be different if there was shear connection.
User’s Manual X6 Release1 e2 167
The spring characteristics library window can also be opened using the Table Browser icon and by
selecting Libraries/Spring Characteristics Library. In addition it is available in several dialog windows for
easier selection. See 4.9.17 Spring
The library has a Default library containing Rigid and Soft spring characteristics to translational and
rotational degrees of freedom. Theese characteristics are automatically added for new models. You can
add aa new library and you modify or expand existing spring characteristics libraries. In case of creating
custom library, its elements will be added to the model only if they are added to the Spring
characteristics (4.9.3 Spring characteristics) in Table Browser or be selected from the library when
defining a new spring element.
This library allows definition of parametric seismic isolators as well.
Properties:
Name Name of the characteristic
Type Node to node spring (cannot be modified )
Degree of freedom Translational/rotational degree of freedom
Model Spring behaviour of the spring (Linear / NL elastic / Plastic)
K Initial stiffness
Kv Vibration stiffness
Damper type Kelvin-Voigt or Maxwell model
C Damping coefficient
Isolator type Rubber bearing, Curved surface slider, Other type
Exy Effective horizontal stiffness
ξ Damping ratio
Ez Vertical stiffness
F1 Yield force
KT Horizontal stiffness in plastic state
µ Friction coefficient
R Equivalent curvature radius
Damping coefficient is shown if at least one spring characteristic has a value greater than zero. Seismic
isolator parameter columns appear only if at least one isolator has been defined.
Changes in the Spring charateristics library does not reflect in models using the modified spring
characteristic.
Define new spring When entering a new spring characteristic, the following dialog is displayed.
characteristic Defining a new spring characteristic or clicking on a table column (except the Name column) a dialog
appears, where all properties of spring charateristic can be defined or changed.
[Ctrl+Ins] Characteristics with the same degree of freedom (translational or rotational) cannot have the same
Change name. If an existing name is entered a number will be appended to the name.
properties
A spring characteristic has the following properties: Name, Type, Degree of freedom, Damping type
168
Clicking on the icon the Browse spring characteristics library dialog is displayed.
Stiffness / Damping In default setting the springs operate on linear elastic principle. In the table the Model column contains
the type of behaviour of the spring characteristic.
Translational degree of freedom
Ku kN/m Initial stiffness
KuV kN/m Vibration stiffness
Cu kN/(m/s) Damping coefficient
Rotational degree of freedom
Kr kNm/rad Initial stiffness
KrV kNm/rad Vibration stiffness
Cr kNm/(rad/s) Angular damping coefficinet
Nonlinear A spring can have a nonlinear behaviour like materials. The parameters of the nonlinear behaviour are
properties available if Nonlinear properties is checked.
The parameters for stiffness and damping can be defined on different tab.
Stiffness
User’s Manual X6 Release1 e2 169
Plastic Irreversibile deformation occurs after reaching the limit force / limit moment.
Force – By parameters: The function of the spring characteristic can be defined in the same way as the nonlinear
displacement nodal support in the previous versions (simple) or as the stress-strain diagram of materials (advanced).
diagram The units depend on the degree of freedom. Plastic model requires advanced definition.
By function: A custom characteristic can be defined using the force-displacement function editor.
Compression only
Tension only
Setting a spring to Compression only or Tension only does not set the inactive stiffness to zero. To avoid
numerical instability a negligibly small stiffness is introduced.
𝐤𝐍 𝐤𝐍𝐦
𝑲𝒖𝟎 = 𝐦𝐢𝐧 {𝟏𝟎−𝟑 𝐦
; 𝑲𝒖 ∙ 𝟏𝟎−𝟖 } ; 𝑲𝒓𝟎 = 𝐦𝐢𝐧 {𝟏𝟎−𝟑 𝐫𝐚𝐝
; 𝑲𝒓 ∙ 𝟏𝟎−𝟖 }
If the deformation of the element is large even this stiffness can result in a significant force. In this case
choose Advanced and define a characteristic with a resistance on the inactive side.
By parameters, In case of advanced definition the tension and compression side can be different. If the spring model is
Advanced plastic, the Ku,T and Ku,C have to be the same, the value of Ku,C cannot be edited.
definition Using Resistance checkbox limit force/moment and tangent stiffness of the diagram after reaching the
limit force can be specified. If the resistance is switched off a simple linear diagram is obtained.
The red dashed line represents the FyC, FyT and MyC, MyT limit values.
170
Advanced definition Advanced mode provides several ways to define a nonlinear elastic character-
of a characteristic istic active on one side only. The simplest solution is to set the function to zero
active only in ten- on the inactive side. However, this ideally one-sided characteristic is numeri-
sion or compression cally the least stable near the active / inactive transition.
𝑲𝒖,𝑻 = 𝟎 ; 𝑭𝒚,𝑻 = 𝟎
Instead of the ideal one, it is better to include a characteristic that has the same
initial stiffness and a negligibly small resistance on the inactive side. This en-
sures numerical stability with a definite appearance of the tension or compres-
sion state, while the resistance can be set so small that it will be displayed as
zero with the given number of decimals. However, numerical stability problems
can occur even with this characteristic. Unexpected behavior of the model may
𝑲𝒖,𝑻 = 𝟎 ; 𝑭𝒚,𝑻 > 𝟎
make the active stiffness fail to ensure the equilibrium. In that case, the calcu-
lation will not converge.
This stability problem can be avoided by setting a very small but not zero slope
in the linear characteristic on the inactive side. This way a sufficiently large de-
formation can ensure the equilibrium on the inactive side as well and unrealis-
tically large deformations will warn us of the unexpected behavior of the
model. However, it should also be considered that if a large-scale deformation
in the inactive direction is the intended behaviour, an unrealistically significant 𝑲𝒖,𝑇 > 0
force may develop on the inactive side.
Advanced definition A characteristic with a given resistance can also be defined as an ideal one
of a characteristic where the stiffness becomes constant after reaching the resistance. However a
with a given zero tangential stiffness may cause numerical instability and an unexpected
resistance behavior of the model may make these springs fail to ensure the equilibrium.
In that case, the calculation will not converge.
𝑲𝒖𝑻,𝑇𝐶 = 𝑲𝒖𝑻,𝑇 = 0
This problem can also be solved by setting a small but not zero slope after
reaching the resistance.
This way a sufficiently large deformation can ensure the equilibrium and unre-
alistically large deformations will warn us of the unexpected behavior of the
model. However, it should also be considered that if a large-scale deformation
is the intended behaviour, forces may considerably exceed the given re-
𝑲𝒖𝑻,𝐶 > 0 ; 𝑲𝒖𝑻,𝑇 > 0
sistance.
Advanced definition Backlash can be simulated by defining a small initial stiffness
of a characteristic followed by a high stiffness nonlinear elastic state. It is recom-
with backlash mended to replace the ideal zero stiffness with a small slope
and small resistance. The ideal characteristic cannot be set par-
ametrically only through then function editor.
Using function editor the parameter values on advanced panel are set on the basis of the function in
the following way (in case of translational components)
Ku,T [kN/m] The slope of the first segment on the tension side
Ku,C [kN/m] The slope of the first segment on the compression side
KuT,T [kN/m] The slope of the last segment on the tension side
KuT,C [kN/m] The slope of the first segment on the compression side
Fy,T [kN] Plastic – First breakpoint on the tension side
NL elastic – Editable
Fy,C [kN] Plastic – First breakpoint on the compression side
NL elastic – Editable
The user can enter the limit values (resistances) only in case of NL elastic spring model and the utilization
will be calculated by comparing to this value.
For plastic spring model the parameter values cannot be edited and Ku,T = Ku,C
The force-displacement diagram must be monotonically increasing.
In case of a plastic characteristic the slope of the diagram must be monotonically decreasing.
Stiffness Initial stiffness The materially linear initial stiffness matrix is used when evaluating the global
stiffness matrix. The calculation is slower (more iterations are necessary) but
the convergence is more stable.
Tangent stiffness The materially nonlinear tangent stiffness matrix is used when evaluating the
global stiffness matrix. The calculation is faster (less iterations are necessary)
but if increments are not small enough convergence problems may occur.
172
Damping A bilinear function can be defined, where the damping coefficients for tension and compression might
differ.
If the analysis uses linear material model the damping coefficient is the average of the two values:
𝐶𝑢 = (𝐶𝑢,𝑇 + 𝐶𝑢,𝐶 )/2
Type Three types of seismic isolators are available: Rubber bearing, Curved surface slider and Other type.
Rubber bearing
µ Friction coefficient
R [mm] Equivalent curvature radius
Ez [kN/m] Vertical stiffness
Other type
3.1.18. Exit
[Ctrl]+ [Q] Exits the program.
174
3.2. Edit
3.2.1. Undo
Alt+BkSp
Undoes the effect of the previous commands. To undo a sequence of actions (more levels), click the down
arrow next to the Undo icon, and then select the actions you want to undo based on the time or type of
the commands.
You can set the number of undo/redo levels (maximum 99) in the Main menu Settings / Preferences /
Data integrity dialog box.
3.2.2. Redo
Shift+Alt+BkSp
Undoes the undo command or goes forward to reverse one or more undo commands. You can select the
actions you want to redo based on the time or type of the commands.
Launches the last command again if it was started by clicking on an icon or pressing a keyboard shortcut.
[Space] The name of the command is also displayed.
3.2.6. Copy
Copies the selected elements of the model to the Clipboard. If nothing is selected but there are active
[Ctrl]+ [C] parts, active parts are copied. If neither selection nor active parts are present the entire model is copied.
This function copies the drawing of the current graphics window to the clipboard like in earlier versions
but this operation can be deactivated.
3.2.7. Paste
Pastes AxisVM elements from the Clipboard. For paste options see next chapter: Copy / paste options.
[Ctrl]+ [V]
Merge loads from all load cases into the current one. This option copies all loads from all clipboard load
cases into the current load case of the model.
Parts
User-defined parts containing the selected elements are also copied to the clipboard. The first option
is to paste elements of parts into all active parts of the model. The second option is to paste the parts
themselves.
Paste position
There are three options.
Paste into original position: pasted elements will get into their original coordinate position.
Drag by the relative origin / Drag by a corner node of the structure: If one of these options are
selected paste position can be defined by clicking the left mouse button. In the first case the clicked
position will become the position of the relative origin in the source model when the elements were
copied. In the other case the clicked position will become the position of an automatically identified
corner of the copied structure.
3.2.9. Delete
Deletes the selected entities. If no elements are selected it brings up the Selection icon bar and then
[Del] the Delete dialog window.
Lets you delete the selected geometric entities.
To delete:
1. Select the geometric entities to be deleted. You can select them by holding the [Shift] key pressed
while you click on the entities with the left mouse button or use the Selection Icon Bar.
2. Press the [Del] key. If there is no selection, the selection toolbar appears and objects can be
selected for deletion. See 2.16.1 Selection.
3. Enable the check-boxes of the entities you want to delete.
4. Press the OK button, to finish and close the dialog window.
In the dialog window the check-boxes are active or inactive according to the contents of the current
selection set (intended for deletion).
User’s Manual X6 Release1 e2 177
Geometry Lets you select geometric entities for deletion. Deleting geometric entities that have assigned finite
elements, will result in the deletion of its finite elements and of the associated loads.
Elements Lets you select finite elements for deletion. Deleting finite elements will not delete the respective
geometric entity, but will delete the loads.
References Lets you select references for deletion. All finite elements that use the deleted references, and the
associated loads will be deleted too.
Mesh Lets you remove mesh from domains.
R.C. Design Lets you select the reinforcement parameters attached to the selected elements for deletion. Footing
parameters are also deleted.
Steel / Timber Lets you select the steel / timber design parameters attached to the selected elements for deletion.
design
Dimensions Lets you select the dimension lines, text boxes etc. for deletion.
You can save drawings from AxisVM in many different contexts: you can save AxisVM main windows,
beam displacement and internal forces diagrams, steel design results, nonlinear calculation results,
Add drawing to reinforced concrete column and beam design diagrams, bolted joint diagrams. In case of a divided view
Gallery [F9]
you can select to save all windows or the active one only.
Drawings Library is another way to store diagrams. While Gallery contains static image files, the
Drawings Library uses associative drawings following changes in the model.
See 2.13 Drawings Library
Which file format Bitmap formats (.BMP, .JPG) store the pixels of the diagram, so Windows metafiles provide higher
to use? resolution when printed. JPG is a compressed format with a slight loss of quality but these files are much
smaller than BMPs.
Windows metafiles (.WMF, .EMF) store a series of drawing commands so they can be scaled and printed
in any size in the same quality. However if you choose hidden line removal or a rendered view drawn
by OpenGL technology metafiles will contain only bitmaps.
To get a high resolution rendered view print the picture directly.
Drawings will be saved to a subfolder Images_modelname automatically created under the folder of the
model file. These pictures can be inserted into a report. Do not modify the name of the subfolder
Images_modelname.
178
The weight of the entire model, selected elements or details can be listed as a table per material, per
[F8] cross-section or surface type. Weights per material also displays weight of the footings.
Weight report / Rebar statistics displays the total amount of reinforcement bars placed in surface
elements, beams and columns. Detailed rebar statistics breaks down the rebar data by type (slab, beam,
column) position and diameter.
AxisVM handles line elements as structural members. It means that Meshing of line elements on the
Shift+A Mesh tab creates finite elements but the line elements themselves are not divided. The Find structural
members menu command joins adjacent line elements into a single element until a breaking point is
found. A breaking point is defined by different local x or z directions, different material, cross-section
or eccentricity, end release or a domain boundary. Line elements must be on the same line or on the
same arc.
The Break apart structural members menu command breaks apart line elements created with the As-
semble structural members command.
Shift+B
This menu item converts selected surface loads distributed over beams into individual distributed beam
loads.
Loads generated by the distribution algorithm from loads of load panels can be converted to individual
loads. After the conversion they can be modified or deleted but cannot be updated form the load panel.
This menu item converts automatic references assigned to line or surface elements into
reference vectors.
If the model contains a parametric ribbed domain or a composite parametric ribbed domain
(See 4.9.6.4 Parametric ribbed plates, 4.9.6.2 Composite ribbed domain) created with actual ribs this
command converts the parametric ribbed domain to a normal one. Ribs are not removed but
become independent of the parametric grid.
180
3.3. Settings
[F11]
User’s Manual X6 Release1 e2 181
The Layer Manager allows you to manage AxisVM layers, imported DXF or ArchiCAD layers. While only
one ArchiCAD layer can be imported, multiple DXF layers are allowed.
If no AxisVM layers are defined AxisVM automatically creates a new layer for dimension lines with the
name Dimensions.
On the left side of the Layer Manager dialog a tree view of the available layers is displayed. If you select
(highlight) a DXF layer in the tree, you can modify its properties in the right side (Name, Color, Style,
Size). If you select the main DXF file entry of the tree, you can modify all the DXF layers at a time.
Properties of AxisVM structural layers cannot be modified.
Apply to All: When using this button, a dialogue window will allow you to select the items in the DXF
layers that will have their properties set based on the layer’s settings.
The visibility of the layers or DXF files can also be set by clicking on the bulb or cursor symbol next to
the layer or file name.
New AxisVM Layer Creates a new AxisVM layer. You can set the layer’s name, color, line style, and width.
Delete More than one layer or group can be selected and deleted by the [Del] key.
Delete Empty Deletes all AxisVM layers that are empty (contain no entities).
AxisVM Layer
Delete Empty DXF Deletes all imported DXF layers that are empty (contain no entities).
Layer
Delete Empty PDF Deletes all imported PDF layers that are empty (contain no entities).
Layer
…Apply to all Applies selected layer properties (color, line style and width) to all objects on the layer.
Layer detection If activated the mouse will detect the underlying objects of the layer.
Equivalent to clicking on the arrow icon.
Show full path If activated, the tree view will display names of imported files with their full path.
Auto Refresh If activated, changing layer properties or their visibility immediately updates the main window.
Refresh All If activated, all views of the main window reflect changes, otherwise only the active view is affected.
3.3.4. Storeys
[F7]
Storeys are to make it easier to overview and edit the model. They can be defined before drawing the
model or assigned to an existing structure.
182
A storey is a workplane parallel to the global X-Y plane, with a given Z position. If a storey is selected
mouse movements will be projected to the plane of the storey even if you find an element at a different
Z position. Coordinates will always be projected to the storey plane to help tracing objects at different
levels.
Storeys are always listed by decreasing Z position, having automatic names. Changing the report lan-
guage makes storey names change.
Elements are considered to be part of a storey if their lowest Z coordinate is greater than or equal to
the storey level but less than the next storey level. Therefore, if a multi-story column or wall was defined
as a single element it will appear only at the lowest level. To change this behaviour the element has to
be cut with storey planes.
New elements will be linked to their storey automatically.
Storeys are logical parts of the model created for editing purposes and they do not affect the
analysis results.
If torsion effects have to be taken into account in seismic analysis, seismic storeys have to be
defined separately in the seismic parameters dialog.
Stories can be managed in the following dialog.
Turn off storeys If this button is down no storeys are displayed. Win-
dows will show the entire structure or the active parts.
storeys can be added or deleted in this state as well.
There can be only one active storey. However, display of neighboring storeys is also possible.
Editing will be constrained to the active storey.
Pick up Click this icon to get back to the model and click one or more nodes to pick up Z coordinates. Close the
process by clicking on an empty area. Z coordinates will be added to the list of storeys.
Enter a new storey Enter the Z coordinate into the edit field and click the + button. A new storey will be added to the list.
Find If you have an existing multi-storey structure with slabs, you can find and add Z coordinates of horizontal
domains to the list with one click. If not the all horizontal domains refer to a real storey you can delete
unnecessary storeys later.
Storey position cannot be changed. Delete the storey and define a new one.
Delete Deletes selected storeys. Remaining storeys will be renamed and storey assignments of the elements
will be updated automatically.
Display the storey Display the storey If any of these buttons is down elements of the storey below/above the active sto-
below the current above the current rey is also displayed to help tracing other objects.
storey storey
If this button is down elements of the storey above the active storey is also displayed to help tracing
other objects.
To display further storeys, open the Parts dialog instead where logical parts of any storey can be
turned on. Choosing a new active storey overrides the parts settings.
Numbering of sto- Numbering of storeys can be controlled with these buttons. If the left one is down (Numbering of storeys
reys from the bottom) the lowest floor will be considered as ground floor and other storeys will have a positive
number. If Signed numbering of storeys is selected the storey closest to the zero level will be the ground
floor. Underground storeys will get a negative number, others will get positive numbers.
Renaming storeys Click the right mouse button over a list item to get to the popup
menu. storeys can be renamed or their default names restored.
Several further options can be selected with checkboxes.
Display only walls The storey above/below the active storey is usually displayed to find the supporting walls and columns.
and columns of This options hides slabs and beams of the neighbouring storey.
neighbouring sto- If a storey is activated direct drawing of a column or a wall automatically starts with the storey height.
rey
Project points to This feature is to project all detected coordinates to the base level of the storey. It is useful when wall
storey level and column positions below the slab must be transferred the floor.
3.3.5. Guidelines
Guidelines
[Ctrl+G] See 2.16.9 Guidelines
Structural Grid
[Shift+G] See 2.16.8 Structural grid
184
Sets the design code to be used in case of code specific tasks. Changing design code changes the
method of calculating critical load combinations therefore all load group parameters but partial factors
will be deleted. Seismic analysis parameters and seismic load cases will also be deleted. As material
properties and certain reinforcement parameters are not the same in different codes it is recommended
to revise the values you have specified.
If Set current settings as default is checked, new models will be created with the current design code.
In case of Eurocode-based design codes, a number of parameters prescribed in National Annexes (NAs)
can be modified in AxisVM. Structural design according to NAs not included in the software can be
performed by the modification of NA parameters of the basic Eurocode.
The table displays the user value, default value, unit, description and code reference for each parameter.
User values are editable. Click OK, to apply them. The current set of user values will be saved/loaded
with the model.
Changing the design code causes the current set of user values to be lost unless it has been saved as
the default set of user values by clicking on the Save button. New models will use these default values.
The Default column of the table displays the value as specified in the NA. Clicking on the Restore button
replaces all user values with the NA defaults.
According to the rules and regulations in NAs, the parameters can be classified in three groups:
• the parameter is given by a number (e.g., partial factors): the value can be edited directly;
• the parameter is given by a formula; in most cases the constants of the formula are specified in
the NA (e.g., slenderness limit for RC columns): in this case, the constants can be edited (note:
constants are often represented by letter c);
• the parameter is given by neither a number nor by a formula; (e.g., method for calculation of
second order eccentricities for RC columns, buckling curves for LTB of steel beams): in this case,
the method or buckling curves can be selected from a dropdown list.
User’s Manual X6 Release1 e2 185
Lets you configure the units (SI and/or Imperial) and formats of variables used throughout the program
(number of decimals used for displaying or exponential format). You can use predefined sets as the SI
set, or create and save your own custom sets.
Storing unit settings The software stores the defined units and formats settings in the Axunits.ini data file which can be found
in the folder c:\Users\'username'\AppData\Roaming\AxisVM\15\.
3.3.9. Gravitation
Lets you set the gravitational acceleration constant and the direction of
gravitation as one of the global coordinate directions or a custom
direction.
If Custom direction is selected the X, Y, Z components of the direction
vector in the global coordinate system must be specified.
If Save as default setting is checked any new model will start with the
entered value of gravitational acceleration.
186
Seismic analysis based on response spectrum analysis according to Eurocode allows using stiffness re-
duction factors (k). These reduction factors can be defined for individual structural members inde-
pendently according to their architectural element type (column, beam, wall, slab, other element).
The radio buttons on the left are to choose the range of definition (Entire model / Displayed parts /
Selected elements). Reduction factors for these elements are listed and can be modified in the table. If
elements with the same type have different reduction factors a * is displayed in the respective cell. Chang-
ing it will assign a common value for the specified range of elements.
Pick up Click on the Pick up button then click on an element to copy its k values to the respective rows of the
table according to the architectural type of the element.
Default The last column displays the default stiffness reduction values for each type of element (these values are
also editable). If new elements are defined these default values will be automatically assigned to them.
Click on the Default button at the bottom to set all k values to their default.
Stiffness reduction is optional in a vibration analysis. If the vibration analysis has been performed with
stiffness reduction and the response spectrum analysis used these modal shapes, AxisVM will perform
linear analysis with the same reduced stiffness (in non-seismic load cases as well).
User’s Manual X6 Release1 e2 187
3.3.11. Preferences
Storing preferences The software stores the settings described below in the data files under the
settings c:\Users\'username'\AppData\Roaming\AxisVM\16 folder.
188
Recent file list Lets you set the number of recently opened AxisVM model files listed in the bottom of the File menu,
and set if you want the last edited file to be opened at startup. The welcome screen
(See 2.2 Protection and installation) will be shown on startup if the Show welcome screen on startup
checkbox is checked.
Save To make sure that you do not lose your work, select the Autosave option by the check box. In the min.
box, enter the interval at which you want to automatically save the opened model (1-99 minutes). You
must still save the model when you exit. A model that is saved automatically is stored in the default
temporary folder of the operating system (by default it is C:\Users\username\AppData\Local\ Temp\) as
~modelname.avm until you perform a save command. When you have to restart AxisVM after a power
failure or due to any other problem that occurred before you saved your work, AxisVM can recover it
from the temporary file stored in the above folder under the name $modelname.avm.
Compressed AXS format
If this checkbox is checked the AXS model file will be saved in a compressed format. The average size
of the compressed file size is about 10% of the original. The larger the model file the more efficient the
compression is. Result files (*.AXE) are not compressed.
Create Backup Copy
If this checkbox is checked and a model is saved after making changes a backup copy is automatically
created from the previous state of the axs file. Name of the backup file is modelname.~AX.
Store previous revisions of the model
If this option is checked subsequent revisions of the model are saved.
Click on the icon to get to the Settings dialog
See 3.1.2 Previous revisions…
Save derivative results
If this checkbox is checked stresses, envelopes, critical combinations and design results will be saved as
well.
User’s Manual X6 Release1 e2 189
Undo The previous state of the model can be retrieved From hard disk or From memory. If you work on big
models and/or your computer is low on memory it is recommended to use the first option (which is a
bit slower).
You can undo your last actions. You have to specify the maximum number of actions you want to undo.
This number must be between 1 and 99.
The Group Undo option allows you to undo the effects of complex commands in a single step. Undo
data can be stored in memory or on hard disk. The first option is faster, the second option leaves more
memory for the program (it may be important if a huge model is calculated).
Work on local If models are opened through a network, the speed of data transfer may reduce the performance of
copies of network AxisVM. This effect can be eliminated by allowing making local copies of network files. Local copies will
files be placed into the folder where the temporary files are stored during the analysis – except when this
folder is set to the model folder. In this case the files are saved to the default folder for temporary files.
The original files will be updated at each save operation.
Network time-out In case off network hardware protection keys, if in a time period set here there is no activity (checks)
with the key, the current AxisVM session is closed.
Disconnecting may also happen in a situation when you get a phone call and you do not use the
program for a time longer than the network time-out. If another user asks for access to the key the
server gives a license to him/her and when you try to continue your work the program displays an error
message and halts at the next key check.
3.3.11.2. Colors
Lets you select graphics area background color (black, dark gray, light gray or white). Labels, numbers,
symbols and elements will automatically change their colors to remain visible
Default number of colors on color legend window: Set the number of colors displayed in the color legend
window. See 2.19.4 Color legend window)
OpenGL settings Turn on Hardware graphics acceleration, Allow caching of OpenGL data and Enable fast coloring for faster
and smoother rotation of the model – if the video card and driver supports it. The results heavily depend
on the configuration details of the operating system, the driver and the graphics card so sometimes
turning of all or some of these settings may improve performance.
Zoom acceleration hides labels and other non-scaled items during zooming.
Reverse mouse wheel direction for zooming. If unchecked the mouse wheel has the default behaviour:
turning the wheel up zooms in, turning the wheel down zooms out the view. If checked these functions
will be reversed.
190
Color and line thickness of graphic symbols, color and size of labels can be customized. Color and
thickness of eccentric axes of beams and ribs can be set.
If the checkbox in the Custom value column is left unchecked the symbol is set to default. If it is checked
click on the Thickness / size value or the Color rectangle to edit the values. The new settings can be
saved to a scheme clicking the Save icon. Schemes can be loaded by selecting from the dropdown list.
From Version X4 colors and line width of the local system axes, result diagrams and applied reinforce-
ment can also be changed.
User’s Manual X6 Release1 e2 191
3.3.11.4. Fonts
This list allows changing fonts used on screen and in printed output. Move in the list with cursor keys
or by clicking on items. Double-click on the selected line or press Enter to change the font.
To change the sample, edit Sample text.
Right click on an item to display the popup menu.
Modify... : change the font.
Default font: reset the selected font to its default.
Click on Default settings to reset all fonts to their default.
Screen Drawing labels Font used for numbering and labeling on model drawings. Dimension lines
and text boxes have their own font, see 2.16.11 Dimension lines, symbols and
labels.
Information windows Font used for information windows on the screen (color legend window, info
window, coordinate window), see 2.19 Information windows.
Dialog windows Font used for dialogs.
Printer Project title Fonts used for the printed page header. See 3.1.10 Page header
Page header
Comment
Labels Font used for numbering and labeling on printed model drawings. Dimen-
sion lines and textboxes have their own font.
Information windows Font used for information windows (color legend window, info window)
when printed on drawings.
Report Table title Font used for titles of printed tables.
Table header row Font used for the first (header) row of tables.
Due to different resolution of the screen and the printer the ratio of label size and drawing is different
on the two devices, especially in case of textboxes. Check / uncheck Adjust font size when printing if
you are not satisfied with the result.
192
If the operating system is Vista or later you can set the Style of Save / Open dialogs used in AxisVM.
Under Windows XP only the first option is available.
If the second option is chosen AxisVM file preview requires a successful registration of a preview library
(this DLL is part of the AxisVM package). Installing AxisVM with administrative rights automatically reg-
isters this library. Without administrative rights this registration fails making the preview unavailable.
The preview library file can be registered later by running !REGISTER_PreviewLib.BAT from the AxisVM
program folder.
Move mouse pointer automatically to dialog windows positions the mouse pointer over the OK button
of dialog windows. Certain mouse drivers provide this functionality without using this option.
When working on high pixel density monitors it is recommended to enlarge toolbars, palettes, dialog
windows and cursors. Main toolbars and pet palettes offers two sizes for toolbar buttons. Choosing Large
icons also scales up certain cursors. Size of dialog windows makes it possible to scale dialogs between
75 and 200%. Windows settings defined in Control Panel / Display can also be applied.
Opening large models with a huge result file can slow down loading considerably. The Display results
after loading a model option controls what happens when a model with results has been opened. Results
will be displayed Always / Never / Always ask (a dialog is displayed to confirm displaying the results).
User’s Manual X6 Release1 e2 193
Change Model to change the default name set for a new model. Select the style of the postfix
(1, 2, 3 ... or A, B, C, ...) appended to generate a unique name.
Default names and postfix styles of new load cases or load groups can be edited the same way.
Clicking Reset resets all names and styles to their defaults.
3.3.11.7. Editing
Circle Closing Angle Parameter for drawing arcs. If the center angle of the arc is smaller than this angle or it is closer to 360°
than this angle then a whole circle will be drawn.
Projection line to Display of projection lines can be turned on/off. Its shows the distance of the cursor from the current
workplane workplane.
Delete unnecessary contour lines after automatic domain intersection
Controls if contour lines are automatically deleted after domain intersection. If this function is turned
off contour lines became internal lines of the union.
Break unmeshed structural members whenever a node is inserted
By default AxisVM uses structural members. These are line elements consisting of one or more finite
elements. Unmeshed members contain one finite element only. If a new node is inserted on an
unmeshed member or the line is divided the structural member remains unaffected but will contain
more than one finite element. Clicking on a structural member selects all finite elements belonging to
the member. This behaviour can be changed by checking this option. Then new nodes inserted on
194
structural members will break apart the member. To break apart existing structural members use Edit
/ Break apart structural members.
Enable selection of finite elements on lines
If activated, finite elements of a structural member can be selected individually. Otherwise only the
whole structural member can be selected.
Enable selection of design members
If activated, design members can be selected instead of structural members. Design members consists
of a group of lines with the same design parameters handled as one entity for steel or timber design
purposes.
Elements of a hidden mesh can be selected
If display of mesh is turned off this field controls if the hidden nodes / lines / surface elements can be
selected or not. This switch also controls if these nodes and elements appear in tables or not.
Show instructions at the cursor
Controls the display of a small tooltip window at the cursor with instructions for the next step of the
current task. If unchecked, instruction messages appear only at the bottom status line.
Make all layers editable when entering the layer editor
If activated, all locked layers will be unlocked when entering the background layer editor.
See 2.16.12 Editing background layers. Otherwise locked layers must be unlocked manually.
See 2.12 Layer Manager and 2.18 Speed Buttons.
Geometry check before running an analysis
If activated a geometry check is automatically performed before analysis.
See 4.8.19 Geometry check
Fit model with structural gridlines into view
If activated and the model is zoomed to fit in view, structural gridlines are also taken into account
when determining the zoom ratio.
Keep the view unchanged in undo operations
If activated undo operations does not affect the view of the model. Otherwise undoing a command
sets the the model as it was displayed before the command.
Highlight elements under the mouse cursor
Element(s) under the mouse cursor will be highlighted (both in wireframe and rendered view). Click
on the color sample to customize highlight color.
3.3.11.8. Meshing
Mesh management One of the following mesh management methods can be chosen.
Remove and create mesh automatically
Any editing performed on a domain deletes its mesh. When launching the analysis missing meshes
will be recreated based on the meshing parameters of the domain.
Keep mesh editable
Meshes can be edited manually.
User’s Manual X6 Release1 e2 195
3.3.11.9. Toolbar
Displaying toolbar If Horizontal toolbars expanded is chosen, all icon appears in a row. Vertical separator lines indicate
different groups of functions.
If Flyout toolbars is chosen, different functional groups will be represented by a single icon. Clicking the
arrow in the right bottom another toolbar flies out showing different tools.
Using domains and automatic meshes make it unnecessary to create individual surface elements and
place loads on them. For compatibility reasons the related functions are not removed but kept hidden.
Their visibility is controlled by the option Show obsolete surface element tools on toolbars.
Pet palette position Pet palette position can be:
Relative
Specify the horizontal (dx) and vertical (dy) distance from the operation in pixels.
Appear in the latest position
Pet palette appears in its latest position.
196
3.3.11.10. Display
3.3.11.11. Parts
Display of load panels when only parts are visible can also be controlled here.
Perform design calculations for visible parts only
If the checkbox is active, the design calculations are performed for the visible parts only. In case of large
models, this can result more efficient calculation process. However, if another part is set to visible, it
takes some time to perform design calculation of the elements that do not have design results yet. If
the checkbox is inactive, the program performs design calculations for every element in the model at
once. In this case, if another part is selected, the design results are already available and the user does
not have to wait for the calculation in every case.
The checkbox only affects the calculation of the following result components:
• required reinforcement
• crack width
• steel structural member utilization
• timber structural member utilization
Settings for envelopes and combinations are also placed here. The content of load cases and
combinations dropdown lists (on the result and design tabs) can be controlled here. AxisVM allows
creating different envelope sets (see Result display options in 6.1 Static). If the first option is selected
only the selected envelope will appear in lists. If the second option is selected only custom envelopes
will be listed. If the third option is selected all standard and custom envelopes will be listed.
Critical results: check to enable finding of critical combinations. If unchecked results will be available of
only for load cases / combinations and their envelopes.
Include combination type in the description of combinations: description of combinations can be
extended to show the combination type (ULS or SLS types).
Show ID number of critical combinations: to make it easier to compare critical combinations the ID of
the critical combination can be displayed as a prefix.
3.3.11.13. Analysis
At the beginning of the analysis AxisVM divides the system of equations into blocks according to the
available physical and virtual memory. It makes analysis more efficient but can considerably slow down
other applications. Set the amount of virtual memory you let AxisVM use during the analysis here.
Enable extended If more than 4 GB of memory is installed with a 32-bit operating system, this option makes it possible
memory access to get more memory for analysis. If this option is disabled it means that memory pages are not locked.
(AWE) See 2.1 Hardware requirements for details.
Using a single Using multiple threads/cores makes AxisVM run analysis on multiple threads. To make the most of this
thread / option it is recommended to use a processor with HT-Hyperthreaded or Dual Core technology.
Using multiple Multi-threading improves speed of calculation. Improvement depends on the available memory and
threads the model size. Linear analysis will be 1.5 times faster, while vibration analysis can be 4 times faster.
Setting Number of parallel threads allows adjusting the software to the capabilities of the hardware it is
running on.
Folder for You can specify the location of temporary files during analysis.
temporary files Select any of these options : Model file folder, Local system temporary folder, Custom
during analysis Create analysis log file: If this option is turned on technical details of the analysis will be logged and
saved to a text file modelname_log.txt.
Message sounds If this option is activated system sounds will be played after completing an analysis or getting an error
during analysis message. Sound card and speakers must be present.
User’s Manual X6 Release1 e2 199
3.3.11.14. Report
Report language Depending on your configuration you can select from the following languages: English, German, French,
Italian, Spanish, Dutch, Hungarian, Russian, Portuguese, Romanian, Serbian.
Table layout If Allow multiple columns is checked, narrow report tables will be printed in a multi-column layout to
reduce the space required. Minimum number of rows per column can be specified to avoid column
breaks for short tables.
The position of table of contents in reports can be selected from a list (Front page, Second page, Third
page).
Printer buffer If a report includes many pictures building the entire report in memory may consume too much system
resources and cause printing problems. In this case set printer buffer to hard disk.
Print page numbers even if page header is turned off
If this option is turned on page numbers appear on printed pages even if headers are disabled in the
printing dialog.
Translate automatic item names if report language is changed
If this option is turned on AxisVM-generated names of Drawings Library or report items will be trans-
lated automatically.
Company logo
200
A company logo can be loaded, saved, deleted using the three toolbar buttons. This image will appear
in the page header of printed drawings, tables, reports and/or on the cover page of the report according
to the settings for position, size, margins.
These settings have no effect on the RTF output as that is based on a separate RTF template file.
See 2.10.2 Report
3.3.11.15. Update
Searching for AxisVM checks regularly if there is an update available on the web. The frequency of update checks can
program update be controlled. If Never is chosen an update process can be launched by clicking AxisVM Web Update.
The date of the latest search is displayed. If internet connection goes through a proxy server, proxy
settings have to be defined after clicking Proxy settings.
AxisVM Click the button to get to the AxisVM Web Update Wizard which is a guide to the download process. If
Web Update download is complete and the Update the program option is checked on the last page, the program
quits and start the installation of the new release.
Proxy settings If the network reaches the web through a proxy server, the configuration data (proxy name, port, user
name and password) can be entered here.
User’s Manual X6 Release1 e2 201
Edit
The tree on the left side displays the available command groups. Use the Search field to search for a
command. Click a command in the list on the right side, then press the desired shortcut. Backspace clears
the assignment.
If the desired shortcut is already in use a Conflicting shortcuts warning is displayed.
Shortcut list
The Shortcut list tab displays a list of all available shortcuts. It can be a Command list or a Shortcut list
(the first is ordered according to the command groups, the second by the shortcuts)
For the entire list for the default configuration see 2.6 Keyboard shortcuts.
3.3.13. Language
The moveable Icon bar will get back to the left side. All flyout toolbars undocked and dragged to a new
position will get back to the Icon bar. Custom combo box and edit field width settings on postprocessing
tabs will be reset to default. See 6.1 Static.
User’s Manual X6 Release1 e2 203
AxisVM remembers the last position of dialogs and display them there. If any problem is detected on
systems with multiple monitors setting dialog boxes to default position can solve it.
3.4. View
3.5. Plugins
If third party applications are installed into the Plugins folder they appear under the Plugins menu.
Add-ons installed into the AddOns folder does not appear as menu items but create additional buttons
on toolbars.
3.6. Window
User’s Manual X6 Release1 e2 205
Property Editor provides the fastest way to change properties of the selected nodes, elements or loads.
All changes are made immediately. If the selection contains different elements it is possible to change
their common properties (e.g., after selecting trusses, beams and ribs their material and cross-section
will be editable). If result or design tabs are active the values are read only.
In certain fields regular mathematical expressions are also accepted.
Available operators and functions are:
(, ), SIN, COS, TAN, EXP, LN, LOG10, LOG2, SINH, COSH, TANH, ARCSIN, ARCCOS, ARCTAN, ARCSINH,
ARCCOSH, ARCTANH, INT, ROUND, FRAC, SQR, SQRT, ABS, SGN.
Few fast operators:
++8 adds 8 to the actual value
--8 deducts 8 from the actual value
Negative numbers within operation have to be in brackets.
In these expressions # substitutes the actual value (For instance #/3 divide it by 3). When entering a
value of nodal coordinates, load values, surface thicknesses you can refer to global coordinates as X, Y,
Z or x, y, z. In case of certain load types variables refer to other load components as well.
For nodal loads or point loads on beams variables Fx, Fy, Fz, Mx, My, Mz refer to force and moment
components. For distributed beam loads px1, py1, pz1, m1, px2, py2, pz2, m2 refer to load components.
Variable names are not case sensitive.
Example 1: If you want to turn selected distributed wind loads with different X components to Y direc-
tion enter 'px1' into field pY1 and 'px2' into field pY2 then enter zero into fields pX1 and pX2.
Example 2: to scale the structure in direction X by 200%, first select all nodes then click the line first
line and enter X*2 as X.
The question mark button turns on/off the help information.
Properties are displayed in a tree-like structure. Clicking a [+] or
[–] symbol before the property name expands or collapses a list
of sub-properties.
If the (...) button appears in a line the property can be changed
using a separate dialog.
If the (>>) button appears in a line the property can be picked up
from another element by clicking it.
Property Editor can be used to modify data but also to select and
filter elements with the same property.
Filter Selecting a property and clicking the filter button you can select all the elements having the same
property value.
Example: changing an existing cross-section in the whole structure.
Select the cross-section property of a rib element, click the Filter button to select all rib elements with
this cross-section then change their cross-section property.
Load components can be filtered the similar way. Select a load, select a load component in the Property
Editor and click on the Filter button. This way you can select all loads with the same component value.
Lets you set the display of the Status, Color coding, Coordinate, and Color legend windows to on or off.
See 2.19 Information windows
206
The submenu makes several options available. An automatically fitted background picture can be
loaded to the main window of AxisVM to show the model in its future environment. Load Background
Picture... submenu item or [Ctrl+B] opens a file browser dialog, Reload Background Picture shows the
most recently used picture files. In multi-window mode each window can have its own background
picture.
Picture in the active window can be turned on and off by clicking Display or by [Ctrl+Alt+B].
Save Background Picture saves the picture in the active window into a file. If the aspect of the picture
differs from the window aspect Shift Background Picture makes it possible to drag the background to a
new position. Remove Background Picture removes the picture in the active window.
Background pictures are saved into the AXS file.
After loading a background picture the model can be set to an appropriate view by zooming out,
zooming in, panning, rotating and setting the perspective.
Splits the graphics window horizontally into two parts. Clicking into one of them makes that window
active. The display settings of each window can be set independently. Different load case or combination
can be selected for each window.
You can maximize or minimize or restore the graphics windows by using the buttons at the top-right of
the windows.
User’s Manual X6 Release1 e2 207
Splits the graphics window vertically into two parts. Clicking into one of them makes that window active.
The display settings of each window can be set independently. Different load case or combination can
be selected for each window.
You can maximize or minimize or restore the graphics windows by using the buttons at the top-right of
the windows. Different load cases can be set in each window but only when displaying results.
Decrease / Increase These two menu items are to change the font size for all labels on diagrams.
label font size
If this option is checked the program prevents labels from overlapping by moving and hiding less im-
portant labels. In the latter case zooming in may increase the number of visible labels. Priorities are
determined by label types and values. If large number of labels are to be displayed this optimization
may be a time-consuming process.
You can also set this option in Display options/Labels. See 2.16.18 Display options
The Drawings Library contains drawings saved in the program. Drawings are not saved pictures but
instructions how to draw a view of the model or parts of it including multi-window settings. Drawings
can be reloaded to restore saved view and display settings. Including drawings into a report makes it
easier to update the report when the model has changed and recalculated as drawings will be updated
automatically like tables.
Drawings Library can store displacement, force, stress diagrams of line elements, diagrams of steel and
bolted joint design, punching analysis, reinforced concrete column check and beam design in an
associative way.
Clicking the arrow beside the tool button an existing drawing can be
selected from a pop-up list, restoring its view and display settings.
Graphic symbols
Display of graphic symbols can be modified in library items. Select one or more items and click
the button on the toolbar. The status of symbols in the selected drawings is displayed and can
be changed. Mixed status is represented by greyed checkmarks.
Text options Text size and appearance can be controlled in the Text options group.
Page options Size and orientation of the PDF document pages can be set in the Page options group.
3D export options Sometimes drawing all edges makes the view a bit blurred. So export of edges can be turned on / off.
By clicking this tool button one or more drawings can be saved into the Drawings Library.
If the current drawing already exists, a Found in the Drawings Library label is displayed in the dialog. It
can be overwritten or the drawing can be renamed. Multiple drawings button opens additional options.
Load cases, load combinations (and result components if results are displayed) can be chosen. AxisVM
creates all combinations (i.e., all selected result components in all selected load cases) and saves them
into the library with the current view and display settings.
If multiple model views have been created by dividing the main window two options are available:
Clicking the Drawings Library button displays the Drawings Library dialog.
User’s Manual X6 Release1 e2 211
3.7. Help
3.7.1. Contents
Opens the present User’s Manual in PDF format. AxisVM X4 Release 2 reintroduces context-sensitive
[F1]
help: pressing F1 in dialog windows opens the manual at the related topic.
3.7.2. Guides
Guides offer insight into some of the design methods used in AxisVM.
SC1 Steel connection design explains how the SC1 module should be used.
The AutoMcr Guide summarizes the calculation method of the lateral-torsional buckling resistance in
the SD1 steel design module. The AutoNcr Guide explains the method of determining the buckling
length in the same module, shows the area of applications and the proper usage.
Custom Stiffness Matrix Guide explains the theoretical background of calculating elements of a custom
stiffness matrix that can be assigned to a domain. See 4.9.6.6 Domain with a custom stiffness matrix.
The FFA Guide is about the theoretical background behind the footfall analysis implemented in the FFA
module.
Visits AxisVM Home Page using the default Internet browser (https://www.axisvm.eu)
3.7.5. About
[Shift]+[Ctrl]+[Z]
See 3.3.3 Layer Manager
[F11]
See 3.3.4 Stor
[F7]
See 2.9 Table Browser
[F12]
See 2.10 Report Maker
[F10]
See 3.6.10 Drawings Library
4. The Preprocessor
The preprocessor lets you create or modify the geometry of the model, in a completely visual way. The
advanced Visual Modeling feature allows quick and reliable modeling and design.
This chapter introduces the AxisVM modeling commands (geometry generation, element / mesh
generation, and load case/combination definition).
4.1. Geometry
Geometry commands let you interactively and graphically create the model geometry in 3D.
The model geometry is defined by nodes (points), mesh lines (lines) between nodes, and surfaces
(triangular or quadrilateral) created from three or four appropriate lines. Later you can define finite
elements based on the geometry constructed here.
In the case of frame structures (beams or trusses) the mesh consists of the axes of the elements.
214
Graphics
Property
area
Editor
Context sensitive
help message
Speed buttons
When AxisVM starts, the graphical user interface is ready for geometry editing. In case of a new model
X-Y, X-Z or perspective view can be set as the default view. In case of an existing model the latest view
settings will be loaded.
Using the horizontal icon toolbar at the top of the graphics area you can apply various commands to
construct geometry meshes describing the geometry of your finite element model.
See 4.8 Geometry Toolbar
Using the vertical icon bar on the left you can apply commands that change the display of the model,
and can configure the working environment of the editor.
See 2.16 The Icon bar
When the model is complex, it is useful to display different views of the model simultaneously on the
screen. AxisVM allows you to split the graphics area horizontally or vertically. Each newly created
graphics window has its own settings, and allows the independent display of the model views. This
feature is also useful when interpreting results. Find split commands in the Window menu.
Split horizontally Splits the active graphics window horizontally into two equal parts. The top window will become the
active window.
See 3.6.4 Split Horizontally
Split vertically Splits the active graphics window vertically into two equal parts. The left window will become the active
window.
See 3.6.5 Split Vertically
Close Window Closes the active window if there are more than one graphics windows in use.
The new default window will be that in which you previously worked.
You can change views during any editing command.
In the perspective view some editing commands cannot be used, or are limited in use.
User’s Manual X6 Release1 e2 215
AxisVM uses different coordinate systems, to describe the model. The global coordinate system is used
to describe the model geometry. Local coordinate systems are mainly used in the element definitions.
The local systems are usually defined by the element geometry and additional references. AxisVM
denotes the axes of the global system with capital letters, and the local axes with small letters.
The geometry can be created using Cartesian, Cylindrical or Spherical coordinate systems.
See 4.3.2 Polar coordinates
Global and relative A new model uses the view selected in the New Model dialog (see 3.1.1 New model).
origin The origin of the coordinate system is shown by a blue X initially located at the left bottom corner of
the editor window.
A fixed (X, Y, Z) and a relative (dX, dY, dZ) global system are used to locate points (nodes) in your
model. The origin of the relative system can be moved anywhere (using [Alt]+[Shift] or [Insert]), at any
time during modeling.
The Coordinate Window displays either the fixed or the relative global coordinates according to its
current settings. If the relative mode is selected, the denotation of axes becomes dX, dY, dZ.
With the help of the Coordinate Window, and according to the movement of the relative origin you
can make measurements on the model (distances, angles).
The nodal displacements and mode shapes refer to the fixed global system.
In the X-Y and Y-Z views the third axis (normal to the view’s plane) is oriented towards you. As
a result, when a copy is made by translation with a positive increment about the respective third
axis, the copies will be placed nearer you. In X-Z view the opposite occurs as the third axis in
this case points to the opposite direction.
See 4.9.22 References
In addition to the Cartesian global coordinate system, you can use either a cylindrical or a spherical
coordinate system. One of the polar coordinate systems can be selected through its corresponding
radio button in Settings / Options / Editing / Polar coordinates.
In the Coordinate Window three variables will be displayed depending on selection:
Cylindrical
h: the value measured from the view plane to a point on the cylinder’s main axis (that is
perpendicular to the view plane) oriented outward from the screen
r: radius that is the distance on the view plane from the projection of the point to the cylinder’s
main axis
a: the angle between the line that joins the point with the origin and the horizontal
Spherical
r: the radius, that is the distance from the point to the sphere’s center (origin)
a: the angle on the view plane between the line that joins the projection of the point with the
origin and the horizontal
b: the angle between the line that joins the point with the origin and the view plane, which is
positive if the point is in front of the view plane (between the user and the view plane).
216
Displays the current absolute and relative values of the cursor position in the global coordinate system
(Cartesian and cylindrical or spherical).
You can switch between absolute and relative coordinate displays, by clicking on the letters d in the
Coordinate Window (delta switch). The display of the d’s also show whether the relative coordinates are
enabled or not.
During editing it is possible to jump into the coordinate window by shortcuts. The default assignment
can be changed, see 3.3.12 Keyboard shortcuts
The relative switch (delta) can be used together with the constrained cursor movements.
See 4.7.4 Constrained cursor movements.
You can enter expressions in the edit fields (e.g.: 12.927+23.439, cos(45), sin(60))
User’s Manual X6 Release1 e2 217
4.5. Grid
Editing tools help the work by several features. See 2.16.19.2 Editing
When you position the cursor over the graphics area, AxisVM finds the entity of the model that is closest
to the center of the cursor from among the entities that are located in or intersect the identification
area. The size of the identification area can be set at Settings / Options / Editing / Cursor identification.
The current shape of the cursor shows what kind of entity was identified. Depending on entity type, the
cursor will have the following shapes:
Node
Mid-side node
Support
Edge hinge
Center of an arc
Arc
Tangent
Bézier curve
References
Line
Surface
Domain
Rigid element
ARBO element
CRET element
218
Guideline
Structural gridline
Intersection
Perpendicular (normal)
Dimension line
If there are several entities at the same location, the program identifies the first entity according to the
ordering of the list above. If there are multiple entities of the same type, the cursor will show a double
symbol.
Use the Coordinate Window to find out which one of the elements was actually identified.
Background The cursor can be set to detect the lines on architecture background layers.
detection
During the model editing, coordinates of the cursor can be specified directly entering the numerical
values into the Coordinate Window. There are two ways to enter the numerical values:
1. by pressing the corresponding character button on the keyboard
2. by clicking with the left button on the desired coordinate value display field, and then typing
in the value.
If the relative mode is enabled (the letter d is pressed), the coordinates you enter will define a point
from the relative origin.
If contradictory values are entered (in case of a constraint), the last entered value will update the others.
You can enter expressions in the edit fields (e.g.: 12.927+23.439, cos(45), sin(60))
The relative origin can be moved at any time to the current mouse cursor position by pressing the Insert
key. So, when drawing a polyline, you can specify the end point coordinates relative to the previous
point.
To draw a line with a given length and direction move to relative origin to the starting point
(using [Alt]+[Shift] or [Insert]), enter the angle at d a[°] and enter the length at d r[m] then press the
Enter button.
User’s Manual X6 Release1 e2 219
The distance between two points or the length of a line can be measured by moving the relative origin
onto the first point and then identifying the second point by positioning the cursor over it. In this case
the value of dL in the Coordinate Window is the distance between the points.
The cursor can be moved to a location relative to a reference point by moving the relative origin onto
the reference point, then entering the angle in the input field da and the distance in the dr input field.
Δα Holding the [Shift] key pressed, the cursor is moving along a line that connects its current positionwith
the origin, and that has an n*Δα angle, where the value of n depends on the current cursor position.
Custom α Holding the [Shift] key pressed, the cursor is moved a line that connects its current position with the
origin, and that has an or + n*90° angle, where the value of n depends on the current cursor position.
and can be set in Settings/Options/Editing/Constraint Angle.
The meaning of origin depends on the d switches of the coordinate palette. Turning off both the origin
will be the global origin. Turning on any of the d switches the origin will be the local origin.
When the cursor identifies a domain or surface element pressing [Shift] makes the cursor move in the
plane of the element.
Intersection point
Perpendicular
Midside point
Geometry tools
The icons of Geometry tools allow you to lock the direction of drawing a line.
See 2.16.10 Geometry tools
220
You can lock the value of a coordinate, allowing for better positioning. A locked coordinate will not
change on cursor motion. Locking can be achieved by using [Ctrl+Alt] + [X], [Y], [Z], [L], [R], [A], [B], [H]
respectively.
A black rectangle over the coordinate input field shows that the coordinate is locked. To cancel
coordinate locking, press the same button combination, that was used to lock it.
Frozen X coordinate Frozen angle Frozen radius
At the intersection point of the lines, a node will be generated and the lines will be bisected. If surfaces
are intersected by lines, they will be split, and the resulting elements will have the same material and
cross-sectional properties as the original. Set the line intersection options in Settings / Options / Editing
/ Auto Intersect. See 2.16.19.2 Editing
If Auto Intersection is on, surfaces will be divided into smaller surfaces if necessary. Surface finite
elements are also divided and the new elements inherit the properties and loads of the original element.
These tool buttons create new geometry or change the existing one..
If you are working on parts and Settings / Options / Editing / Auto / Part Management option is
checked then all new geometric entities will be added to the active parts.
The geometric entities can be selected prior to applying the geometry construction commands, as well.
When working on parts with Settings / Options / Editing / Auto / Part Management turned on all
geometric entities created will be automatically added to the active parts.
4.8.2. Line
The Line tool is to construct lines or other simple shapes. The line type can be chosen by clicking on the
arrow at the bottom-right corner of the currently used Line tool icon, and then clicking on the desired
icon.
The Line tool offers the following options to draw simple shapes:
Line Constructs straight lines by defining their end points (nodes). You must graphically or numerically (by
the Coordinate Window) specify the end points (nodes). The command lets you generate one or more
independent lines. You can cancel the process by pressing the [Esc] key or the right mouse button.
In perspective view lines are drawn on the Z = 0 plane by default. To draw lines in perspective in a
different plane workplanes can be used.
See 2.16.7 Workplanes.
Polyline Constructs a series of connected straight lines (a polyline). You must specify the vertices.
Exit current polyline by any of the following
1. [Esc] key
2. [Esc] key a second time will exit polyline drawing mode.
3. right button & popup menu/Cancel
4. left button while pointing to the last point (node) of the current polyline.
Rectangle Constructs a rectangle (its corner points (nodes) and edge lines). You must specify two opposite corner
points.
Skewed rectangle Constructs a skewed rectangle (its corner points (nodes) and edge lines). You must specify one of its
sides (by its end points), and then the other side.
Polygon Number of sides has to be defined in a dialog. Polygon has to be defined by entering a centerpoint and
2 polygon points.
Polygon Number of sides has to be defined in a dialog. Polygon has to be defined by entering three points of
the arc.
222
4.8.3. Arc
Draws an arc or a circle. Arcs and circles will be displayed as polygons according to the Arc resolution
set in Settings / Preferences / Display.
[Esc] cancels the command.
Defining an arc by its radius, and starting and ending points.
2nd point
3rd point arc
Defining an arc by three points. The command can be applied in perspective setting as well.
2nd point
3rd point
arc
1st point
End point
This function creates a horizontal divider line passing through the cursor position. This line is in a plane
parallel to the X-Y, X-Z or Y-Z plane depending on the actual view (or parallel to the workplane if a
workplane is used). Creates new nodes at the intersections. If finite elements are intersected new
elements inherit properties and loads of the original element.
User’s Manual X6 Release1 e2 223
This function creates a vertical divider line passing through the cursor position. This line is in a plane
parallel to the X-Y, X-Z or Y-Z plane depending on the actual view (or parallel to the workplane if a
workplane is used). Creates new nodes at the intersections. If finite elements are intersected new
elements inherit properties and loads of the original element.
The quad and the mesh are displayed with solid grey lines.
If the mesh leads to quad subdivisions that are distorted (have an angle smaller than 30° or greater than
150°), the quad is displayed with grey dotted lines.
If a quad shape is entered that is not allowed (e.g., concave), the quad is displayed with red dotted lines.
If a quad shape is entered that is not allowed (e.g., concave), the quad is displayed with red dotted lines.
The triangle and the mesh are displayed with solid grey lines.
If the mesh leads to quad subdivisions that are distorted (have an angle smaller than 30º
or greater than 150º), or to triangle subdivisions that are too distorted (has an angle smaller than 15º
or greater than 165º), the triangle is displayed with grey dotted lines.
If a quad shape is entered that is not allowed (e.g., three collinear corners), the triangle is displayed with
red dotted lines.
4.8.7. Intersect
If the Settings / Options / Editing / Auto / Intersect checkbox was not enabled in the dialog window at
the time of creating the geometric entity, using this command you can intersect the selected lines.
Intersection can be filtered by element types in a dialog. Selected lines will be divided by creating nodes
(points) at their intersections.
If finite elements are assigned to the lines, finite elements are also divided and inherit the properties
and loads of the original element..
Removes the selected nodes at the intersections of lines. It makes it easier to construct trusses crossing
but not intersecting each other or to remove unnecessary division points along a line.
Intersection nodes can be removed only if the number of connecting lines are even and lines can
be joined.
Removes unnecessary intermediate nodes on lines. Nodes with two connecting lines are removed
provided they can be joined.
Extends existing straight lines or arcs to meet another straight line, arc or plane.
Extension modes Extend lines to meet another line
The first step is to specify the boundary: click on an existing line or arc or enter two points of
an imaginary boundary line. Click on lines near their end point to extend.
Extend lines to meet more than one line
The first step is to click on multiple lines defining the boundary. Click OK on the selection
toolbar or press Enter. Select the lines to extend. Click OK on the selection toolbar or press
Enter. Both ends of the lines will be extended to meet the boundary.
Extend lines to meet a plane
User’s Manual X6 Release1 e2 225
The first step is the specify the boundary plane: click on an existing domain or enter three non-
collinear points of an imaginary boundary plane. Click on lines near their end point to extend.
Extend lines to meet more than one plane
The first step is to click on multiple domains defining the boundary. Click OK on the selection
toolbar or press Enter. Select the lines to extend. Click OK on the selection toolbar or press
Enter. Both ends of the lines will be extended to meet the boundary.
Settings Extend lines even if there is no actual intersection
This is a two-state button. If it is down (its background turns white) lines will be extended even
if the point of intersection does not fall between the start and end point of the boundary line.
If it is up (its background is blue) lines will be extended only if the point of intersection is on
the boundary line.
The last three buttons control the behaviour of lines connected to the extended point.
Move free ends, add splice if other elements are connected
The program automatically chooses from the two options listed below. If no lines are con-
nected to the extended point, it chooses the first option, otherwise it chooses the second one.
Always move line ends
The node of the extended point is moved to the point of intersection. The direction of lines
connected to this node will change.
Always splice line ends
The node of the extended point is not moved. A new node will be created at the point of
intersection and a new line (a splice) is inserted between the two nodes. Connected lines re-
main unchanged.
Trims existing straight lines or arcs to meet other straight lines, arcs or planes.
Trimming modes Trim lines to meet another line
The first step is to specify the cutting line: click on an existing line or arc or enter two points
of an imaginary cutting line. Click on the line segments to be removed.
Trim lines to meet other lines
The first step is to click on multiple lines defining the cutting lines. Click OK on the selection
toolbar or press Enter. Select the lines to trim. Click OK on the selection toolbar or press Enter.
Click on that side of the cutting lines where you want to remove the line segments.
Trim lines to meet a plane
The first step is the specify the cutting plane: click on an existing domain or enter three non-
collinear points of an imaginary cutting plane. If the cutting plane intersects the lines click on
the line segments to be removed.
Trim lines to meet more than one plane
The first step is to click on multiple domains defining the cutting planes. Click OK on the se-
lection toolbar or press Enter. Select the lines to trim. Click OK on the selection toolbar or press
Enter. Click on that side of the cutting planes where you want to remove the line segments.
Settings Trim lines even if there is no actual intersection
This is a two-state button. If it is down (its background turns white) lines will be trimmed even
if the point of intersection does not fall between the start and end point of the boundary line.
If it is up (its background is blue) lines will be trimmed only if the point of intersection is on
the boundary line.
The last three buttons control the behaviour of lines connected to end point of trimmed lines.
The trimmed line always gets a new end point at the point of intersection.
Move free ends, remove segment if other elements are connected
The program automatically chooses from the two options listed below. If lines are connected
to the extended point, it chooses the second option.
226
Extend/trim selected lines to the point of intersection with the first one
Extends/trims consecutively selected lines to their point of intersection with the first line.
The next two buttons control the behaviour of arcs where there may be two points of intersection.
Settings Extend the arc end where the extension is shorter
The program always chooses the end where the point of intersection is closer.
The last three buttons control the behaviour of connected lines according to the above description.
By Ratio: Lets you divide the selected lines into two segments. You
must specify the parameter a of the location of the inserted
node relative to the first node (i). The parameter a must be
between 0 and 1. a = 0.5 represents a division of the selected
lines into two equal segments.
By Length: Lets you divide the selected lines into two segments. You
must specify the length (d) of the segment corresponding to the
first node (i end). The parameter d must be between 0 and the
total length.
Into N equal parts: Lets you divide the selected lines into several
equal-length segments. Specify the number of segments (N).
Uniform by length: Lets you divide the selected lines into several
equal-length segments. Specify the length of segments (d).
If finite elements are divided the new elements inherit properties and loads of the original elements.
After defining the intersecting plane intersection lines and nodes will be added to the model. Domains,
beams and ribs will be divided.
4.8.16. Intersect plane with the model and remove half space
This operation is similar to Intersect plane with the model, but after defining the plane a half space can
be selected. Elements within that half space will be deleted.
This function removes nodes not connected to any element (line element, domain, support, etc.) and
lines without element properties assigned. Possible options are
From the selection: Only selected nodes and lines will be checked.
Also check nodes and lines within domains: When mesh-independent loads are placed on the domain
additional nodes and lines may be created within the domain to guide the meshing algorithm. Deleting
the mesh does not remove these nodes and lines. As these nodes and lines have no physical role in the
model they are detected during purging. Leaving this option unchecked keeps them.
Select but don’t remove: Unnecessary lines and nodes will be selected and not removed.
The function fixes domain contours forcing contour segments into the same plane checks the openings
and adjusts arcs if radius is not the same at the start point and the end point.
Checking domain If checked the program removes zigzags from nearly straight segments of domain contours if the
contours deviation from the line is below the Tolerance set here.
Check all loads This function recreates all loads that depend on the geometry (like surface loads distributed over beams
or load panel loads). Rebuilding of loads may cause loss if results.
4.8.20. Surface
In any cases when you wish to model surfaces (plates, membranes, or shells) you have to create a mesh
that consists of triangles and convex flat quadrilaterals. The mesh then can be refined. The command
searches all triangles and quads in the selected mesh of lines. You must select all surface edges when
applying the command.
Introduction of domains and automatic meshing rendered this method obsolete.
Quads have to be flat. AxisVM takes into account only those surfaces that have an out-of-plane
measurement smaller that the tolerance entered in the Settings / Options / Editing / Editing
Tolerance.
Moving selected nodes into the same plane: if the plane is a global one you can move selected nodes
into this plane easily.
1. Click on any of the selected nodes.
2. Select the entire column of the respective coordinate.
3. Use Edit / Set common value to set a common coordinate value.
Using pet palettes Depending on the type of the dragged element different pet palettes appear on the screen. Their
position can be set in Settings / Preferences / Toolbar. See 3.3.11 Preferences
User’s Manual X6 Release1 e2 229
Dragging nodes
Chamfer
Fillet
The new arc is defined by the dragged node, the start point and midpoint of the original arc.
Enabled only in detaching mode. It pops up a list of prop-
erties to be copied.
Entering node coordinates: Clicking a node the table of nodes appears where coordinates can be
changed. After selecting one or more nodes their coordinates can be edited in the property editor as
well.
Examples of aligning nodes to a plane if this plane is parallel to one of the global coordinate plane:
Dragging lines
Fillet
Converting to arc
Replacing a straight line with an arc based on two end point tangents.
See Dragging nodes
Modifying arcs
Fillet
Converting to line
4.8.22. Delete
[Del] See in detail 3.2.9 Delete
The commands related to the definition of the finite elements are described below.
The commands associated with the icons let you define the finite elements used for modeling. In the
definition process you must define and assign different property sets.
Properties of finite Depending on the type of finite element, you have to define the following properties:
elements
o: optional
Note that some elements like springs and gaps can have nonlinear elastic stiffness properties that are
taken into account only in a nonlinear analysis. In a linear analysis the initial stiffness is taken into
account for the spring element, and the active or inactive stiffness depending on its initial opening for
the gap element.
User’s Manual X6 Release1 e2 231
4.9.1. Materials
Define Materials
For color coding of table cells and sorting by columns see 2.9 Table Browser.
Lets you define and save material property sets or load them from a material library.
If you delete a material the definition of the elements with the respective material will be deleted.
If a material type is deleted all elements made of this material will be deleted.
Material Properties Depending on the type of the finite element you must define the following material properties:
Finite Element E ν α ρ
Truss • • •
Beam • • •
Rib • • •
Membrane • • • •
Plate • • • •
Shell • • • •
Support
Rigid
Diaphragm
Spring
Gap
Link
232
In AxisVM all the materials are considered to be linear elastic (Hooke’s Law) or plastic, and
uniform isotropic or orthotropic (for beam, rib, membrane, plate, and shell elements).
Some elements can have nonlinear elastic material (truss), or stiffness (support, gap, link, spring
elements).
Nonlinear material models are taken into account only in a nonlinear analysis.
In a linear analysis the initial stiffness is taken into account for the nonlinear elements.
4.9.2. Cross-sections
Define Cross-
sections
For color coding of table cells and sorting by columns see 2.9 Table Browser.
Lets you define and save cross-sections or load them from a cross-section library (see 4.9.2.1 Browse
cross-section libraries). The beam, truss, and rib elements require a cross-section. The properties are
interpreted in the element’s local coordinate system.
For cross-section properties see 3.1.16 Cross-Section Library
If you delete a cross-section, the definition of the elements to which it was assigned will also be
deleted. The lines will not be deleted.
You must enter values for all properties.
Cross section properties are defined in the coordinate system of a truss / beam / rib element.
User’s Manual X6 Release1 e2 233
A cross-section can be imported from the database of the Library or can be defined by entering geo-
metric parameters describing Parametric shapes.
Library Choose thin-walled, thick-walled or composite cross-sections and click on the desired shape on the
toolbar. The tree on the left will list the available tables for the selected shape.
The list in the middle shows the cross-sections of the selected table. Clicking on column header sorts the
list into ascending/descending order by the respective parameter: Name, Height, Width, Ax (cross-section
area). One or more cross-sections can be selected. Cross-section parameters and drawing appears on
the right.
For cross-section properties see 3.1.16 Cross-Section Library
If a button is disabled the library does not contain cross-section tables for the respective shape. If
you define a custom cross-section table with that section type, and open this dialog the button
will be enabled.
Click on the Replace button to replace the the selected row in the table of cross-sections with the item
selected here. Click the Add button to append selected items to the table of cross-sections. If multiple
cross-sections are selected Replace will be disabled.
Parametric shapes Choose thin-walled, thick-walled or composite cross-sections and click on the desired shape on the
toolbar to open a dialog. Set the parameters describing the geometry and click OK to add the cross-
section to the list of parametric cross-sections with a generated name. You can add different parametric
shapes to the list.
Double-click on the name to edit it. Press the Delete key to remove a selected cross-section(s) from the
list.
If there is only one cross-section in the list click on the Replace button to replace the the selected row in
the table of cross-sections. Click the Add button to append all cross-sections in the list to the table of
cross-sections. If the list contains more than one cross-section Replace will be disabled.
In case of parametric shapes there is no need to select cross-sections in the list to import them.
The entire list is imported.
User’s Manual X6 Release1 e2 235
Spring characteristic can be selected from the Spring characteristic library. A window with libraries and
the characteristics is displayed. The spring characteristics are listed depending on the selected degree
Browse Spring char- of freedom (translational, rotational, all). The properties of selected characteristic are shown in the third
acteristic column.
library
Specify a new spring characteristic is described in 3.1.17 Spring characteristics library section.
New spring
characteristic
236
Properties of spring Existing spring characteristics can be modifeid by clicking to a non-editable column. In the appearing
characteristics dialog all properties of the spring carateristic can be defined or changed.
The properties of spring characteristics are described in 3.1.17 Spring characteristics library section.
Top toolbar
Property fields
Bottom toolbar
After clicking the icon, a direct drawing toolbar and property editor appears. With the help of this win-
dow columns, beams, walls, slabs and holes can be drawn. Their properties can be set previously and
changed any time during the drawing.
The top toolbar shows the type of the object to draw and the orientation of the object
(for columns and walls). Property fields can be edited like in the Property Editor.
The bottom toolbar shows the drawing methods available for the object (one segment,
polyline, polygon, rectangle, etc.).
Clicking a domain contour before drawing holes forces the drawing into the plane of the domain.
Beam (spatial)
Wall (always vertical with a constant height, i.e., its normal and upper/lower edges are parallel
to the global X-Y plane)
Slab domain (parallel to the global X-Y plane)
Hole
Wall justification: during drawing walls it can be set whether the line drawn is the
centerline or left / right side of the wall (or interior / exterior side in case of arcs
and rectangles).
Column
Object geometry
Walls on a rectangle
This function allows direct drawing of nodal and line supports with
predefined properties.
Nodal support
4.9.6. Domain
A domain is a planar structural element with a complex geometric shape described by a closed polygon
made of lines and arcs. A domain can contain holes, internal lines and points.
Polygon vertices, holes and internal lines must be in same plane.
A domain has the following parameters:
Element type (membrane, plate, shell)
Material
238
Thickness
Eccentricity
Local coordinate system
Custom color for rendered view
The following parameters can be assigned to the polygon, hole edges, internal lines and points of a
domain:
point, line, and surface support
rib element
distributed load
dead load
thermal load
nodal degrees of freedom (DOF)
Domains can be defined for floors, walls, and any other complex structural surface element.
The domain can be meshed automatically. See 4.11.1.2 Meshing of domain
More than one domain can be used to model a structural element.
2nd domain
3rd domain
Finite element Domains can be modeled with different type of finite elements (membrane, plate, shell). To understand
the meaning of these types, see 4.9.10 Surface elements.
Select lines on the contour of the domains you want to define. If you select more lines or lines from
different planes, AxisVM will find the planes and the contour polygons of the set. The program applies
the parameters you entered in a dialog window.
User’s Manual X6 Release1 e2 239
Tapered domains and/or domains with variable eccentricity can be defined by clicking the
button beside the edit fields.
k,torsion If the material is concrete, it is possible to set reduced torsional strength of RC plate or shell.
k,torsion factor must be between 0.1 and 1, i.e., the torsional strength of the RC plate or shell can be
10%-100% compared to the elastic, isotropic material.
k,bending If the material is concrete, it is possible to set reduced bending strength of RC plate or shell.
k,bending factor must be between 0.1 and 1, i.e., the bending strength of the RC plate or shell can be
10%-100% compared to the elastic, isotropic material.
In case of concrete surfaces with reinforcement, nonlinear analysis does not take into account k,shear
and k,torsion factors if Use reinforcement in calculation is checked. (See 6.5.6 Nonlinear analysis of RC
surfaces)
If domain thickness would be reduced to zero on certain points a warning will appear.
The reference plane of the static model is the middle plane of the domain. Eccentricity is relative
to this plane.
Eccentricity Setting eccentricity is optional. Choosing Constant eccentricity from the list or setting e = 0, the midplane
of the domain will be the same as the plane of the statical framework. Other options are
Constant eccentricity: Midplane of the domain gets an offset of e in the
local z direction.
One-way eccentric: Enter e1, e2, click on the button then click on two
eccentricity reference points on the model. The eccentricity of the mid-
plane will change linearly between reference points, being e1 at the
first point and e2 at the second one.
Two-way eccentric: Enter e1, e2, e3, click on the button then click on
three reference points on the model. The eccentricity of the midplane
will change linearly between reference points, being e1 at the first
point, e2 at the second one and e3 at the third one.
If multiple domains are selected and the Eccentricity group option is activated domain eccentricities will
be set to make the upper or lower plane of the domains align. Changing the thickness of any domain in
the group will update the eccentricity of other domains within the group to keep the planes aligned.
The Pick up button is to pick up thickness and eccentricity from another domain. As reference points are
also picked up the thickness change will follow the plane defined by the domain.
Eccentricities and eccentricity groups can serve as a base for color coding (see 2.19.3 Color coding) and
eccentricity groups can be displayed as parts (see 2.16.14 Parts).
Color Domains can have their own fill and outline color used in rendered display mode. The default values are
taken from the material colors. If a color coding is applied the domain color is determined by the color
coding both in wireframe and rendered modes.
See 2.16.5 Color coding
Modify a domain Select the domain (click on the contour line of the domain) you want to modify and make the changes
in the dialog displayed.
Delete a domain Press the [Del] button, select the domains (click on the contour line of the domain) you want to delete
and click OK in the dialog.
User’s Manual X6 Release1 e2 241
Defining a composite ribbed domain is an easier way to place a system of ribs over a domain. Unlike the
parametric ribbed slabs (1) the ribs can run in local x or y direction but not both, (2) their material can be
different from that of the domain and (3) they can have any cross-section taken from the cross-section
library or created in the cross-section editor.
Beyond basic domain parameters (see 4.9.6.1 Defining a normal domain) the following parameters can be
specified:
Composite ribbed domains require plate or shell element type.
Origin of the rib grid Ribs can run in the local x or y direction. The
origin of the rib grid can be entered numer-
ically or picked up from the model.
Create actual ribs In case of steel ribs cast partially or entirely in concrete Create actual ribs must be left unchecked. If upper
or lower ribs are attached to the surface of the slab or the spacing is high compared to the dimensions of
the domain it is recommended to activate this option.
The reference plane of the static model is the middle plane of the concrete plate.
Application of logi- If Create actual ribs is left unchecked, composite ribs will be represented as logical entities, no finite ele-
cal ribs ments will be created. Ribs will be taken into account when creating the material stiffness matrix of the
plate. From the 3D geometry we calculate the effective orthotropic elastic constants of the material, suit-
able to represent the stiffness of a periodic body with a 2D model. This is carried out according the theory
of asymptotic homogenization. The effectiveness of such a method is mainly influenced by the ratio of
the length of the period and the dominant size of the domain being modelled (in this case the ratio of
parameter d and the span of the domain along the rib-perpendicular direction).
As a general guideline we can say that the spacing of the ribs be way smaller than the dominant in-
plane size of the problem domain.
The material stiffness matrix of a general shell element:
{𝑁} 𝒜 ℬ {𝜖0 }
{ }=[ T ]{ },
{𝑀} ℬ 𝒟 {𝜅}
𝑄𝑥 𝐴 𝐴45 𝛾𝑥
{ } = 𝐾𝑠 [ 44 ]{ }
𝑄𝑦 𝐴45 𝐴55 𝛾𝑦
where matrices
𝐴11 𝐴12 𝐴16 𝐵11 𝐵12 𝐵16 𝐷11 𝐷12 𝐷16
𝒜 = [𝐴12 𝐴22 𝐴26 ] ; ℬ = [𝐵12 𝐵22 𝐵26 ] ; 𝒟 = [𝐷12 𝐷22 𝐷26 ]
𝐴16 𝐴26 𝐴66 𝐵16 𝐵26 𝐵66 𝐷16 𝐷26 𝐷66
242
can be originated from the 6x6 stiffness matrix of general Hooke’s law. Matrix ℬ represents the coupling
effect between bending and membrane or in other words the eccentric effects. This matrix can be zero if
and only if the ribs both have zero eccentricity. In every other case the structure must be modelled using
shell elements to avoid a serious loss of information. The entries of the matrices are determined by ap-
plying independent strain states on a 3D model, and retrofitting the obtained stress field. Since a shell has
8 generalized strain components, we need to perform 8 finite element calculations in the background. A
shell equipped with the so calculated stiffness properties is equivalent to the original problem in a way
that the work needed to deform the body to a certain level remains the same, but it doesn’t entail the
equivalence of all displacement components at all points.
Logical ribs are most effective when the spacing of the ribs is dense or the plate is mostly loaded
by distributed loads. This is due to the local effects of concentrated forces.
Tubular Rectangular
Origin of the grid Voids are running in local x or y direction
along a grid.
The origin of this grid can be specified nu-
merically or picked up from the model.
The reference plane of the static model is the middle plane of the concrete plate.
Tubular voids Parameters: Ø is the diameter of the void, d is the distance between the centers of the
voids.
Rectangular voids Parameters: b is the width of the void, h is the height of the void, d is the distance be-
tween the centers of the voids.
Calculation The calculation method is the same what can be seen at Composite ribbed domain, since present config-
urations can be considered as special cases of that. Consequently, all the properties of the method written
there applies for this domain type too.
As a general guideline we can say that the spacing of the voids be way smaller than the dominant
in-plane sizes of the problem domain.
In the context of elastic anisotropy, the material stiffness (ABD matrix) of a general flat shell element can
be described with the following system of equations
{𝑁} 𝒜 𝓑 {𝜖0 } 𝑄𝑦 𝐴 𝐴45 𝛾𝑦
{ }=[ T ]{ } ; { } = 𝐾𝑠 [ 44 ]{ }
{𝑀} ℬ 𝓓 {𝜅} 𝑄𝑥 𝐴45 𝐴55 𝛾𝑥
where
𝐴11 𝐴12 𝐴16 𝐵11 𝐵12 𝐵16 𝐷11 𝐷12 𝐷16
𝒜 = [𝐴12 𝐴22 𝐴26 ] ; ℬ = [𝐵12 𝐵22 𝐵26 ] ; 𝒟 = [𝐷12 𝐷22 𝐷26 ]
𝐴16 𝐴26 𝐴66 𝐵16 𝐵26 𝐵66 𝐷16 𝐷26 𝐷66
matrices can be derived from the 6x6 stiffness matrix of the generalized Hooke’s law. ℬ represents the
material relation between normal forces and bending, which has only zero values int the present case.
Nonzero stiffness values are determined by the calculation of average stresses, due to pure strain states
applied on the representative volume element of the material.
User’s Manual X6 Release1 e2 243
Rib type
Custom
Bottom rib Top rib eccentricity
Origin of the rib Ribs follow a grid of lines in the local x and
grid y directions. The origin of this grid can be
entered numerically or picked up from the
model. The ribs may be modelled with
logical or actual ribs.
Ribs in x / y Ribs running in x or y direction has the following geometric parameters h is rib height b is rib width, d
direction is distance between ribs, exc is the eccentricity (if custom eccentricity is selected).
The maximum value of the rib eccentricity = (plate thickness + h) / 2
The reference plane of the static model is the middle plane of the concrete plate.
Calculation The calculation method is the same what can be seen at Composite ribbed domain, since present con-
figurations can be considered as special cases of that. Consequently, all the properties of the method
written there applies for this domain type too.
As a general guideline we can say that the spacing of the ribs be way smaller than the dominant
in-plane size of the problem domain.
Another consequence of the homogenization is that stiffness peaks at the rib intersections will be
smoothed, which may lead to insufficient design. The limits of application regarding a ribbed plate as a
homogeneous one is collected in EN 1992-1-1:2010, chapter 5.3.1. The program automatically detects
is these suggestions are not met.
In case of violating geometric restriction, the user is informed with a message window.
Create The homogenization method is not applicable in nonlinear analysis. To get proper nonlinear results
actual ribs actual ribs must be created. This option allows automatic generation of rib elements according to the
parameters.
For the definition of basic parameters see 4.9.6.1 Defining a normal domain.
The trapezoidal plate must be steel, the infill must be concrete.
Infill If this field is unchecked the domain is just a trapezoidal steel plate. If checked it is a composite plate
where a steel deck is filled with concrete.
Origin of the grid Ribs of the trapezoidal plate can run in the local x or y direction of the domain depending on the status
of the Rib direction switch. Enter X, Y, Z global coordinates to define the origin of the rib system or click
the Pick up button to define it by clicking.
244
Parameters The figure above shows the main geometrical parameters of the composite slab. Rolled mechanical
embossments and end anchorages typically ensure the shear connection between the trapezoidal
sheet and concrete. The parameters η and p are the degree/effectiveness of shear connection and the
width of rolled mechanical embossments, respectively. The software ignores the rolled mechanical
embossments in the course of calculation of the bending resistance for trapezoidal sheet.
If only rolled embossments are applied for shear connection, the degree of shear connection is 0 at the
supports and it increases with the distance from the supports. In such case, it is recommended to split
the domain parallel to the supports and to define different discrete η values for subdomains. The exact
values can be calculated from the strength of shear connection (τu,Rd) that depends significantly on the
geometry of embossments thus, it may vary for different trapezoidal sheets. If τu,Rd is known, the degree
of shear connection can be calculated in the following way:
𝜏𝑢,𝑅𝑑 𝐿
𝜂= ≤ 1,
𝑁𝑐,𝑓
where 𝐿 is the distance from the supports.
If only end anchorages are applied, the
degree of shear connection can be cal- M
culated in the following way for the en- Nc= Nc,f
Mpl,Rd
tire domain:
𝑉𝑙𝑑
𝜂= ≤ 1,
𝑁𝑐,𝑓 Nc= Nc,r
Bending failure
where 𝑉𝑙𝑑 is the resistance of end an-
chorages. Nc
Nc=0 L
If both end anchorages and rolled em-
bossments are applied, the degree of Longitudinal shear failure
shear connection can be calculated as
the sum of η values shown above. L
If no infill is applied the reference plane of the static model is the middle plane of the steel plate
(t/2) otherwise the middle plane of the total composite slab (h/2).
Analysis The calculation method is the same what can be seen at Composite ribbed domain, since present config-
urations can be considered as special cases of that. Consequently, all the properties of the method writ-
ten there applies for this domain type too.
User’s Manual X6 Release1 e2 245
Entering a custom stiffness matrix provides full control over the relations between internal force
components and strains. This solution is suggested in every case when other modelling solutions are not
possible and the user is capable of generating the necessary values, or is in posession of suggested values
from any source. The common term for this matrix is ABD matrix. ABD matrices must be positive definite,
this criterion is checked by the program.
Indices of matrix elements follow the Voigt notation with the slight modification of interchanging the
incices of 𝜏𝑥𝑧 and 𝜏𝑦𝑧 . Therefore
𝜎1 = 𝜎11 , 𝜎2 = 𝜎22 , 𝜎3 = 𝜎33 , 𝜎4 = 𝜎23 , 𝜎5 = 𝜎13 , 𝜎6 = 𝜎12
and
𝜀1 = 𝜀11 , 𝜀2 = 𝜀22 , 𝜀3 = 𝜀33 , 𝜀4 = 𝜀23 , 𝜀5 = 𝜀13 , 𝜀6 = 𝜀12 .
Hence the mapping from strains to stresses is given with the general form of Hooke’s law for anysotropic
materials:
𝜎1 𝐶11 𝐶12 𝐶13 𝐶14 𝐶15 𝐶16 𝜀1
𝜎2 𝐶21 𝐶22 𝐶23 𝐶24 𝐶25 𝐶26 𝜀2
𝜎3 𝐶31 𝐶32 𝐶33 𝐶34 𝐶35 𝐶36 𝜀3
𝜎4 = 𝐶41 𝐶42 𝐶43 𝐶44 𝐶45 𝐶46 𝜀4
𝜎5 𝐶51 𝐶52 𝐶53 𝐶54 𝐶55 𝐶56 𝜀5
{𝜎6 } [𝐶 𝐶62 𝐶63 𝐶64 𝐶65 𝐶66 ] {𝜀6 }
61
where a pair of indices refer to the element of the material stiffness matrix which is related to the given
stiffness component.
The meaning of blocks can be described as follows
𝐴𝑖𝑗 – membrane stiffness (normal forces – relative strain)
𝐷𝑖𝑗 – plate stiffness (moments – bending)
𝑆𝑖𝑗 – adjusted shear stiffness (shear forces – relative shear strain)
𝐵𝑖𝑗 – connecting stiffness (normal forces – bending or moments – relative strain)
The basic properties of the domain can be entered as for normal domains. Those parameters will affect
only the visual representation as the physical properties are controlled by the stiffness matrix.
Corrected shear stiffness components must include the shear correction factors as described in the
chapter about XLAM domains. In a general case these may be other than 5/6, used for
homogeneous sections and may be different in local directions. Determining shear corretion factors
is the responsibility of the user.
The meaning of the different stiffness parameters is explained in a dedicated theory guide.
Pick up
stiffness matrix
Service Class This is a classification based on the moisture content of the material and the relative humidity. For details
see Service class in 6.7.1 Timber beam design – TD1 module
User’s Manual X6 Release1 e2 247
Top layer grain Grain direction of the topmost layer must be specified it can be the local x or y direction.
direction
ksys When checked, the system strength factor is taken into consideration when calculating the strength
properties of the member. For further information see the “X-LAM (CLT) Theory and Design Guide”.
kfin When checked, the flexural strength is taken into account with reduced value. Check it on if the lamellas
are finger-jointed. For further information see the “X-LAM (CLT) Theory and Design Guide”.
Calculation Layered structures built from homogenous layers an equivalent orthotropic material stiffness matrix can
be calculated. This method converts the geometric inhomogeneity into material orthotropy
Material stiffness of a general shell element can be described with the following system of equations
𝐿
1
𝐵𝑖𝑗 = ∑(𝑄̅𝑖𝑗 )(𝑘) (𝑧𝑘+1 2 − 𝑧𝑘 2 ) = 0,
2
𝑘=1
𝐿
1 𝐿𝑎𝑚𝑖𝑛𝑎𝑡𝑖𝑜𝑛 𝑝𝑎𝑡𝑡𝑒𝑟𝑛. (Reddy, J. Mechanics
𝐷𝑖𝑗 = ∑(𝑄̅𝑖𝑗 )(𝑘) (𝑧𝑘+1 3 − 𝑧𝑘 3 ).
3 of Laminated Composite Plates and Shells.
𝑘=1
CRC Press, 2004.)
where 𝑄̅𝑖𝑗 is the Hooke-modell value transformed into the panel coordinate-system.
For further details on the stiffness calculation procedure, check the X-LAM (CLT) Theory and Design
Guide.
4.9.7. Hole
Holes can be defined in domains. Holes have to be inside the domain and in the domain’s plane.
Domain contours can be changed, cut and a union of domains can be calculated.
Change domain 1. Click the Change domain contour icon on the toolbar.
contour 2. Select a domain to change. Domain countour will be selected.
3. Change selection to modify domain contour and click OK on the selection toolbar.
Before After
Domain properties (material, thickness, local system) will be retained but the existing mesh will
be removed.
If loaded areas are removed from the domain, loads will automatically be removed.
Before After
Before After
User’s Manual X6 Release1 e2 249
Line elements are defined and modified in a common dialog. After choosing the element type specific
truss / beam / rib element parameters can be set.
Line elements are handled as structural members and not as finite elements. Meshing a line element
divides a beam or a rib into finite elements. Existing line elements can be joined to form a single element
if the geometry and their properties allow it (Edit / Find structural members). Numbering, labeling, listing
functions will consider it to be a single structural member. Structural members can be broken apart by
Edit / Break apart structural members) See 3.2.14 Assemble structural members, 3.2.15 Break apart structural
members
Color Elements can have their own fill and outline color used in rendered display mode. Default colors are
taken from the material colors. If a color coding is applied the color of a line element is determined by
the color coding both in wireframe and rendered modes.
See 2.16.5 Color coding
4.9.9.1. Truss
Cross-section Editor
Truss elements can be used to model truss structures. Trusses are two
node, straight elements with a constant cross-section along the truss
length. A maximum of three translational degrees of freedom are
defined for each node of the elements. The elements are pin-ended
(spherical hinges). Only the Nx axial internal force is calculated. It is
constant along the element.
i denotes the truss end with the lower node index (first node). By default the element x axis goes from
the node (i), to the node (j). It can be changed by selecting the other orientation from Local x
orientation.
Define You must select the lines to which you want to assign the same material and cross-sectional properties
in order to define truss elements.
If elements of different type are selected element definition will be activated.
Defining materials Materials and cross-sections can be selected from built-in libraries or from a list of the materials/cross-
and cross-sections sections already defined.
Allows browsing of the material library to assign a material to the element. The material selected will
be added to the material table of the model.
250
Allows browsing of the cross-section library to assign a cross-section to the element. For details see
4.9.2.1 Browse cross-section libraries. Only one cross-section can be selected from the library or only
one parametric shape can be created.
Launches the Cross-section Editor. The cross-section created in the Editor will be registered in the list
of model cross-sections.
i → j : local x axis is directed from the end node with a lower number to the node with the higher one
j → i : local x axis is directed from the end node with a higher number to the node with the lower one
Setting this parameter to automatic means that the program determines this orientation based on the
end point coordinates.
The orientation can be reversed any time using the shortcut [Ctrl+E] or in the dialog or in the property
editor window.
Cross-section In the calculation of the element stiffness, only the cross-sectional area Ax is considered from the cross-
sectional properties.
Local z Reference A reference point can be assigned to define the element orientation.
This allows a correct display of the cross-section on the screen. In case of selecting Auto the reference(s)
will be set by the program. Affects only the display of references.
See 4.9.22 References
Reference angle Rotation of cross-sections is made easy by the reference angle. The automatic local coordinate system
(and the cross-section) can be rotated around the element axis by a custom angle. If the element is
parallel to the global Z direction, the angle is relative to the global X axis. In any other case the angle is
relative to the global Z axis.
Nonlinear In a nonlinear analysis you can specify that a truss has stiffness only if it is in tension or compression.
parameters You can optionally enter a resistance value as well. A nonlinear elastic behavior is assumed for the
nonlinear truss elements.
The nonlinear parameters are taken into account only in a nonlinear analysis.
The initial elastic stiffness of a truss element is taken into account if a linear static, vibration, or
buckling analysis is performed, disregarding any nonlinear parameter entered.
4.9.9.2. Beam
Material, cross- Defining material, cross-section and local direction X are similar to truss elements.
section, local x If Variable cross-section is checked the start and end cross-sections can differ. To calculate proper
orientation intermediate cross-sections certain criteria must be met.
1. Both cross-sections must be thick-walled or thin-walled.
2. Thick-walled sections must have the same number of openings and the respective polygons must have
the same number of points.
3. Thin-walled sections must be of the same type.
Automatic The reference vector will be generated by the program according to the section 4.9.22 References.
reference
The orientation of the local x axis of the element can be reversed or can be set to Auto which means that
local x directions will be set automatically based on the beam end coordinates.
Reference angle Rotation of cross-sections is made easy by the reference angle. The automatic local coordinate system
(and the cross-section) can be rotated around the element axis by a custom angle. If the element is
parallel to the global Z direction, the angle is relative to the global X axis. In any other case the angle is
relative to the global Z axis.
End releases Click on the Setup… button to set end releases of the selected elements. End releases remove the
connection between the selected elements’ degrees of freedom (in the local coordinate system) and the
nodes. The end-releases are defined by six or seven (for 7DOF beams, see 4.9.9.3 7 DOF beam element –
7 DOF module) codes for each end: ex, ey, ez, θx, θy, θz, ew. Each code corresponds to one local internal
force component, ew represents warping. By default the beam ends are considered rigidly connected (all
codes are of rigid connection) to the nodes. Setting a code as hinged connection will result in the
corresponding internal force component of the respective end to be released. A semi-rigid connection
code can be assigned to the in-plane rotation components of the beam ends.
End release types are displayed in the table of Beams or Ribs, but separate tables are also created in the
Table Browser for Beam end releases and Rib end releases with all details.
252
At the top a list of predefined typical configurations is displayed in two dropdown lists, one for the start
point and one for the endpoint but end releases can be customized further.
Graphical symbol of a rigid connection code (the corresponding local displacement or warping
component of the beam end is transferred to the node)
Graphical symbol of a hinged connection code (the corresponding local displacement or warping
component of the beam end is not transferred to the node)
Graphical symbol of a semi-rigid connection code (the corresponding local displacement or warping
component of the beam end is partially transferred to the node). The semi-rigid connection requires
spring characteristic, as in case of springs and nodal support elements.
The drop-down menu contains just the applicable spring characteristics. New characteristic can be
defined by clicking on the spring icon, see 3.1.17 Spring characteristics library. Characteristics can be any
monotonic increasing function.
Graphical symbol of a steel plastic connection: the maximum value of the moment at the end points is
calculated from the material and cross-section properties. Available only for steel elements and only for
the ex, θy, θz components.
Graphical symbol of a customizable pushover hinge: the corresponding moment-rotation relationship is
defined by the user. Click the double arrow to select, load or edit the moment-rotation curve. Available
only for the θy, θz components.
User’s Manual X6 Release1 e2 253
The table below demonstrates the use of end releases for some typical configurations:
Care must be taken not to release an element or group of elements such that rigid body
translations or rotations are introduced.
For example, if you specify spherical hinges at both ends (code: 000111), a rigid body rotation about
element axis is introduced. In this case at one of the ends you may not release the element degree of
freedom corresponding to the rotation about local x axis (e.g., i end numerical code: 000011; j end
numerical code: 000111).
Kr rotational stiffnesses:
4 ∙ 𝐸𝑥 ∙ 𝐼𝑦 4 ∙ 𝐸𝑥 ∙ 𝐼𝑧
𝑦𝑦: 𝐾𝑟 = 1000 ∙ , 𝑧𝑧: 𝐾𝑟 = 1000 ∙
ℓ ℓ
where Iy and Iz are the flexural inertias of the cross-section.
The program does not calculate a default value for y and z shear deformation components and for the
torsional component.
The stiffness of the joints has to be determined according to its local coordinate system. Generally, the
initial stiffness (or its certain quotient) of the nonlinear characteristics is given.
254
The following example demonstrates how to interpret the rotational stiffness of a steel joint:
Connection: Model: Moment - Relative Rotation Diagram
For example, in the case of steel frame structures, Eurocode EN 1993-1-8 gives the details of
application.
Resistance To semi-rigid connections a resistance can be assigned, that is the maximum force/moment that can
develop in the connection. The value can be defined in the window of spring characteristics.
In case of steel material, the default values for Fu, My and Mz resistance values are calculated
based on the cross-sectional characteristics and the material yield strength. In case of any other
material, no default resistance value is calculated by the program.
Steel plastic hinge To define steel plastic hinges choose Plastic from the dropdown list.
Plastic hinge can only be set for x directional displacement and y, z
rotational components. The calculated axial or moment resistance will
be displayed but cannot be edited.
If elements with different materials or cross-sections are selected no value will appear in the edit field
but hinges will be defined with the appropriate resistance.
After completing the nonlinear analysis and displaying beam internal force diagrams hinges that got
into plastic state at the current load step become red. The number beside the hinge shows the order of
getting into a plastic state. Hinge with number 1 is the hinge getting plastic first. Where hinges are not
red, plastic limit moment is not reached yet.
Steel plastic hinges can only be used with steel beams but are considered only in nonlinear
static/dynamic analysis.
Pushover hinge To define a pushover hinge choose Pushover from the dropdown
list then select a function from the dropdown list, load a previously
saved function from the library or edit the moment-rotation
diagram by clicking on the Function editor button. To modify a
pushover hinge click the double arrow to show the buttons and the
second dropdown list.
A total of five points can be defined for both directions of the moment-rotation diagram. This allows
modeling of complex connection behaviour including the possibility of hardening, softening and
strength degradation. Behaviour after the last point is extrapolated based on points D and E. The
diagram is defined by specifying the corresponding moment and rotation coordinates in the table on
the left side of the window. The created digaram is symmetric by default, but this can be overriden by
clicking on the Symmetrical function button. The created diagrams can be saved and used for other
elements in the model.
To facilitate numerical analysis and prevent convergence difficulties, it is recommended to avoid sudden
drops of capacity and perfectly plastic sections in the diagram. Instead, relax the diagram, by making
sure that there is at least a small difference in both coordinates of consecutive points. This does not
influence the results, but improves numerical stability significantly.
User’s Manual X6 Release1 e2 255
After completing the nonlinear analysis and displaying beam internal force diagrams hinges that have
got into plastic state by the current load step become red. The number beside the hinge shows the
order of getting into a plastic state. Hinge with number 1 is the hinge getting plastic first. Where hinges
are not red, plastic limit moment has not been reached yet.
If any beam end release code is of a hinged connection, the beam end is displayed on the screen as a
blue circle. If it has a stiffness value a blue cross is inscribed. If the end release corresponds to a spherical
hinge, it is displayed as a red circle.
The plastic hinges are displayed as solid circles.
The defined beams appear as dark blue lines.
Beam or independent rib elements can have eccentricity in the local y and z directions relative to the
logical axis connecting the end nodes. Eccentricities assigned to the endpoints can differ, in this case the
eccentricity of the intermediate cross-sections changes linearly between the two values.
Eccentricity can be edited in two ways: either from the beam dialog or directly from the toolbar on the
Elements tab.
The dialog has two tabs. Eccentricity is to define an ey and ez eccentricity, while Connection is to fine-
tune the properties of the eccentric connection.
256
Eccentricity
Methods of The appropriate method can be selected from the dropdown list.
definition
• No eccentricity
• Uniform offset
• Aligned to eccentricity group
• Aligned to another structural member
• Different eccentricity at endpoints
Aligned to The selected structural members form an eccentricity group just like domains. The ey and ez eccentricity
eccentricity group of the element smallest in the respective direction will be set to zero, while other elements will shift so
that their selected point on the bounding rectangle remains aligned.
Aligned to another It is a common situation when a beam rest on another one. Choose Aligned to another structural member
structural member to set the appropriate eccentricity.
User’s Manual X6 Release1 e2 257
Set the point of alignment at the start point and endpoint of the structural member at the top then set
the point of alignment for the reference elements connected to the respective points.
By clicking on the Select reference elements button the selection is retained but trusses, beams or ribs
that can be used as reference elements can also be selected. A reference element must have a common
node with the elements to modify. The reference element may also be eccentric, but it must not be
aligned to another structural member.
If the selected reference elements meet the above criteria the eccentricity of the element is adjusted
matching the specified points of the two bounding rectangles.
Changing the eccentricity or cross-section of a reference element immediately updates the eccentricity
of the elements aligned to it.
Different Allows to define a different eccentricity at the endpoints either by setting an Offset or making the end-
eccentricity point Aligned to another structural member.
at endpoints For details see the description of these methods above.
Eccentricity of ribs In case of ribs connected to surface elements the last option on the toolbar is Custom value, where lo-
connected to cal y and z eccentricities can be set independently at the endpoints. In other cases, ey = 0. See 4.9.9.4.
surface elements Rib.
258
Connection An eccentric connection can be rigid or hinged according to each displacement or rotation component.
Signs: ■ = rigid connection, ○ = free displacement or rotation.
If the connection is not rigid, a link element is automatically generated between the end node and the
eccentric endpoint.
The position of the link element can be defined By length or By ratio. If the element is aligned to a
reference element a third option is also available: Calculate the position. Knowing the cross-section of
the two structural members the program calculates the position of the link element.
Pick up Beam and rib eccentricites can be copied from another beam or rib if that eccentricity is not
determined from alignment to a reference element.
End releases End releases can be defined for the warping degree of freedom (ew).
Fixed: Warping degree of freedom at the beam end is the same as that of the node.
The value of WF depends on the joint type. The most common types are:
A diagonal joint provides complete and direct warping transmission, a box joint provides a complete and
inverse warping transmission, while a box-diagonal joint does not provide transmission.
The respective WF coefficients are
For other joint types WF can be determined by building a shell model of the joint.
4.9.9.4. Rib
Automatic The reference vector will be generated by the program according to the section 4.9.22 References
reference
Reference Independent rib: Reference point
The local coordinate system is defined as follows: the
element axis defines the x local axis; the local
z axis is defined by the reference point or vector; the
y local axis is according to the right-hand rule.
Reference point
Reference angle The automatic local coordinate system (and the cross-section) can be rotated around the element axis
by a custom angle. If the element is parallel to the global Z direction, the angle is relative to the global
X axis. In any other case the angle is relative to the global Z axis.
End releases End releases can be defined for ribs the same way as for beams. By default both ends are fixed.
End release types are displayed in the table of Ribs, but a separate table is also created in the Table
Browser for Rib end releases with all details.
Eccentricity of ribs Ribs are used mostly connected to surface elements (domains) but they can be defined as independent
line elements as well. In the latter case their eccentricity can be defined the same way as for beams (see
4.9.9.2.1 Eccentricity of beams and ribs). If ribs are connected to surface elements (domains) their eccen-
tricity can be set as follows.
If more than two surfaces are connected to the edge also select one or two of them to define eccentricity
for the rib.
User’s Manual X6 Release1 e2 261
The eccentricity of a rib is given by the distance of the center of gravity of its cross-section to the plane
of the model of the surface (neutral plane). It is positive if the center of gravity is on the positive direction
of its local z axis.
There are standard options to set the rib eccentricity.
Rib types in case of
concrete slab and
concrete rib
Bottom rib Bottom rib Rib in midplane Top rib Top rib Custom
(embedded) (embedded) eccentricity
In the first three cases the actual eccentricity is calculated from the rib cross-section and the plate thick-
ness. If the rib is made of concrete the definition of top and bottom ribs is different, so button pictures
change according to the rib material. If rib cross-section or plate thickness changes the eccentricity is
automatically recalculated.
If the rib is made of steel or timber, connected to a shell and is defined as a top or bottom rib, an
additional axial connection stiffness can be defined.
In case of reinforced concrete plate-rib connections, rib cross-section must include the plate
thickness. In other cases (steel or timber structures), the cross-section is attached to the top or
bottom plane of the plate.
For plates, the eccentricity of the rib will modify the flexural inertia of the rib as follows:
𝑰∗𝒚 = 𝑰𝒚 + 𝑨 ⋅ 𝒆𝒙𝒄𝟐
For shells, due to the eccentric connection of the rib to the shell, axial forces will appear in the rib
and shell.
Shear connection In case of top and bottom ribs (except for concrete structure), the shear connection between the slab
and rib can be set, which can be Rigid or Elastic. The elastic stiffness – kx component - has to be set by
the user, it is not calculated by the program (the default value is 1E+6 kN/m/m).
Using Elastic setting, the specific longitudinal shear connection force (Vxz component) is also calculated
(this is not calculated with rigid settings).
262
Stifffness reduction Using Eurocode standard, stiffness reduction can be set for certain architectural elements (such as
columns, beams, walls, slabs and other type of elements) for modal response spectrum analyis. For
details see chapter 3.3.10 Stiffness reduction. In case of concrete ribs two kind of reduction factor can
be set: kA and kI.
Factor kA lowers the cross-sectional area of the concrete. It has an effect only on the elongation stiffness
and has no effect on the shear stiffness. Factor kI lowers only the flexural stiffness of the cross-section.
Pick up>> Properties of another element can be picked up and assigned to the selected elements. Clicking the
Pick up button closes the dialog. Clicking an element picks up the value and shows the dialog again.
Only those properties will be copied where the checkbox is checked.
Popup menu If the Element tab is selected click the right mouse button on line
elements to find further useful tools operating on selected line
elements. If no element has been selected these functions will
change the element you clicked on.
Reverse local Sets the local x direction of the selected line element in the reverse direction. Start and end cross-
coordinate system sections of tapered beams/ribs and their end releases will remain in their original position only their
interpretation will be different.
Swap start and end Swaps the start and end cross-sections of the selected variable cross-section beams/ribs.
cross-section The local x direction remains unchanged.
Swap end releases Swaps the end releases of the selected beams/ribs.
of start and The local x direction remains unchanged.
end point
Surface elements can be used to model membranes (membrane element), thin and thick plates (plate
element) and shells (shell element) assuming that the displacements are small.
As surface elements you can use a six node triangular or eight/nine node quadrilateral finite elements,
formulated in an isoparametric approach. The surface elements are flat and have constant thickness
within the elements.
Introduction of domains and automatic meshing rendered the definition of surface elements
obsolete.
It is preferable for the element thickness not to exceed one tenth of the smallest characteristic
size of the modeled structural element, and the deflection (w) of a plate or shell structural
element is less than 20% of its thickness (displacements are small compared to the plate
thickness).
Use of elements with the ratio of the longest to shortest element side lengths larger than 5, or with the
ratio of the longest structural element side length to the thickness larger than 100 are not
recommended.
In some cases when the elements are used (that are flat with straight edges) to approximate curved
surfaces or boundaries, poor results may be obtained.
4.9.10.1. Membrane
Membrane elements may be used to model flat structures whose behavior is dominated by in-plane
membrane effects. Membrane elements incorporate in-plane (membrane) behavior only (they include
no bending behavior).
Automatic reference:
The axis of element local directions x and z can be determined by reference elements,
see part 4.9.22 References or can be set automatically.
4.9.10.2. Plate
Plate elements may be used to model flat structures whose behavior is dominated by flexural effects.
AxisVM uses an eight/nine node Heterosis finite element as plate element, that is based on Mindlin-
Reissner plate theory that allows for transverse shear deformation effects). This element is suitable for
modeling thin and thick plates as well.
Plate elements incorporate flexural (plate) behavior only (they include no in-plane behavior).
The plate internal forces are: 𝑚𝑥 , 𝑚𝑦 , 𝑚𝑥𝑦 moments, and 𝑣𝑥 , 𝑣𝑦 shear forces (normal to the plane of the
element). In addition, the principal internal forces: 𝑚1 , 𝑚2 , the angle 𝛼𝑚 and the resultant shear force
𝑞𝑅 are calculated.
The variation of internal forces within an element can be regarded as linear.
The following parameters should be specified:
1. Material
2. Thickness
3. Reference (point/vector/axis/plane) for local x axis
4. Reference (point/vector) for local z axis
Allows browsing of the material library to assign a material to the element. The material selected will
be added to the material table of the model.
Automatic reference:
The axis of element local directions x and z can be determined by reference elements, see part 4.9.22
References or can be set automatically.
4.9.10.3. Shell
Shell elements may be used to model structures with behavior that is dependent upon both in-plane
(membrane) and flexural (plate) effects.
The shell element consists of a superimposed membrane and plate element. The element is flat, so the
membrane and plate effects are independent (first order analysis).
The element can be loaded in its plane and perpendicular to its plane.
The shell internal forces are: 𝑛𝑥 , 𝑛𝑦 , 𝑛𝑥𝑦 forces (membrane components), 𝑚𝑥 , 𝑚𝑦 , 𝑚𝑥𝑦 moments, and
𝑣𝑥 , 𝑣𝑦 shear forces (plate components). In addition, the principal internal forces and moments
𝑛1 , 𝑛2 , 𝑚1 , 𝑚2 , the angles 𝛼𝑛 , 𝛼𝑚 and the resultant shear force 𝑣𝑆𝑧 are calculated.
The variation of internal forces within an element can be regarded as linear.
Allows browsing of the material library to assign a material to the element. The material selected will
be added to the material table of the model.
Automatic reference:
The axis of element local directions x and z can be determined by reference elements, see part 4.9.22
References or can be set automatically.
Modifying Selecting elements of the same type Modify will be activated. Checked properties can be changed or
picked up from another element. Selecting elements of different types Define will be activated.
Pick up>> See Pick up at 4.9.9 Line elements
266
The rotational DOF about the axis normal to the surface is a special
rigid elements
DOF. It is the 3rd DOF of membrane elements and the 6th DOF of
shells. It is not coupled with the planar translations. Only the drilling
DOF of the nodes are connected with a small stiffness. In case of
shells the main goal is to avoid singularities at 3D connections (L or
T shape connection of flange and web or modeling curved shells).
In case of membranes the main purpose is to obtain the same
results like in-plane loaded shells see References [2].)
Since the drilling DOF has only a small stiffness a directly loaded
node of a membrane or shell element works as if it was released
about the axis normal to the surface. Using rigid elements the
moment resistance of this node can be increased.
Nodal support elements may be used to model the point support conditions of a structure. Nodal
support elements elastically support nodes, while the internal forces are the support reactions. Midside
nodes of surface edges cannot be supported. References are used to arbitrarily orient the x and z axes
of the element.The x axis is directed from a reference point to the attachment node (the node to which
it is attached).
You can specify the translational and/or rotational (torsional) properties about the element axes by
spring characteristic. The values of initial and vibration stiffness and the damping coefficient are also
displayed.
Components with a linear spring characteristics have editable fields to enter stiffness and damping
directly. Closing the dialog will create a new spring characteristics with the new values automatically.
Nonlinear parameters can be assigned to each direction. To change the characteristics click the spring
characteristic icon (see 3.1.17 Spring characteristics library). The values of damping coefficient are visible
if at least one of them is not zero.
You can define only one global support for a node. You cannot define nodal support for a midside
node of a surface element.
Local Minimum two applicable reference is necessary to define local nodal
support element. This kind of nodal support is actually a rotated
version of the global one. The firs reference is the direction of local
x-axis and the second reference together with the first one
deterimines the plane of the local z-axis. Beside the references the
corresponding stiffness (spring characteristic) must be specified.
Reference Defines nodal support elements in the direction of a reference (point or vector). You must select the
nodes that are identically supported, and specify the corresponding stiffness (translational 𝑲𝑥 , and
rotational 𝐾𝑥𝑥 ).
The direction of the reference vector is
defined by the element node and its
reference point or reference vector in the
following way:
Beam/rib relative Defines nodal support elements about local coordinate axes of
beam / rib elements. You must select the beam / rib elements
and the nodes that are identically supported, and specify the
corresponding translational (𝐾𝑋 , 𝐾𝑌 , 𝐾𝑍 ) and rotational
(𝐾𝑋𝑋 , 𝐾𝑌𝑌 , 𝐾𝑍𝑍 ) stifnesses.
Edge relative Defines nodal support elements about local coordinate axes of
surface element edges. You must select the surface elements
and the nodes that are identically supported, and specify the
corresponding translational (𝐾𝑋 , 𝐾𝑌 , 𝐾𝑍 ) and rotational
(𝐾𝑋𝑋 , 𝐾𝑌𝑌 , 𝐾𝑍𝑍 ) stifnesses.
If one surface is connected to the edge the local coordinate axes of the edge are:
x = the axis of the edge
y = the axis is oriented toward inside of the surface element in its plane
z = parallel to the z local axis of the surface element
If two surfaces are connected to the edge the local z-axis direction is bisecting the angle of surfaces.
The y-axis is determined according to the right hand-rule.
If more than two surfaces are connected to the edge and you select one or two of them then support
local system will be determined based on the selected surfaces.
Positive displacement elongates the support spring that causes tensile force. Negative displacement
shorten the spring that causes compressive force.
268
Nonlinear characteristics is taken into account only in static or dynamic analysis with material
nonlinearity. Linear static, linear dynamic and buckling analyses take into account the Initial
stiffness only, while vibration analysis takes into account the Vibration stiffness only.
The damping coefficients are considered only in dynamic analysis.
Nodal supports appear as brown (𝐾𝑋 , 𝐾𝑌 , 𝐾𝑍 ) and orange (𝐾𝑋𝑋 , 𝐾𝑌𝑌 , 𝐾𝑍𝑍 ) pegs in 3 orthogonal direction.
Support stiffness
calculation Load from material li-
brary
Use the Calculation... button on the support form to calculate the support stiffness (including the
rotational stiffness) due to a column type support. The support stiffnesses are determined based on the
end releases, material, and geometry of the column. Using the parameters, spring characteristics are
generated by the program.
Calculating nodal support stiffness, a column below and a column above the node can be specified
separately. These column parameters can also be used in punching analysis (especially in the case of
intermediate slabs). The columns and walls modeling the supports also appear in rendered view and
the cursor can identify them.
Modifying Selecting elements of the same type Modifying will be activated. Checked properties can be changed
or picked up from another element. Selecting elements of different types Definiton will be activated.
Seismic isolator
Line support elements may be used to model the line support conditions of a structure. Line support
elements are elastically supporting beams, ribs, or surface (domain) edges. The internal forces are the
support reactions.
Select domain edges, beams or ribs then enter the translational and rotational stiffness values. Nonlinear
parameters can be assigned to each direction. To change the characteristics click one the three buttons
(bidirectional, compression only, tension only) and set the resistance checkbox and specify a value if
necessary.
The support can be defined in the following systems: Global, Beam/rib relative, Edge relative, Referential
(taking the direction from a reference point or reference vector).
Two types of supports are available: Winkler or Winkler-Pasternak.
Winkler type These support elements represent a Winkler type elastic foundation. Select domain edges, beams or ribs
supports then enter the translational and rotational stiffness values.
270
Construction
Pasternak layer
Winkler foundation
Both line element or domain edge supports are considered to be structural elements, i.e., when a
mesh is created these elements will be automatically divided internally but tables and labels always
refer to the structural elements.
Global Defines line support elements parallel to global coordinate axes. You must specify the corresponding
translational 𝐾𝑋 , 𝐾𝑌 , 𝐾𝑍 and rotational 𝐾𝑋𝑋 , 𝐾𝑌𝑌 , 𝐾𝑍𝑍 stiffnesses.
Beam/Rib relative Defines line support elements for beam/rib elements in their local coordinate system acting as an elastic
foundation. You must specify the corresponding translational 𝐾𝑥 , 𝐾𝑦 , 𝐾𝑧 and rotational 𝐾𝑥𝑥 , 𝐾𝑦𝑦 , 𝐾𝑧𝑧
stifnesses.
The beams/ribs with line supports must be divided into at least four elements.
In addition, the following condition must be satisfied:
1 4 4𝐸𝑥 𝐼𝑧 4 4𝐸𝑥 𝐼𝑦
𝐿 ≤ 𝑙𝑘 = min ( √ ,√ )
2 𝑘𝑦 𝑘𝑧
where L is the beam / rib length.
AxisVM warns you if the condition is not satisfied (by one or more elements). In this case the
Winkler’s modulus of the defined elements are set to zero, therefore you can divide the elements
and repeat the definition/modification process.
If you specify line supports the internal forces are linearly interpolated between the ends of the
element, therefore the division of the elements is required.
Edge relative Defines edge support elements relative to local coordinate axes of the edges. You must specify the
corresponding stiffness (translational 𝐾𝑥 , 𝐾𝑦 , 𝐾𝑧 and rotational 𝐾𝑥𝑥 , 𝐾𝑦𝑦 , 𝐾𝑧𝑧 ).
If one surface is connected to the edge the local coordinate axes of the edge are:
Positive displacement stretches the support spring creating tensile force. Negative displacement
compresses the spring creating compressive force.
User’s Manual X6 Release1 e2 271
Nonlinear behavior Nonlinear force-displacement characteristics can be specified for this element as follows: compression
only (very small stiffness in tension), tension only (very small stiffness in compression). A resistance value
can also be entered.
The nonlinear parameters are taken into account only in a nonlinear analysis. In any other case in
the analysis (Linear static, Vibration I/II, Buckling) the initial stiffnesses are taken into account.
The nonlinear behaviour of Winkler-Pasternak supports in local z direction derives from the nonlinear
characteristic of the line-to-line link layer. In case of tension only supports the link layer stretches freely
letting the construction rise from the ground. The Rz reaction force will be zero but due to the Pasternak
layer the Winkler foundation will remain loaded.
Line supports appear as brown (𝐾𝑋 , 𝐾𝑌 , 𝐾𝑍 ) and orange (𝐾𝑋𝑋 , 𝐾𝑌𝑌 , 𝐾𝑍𝑍 ) lines.
Support stiffness Use the Calculation... button to calculate the global or edge-
calculation relative line support stiffness (including the rotational
stiffness) due to a wall type support. The support stiffnesses
are determined based on the end releases, material, and
geometry of the wall.
Surface supports provide support for surface elements and domains, with stiffness component
interpreted in the local coordinate system of the surface or domain.
Winkler Defines a surface support element (Winkler type elastic foundation) to surface elements. You must
specify a translational stiffness in the surface element local coordinate system. The surface support
behaves identically in tension and compression and is considered constant within the element.
You must specify the support stiffness 𝐾𝑥 , 𝐾𝑦 , 𝐾𝑧 (Winkler’s modulus) about the surface element local x,
y, and z axes.
Winkler-Pasternak These support elements represent a Winkler type elastic foundation against rotation and displacements
type supports in local x and y directions but modeling of the support against displacements in local z direction is more
sophisticated. A new Pasternak layer modeling the vertical shear stiffness of the soil is placed above the
Winkler foundation. It has a constant RG shear resistance. Above the Pasternak layer a line-to-line link
layer is generated with a stiffness of 1000 · Kz. In linear analysis the only function of this layer is to
transfer the Rz support reaction between the soil and the structure.
Construction
Pasternak layer
Winkler foundation
Positive displacement stretches the support spring creating tensile force. Negative displacement
compresses the spring creating compressive force.
Nonlinear behavior Nonlinear force-displacement characteristics can be specified for this element as follows: compression
only (very small stiffness in tension), tension only (very small stiffness in compression), or with resistance
(the same stiffness for compression and tension).
The nonlinear parameters are taken into account only in a nonlinear analysis. In any other case
in the analysis (Linear static, Vibration I/II, Buckling) the initial stiffnesses are taken into account.
The nonlinear behaviour of Winkler-Pasternak supports in local z direction derives from the nonlinear
characteristic of the line-to-line link layer. In case of tension only supports the link layer stretches freely
letting the construction rise from the ground. The Rz reaction force will be zero but due to the Pasternak
layer the Winkler foundation will remain loaded.
Edge hinge can be defined between domain edges or between a rib and a domain edge. Select edge
and a domain. Hinge stifness can be defined in the local system of the edge of the selected domain.
Rigid elements may be used to model parts with a rigid behavior relative to other parts of the structure.
The elements can be defined by selecting the lines that connect its nodes. The selected lines that have
common nodes define the same rigid element. There is no limit to the number of nodes of any element.
The degrees of freedom of the nodes of a rigid element cannot be constrained (fixed).
User’s Manual X6 Release1 e2 273
Define Lets you define rigid elements. You must select the lines that connect the nodes attached to rigid
elements. Recall that the lines with common nodes define the same rigid element.
You can join or split rigid elements using the modify command.
If you select lines that connect nodes of different rigid elements, the elements will be joined. If you
deselect lines of rigid elements interrupting their continuity, the respective elements will be split.
A finite element cannot have all of its lines assigned to the same rigid body.
If we want to calculate the mass of the body in a vibration analysis, place a node to the center of
gravity, connect it to the body and make this line a part of the rigid body. Assign the mass of the
body to this node.
The rigid elements are displayed on the screen with thick black lines.
4.9.16. Diaphragm
Using diaphragms means simplifying the model. Diaphragms are special rigid bodies where the relative
position of the element nodes remain constant in a global plane. Diaphragms considerably reduce the
amount of calculation. It can be an advantage running vibration analysis of big models. Diaphragms can
represent plates totally rigid in their planes.
Definition Select lines to define diaphragms. Each set of connecting lines will form a diaphragm.
4.9.17. Spring
The spring element connects the degrees of freedoms of two nodes of the model. The element has its
own coordinate system. You can specify the translational (X, Y, Z) and/or rotational (XX, YY, ZZ) spring
characteristics. The values of initial and vibration stiffness and the damping coefficient are also displayed.
Nonlinear parameters can be assigned to each direction. To change the characteristics click the spring
characteristic icon (see 3.1.17 Spring characteristics library)
Springs can be defined in the following systems: Global, By geometry, By reference, Element relative,
Node relative, or as a Seismic isolator.
Node relative Spring element with node relative is determined by two applicable
nodes in addition to the selected line. The local x-reference node
with the lower number node of the spring line specifies the local
x-axis of the spring. The local z-reference node with the local x-
direction determines the plane that contains the local z-axis.
The model requires at least two applicable nodes to be specified.
Spring character- Spring characteristic can be imported from Spring characteristics library see 3.1.17 Spring characteristics
istics library library. The program automatically filters out the spring properties corresponding to the degree of
freedom.
Spring character- The components can be edited with the spring characteristic icon.
istic editor See 3.1.17 Spring characteristics library.
Components with a linear spring characteristics have editable fields to enter stiffness and damping
directly. Closing the dialog will create a new spring characteristics with the new values automatically.
The values of damping coefficient are visible if at least one of them is not zero.
Nonlinear spring characteristics are taken into account only in case of materially nonlinear static
and materially nonlinear dynamic analyses.
Initial stiffnesses are taken into account in materially linear static, materially linear dynamic and
buckling analyses even if the spring properties are nonlinear. In case of Vibration I/II the vibration
stiffnesses are taken into account, that values remain constant during the analyses.
The damping coefficients are considered only in dynamic analysis.
The spring elements are not real structural elements. They connect the nodal DOFs of the end
points one by one. There is no geometric coupling between the different DOFs (translation-
rotation, force-moment).
Seismic isolator
4.9.18. Gap
As the inactive state has a small but nonzero stiffness a gap element is an approximate model of this
type of connection. But using this element the stiffness matrix remains a band matrix and any number of
gaps can be used in a model.
The gap element is a nonlinear element that can impose difficulties to the solution of the nonlinear
problem, due to large changes of element stiffness when it changes status (active/inactive).
If numerical convergence problem occurs, it may be desirable to reduce the ratio of active and inactive
stiffness, thereby mitigating the nonlinearity induced by the contact element. This has to be checked by
the user, that it is possible or not.
If the gap element is used in an analysis different from a nonlinear static analysis, the element will
be taken into account as a spring with a stiffness corresponding to its initial opening. If the initial
opening is zero, the active stiffness will be taken into account.
During different analyses the seismic isolators are taken into account in the following way.
The seismic isolator system is considered as a single degree of freedom model, with the following
parameters:
Seismic loads Tis the natural period of the system
SE1 module ξesi damping ratio of the system
Kesi effective horizontal stiffness
Kv effective vertical stiffness
User’s Manual X6 Release1 e2 277
dcx, dcy the coordinates of the centre of the isolator system on the X-Y plane
The effective hori- In order to determine the effective horizontal stiffness of the nonlinear isolators one must know the
zontal stiffness and ultimate limit displacements of the isolators (D2, see 4.9.11 Nodal support, 4.9.17 Spring).
damping ratio As displacements of the isolators depend on the seismic load the solution can only be provided iteratively.
At the end of the procedure the relative difference between the design displacement and the actual
maximum displacement must be less than 5%.
The effective stiffnesses and damping ratios can be determined in the following way for different type of
isolators:
Parameters: 1 µ
𝐾𝑒𝑓𝑓 = 𝑁 ( + )
𝑅 𝐷2
𝐷1 = 𝐹1 /𝐾1
𝐹2 = 𝐹1 + 𝐾𝑇 (𝐷2 − 𝐷1 ) 2 1
𝐾𝑒𝑓𝑓 = 𝐹2 /𝐷2 𝜉𝑒𝑓𝑓 = ( )
𝜋 1 + 𝐷2
2 𝐹1 𝐷1 µ𝑅
𝜉𝑒𝑓𝑓 = ( − )
𝜋 𝐹2 𝐷2
where N is the normal force on
the isolator element
The vertical and 𝐾𝑣 = ∑ 𝐸𝑧,𝑖 , 𝐾𝑒𝑠𝑖 = ∑ 𝐾𝑒𝑓𝑓,𝑖
horizontal stiffness
of the isolator
system
2𝜋
The natural period 𝑇𝑖𝑠 = where 𝑀𝑖𝑠 is the mass of the isolated structure
√𝐾𝑒𝑠𝑖 /𝑀𝑖𝑠
of the isolator
system
Damping ratio of ∑ 𝜉𝑒𝑓𝑓,𝑖 𝐾𝑒𝑓𝑓,𝑖
𝜉𝑒𝑠𝑖 =
the system 𝐾𝑒𝑠𝑖
The stiffness centre ∑ 𝑥𝑖 𝐾𝑒𝑓𝑓,𝑖 ∑ 𝑦𝑖 𝐾𝑒𝑓𝑓,𝑖
𝑑𝑐𝑥 = , 𝑑𝑐𝑦 =
𝐾𝑒𝑠𝑖 𝐾𝑒𝑠𝑖
of the isolator
system
During response spectrum analysis isolators are considered in the following way (available only if NTC,
Italian code is selected)
In this case the program takes into account the effective stiffness of the isolator (𝐾𝑒𝑓𝑓,𝑖 ), which is
introduced above.
Linear static
analysis
For nonlinear static analysis (pushover) the isolators are considered with a bilinear constitutive law.
Nonlinear static
analysis
For first and second order analysis the effective stiffness of the isolator (𝐾𝑒𝑓𝑓,𝑖 ) is taken into account, which
is introduced above. If the model contains isolator elements, first a linear static analysis is carried out to
Vibration
determine the mass of the isolated system.
In buckling analysis the elements are taken into account with initial stiffness (𝐾1 ). In the case of curved
surface slider the horizontal stiffness is set to 1e10 kN/m.
Buckling
Without following nonlinear The effective (𝐾𝑒𝑓𝑓,𝑖 ) stiffnesses of isolators are considered, which are
behaviour of materials and fi- introduced above. Furthermore, the program takes into account the
Dynamic nite elements dissipated energy due to the elastic-plastic constitutive behaviour in the
i. isolator element:
𝑊𝑑,𝑖
𝑐𝑖 =
𝜋𝜔𝐷2,𝑖 2
2𝜋
where 𝜔𝑖𝑠 = , 𝑊𝑑,𝑖 is the dissipated energy during one cycle, which
𝑇𝑖𝑠
is 𝑊𝑑 = 4(𝐹1 𝐷2 − 𝐹2 𝐷1 ) for rubber element and 𝑊𝑑 = 4µ𝑁𝐷2 for slider
element.
With following nonlinear be- Rubber bearing isolators and curved surface sliders are taken into
haviour of materials and finite account with a bilinear constitutive law.
elements
Note: if a curved surface slider is defined in the model, the analysis begins with a linear static analysis in
order to determine the normal force on the slider element for the selected load case. The program
neglects the oscillation of the pressure force on the isolator element.
The nonlinear behaviour of the isolators are represented by separate springs in x and y direction. No
interaction is considered.
User’s Manual X6 Release1 e2 279
4.9.20. Link
Link elements Link elements connect two nodes (N-N) or two lines (L-L) and have six stiffness components (defined in
their coordinate system) that are concetrated on an interface (located between the connected
nodes/lines). Its position can be entered relative to one node/line that is considered as reference.
Link elements can have a nonlinear parameter called limit resistance that limits the force they are able to
transfer.
Connects two nodes. The stiffness components are defined in the global coordinate system. The position
of the interface can vary from 0 to 1 relative to the master node (selected by the user). If the location of
the interface is = 0 the interface is at the master node. If it is = 1 the interface is at the opposite node.
For any value greater than 0 or lower than 1 the reference is between the nodes.
Typical applications are: main girder-purlin connection; some types of grillage connections; bracing con-
nections; etc.
Example: A main girder-purlin connection (see SteelFrame.axs in the examples folder)
Let assume that the vertical axis is Z being parallel to the local z axis. The main girder is an IPE-400 in X-
Z plane, the purlin is an I-200. You would like to transfer forces from the purlin to the main girder but
not the moments.
280
These elements are represented by their line of gravity. The link has to be placed between these two axes
at their point of intersection (if seen from above). Therefore, this link has to be assigned to a vertical line
having a length equal to the distance of axes i.e., 30 cm (40/2 + 20/2). Select the node on the main girder
to be the master node of the link. The inter-face always has to be placed at the actual point of contact.
In this case the interface is located 20 cm far (40/2) from the master node (i.e., the main girder axis). So,
the interface position is 20/30 = 0.666. You assume that the connection is fixed against displacements
but can rotate. Therefore, you enter 1E10 for translational stiffnesses and 0 for rotational ones. If the
purlins are supported only by these links you have to enter KYY=0.001 or a similar small value to eliminate
rotation around the main girder axis.
Nonlinear Nonlinear parameters can be assigned to each nonzero stiffness component. To change the character-
parameters istics click one the three buttons (bidirectional, compression only, tension only) and set the resistance
checkbox and specify a value if necessary.
Line-to-Line Link
The position of the interface can vary from 0 to 1 relative to the master line (selected by the user).
If the location of the interface is 0, the interface is at the master line (at the start point of the arrow).
If it is 1 the interface is at opposite line (at the end point of the arrow). For any value greater than 0 or
lower than 1 the interface is between the lines.
Typical applications are: floor-wall hinged connections; semi-composite / full-composite layered beams;
Semi-rigid rib-shell connections; etc.
User’s Manual X6 Release1 e2 281
Nonlinear A limit resistance can be specified for each corresponding component with non-zero stiffness.
parameters
When used in conjunction with domains the following steps can be followed to define line-to-line link
elements:
Lets you constrain the six nodal degrees of freedom that are: translations (eX, eY, eZ and rotations (X, Y
and Z).
In the default setting no nodes have constrained degrees of freedom.
During the analysis equilibrium equations will only be written in the direction of the free displacements
(translations/rotations).
Any combination of the six nodal degrees of freedom (eX, eY, eZ, X, Y and Z) can be selected. However,
in many cases typical combinations of degrees of freedom can be used. In these situations, you can
quickly apply a predefined setting by selecting it from the list box.
The following particular structures are listed:
Plane truss girder / Space truss / Plane frame/ Grillage / Membrane / Plate
Define a nodal DOF Use the buttons to set the degrees of freedom. Button captions will reflect the current value. Changes
will be applied only to those nodal DOF which have their corresponding check-box checked. Unchecked
components will retain their original values in the selection.
You have two options to change nodal DOF:
Overwrite
The new setting overwrites the existing degrees of freedom settings of the selected nodes.
Union
Sets a degree of freedom constrained if either the existing or the new value is constrained. This option
is useful in the definition of symmetry conditions.
Example of union ex eY eZ ΘX ΘY ΘZ
initial value free constr. free constr. free constr.
new value free free free constr. constr. constr.
result free constr. free constr. constr. constr.
The six nodal degrees of freedom (𝑒𝑋 , 𝑒𝑌 , 𝑒𝑍 , 𝜃𝑋 , 𝜃𝑌 , 𝜃𝑍 ) are set by a six digit code comprised of f (free)
and c (constrained) symbols.
Each digit corresponds to one degree of freedom component. By default the nodes are considered free
(all digits are f-free symbols). By setting a digit to c (constrained) the corresponding degree of freedom
component is constrained. The default DOF code of a node is [f f f f f f].
The loads that apply in the direction of a constrained degree of freedom are not taken into
account. Loads in the direction of the constrained degrees of freedom will appear in the table of
unbalanced loads.
The nodes with DOF different from [f f f f f f] are displayed on the screen in cyan.
Truss girders
Truss girder in Truss girder in
X-Y plane X-Z plane
Frames
X-Y plane frame X-Z plane frame
Grillages
Grillage in X-Y plane Grillage in X-Z plane
Membranes
Membrane in X-Y Membrane in X-Z plane
plane
Membrane in Y-Z
plane
Plates
Plate in X-Y plane Plate in X-Z plane
Symmetry
X-Y X-Z
symmetry plane symmetry plane
Y-Z
symmetry plane
Pick up>> Degrees of freedom can be picked up from another node and assigned to the selected nodes.
284
4.9.22. References
Lets you define reference points, vectors or axes, and planes. The references determine the orientation
of the local coordinate systems of the finite elements in the 3D space. The local coordinate system of the
elements defined with the references is used to define cross-sectional properties and to interpret results.
The element properties are defined and the internal forces (Nx, Vy, Vz, Tx, My, Mz for beams, mx, my, mxy
for plates, nx, ny, nxy for membranes, etc.) are computed in that local system.
Quick modify: Clicking on the symbol of a reference the Table Browser is invoked displaying the table
of the references. The reference vector and axis can be defined by two points, the reference plane by
three points. When closing the table the reference vectors, and axes are normalized with respect to 1.
Color codes: x = red, y = yellow, z = green.
The Edit / Convert automatic references menu item converts automatic references into reference vectors.
Reference point Reference point is used to define the orientation (local coordinate system) of beam, rib, support, and
spring elements or to define the positive local x and z axes of surface elements.
The reference points are defined (by its coordinates) in the global coordinate system.
The reference points are displayed on the screen as small red + symbols.
User’s Manual X6 Release1 e2 285
Reference
Referencepoint
point
Surface elements:
The positive local z axis is oriented toward the half-space in which the reference point is located, and is
perpendicular to the element’s plane. Once the local x-axis is defined local y-axis is determined according
to the right hand-rule.
The local x axis will be oriented in the direction of the reference point.
In the case of a surface element the reference point must be located in the plane of the element.
Reference point
Supports:
In the case of a support element you can use a reference point to define local x axis.
Reference point
Reference vector Lets you define the local x axis for surface, support, and spring elements. Also defines the orientation of
local z coordinate axis of beam, rib and spring elements.
Surfaces:
The local x axis will be parallel to the reference vector. In the case of a surface element the reference
vector must be parallel to the plane of the element.
The orientation of local z-axis can also be defined by a reference vector.
286
Supports:
In the case of a support element you can use a reference vector to define local x axis.
Reference vector
Reference vector
Reference axis Reference axis is used to define the local x-axis of surface elements, that will be oriented towards the
reference axis. The reference axis must not include element centerpoint.
Reference plane Reference plane is used to define the local x axis of surface elements, that will be parallel to the
intersection line of the reference plane and the plane of the element. The reference plane must not be
parallel to the plane of the element.
User’s Manual X6 Release1 e2 287
Reference angle Rotation of truss / beam / rib cross-sections is made easy by the reference angle. The automatic local
coordinate system (and the cross-section) can be rotated around the element axis by a custom angle.
If the element is parallel to the global Z direction, the angle is relative to the global X axis. In any other
case the angle is relative to the global Z axis.
Unused references can be deleted in the Table Browser by selecting References and choosing Edit
/ Delete unused references from the menu.
Display Select architectural project storeys and element types you want to be displayed.
If you create model framework or delete objects and nothing is selected the Selection Toolbar appears.
Click the Property Filter icon to select beams and columns within a certain range of section size
according to their minimum side length or select walls or slabs within a certain range of thickness.
If you want to restore the whole range click the button at left bottom.
If the Only objects without static model is checked only elements not having static model will be selected.
Refresh all Makes the architectural model visible in all windows.
Import footings If activated, footings in the IFC file are also processed.
Deleting an architectural object having a static model will not delete its associated static model.
Create Model Model framework will be created from selected layer elements. Columns will be reduced to their axis,
Framework walls, slabs and roofs will be reduced to their center plane. Framework nodes and lines become part of
the AxisVM model and are independent of the background layer.
When importing IFC files, it is not always possible to detect the statical frames of objects. Those objects
whose statical frame detection failed, are drawn to the screen with a dashed line. User now has many
possibilities to create, modify or delete statical model though:
288
Automated detection of statical frame: in as much as the automated detection method has
been successful while importing, the program can use those precalculated data to create stat-
ical frame. More than one item can be selected for this operation.
Deleting selected IFC objects: with this function user can delete the selected IFC objects from
the model. It is important to note, though, that if we delete objects, whose statical frame
has been created before, this frame will not be deleted with the object. More than one
item can be selected for this operation.
Clipping beam-like objects by clipping plane: the selected objects are clipped by the plane the
user has supplied by giving 3 points. Then the algorithm defines the sections to all the se-
lected elements, and calculates their normals and centres of gravity. These values give us lines
imprinting the selected bodies. The resulting segments shall than be treated as statical frames
of the objects, and the calculated polygons as their cross-sections. More than one item can
be selected for this operation.
Handling beam-like objects by supplying their middle lines: user must support two points that
will form a line. This line then is imprinted to the selected bodies, and the resulting segments
will serve as the middle lines of the objects. Taking the middle points of these segments and
their direction vectors, clipping planes can be obtained. At the final step, the segments are
moved to the centres of gravity of the cross-sections. More than one item can be selected
for this operation.
Handling beam-like objects by supplying their middle lines and position of the cross-section:
almost the same like the previous operation, except for that here, the position of the cross-
section must be given by the user, too. Then the center line of the selected body is modified
in a way, that the start point is left unchanged, and the end point is the mirror image of the
start point to the cross section. Only one item can be selected for this operation.
Handling circular beam-like elements: user must supply an arc by 3 points. Then the algorithm
calculates the intersections of this arc and the selected body. If more than 2 intersections are
obtained, then it takes the two extremes. The middle point of the resulting arc and the normal
of the arc provide a clipping plane that the algorithm uses to get the cross-section. More
than one item can be selected for this operation.
Clipping plates by clipping plane: the selected domains are clipped by the plane given by 3
points, and the resulting polygons will be the statical frames of the objects. More than one
item can be selected for this operation.
Clipping cylindrical swept bodies by a clipping cylinder: the user must supply a cylinder that
the algorithm will use for clipping. This cylinder is adjusted in a two-step operation. First, an
arc is obtained by 3 points. Secondly, this circular arc is offsetted when moving the cursor.
This resulting arc shall serve as the base of the cylinder. Its height is calculated automatically
from the geometric parameters of the selected body. Finally, the statical frame is the result of
the intersection of the cylinder and the selected body. Only one item can be selected for
this operation.
In all editing operations, user is given a real-time visual feedback about the current statical frame, thus
it is possible to edit the correct statical model of the object. Sometimes some pre-editing can be nec-
essary, though. An optional filter is given to filter out IFC objects that already have statical models
calculated. However, if we edit an object with an existing statical model, the result of the operation will
overwrite the former statical frame.
Parts will automatically be created for levels and object types and the elements created for the static
model will be included in the appropriate parts.
Hinged wall connections can be modeled using edge hinges when creating a model framework from
the architectural model.
If the Material field is set to Automatic, and the IFC file includes material data and assigments the model
will import them.
You can assign properties to the selected architectural objects as follows:
User’s Manual X6 Release1 e2 289
Convert walls to supports: You can convert wall objects to supports by enabling this checkbox.
The support will be placed at the top edge of the corresponding wall. The support stiffness will be
computed based on the top and bottom end releases.
Graphical
Cross Section
Editor
Cross-section
Library
Convert columns to supports: The selected column objects can be converted to supports. Support
stiffness is established based on the end releases. Supports will be placed at the top of the column.
290
You can select the layers that you want to take into
account.
4.9.24. Modify
In fact, the modification is similar to the element definition, but does not assign properties to
undefined geometrical elements and allows access to a specific property without altering others.
You can switch to the element definition radio button to define all properties of all the selected
elements, lines or surfaces.
Immediate mode If the Geometry or Elements tab is active click a finite element to modify its properties. If more finite
elements have been selected they can be immediately modified by clicking one of them. If you click an
element which is not selected, selection disappears and you can modify the element you clicked. If you
click on a node its nodal degrees of freedom can be edited immediately.
You can also modify the properties using of Property Editor.
See 3.6.1 Property Editor
User’s Manual X6 Release1 e2 291
4.9.25. Delete
[Del] See 3.2.9 Delete
4.10. Loads
Lets you apply various static loads for static, dynamic and buckling analysis, and define concentrated
masses for vibration analysis. The current load case can be selected from the dropdown list.
New Case You must assign a different name to each case. These are the possible types of
load cases:
1. Static
The static load case can be applied to static, vibration, and buckling analysis. In case of vibration
analysis, the loads can also be taken into account as masses.
The load case can be included into a load group. When calculating the critical load combination, the
load case will be taken into account according to the parameters of the load group to which it
belongs.
Critical combination can be determined only from the results of a linear static analysis.
2. Influence line
Lets you apply a relative displacement load to obtain the influence line of a result component, of a
truss or beam element.
When the influence line load case type is selected you can apply only the influence line load.
3. Moving load
In this type of load case only moving (line or surface) loads can be defined. When defining a moving
load, a group of new load cases will be created. The number of these load cases is equal to the
number of steps specified in the moving load definition dialog. Their name is created automatically
like MOV_xx. As they get into a load group the most unfavorable effect of the moving load can be
checked displaying the result of the critical combination. These auto-created load cases can be moved
together only and only into another moving load group.
If more than one moving load is applied in the same load case the number of steps (and auto-created
load cases) will be equal to the maximum number of steps specified. If the maximum number of steps
is k, and another moving load has i steps (i < k), then this load will remain at the end of the path in
steps i+1, i+2, …, k.
See details 4.10.29 Moving loads
Moving load cases can be placed only in moving load groups. Static loads can also be defined in
moving load cases. They will be present in all steps but only in moving load cases.
4. Seismic (SE1 module is required)
When selecting seismic load case type you can specify the parameters for calculation of earthquake
loads. Prior to creating an seismic load case, you must perform a vibration analysis. Based on the
mode shapes, and on the structural masses, AxisVM generates seismic loads case, in a k+2 number,
where k is the number of available smallest frequencies. The two additional cases corresponds to
the signs +, and -, that contain the critical combinations.
Multiple seismic actions can be defined for the structure. Seismic actions cannot be combined.
See details 4.10.24 Seismic loads – SE1 module
When selecting seismic load case the only icon available on the Toolbar will be Seismic para-
meters.
5. Pushover (SE2 module is required)
When selecting pushover load case type, you can specify parameters for generating load distributions
that can be used in pushover analyses. Prior to creating a pushover load case, you must perform
vibration analysis. Based on specified mode shapes AxisVM generates nodal forces on each node of
the model. A total of four load cases are generated initially. They represent a uniform (U) and a modal
(M) distribution in the direction of each of the horizontal axes (X and Y by default). The uniform load
distribution option generates nodal forces proportional to the masses assigned to each node in the
model. The modal load distribution uses the mode shape weighed by the masses at each node to
generate the nodal force distribution. In both cases the sum of forces generated is 1 kN in the same
horizontal direction.
See details 4.10.25 Pushover loads – SE2 module
When selecting pushover load case the only icon available on the Toolbar will be Pushover
parameters.
User’s Manual X6 Release1 e2 293
Imperfection load cases does not contribute to critical combinations of linear analysis results.
When selecting tensioning load case the only icon available on the Toolbar will be Tensioning.
Dynamic load cases cannot be included in load groups and load combinations. Loads within dy-
namic load cases will be applied only in Dynamic analysis.
9. Snow load cases (SWG module is required)
AxisVM can calculate and apply snow loads on the structure. The limits of automatic snow load gen-
eration in the program are explained in 4.10.13 Snow load – SWG module.
Snow loads can be placed on flat load panels in arbitrary planes. As a first step a temporary snow
load case is created in the snow load group and its name can be set. If the design code requires
verification for exceptional snow load, an exceptional snow load case is also created in an exceptional
snow load group (except for exceptional snow loads as per Annex B of Eurocode 1-3). After defining
load panels and setting the snow load parameters the program replaces the temporary load case
with automatically generated snow load cases.
The undrifted load case receives a UD suffix. Drifted load cases receive a D suffix with 2-4 extra char-
acters. These characters indicate the wind direction (X+, X–, Y+, Y–). Besides winds parallel to the
coordinate axes, winds in 45°+n*90° directions (where n = 0, 1, 2, 3) are also taken into account and
are indicated by the quadrant where the wind speed vector points to. X+Y– refers to a wind in the
315° direction for example (0° being defined by the positive X axis, and angles measured counter
clockwise).
Only the necessary load cases are created by the algorithm, therefore the number of load cases and
their type depends on the type of structure under consideration and the snow load parameters given
by the user. Load parameters shall be defined after closing the Load Cases Dialogue by clicking on
the Snow Load icon in the Toolbar. Before this step it is recommended to define load panels of the
roof using the Load Panels icon in the Toolbar.
For details of snow load generation see 4.10.13 Snow load – SWG module.
If a snow load case is selected only two buttons are enabled on the Loads toolbar. These are the
load panel and snow load definition.
10. Wind load cases (SWG module is required)
During parametric definition of wind loads it is necessary define one or more substructures exposed
to wind. If Separate load cases for each substructure is checked wind loads of different substructures
will be generated into different load cases. Otherwise load cases will contain wind loads for more
substructures.
Calculated wind loads are reliable only for certain building types as described in the design code. It
is recommended to use these automatic methods only for geometries within the limits explained in
detail at 4.10.14 Wind load – SWG module. Wind loads can be placed on flat load panels in arbitrary
planes. As a first step a temporary wind load case is created in the wind load group and the name of
294
the wind load case can be set. After defining load panels and setting the wind load parameters the
program replaces the temporary load case with automatically generated wind load cases.
Wind load cases are generated with a name code corresponding to the loading situation.
The first two characters after the name of the load case describe the wind direction (X+, X–, Y+, Y–).
The logic behind this notation is the same as for snow loads. The next one or two characters denote
the type of wind action.
P and S denote pressure and suction, respectively. For pitched roofs the design code requires to check
all possible combinations of wind actions on the two sides of the roof. Therefore, for pitched roofs
Pp, Ps, Sp and Ss load cases are created. Here the first character refers to the windswept side, the
second one refers to the sheltered side. For special torsion actions T+ and T– are used referring to
the two different torsion directions.
The last character denotes the type of internal action. The internal wind actions are ignored for O; P
stands for internal pressure, S stands for internal suction. These load cases are needed only if no
further information is available on internal pressure so a positive and a negative critical value must
be used as an approximation on the conservative side. If the last character is C, the load case was
created based on a user-defined μ value which depends on the layout of openings of the structure.
If a C load case is created for a given direction P and S load cases are not required, hence not created.
Only the necessary load cases are created by the algorithm, therefore the number of load cases and
their type depends on the type of structure under consideration and the wind load parameters given
by the user. Load parameters shall be defined after closing the Load Cases Dialogue by clicking on
the Wind Load icon in the Toolbar. Before this step it is recommended to define load panels of the
building walls and roof using the Load Panels icon in the Toolbar.
For details of wind load definition see 4.10.14 Wind load – SWG module.
If a wind load case is selected only two buttons are enabled on the Loads toolbar. These are the
Spa load panel and wind load definition.
11. Fire load cases (SD8 or TD8 module is required)
Fire effects can be assigned to steel and timber line elements.
Fire load groups contain only fire load cases. This load group has no parameters and is automatically
deleted if their load cases are deleted. In the process of finding critical combinations AxisVM takes
into account fire load cases only in accidental load combinations. However, it is also possible to
consider fire load cases in user specified load combinations.
The definition of fire effects is discussed in 4.10.31 Fire effect on steel elements – SD8 module and
4.10.32 Fire effect on timber elements – TD8 module
If a fire load case is selected only the Fire effect on steel/timber elements button is enabled on
the Loads toolbar.
Load-duration Timber design module requires information on the load duration. So, if a timber material has been de-
class fined in the model load case duration class can be entered. (Permanent: > 10 years; Long term: 6 months–
10 years; Medium term: 1 week–6 months; Short term: < 1 week; Instantaneous; Undefined)
Duplicate Lets you make a copy of the selected load case under another name. You must specify the new name,
and a factor that will multiply the loads while copying. The factor can be a negative number as well.
Selected loads can be copied or moved to another load case by changing load case during the
copy or move process.
Conversion Generated snow or wind load cases can be converted to regular load cases by clicking the Conversion
button. All load cases of the selected type will be converted. After conversion converted loads and load
cases can be deleted and modified.
Delete Lets you delete the selected load case or load group. Any related load cases and load groups will also
be deleted (e.g., deleting a seismic load group removes all generated seismic loads and their load group
as well).
User’s Manual X6 Release1 e2 295
You can change the current load current case by selecting from the drop down
list near the load case icon. Selection can be moved using the up and down
arrow keys. This is the best way to overview moving load cases.
The default keyboard shortcuts for the next / previous load case are Ctrl+PgUp
and Ctrl+PgDn.
The name of the selected load case will appear in the Info window and the
loads you define will get to this load case.
In case of choosing Tensioning load case only the Tensioning icon will be active
on the toolbar.
Click on it, then select the proper beam or rib elements, so the Tensioning
dialog will appear.
See 4.10.28 Tensioning - PS1 module
Click right mouse button over the list, select Order of load cases to get to a dialog setting the load case
order. This dialog is also available in the Table Browser (Format / Order of load cases).
Load Groups Load groups are used when generating of critical (design) values of the results.
New Group Lets you define a new load group. You must specify the name and the type (permanent, incidental,
exceptional) of the load group, and the corresponding factors according to the current design code.
Click on these buttons to select from the list of recommended values according to the current design
code.
Later you can specify which load cases belong to a specific load group. Clicking any icon within the New
Group group box will create a new group in the tree and you can specify a name for it. Existing load
group names will be rejected. After creating a load group you have to specify the value of its paremeters
(like the partial factor, dynamic factor, simultaneity factor, etc.). A load case can be assigned to a load
group by choosing a group from the dropdown list or dragging the load case under a load group in
the tree. See 4.10.2 Load combinations
The default behaviour of new load cases can be set in Settings / Preferences / Load group.
See 3.3.11 Preferences
The following load groups are allowed depending on the design code:
Permanent
Includes dead load, permanent features on the structure...
Include all load cases in combinations
All load cases from the group will be taken into account in all load combinations with their upper
or lower partial factor.
Exceptional
Includes earthquake, support settlements, explosion, collision... Only one load case from the group
will be taken into account in a load combination at one time. That load case must have the
simultaneity factor of zero.
296
Critical Critical combinations are determined according to the load groups. Certain exclusive loading situations
load group (like snow and exceptional snow) are detected (so critical combinations do not include both snow and
combinations exclusive snow loads). If it is required to ensure exclusivity or linkage between load groups it is possible
to control this through critical load group combinations.
Go to the Table Browser and find Critical load group combinations.
Columns of the table list the load groups defined in the model. Each row describes a rule for generating
critical combinations.
Click on the New row icon to add a new rule. The Active status of permanent load groups cannot be
changed. In case of variable load groups, the way of contributing to critical combinations can be se-
lected from the dropdown list.
Active Load cases of the load group can be included in the automatic combinations built internally.
Excluded Load cases of the load group cannot be included in the automatic combinations built internally. Mutu-
ally exclusive load groups can be created by adding as many rules as the number of exclusive load
groups, and only one group must be set to Active in each row, leaving other groups Excluded.
Simultaneous Load groups acting simultaneously as simultaneity groups. Load groups having the same status always
(1), (2), … act together, i.e., they are included in automatic combinations together. More than one simultaneity
group can be defined. Having two load groups with a status of Simultaneous (1) makes Simultaneous
(2) also available automatically in the dropdown list, and so on.
Leading Load groups acting simultaneously as leading actions. Load groups having the same status always act
(1), (2), … together, i.e., they are included in automatic combinations together as leading action. More than one
leading group can be defined. Having two load groups with a status of Leading (1) makes Leading (2)
also available automatically in the dropdown list, and so on.
AxisVM builds automatic combinations internally according to the rules defined in this table then finds
the most unfavourable combination as critical combination.
The table of Custom load combinations allow generation of these automatic combinations to check the
combination building process (the total number of combinations can be huge depending on the num-
ber of load groups) but generation of the actual combinations are not necessary to obtain critical forces.
User’s Manual X6 Release1 e2 297
Load Element
Lets you define load combinations of the defined load cases. You can specify a factor for each load case
in a load combination.
The results of a load combination will be computed as a linear combination of the load cases taking
into account the specified load case factors. A zero factor means that the respective load case does not
participate in the load combination. To find the most unfavourable of the custom combinations defined
here create an envelope for the combinations.
Load combinations can be listed, defined and deleted by load cases or by load groups (the second
option is easier to overview).
In the former case a factor must be entered for each load case.
In the latter a factor must be entered for each load group. The actual load combinations will be created
using these factors and according to the load group properties (e.g., if the load cases within the load
groups are exclusive or not. See the previous chapter for details). Combining load groups with many
load cases may result in huge number of individual load combinations.
Calculates all critical combinations based on load groups and transfers them into the load combination
table.
The option Include imperfections is available only if the model contains an imperfection load group.
If Replace critical combinations is checked all previously generated critical combinations will be deleted
and replaced with the new combinations. If Replace only combinations of the same type is checked only
combinations from the selected ULS/SLS combination types will be replaced.
Checking Create load combinations with descriptive names changes the naming convention, so gener-
ated combination names will be the description of the combination (like 1.1*ST1+0.9*ST2) instead of
Co. #1, Co. #2, etc.
You can also define load combinations after you have completed a linear static analysis. The
postprocessor computes the results of these load combinations when required
In case of nonlinear static analysis, AxisVM first generates the combination case, and then
performs the analysis (one load combination at a time).
Automatic load The program investigates all possible combinations depending on the load group parameters and the
combination equations of the current design code.
The minimum and maximum result values of these combinations are selected as critical (design) values.
Critical combinations for Eurocode, SIA26x, STAS, DIN, Italian code are assembled according to the
following schemes:
User’s Manual X6 Release1 e2 299
In certain countries (e.g., Austria) the national annex does not allow to use (6.10.a) and (6.10.b). In this
case the following formula is used:
In case of geotechnical combinations, the program uses the above formulae, as well. The partial factors
for actions are selected from the groups of values A1 and A2, which are categorized according to annex
A of EN 1997-1.
Set
Action Symbol
A1 A2
Permanent unfavourable 1,35 1,0
𝛾𝐺
favourable 1,0 1,0
Variable unfavourable 1,5 1,3
𝛾𝑄
favourable 0 0
Remark: the values may be specified differently by the National Annex. The specifications of the se-
lected NA are considered.
In case of verification of static equilibrium, the following partial factors are considered (annex A of EN
1997-1):
ULS 2 - Seismic
ULS 3 - Exceptional
Eurocode ∑ 𝐺𝑘,𝑗 + 𝑃𝑘 + 𝐴𝑑 + Ψ1,1 𝑄𝑘,1 ∑ 𝛹2,𝑖 𝑄𝑘,𝑖
and other codes 𝑗≥1 𝑖≠1
SLS 1 - Characteristic
Eurocode ∑ 𝐺𝑘,𝑗 + 𝑃𝑘 + 𝑄𝑘,1 + ∑ 𝛹0,𝑖 𝑄𝑘,𝑖
and other codes 𝑗≥1 𝑖≠1
SLS 2 – Frequent
Eurocode ∑ 𝐺𝑘,𝑗 + 𝑃𝑘 + Ψ1,1 𝑄𝑘,1 ∑ 𝛹2,𝑖 𝑄𝑘,𝑖
and other codes 𝑗≥1 𝑖≠1
SLS 3 - Quasipermanent
Eurocode ∑ 𝐺𝑘,𝑗 + 𝑃𝑘 + ∑ 𝛹2,𝑖 𝑄𝑘,𝑖
and other codes 𝑗≥1 𝑖≥1
Critical load combination method for internal forces and for displacements are selected automatically.
Critical load combination method for displacements depends on the type of structure you are modeling.
Click Result display parameters on the Static toolbar to set the critical combination formula.
Italian code Combination of seismic loads with other load types: ∑𝐺𝐾 + 𝛾𝐼 ⋅ 𝐸 + ∑Ψ𝑗𝑖 𝑄𝐾𝑖
The positive directions are according to the positive directions of global coordinate axes.
Modify nodal loads You can select, move, copy or modify the load independently of the node.
Modify position 1. Select the loads you want to move together.
2. Grab any of them by pressing the left mouse button.
3. Move them to their new position.
4. Click the left mouse button or use a command button. (Enter or Space).
Modify value 1. Select the load.
2. Click the Nodal Load icon on the toolbar.
3. Change the values
Load components applied in the direction of a constrained degree of freedom will be not taken
into account in the analysis.
The forces are displayed on the screen as yellow arrows, the moments as green double arrows.
Choose Define load eccentricity, then click on a radio button representing a point of the bounding
rectangle of the cross-section or enter custom values.
302
If you apply a concentrated load to a node that is already loaded, you can Overwrite or Add it to the
existing load.
Concentrated loads can be selected, moved, copied, modified independently of the beam. Modify load
values like in case of nodal loads
The positive directions are in accord with the positive directions of the local or global coordinate axes.
If only some part of the structural member is selected (i.e., certain finite elements) then loads will
be interpreted in the local system of finite elements. In this case the same load will be applied to
all selected finite elements.
The forces are displayed on the screen as yellow arrows, the moments as green double arrows.
Interpretation of In case of eccentric beams or ribs, the actual axis of the element is not the same as the line connecting
the loads of the end nodes. Load eccentricity is relative to the actual (eccentric) axis of the element.
eccentric elements
Applies a point (concentrated) load at the location of the cursor if it is over a domain or a load panel.
You can also enter the location of the load by its coordinates. Place loads by clicking the left mouse
button or pressing any of the command buttons.
See 4.7.2 Entering coordinates numerically
The direction of the load can be:
Global (with respect to the global coordinate system)
Local (with respect to the local (element) coordinate system)
Reference (with respect to a reference)
Modify point load You can modify the location and value (intensity) of the load:
on domain
Modify position 1. Select the load with the cursor (a load symbol appears beside the cursor).
2. Keep left mouse button depressed.
3. Move the mouse or enter the relative coordinates to move the load to a new location.
4. Release left mouse button to set the load in its new location.
Modifying domain mesh leaves the concentrated loads (applied on the domains) unchanged.
User’s Manual X6 Release1 e2 303
Lets you apply constant or linearly distributed forces and moments to the selected beam / rib elements.
You can apply multiple distributed loads to a beam/rib in the same load case.
Instead of defining moment components it is also possible to set the ey and ez eccentricity of the load
in the local coordinate system. Choose Define load eccentricity, then click on a radio button representing
a point of the bounding rectangle of the cross-section or enter custom values.
Moment and eccentricity is always defined in the local system of the element even if the Direction of
force components are set to Global.
Line loads can be selected, moved, copied, modified independently of the beam or rib. Modify load
values like in case of nodal loads. If you apply a concentrated load to the same position that is already
loaded, you can Overwrite or Add it to the existing load.
If only some part of the structural member is selected (i.e., certain finite elements) then loads
will be interpreted in the local system of finite elements. In this case the same load will be applied
to all selected finite elements.
You must specify the distribution, the location and the values of the load components in the local or
global coordinate system as follows:
304
Type
segment trapezoid triangle
Interpretation of In case of eccentric beams or ribs, the actual axis of the element is not the same as the line connecting
the loads of the end nodes. Load eccentricity is relative to the actual (eccentric) axis of the element.
eccentric elements
- -
x
Membrane
- -
y
- -
Plate z
X
x
Shell Y
y
Z
z
In the case of shell elements, the load that is applied in global coordinate directions can have a projective
distribution. If the load p is projective, the value of the load that is applied to the shell is p cos , where
is the angle of the load direction and the element plane normal.
306
Applies a uniform or linear distributed line load over a domain or a load panel.
The direction of the load can be global projective, global along element, edge
relative or surface relative.
mx is always the torsional moment (around the application line of the load). Set
load components and placement method then draw the load (or click the lines)
to place it.
In case of Edge relative mode, the local coordinate system of the line load follows the direction of the
specified edge (line, arc or polyline). The x axis is always the axis of the specified line pointing towards
the drawing direction. The local z axis of the load is parallel with the local z axis of the domain. When
placing a line load on an existing edge of the model, always check the direction of the resulting load, as
the drawing direction of the selected line cannot be determined from node numbers.
If the line load is defined as Surface relative, then the local coordinate system of the load (x, y and z axes)
follows the local coordinate system of the domain
Line load along an arc defined by its center point and two points
Line load along an arc polygon defined by its center point and two points
Line load along a complex polygon. Complex polygons can contain arcs
During definition of a complex polygon a pet palette appears with several geometry functions. These
are: drawing a line, drawing a line as a tangent of an arc, drawing an arc with center point, drawing an
arc with a midpoint, drawing an arc with the tangent of the previous polygon segment, drawing an arc
with a given tangent, picking up an existing line.
You can modify the location and value (intensity) and any vertex of the load polyline:
Modify shape 1. Move the cursor above the vertex (a load polyline vertex symbol appears beside the cursor).
2. Click the left mouse button
3. Drag the vertex to its new position after pressing the left mouse button.
4. Click the left mouse button.
Modify value 1. Select the load with the cursor (a load symbol appears beside the cursor).
2. Click the left mouse button.
3. Enter new load values in the dialogue window.
4. Click on the Modify button to apply the changes and close the window.
Delete Select the loads you want to delete and press Delete.
Modifying domain mesh leaves line loads (applied on the domains) unchanged.
308
x - -
Membrane
y - -
z - -
Plate
x X
Shell y Y
z Z
User’s Manual X6 Release1 e2 309
The domain element type determines the load type and direction as follows.
For a membrane domain the load must be in the plane of the domain. For a plate domain the load must
be perpendicular to the plane of the domain. For a shell domain any load direction is acceptable. Load
panels accept loads in any direction but if they transfer loads to membrane or plate domains the above
limitations apply.
The load can be a global load on surface, a global projective load or a local load and the components
will be interpreted accordingly.
You can select between constant or linear load intensities and set if loads disapper over holes or are
distributed on the edge of the hole.
The first icon represents the option that loads over holes are not applied to the structure. The second
one represents the option that loads over holes are distributed on the edge of the hole.
Loads disappear/
allowed on holes
During definition of a complex polygon a pet palette appears with several geometry functions. These
are: drawing a line, drawing a line as a tangent of an arc, drawing an arc with center-point, drawing an
arc with a midpoint, drawing an arc with the tangent of the previous polygon segment, drawing an arc
with a given tangent, picking up an existing line.
1. Enter load components (pX, pY, pZ)
2. Click on the domain
Distributed domain
load The load will be distributed over the domain. The shape of this type of load will automatically follow
any change in the domain geometry. Within a load case you can apply only one load of this type on a
domain. New distributed domain load definition always overwrites the previous one.
The plane of the load intensity can be specified by load intensity values (p1, p2, p3) at three points [(1),
(2), (3)] in the plane of the domain. These points are the load value reference points. If you want to use
the same reference points and values to many loads of different shape and position you can lock the
reference points and values by clicking the Lock button. Loads are applied by entering an area.
During definition of a complex polygon a pet palette appears with several geometry functions. These
are: drawing a line, drawing a line as a tangent of an arc, drawing an arc with center-point, drawing an
arc with a midpoint, drawing an arc with the tangent of the previous polygon segment, drawing an arc
with a given tangent, picking up an existing line.
1. Enter load values at the reference points (p1, p2, p3).
2. Click on the domain.
Distributed domain
load 3. Enter three reference points by clicking or by coordinates.
Within a load case you can apply only one load of this type on a domain. New distributed domain load
definition always overwrites the previous one.
Modify area load The position, shape and intensity of a mesh-independent area load can be changed.
Modify position 1. Place the mouse above the load contour (the cursor will identify the load).
2. Press the left mouse button and move the mouse.
3. Find the new load position by moving the mouse or by coordinates.
4. Drop the load by clicking the left mouse button or pressing the Space or Enter key.
Modify shape 1. Place the mouse above a vertex of the load polygon (the cursor will identify the load polygon vertex
as a corner).
2. Press the left mouse button and move the mouse.
3. Find the new vertex position by moving the mouse or by coordinates.
4. Place the vertex by clicking the left mouse button or pressing the Space or Enter key.
The load shape will change.
Modify intensity 1. Place the mouse above the load contour (the cursor will identify the load).
2. Click the left mouse button. The area load windows appears.
3. Change the load intensity values.
4. Click on the Modify button to confirm the changes.
Multiple loads can be selected and modified this way.
Area load intensity and shape can also be changed in the Table Browser by changing the appropriate
values in the load table.
Delete Select the loads to delete and press [Del]
Mesh-independent loads are not affected by removing meshes or re-creating meshes on domains.
Homogenous surface load can be placed over line elements (trusses, beams and ribs).
Loads over trusses will be converted into loads on the truss end nodes.
1. Click the icon and select the load distribution range in the dialog.
Auto distributes the load over the elements under the load. Any new
truss, beam or rib defined under the load will redistribute the load.
To selected elements only distributes the load over the selected ele-
ments only. Select lines using the selection toolbar. Distribution re-
mains the same if a new beam or rib is defined under the load.
312
2. Define load polygon the same way as for a constant or linearly changing domain area load.
Load direction can be global on surface, global projective or local. Local directions are
defined like automatic references for domains See 4.9.22 References.
Enter load values into the edit fields. (pX, pY, pZ)
Edit / Convert surface loads distributed over beams menu item converts
loads created this way to individual beam loads.
Load panels are used to transfer nodal loads, distributed line and surface loads to the structural elements.
It is also necessary to define load panels, if snow and wind loads are generated automatically by the
software (SWG module). They have no structural role, the only function of a load panel is to distribute
loads over the domains, beam and rib elements. Loads can be distributed to all domains, ribs and beams
automatically, that are in the plane of the panel. It is also possible to specify manually the load bearing
elements within or out of the plane of the panel controlling the load transfer. Their projection to the load
panel should be within the closed contour of the load panel.
Shape tools
Rectangular
Slanted rectangle
Complex polygon
Special
selection tools
Clicking on a domain creates a load panel over the domain
This function creates load panels based on selected lines. Load panel outlines are determined from the
outlines of coplanar subsets of selected lines.
User’s Manual X6 Release1 e2 313
Load distribution
modes
Select this option to distribute the load on all domains, ribs and beams under the load panel.
Select this option to distribute the load on domains, ribs and beams in active parts under the load
panel.
Load options for If the first icon is selected the load panel applies loads over the openings (interpreting them as windows),
openings otherwise it ignores the loads over the openings (interpreting them as actual openings).
Loads from the selected load panels can be converted to the individual loads. See 3.2.19 Convert loads of
the selected load panels to individual loads.
Selected beams / ribs / domains can be out of the plane of the load panel.
If multiple domains are selected and their outlines projected to the plane of the load panel over-
lap, loads will be transferred only to the domain closest to the load panel.
Concentrated loads on columns from transferring a distributed load on the load panel
Method of load dis- Distributing of a concentrated load models the load panel as a rigid sheet supported by three springs at
tribution each load distribution point. The three springs have the same stiffness inversely proportional to the
square of the distance from the point of action.
s
10 4 2
F d max d
s = max
d
di 1 d
d max = max d i d max d max
100
Loads distributed along lines are converted to a series of point loads and the point loads are distributed
with the above method. Surface loads are distributed using the method in 4.10.11 Surface load distributed
over line elements.
314
Defining loads Point, line and surface loads can be placed on load panels the same way as on domains. An example:
Choose the Loads tab and the Distributed loads on domain tool
Enter load value components and choose Distributed domain load from the palette.
Click into any number of domains or load panels. The load will be applied to them.
EC German EN 1991-1-3:2003
DIN EN 1991-1-3/NA December 2010
EC Dutch EN 1991-1-3:2003
NEN EN 1991-1-3/NB November 2007
EC Hungarian EN 1991-1-3:2003
MSZ EN 1991-1-3:2016/NA September 2016
EC Romanian CR 1-1-3/2012
conform with SR EN 1991-1-3
EC Czech EN 1991-1-3:2003
CSN EN 1991-1-3/NA July 2012
EC Slovakian EN 1991-1-3:2003
STN EN 1991-1-3/NA December 2004
EC Polish EN 1991-1-3:2003
PN EN 1991-1-3/NA September 2005
EC Danish EN 1991-1-3:2003
DS/EN 1991-1-3 DK NA October 2015
EC Belgian EC 1991-1-3:2003
NBN EN 1991-1-3/ANB October 2007
EC Austrian EN 1991-1-3:2003
ÖNORM B 1991-1-3 September 2013
EC Finnish EN 1991-1-3:2003
SFS-EN 1991-1-3/NA November 2007
EC Norwegian EN 1991-1-3:2003
NS-EN 1991-1-3/NA 2018
EC Greek EN1991-1-3:2003
ELOT EN 1991-1-3
User’s Manual X6 Release1 e2 315
Eurocode • The algorithm generates snow loads on building roofs. It is not recommended for
general snow load generation on other types of structures, such as bridges.
• The algorithm is applicable to various roof geometries. It takes the effect of ridges
and troughs on snow accumulation into account when calculating the snow effect
on roof panels adjacent to the ridge or trough. It does not take into account the
effect of local snow accumulation on distant (i.e., not adjacent) roof panels, there-
fore, it is not recommended for complex roof geometries where such effect is ex-
pected to have significant influence on the characteristic snow load. Note that nei-
ther does EC 1-3 specify snow load calculation for the latter cases.
• Building altitude shall be less than 1500 m.
• Snow impact when snow falls off a higher roof is not considered; snow load is al-
ways classified as static action.
• Ice loading is not considered.
• Lateral loading from snow is not considered.
• Exceptional snow drifts as per Annex B are not considered.
• Exceptional snow loads are not considered.
• Sliding of snow off the roof is assumed not to be prevented.
• Local snow accumulation in the vicinity of taller construction works and smaller
projections on the roof are taken into account.
• Snow overhanging the edge of the roof is considered.
• Snow load on snowguards is not considered.
• The program can generate snow load corresponding to any arbitrary return period.
EC German • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well with the extensions/modifications listed below.
• The methodology in 5.3.4 for multi-span roofs is not implemented, because it leads
to ambiguous load generation in 3D cases.
• Exceptional snow load with the same distribution is taken into account
NTC Italian • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
EC Dutch • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well with the extensions/modifications listed below.
• No altitudes above 1500 m can be found in the Netherlands, thus the limit on alti-
tude is not applicable.
EC Hungarian • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
EC Belgian • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well with the extensions/modifications listed below.
• No altitudes above 1500 m can be found in Belgium, thus the limit on altitude is
not applicable.
EC Romanian • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
EC Czech • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
EC Slovak • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
• Exceptional snow load with the same distribution is taken into account.
EC Polish • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
EC Danish • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
316
EC Austrian • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well with the extensions/modifications listed below.
• The algorithm provides reliable results only for altitudes below 1500 m.
For higher altitudes the snow load shall be calculated as per 5.1 in EC 1-3 NA.
• The effect of solar panels as per Annex B in EC 1-3 NA is not considered.
• The influence area for snow load is assumed to be greater than 10 m2. For structural
members affected by snow from smaller areas see 9.2.1.1 in EC 1-3 NA and use an
increased value of sk for the analysis.
• The total area of the roof is assumed to be less than 2000 m2. For larger roof areas
see 9.2.1.2 in EC 1-3 NA and apply an appropriate increase in sk for the analysis.
• The effect of short eaves (h < 0.5 m) as per 9.2.1.3 in EC 1-3 NA is not considered
automatically. This effect can be considered by an appropriately extended model
that includes the ground surface around building as an extension of the roof.
• The effect of graves (as per 9.2.5.4 in EC 1-3 NA) is not considered.
EC Finnish • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
EC Norwegian • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
• The characteristic value of snow load is calculated according to NA 4.1 even for
altitudes above 1500 m.
• Exceptional snow load is not taken into account.
EC Greek • Because calculation is based on the general Eurocode, the limits presented there
apply for this standard as well.
• The characteristic value of snow load is calculated according to NA 4.1.
Swiss • The algorithm generates snow loads on building roofs. It is not recommended
for snow load generation on other types of structures, such as bridges.
• The algorithm is applicable to various roof geometries. It takes the effect of
ridges and troughs on snow accumulation into account when calculating the
snow effect on roof panels adjacent to the ridge or trough. It does not take into
account the effect of local snow accumulation on distant (i.e., not adjacent) roof
panels, therefore, it is not recommended for complex roof geometries where
such effect is expected to have significant influence on the characteristic snow
load. Note that neither does SIA261 specify snow load calculation for the latter
cases.
• Building altitude shall be less than 2000 m.
• Exceptional snow loads are not considered.
• Sliding of snow off the roof is assumed not to be prevented.
• Local snow accumulation in the vicinity of taller construction works and smaller
projections on the roof are taken into account.
• Snow overhanging the edge of the roof is considered.
• The software can consider the effect of snow guards.
• According to the settings, the program calculates increased snow loads of large
low-slope roofs. If Based on maximum size option is selected, the reference size
needs to be given by the user. In this case, increased snow load is considered
on every load panels. If the By contiguous load panels option is selected, the
software analyses the size and geometry of load panel groups. Increased snow
load is considered separately based on the geometry of load panel groups. This
option works only for load groups that have rectangular or circular shape.
Calculation details The logic of snow load calculation is explained below for each standard in the program
Eurocode • The recommended values are assumed for all coefficients unless specified other-
general wise by the user. None of the recommendations in National Annexes is assumed.
• The factors for combination value, frequent value and quasi-permanent value of
the snow load are taken according to Table 4.1 of EC 1-3.
• The characteristic value of snow load on the ground shall be specified by the user.
• Snow load on the roof is calculated using Eq. (5.1) in EC 1-3.
User’s Manual X6 Release1 e2 317
• The exposure coefficient is based on the topography selected by the user. The Ce
values that correspond to each topography type are taken from Table 5.1 in EC 1-
3.
• The thermal coefficient is taken as 1.0 by default and shall be modified by the user
if the application of a different value is justified.
• The snow load shape coefficient for roofs composed of planar panels is calculated
as per Section 5.3.1 - 5.3.4 in EC 1-3.
• Snow load shape coefficients for the undrifted load case are based on μ1 in Table
5.2 in EC 1-3. Each panel has its own μ1 value that is calculated using the slope of
the panel.
• Snow load shape coefficients for the drifted load case are based on μ2 in Table 5.2.
μ2 coefficients for troughs are calculated as per Figure 5.4 in EC 1-3 using the slopes
of the connecting roof panels in the wind direction of the given drifted snow load
case. (e.g.: A horizontal trough in the X direction results in no snow accumulation
when the wind blows in the X direction because the slopes of the connecting roof
panels in the X direction are 0°.) When there are no troughs on a roof, the drifted
load arrangements in Figure 5.3 in EC 1-3 are considered. The reduced snow inten-
sity is always assumed on the windswept side of the roof.
• Snow load shape coefficients for cylindrical roofs are calculated as per 5.3.5. in EC
1-3. In order to achieve sufficient accuracy in the load shape, it is recommended to
approximate the cylindrical roof with at least 20 planar segments.
• The effect of taller construction works and obstructions on the roof are considered
as per 5.3.6 and 6.2 in EC 1-3. Their influence is only taken into account in the drifted
load cases. Snow is assumed to be drifted only if the wall or obstacle is not parallel
to the wind direction.
• Snow load overhanging the edge of the roof is considered as per 6.3 in EC 1-3.
• Snow load corresponding to a user-defined return period is calculated using the
procedure in Annex D of EC 1-3.
EC German • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below
• The characteristic value of snow load on the ground is automatically calculated
based on 4.1 in EC 1-3 NA using the zone and the altitude specified by the user.
• Exceptional snow load is generated. The exceptional snow load coefficient is as-
sumed 2.0 by default, but it shall be overwritten by the appropriate value by the
user for Northern Germany as per 4.3 in EC 1-3 NA.
• The effect of taller construction works is calculated as per 5.3.6 in EC 1-3 NA.
NTC Italian • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below.
• The characteristic value of snow load on the ground is automatically calculated
based on 4.1 in EC 1-3 NA using the zone and the altitude specified by the user.
• The value of the exposure coefficient is based on 5.2 (7) in EC 1-3 NA.
EC Dutch • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below.
• The characteristic value of snow load on the ground is taken as 0.7 kN/m2 as per
4.1 (1) in EC 1-3 NA.
• The factors for combination value, frequent value and quasi-permanent value of
the snow load are taken as 0, 0.2, and 0, respectively as per 4.2 (1) in EC 1-3 NA.
EC Belgian • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below.
• The characteristic value of snow load on the ground is automatically calculated
based on 4.1 (1) in EC 1-3 NA as a function of the altitude specified by the user.
• The combination factors for frequent and quasi-permanent load combinations are
based on EC 1-3 NA 4.2 (1).
318
• The exposure and temperature coefficients are taken as 1.0 according to the regu-
lations in EC 1-3 NA 5.2 (7) and (8).
• When calculating the effect of taller construction works, the upper limit of μw is 2.0
as per EC 1-3 NA 5.3.6 (1).
• The effect of snow overhanging the roof is not considered (EC 1-3 NA 6.3).
EC Hungarian • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below.
• The characteristic value of snow load on the ground is automatically calculated
based on NA1.6 in EC 1-3 NA using the altitude specified by the user.
• Exceptional snow load is generated. An exceptional snow load coefficient of 2.0 is
used as per NA1.8 in EC 1-3 NA.
EC Romanian • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below
• The characteristic value of snow load on the ground is calculated automatically
based on 3.1 in CR 1-3 using the basic sk value and the altitude specified by the
user.
• Snow load on the roof is calculated using Eq. (4.1) in CR 1-3.
• The importance factor shall be selected by the user; the list of recommended values
is based on Table 4.2 in CR 1-3.
• The factors for combination value, frequent value and quasi-permanent value of
the snow load are taken as 0.7, 0.5, and 0.4, respectively as per Table 4.4 in CR 1-3.
EC Czech • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below
• The characteristic value of snow load on the ground is based on the map cited in
2.7 in EC 1-3 NA (www.snehovamapa.cz)
• Snow load shape coefficients for cylindrical roofs are calculated as per 2.19 and
Figure NA.1 in EC 1-3 NA
• The effect of taller construction works is taken into account as per 2.20 in EC 1-3
NA with the following assumption: b1,s = 0.5b1.
EC Polish • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below
• The characteristic value of snow load on the ground is calculated automatically
based on 1.7, Figure 1 and Table 1 in EC 1-3 NA using the zone and altitude speci-
fied by the user.
EC Danish • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below.
• The characteristic value of snow load on the ground is taken as 1.0 kN/m2 as per
4.1 (1) NOTE 1 in EC 1-3 NA.
• Default snow load exposure coefficients are based on the assumption of Cs = 1.0
as per 5.2(7) in EC 1-3 NA. When a different size coefficient needs the be applied,
the user shall calculate the resulting exposure coefficient and define it as custom.
• The additional snow load arrangement in 5.3.3(4) in EC 1-3 NA is not considered.
• Snow load shape coefficients for cylindrical roofs are calculated as per 5.3.5. (3) in
EC 1-3 NA
• The effect of taller construction works is taken into account as per 5.3.6 in EC 1-3
NA with the following assumptions: (i) the shelter is global; (ii) bsl = 0.5b1; (iii) the
shelter is on the leeward side.
EC Austrian • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below.
User’s Manual X6 Release1 e2 319
• Snow load shape coefficients for multi-span roofs are calculated as per 9.2.3 in EC
1-3 NA.
• Snow load shape coefficients for vaulted roofs are calculated as per 9.2.4.3 in EC 1-
3 NA.
• The modifications for calculating the effect of taller construction works in 9.2.5.1 in
EC 1-3 NA are taken into account.
EC Finnish • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below.
• The characteristic value of snow load on the ground shall be specified by the user
based on the map in Figure 4.1 in EC 1-3 NA.
• Exposure coefficients are based on Table 5.1 in EC 1-3 NA.
• The snow load arrangement of cylindrical roofs is calculated as per EC 1-3. The
unique load arrangement introduced in Figure 5.6 in EC 1-3 NA is not applied. This
might lead to considerable error for shallow cylindrical roofs.
• The modifications for calculating the effect of taller construction works in 5.3.6(1)
in EC 1-3 NA are taken into account. The area of the roof is assumed to exceed 6
m 2.
EC Norwegian • Because calculation is based on the general Eurocode, the procedure presented
there is applied for this standard as well with the modifications/extensions listed
below.
• Drifted load arrangement for cylindrical roofs is calculated according to NA 5.3.5,
but special load case for local snow loads along the eaves are not generated.
• The calculation of snow load of roofs abutting and close to taller constructions does
not consider case II of NA 5.3.6(1).
Swiss • The recommended values are assumed for all coefficients unless specified other-
wise by the user.
• The characteristic value of snow load on the ground is automatically calculated
based on the reference height. The reference height (altitude) shall be specified by
the user based on the map in Annex D.
• Snow load on the roof is calculated using Eq. (9) in SIA261.
• The exposure coefficient is based on the topography selected by the user. The Ce
values that correspond to each topography type are taken from 5.2.4 in SIA261.
• The thermal coefficient is taken as 1.0 by default and shall be modified by the user
if the application of a different value is justified.
• The snow load shape coefficient for roofs is calculated as per Section 5.3 and Fig-
ures 2 and 3 in SIA261. Each panel has its own μ value that is calculated using the
slope of the panel.
• μ2 coefficients for troughs are calculated as per the second column of Figure 3 in
SIA261 using the slopes of the connecting roof panels in the wind direction of the
given drifted snow load case. (A trough in the X direction for example, results in no
snow accumulation when the wind blows in the X direction if the slopes of the con-
necting roof panels in the X direction are 0°.) When there are no troughs on a roof,
the drifted load arrangements in the first column of Figure 3 in SIA261 are consid-
ered. The reduced snow intensity is always assumed on the windswept side of the
roof.
• Snow load shape coefficients for cylindrical roofs are calculated as per the third
column of Figure 3 in SIA261. In order to achieve sufficient accuracy in the load
shape, it is recommended to approximate the cylindrical roof with at least 20 planar
segments.
• The effect of taller construction works and obstructions on the roof are considered
as per 5.3.6 in SIA261 taking into account the recommended values for snow weight
in 5.4.1 in SIA261. Their influence is only taken into account in the drifted load cases.
Snow is assumed to be drifted only if the wall or obstacle is not parallel to the wind
direction.
320
• Increased snow load is calculated for large low-slope roof according to 5.3.5 of
SIA261.
• Effects of overhanging snow is taken into account according to EC 1-3 6.3.
Usage The following paragraphs explain the usage of the automatic snow load generator module.
To apply snow loads first click on the Load cases / load groups button and
define a snow load case by clicking on the snow load case button. The snow
load group will be created automatically. An exceptional snow load group is
also created if the checking exceptional snow is required and available in the
given design code.
A temporary snow load case is automatically created in the snow load group.
After the snow load parameters are specified, that load case is replaced by
the generated snow load cases. The algorithm handles both undrifted and
drifted load cases. Wind directions X+, X–, Y+, Y– and in 45°+n*90° directions
(where n = 0, 1, 2, 3) are taken into account for details and naming conven-
tions see 4.10.1 Load cases, load groups.
To set the snow load parameters select one of the snow loads cases as the
current load case. It makes the snow load icon enabled in the Loads tab.
If no load panels have been created before, draw the load panels according to 4.10.12 Load panels.
To enter snow load parameters, click the snow load icon on the Loads tab.
The parameters dialog allows choosing pitched (or flat) roof or cylindrical roof load panels for snow
load and setting the load parameters.
Snow load Two roof types can be assigned to load panels. Click
parameters on the icon and select load panels belonging to the
roof.
Cylindrical roof
Exposure coefficient Ce
In case of special circumstances an exposure factor
other than 1.0 can be set depending on the topogra-
phy (windswept, normal, sheltered or other).
A custom Ce value can also be specified. In this case
the program asks for confirmation then uses the cus-
tom value when calculating snow and exceptional
snow load intensity.
Thermal coefficient Ct
The Ct thermal coefficient can be set to a value other than 1.0 only if the engineer performed thermal
transmittance calculations for the roof (heat loss can cause melting). In this case the program asks for
confirmation then uses the custom value
Zone
In countries where the characteristic snow load depends on the geographical location the national annex
divides the country into zones. The zone selected affects the characteristic snow load.
Importance factor
An importance factor can be entered depending on the classification of the building if it is required by
the design code. Nonstandard values can also be used after confirmation.
AxisVM calculates snow load shape coefficients for roofs abutting and close to taller construction works
or having a parapet which acts as an obstruction. It can take into account the effect of overhanging
snow on roof edges.
Parameters are stored with the edges so different roof edges can have different parameters.
Delete
Edge properties can be deleted by clicking the delete icon and selecting the edges.
Abutting wall height [m]
hw height of the abutting wall relative to the roof level
Angle of the roof above the abutting wall
α is the angle of the roof above the abutting wall. It
determines the amount of snow falling from the
higher roof.
Width of the taller construction
b1 is the width of the taller construction measured per-
pendicularly to the wall
Parapet height
hp is the height of the parapet or other obstruction relative to the roof level.
322
EC German EC 1991-1-4:2005
DIN EN 1991-1-4/NA December 2010
EC Dutch EC 1991-1-4:2005
NEN EN 1991-1-4/NB November 2007
EC Hungarian EC 1991-1-4:2005
MSZ EN 1991-1-4/NA January 2016
EC Romanian CR 1-1-4/2012
conform with SR EN 1991-1-4
EC Czech EC 1991-1-4:2005
CSN EN 1991-1-4/NA July 2013
EC Slovakian EC 1991-1-4:2005
STN EN 1991-1-4/NA July 2008
EC Belgian EC 1991-1-4:2005
NBN EN 1991-1-4/ANB December 2010
EC Polish EC 1991-1-4:2005
PN EN 1991-1-4/NA October 2008
EC Danish EC 1991-1-4:2005
DS/EN 1991-1-4 DK NA July 2015
EC Austrian EN 1991-1-4:2005
ÖNORM B 1991-1-4 May 2013
EC Finnish EN 1991-1-4:2005
SFS-EN 1991-1-4/NA November 2007
EC Norwegian EN 1991-1-4:2005
SN-EN 1991-1-4/NA 2009.
EC Greek EN 1991-1-4:2005
ELOT EN 1991-1-4
Assumptions Applicability of the algorithm in the program is limited by the applicability of the specifications in the
Application limits standards it is based on. Following is a list of such limitations for each standard in the program.
Eurocode • The algorithm can determine the wind load of rectangular-plan buildings, rectangu-
general lar-plan canopies, free-standing walls and signboards. Buildings are structures with an
internal empty space surrounded by a closed line of walls and covered by a roof.
• The following types of roofs are handled by the algorithm for buildings: flat, mono-
pitch, duopitch, hipped, vaulted.
• The following types of canopy roofs are handled by the algorithm: monopitch and
duopitch.
• The height and the width of the building shall not exceed 200 m.
• The global wind effect is determined for the structure. Hence, pressure coefficients
are calculated with the assumption of at least 10 m2 of loaded area for buildings. The
wind load of canopies and free-standing objects is based on force coefficients.
• The influence of neighboring structures and obstacles is assumed negligible.
• When calculating the internal pressure coefficients, the building is assumed not to
have a dominant external opening. Internal pressure is only calculated for buildings.
• The structural factor cscd is taken as 1.0 (corresponding applicability limits are listed in
6.2 (1) of EC 1-4).
• the influence of ice and snow on wind load is assumed negligible
EC German • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well with the extensions/modifications listed below
• building height < 300 m
• building altitude < 1100 m
NTC Italian • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well with the extensions/modifications listed below
• building altitude < 1500 m
EC Dutch • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
EC Hungarian • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
EC Romanian • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well with the extensions/modifications listed below
• building altitude < 1000 m (for buildings in southwest Banat and in areas with an
altitude above 1000 m special consideration is required and the basic wind velocity
shall be specified by the user)
EC Czech • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
EC Slovakian • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
EC Belgian • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
EC Polish • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
EC Danish • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
EC Austrian • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
EC Finnish • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
324
EC Norwegian • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
EC Greek • because calculation is based on the general Eurocode, the limits presented there apply
for this standard as well
Swiss • the algorithm is only applicable to buildings with a rectangular plan; an internal empty
space surrounded by a closed line of walls and covered with a roof
• building height < 200 m
• pressure coefficients are calculated as per EC 1-4; roofs of the following types are cov-
ered: flat, monopitch, duopitch, hipped, and vaulted
• wind effects are calculated for the overall load bearing structure, hence a loaded area
of at least 10 m2 is assumed
• the influence of wind friction is assumed negligible
• the building is assumed not to have a dominant face
• the influence of neighboring structures and obstacles is assumed negligible
Calculation details The logic of wind effect calculation is explained below for each standard in the program
• The recommended values are assumed for all coefficients unless specified otherwise
Eurocode
by the user. None of the recommendations in National Annexes is assumed.
general
• Basic wind velocity is calculated as per Eq. (4.1) in 4.2 (2) of EC 1-4
• Terrain roughness is calculated as per Eq. (4.4) in 4.3.2 (1) of EC 1-4
• Terrain orography is taken into account with the orography factor co, but the calcula-
tion of co shall be performed by the user.
• Mean wind velocity is calculated as per Eq. (4.3) in 4.3.1 (1) of EC 1-4
• Wind turbulence intensity is calculated as per Eq. (4.7) in 4.4 (1) of EC 1-4
• Reference height is calculated as per 7.2.2 (1) in EC 1-4
• Peak velocity pressure is calculated as per Eq. (4.8) in 4.5 (1) of EC 1-4
• Pressure coefficients for walls of buildings are calculated as per 7.2.2 (2) in EC 1-4. As
a conservative assumption, the lack of correlation between wind pressures between
the windward and leeward side is not taken into account.
• External pressure coefficients for roofs of buildings are calculated as per 7.2.3 – 7.2.6
and 7.2.8 in EC 1-4.
• Internal pressure coefficients are calculated based on either a μ value (determined us-
ing Eq. (7.3) in EC 1-4 and specified by the user) using Fig. 7.13 in EC 1-4, or custom
cpi values. Should the user decide not to specify μ nor cpi, two different cases are
considered with cpi = +0.2 and cpi = -0.3 as per 7.2.9 (6) NOTE 2 in EC 1-4.
• Torsional effects are calculated as per 7.1.2 of EC 1-4.
• Pressure on the underside of protruding roof corners is considered as per EC 1-4 7.2.1
using the pressure from directly connected walls.
• Wind load of canopies is calculated as per EC 1-4 7.3. The generated distributed load
is compatible with the pressure coefficients for zones A, B, C, D and produces a wind
load with the magnitude and position defined by the general force coefficient.
• Wind load of free-standing walls is calculated as per EC 1-4 7.4.1-7.4.2.
• Wind load of signboards is calculated as per EC 1-4 7.4.3.
• The effect of wind friction is determined as per EC 1-4 7.5.
EC German • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below
• Basic wind velocity is based on Fig. NA.A.1 in EC 1-4 NA. The altitude of the building
is taken into account as per A.2in EC 1-4 NA.
• Mean wind velocity and wind turbulence intensity are calculated as per Table NA.B.2
in EC 1-4 NA.
• Peak velocity pressure is calculated as per Eq. (NA.B.11) in NA.B.4.1 (4) in EC 1-4 NA
• Pressure coefficients for walls are calculated as per Table NA.1 in EC 1-4 NA.
• Pressure coefficients for flat roofs are modified according to 7.2.3 in EC 1-4 NA
User’s Manual X6 Release1 e2 325
NTC Italian • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below.
• Basic wind velocity is calculated using the parameters in Table N.A.1 in EC 1-4 NA.
• Terrain roughness is calculated using 3.3.7 in NTC 2018.
• The wind load distribution for walls of slender buildings is not variable as per EC 1-4,
but uniform according to the regulations in 3.3 in NTC 2018.
• Pressure coefficients for walls are based on Table IV. in 3.3 in NTC 2018.
• Load zone geometry and corresponding pressure coefficients for roofs are based on
3.3 in NTC 2018.
• Global pressure coefficients as per NTC 2018 are not considered.
• Automatic internal pressure calculation is based on 3.3 in NTC 2018 using c pi =±0.2.
EC Dutch • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below.
• Basic wind velocity is proposed based on the zones in Figure NB.1 and the correspond-
ing values in Table NB.1 in EC 1-4 NA.
• Terrain roughness is calculated as per Eq. (4.5) and Table 4.1 in EC 1-4 NA
• Pressure coefficients for walls are calculated as per Table 7.1 in EC 1-4 NA.
The lack of correlation between wind pressures between the windward and leeward
side is always taken into account by multiplying the pressure intensities with 0.85.
• External pressure coefficients for roofs are calculated as per Tables 7.2, 7.3a, 7.3b, 7.4a,
7.4b, and 7.5 in EC 1-4 NA.
EC Hungarian • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below
• Basic wind velocity is taken as 23.6 m/s2 as per NA4.1 in EC 1-4 NA.
• The recommended value of cdir = 1.00 in NA4.2 in EC 1-4 NA is taken into account.
EC Romanian • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below
• Basic wind velocity is calculated using Eq. (2.2) and Fig. 2.1 in CR 1-4.
• The importance factor is taken into account when calculating the wind pressure inten-
sity as per Eq. (3.1) and (3.2) in CR 1-4.
EC Czech • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below
• Basic wind velocity is based on the wind map of the Czech Republic in Fig. NA.4.1 in
EC 1-4 NA.
EC Slovakian • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below
• Basic wind velocity is calculated as per Table EC 1-4 NA NB1.
• External pressure coefficients for roofs are modified as per EC 1-4 NA 2.27–2.30.
EC Belgian • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below.
• Basic wind velocity is based on the wind map of Belgium in Fig. 4.3 in EC 1-4 NA.
• The cdir and cseason coefficients recommended in EC 1-4 NA can be taken into account
by the user, but the default values for both parameters are 1.0.
• The turbulence factor kl is calculated as per 4.4 in EC 1-4 NA.
EC Polish • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below
• Basic wind velocity is calculated as per Table NA.1 in EC 1-4 NA.
• The cdir coefficients recommended in Table NA.2 in EC 1-4 NA can be taken into ac-
count by the user, but the default values for the parameter is 1.0.
• Terrain roughness is calculated as per Table NA.3 in EC 1-4 NA.
326
EC Danish • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below.
• Basic wind velocity is taken as 24 m/s as per 4.2 (1)P NOTE 2 in EC 1-4 NA. Different
values (such as for areas less than 25 km from the North Sea) shall be specified by the
user.
• The cdir and cseason coefficients recommended in 4.2 (2)P in EC 1-4 NA can be taken into
account by the user. The default values for both parameters are 1.0.
• The pressure coefficient for zone I of flat roofs is calculated as per 7.2.3(4) in
EC 1-4 NA.
EC Austrian • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below
• Basic velocity pressure (qb) shall be specified by the user based on Annex A of
EC 1-4 NA.
• Peak velocity pressure is calculated as per Table 1 in EC 1-4 NA
• The simplified approaches for pressure coefficients in EC 1-4 NA are not applied. The
pressure coefficients are calculated with the more accurate complex approaches at all
times.
EC Austrian • Pressure coefficients for walls are calculated as per Table 3a, 3b and 4 in
EC 1-4 NA.
• Zones F and G are not used for roofs when their cumulative area is less than 20% of
the total roof area as per 9.2 in EC 1-4 NA.
EC Finnish • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below.
• Basic wind velocity for various parts of the country is based on 4.2(1)P in EC 1-4 NA.
• The unique value for kr in 4.3.2(1) in EC 1-4 NA is taken into consideration when terrain
roughness is calculated.
EC Norwegian • Because calculation is based on the general Eurocode, the procedure presented there
is applied for this standard as well with the modifications/extensions listed below.
• The fundamental value of basic wind velocity (vb0) shall be specified by the user con-
sidering.
• The basic wind velocity is calculated with the consideration of the altitude factor (c alt).
The altitude factor shall be specified by the user, the default value is 1.
• The turbulence factor (kI) can be set individually, the default value is 1. The turbulence
factor is needed for the calculation of turbulence intensity Iv(z).
• In the calculation of roughness factor, the module takes into account the specifications
for kr, z0, zmin specified in EC 1-4 NA 4.3.2 (1).
• In the case of vaulted roofs and domes, the specifications for G and F zones of
NA.7.2.8(1) are considered.
• The special cases described in NA.4.3.2(2) and NA.4.3.3 cannot be handled in the mod-
ule.
Swiss • The recommended values are assumed for all coefficients unless specified otherwise
by the user.
• The basic value of velocity pressure (qp0) shall be determined by the user based on
Appendix E of SIA 261
• Peak velocity pressure is calculated as per Eq. (11) in 6.2.1.1 in SIA 261
• The wind profile coefficient ch is calculated as per Eq. (12) in 6.2.1.2 in SIA 261 accord-
ing to parameters in Table 4 corresponding to the terrain category selected by the
user.
• Wind pressure is calculated as per Eq. (13) (external) and Eq. (14) (internal) in 6.2.2.1 in
SIA 261. The corresponding pressure coefficients are not taken from Appendix C, but
from Tables 7.2.3 – 7.2.6 and 7.2.8 in EC 1-4 to provide a more generally applicable
solution that is also in line with Swiss design practice.
• The cred reduction factor is assumed 1.0 in all cases.
• The cd dynamic factor is assumed 1.0 in all cases. The corresponding limitations are
listed in 6.3.5 in SIA 261.
User’s Manual X6 Release1 e2 327
Usage The following paragraphs explain the usage of the automatic wind load generator module.
To apply wind loads according to Eurocode, first click on the Load cases / load
groups button and define a wind load case by clicking on the wind load case
button. A wind load group will be created automatically. As a first step a tem-
porary wind load case is created in the wind load group and its name can be
set. After defining load panels and setting the wind load parameters the pro-
gram replaces the temporary load case with the necessary wind load cases.
For details on the naming convention for wind load cases see
4.10.1 Load cases, load groups
To specify wind load parameters, select a wind load case. It enables the wind load icon on the Loads tab.
If no load panels have been created draw the load panels for walls and roofs according to
4.10.12 Load panels. Click on the icon to open the wind load parameters dialog.
Wind load The first tab of the Wind load parameters dialog corresponds to the general parameters, while the
parameters second tab is used to define structure-specific parameters.
General parameters
328
The following parameters can be specified (certain parameters are not required by every design code)
Importance factor
An importance factor can be entered depending on the classification of the building if it is required
by the design code. Nonstandard values can also be entered with confirmation.
Zone
In countries where the characteristic wind load depends on the geographical location the national
annex divides the country into zones. The zone selected automatically defines the basic wind velocity.
Basic wind velocity vb0 [m/s]
vb0 is automatically calculated from the above parameters. This value can be replaced with a custom
value if desired.
Season factor cseason
The design code may allow reducing wind action through a cseason factor for temporary structures. It
takes into account that the calculated wind velocity does not occur during the lifetime of the
structure. The actual value is left to the designer’s judgement and responsibility.
Altitude factor calt
The factor can be set only in the case of the Norwegian design code, the default value is 1. The
parameter is considered in the calculation of basic wind velocity. The parameter shall be set by the
guidance of the national annex.
Turbulence factor kl
The factor can be set only in case of the Norwegian design code, the default value is 1. The
parameter is needed for the calculation of turbulence intensity Iv(z). In case of the other standards,
the module considers the constant values specified in the national annexes.
Orography factor c0
The c0 factor takes into account the effect of orography (hills etc.) to the wind velocities. The design
code gives recommendations on when and how to use this factor.
Terrain category
Select a terrain category from the dropdown list. Each category corresponds to the areas listed below:
0 Sea, coastal area exposed to the open sea
I Lakes or flat and horizontal area with negligible vegetation and without obstacles
II Area with low vegetation such as grass and isolated obstacles (trees, buildings) with sepa-
rations of at least 20 obstacle heights
III Area with regular cover of vegetation or buildings or with isolated obstacles with separa-
tions of maximum 20 obstacle heights (such as villages, suburban terrain, permanent for-
est)
IV Area in which at least 15 % of the surface is covered with buildings and their average
height exceeds 15 m
Checking the option Custom directional factors enables four cdir directional factors which take into
account a dominant wind direction on the site, thus the wind speed is not identical in all directions.
The default factor is 1.0 in all directions.
User’s Manual X6 Release1 e2 329
Substructures ex-
posed to wind
Monopitch Duopitch
After the roof geometry is selected, the following parameters can be specified:
Blockage factors
The p blockage factor describes the number of obstacles present under the canopy roof that can block
the free flow of air in the given direction. The value of 0.0 represents an empty canopy, while the value
of 1.0 represents a canopy fully blocked in the corresponding direction. A blockage factor can be spec-
ified for each of the four main directions. Note that if the windward side of a canopy is fully blocked
by an obstacle it is often worth considering the application of several substructures and a combined
wind load. The fully blocked side in that case can be modeled as a building.
User’s Manual X6 Release1 e2 331
Multi-span canopies
Wind load of mutli-span canopies shall be modeled through the combination of the wind load on a
set of structures. The position of each structure in the multi-span building shall be specified (end bay,
second bay from either end or any of the other bays) using the icons at the bottom left. In order to get
a general wind load for the whole structure, it is important to turn the Separate load cases for each
substructure off.
Elevation relative to terrain level
If the load panels selected to describe the geometry of the canopy roof do not include any wall ele-
ments (this is the typical scenario), then the wind generator algorithm will define the terrain level as
the lowest point of the roof by default. This is an incorrect assumption in most cases that can be
corrected by specifying the elevation of the lowest point of the roof from the terrain level.
Free-standing The following types of free-standing structures are available for automatic wind load generation:
structures
Free-standing wall Signboards
After the type of structure is selected, the following parameters can be specified:
Free-standing wall Shelter factors
Shelter factors consider additional walls or fences upwind that can reduce the wind pressure acting on
the modeled wall. Different shelter factors can be specified for the four main directions.
Solidity
The solidity ratio is proportional to the number of openings on the wall. 1.0 corresponds to a solid
wall. Walls with a solidity ratio below 0.8 shall be calculated as fences and are out of the scope of the
wind load generation algorithm.
Elevation relative to terrain level
Signboards that are separated from the ground by a height less than one quarter of their own height
shall be modeled as free-standing walls and their elevation shall be specified here.
Edges with return corner
The wind load of a wall with return corners depends on the length of the return corners on each end
of the wall. The length of the return corner shall be specified first and then – after clicking on the return
corner icon – the edge of the wall corresponding to the return corner shall be selected. Only the length
of the return corner has any effect the calculated wind load, its direction does not.
Remove all return corners by clicking the Delete return corners from edges button
332
After clicking the OK button, the wind loads are generated automatically for the selected load panels.
Wind load data
After the wind loads have been generated all the data related to their calculation is available in the
Table Browser under Loads/ Wind load on substructures.
Wind load parameters summarizes the values that are not load case specific. Default wind load settings
lead to direction independent wind loading, so identical parameters in all four directions. Should the
user specify direction dependent terrain categories and/or custom direction factors, these values be-
come different in each direction.
Wind load case parameters summarizes load case specific parameters for each wind load case. These
parameters are grouped by the zones generated on the selected load panels. A number after the zone
letter indicates that there are more than one of the given zone types on the model in the selected load
case. Wall zones (A-E) are divided to two areas for tall buildings. “1” always corresponds to the bottom,
“2” to the top area for walls. Multiple roof zones of the same type are available for monopitch (F) and
duopitch (F-I) roofs. Zones on the windward side of canopies are marked with “1” and zones on their
leeward side are marked with “2”. Zones with “S” in their name in free-standing walls are the ones
where the effect of shelters is taken into consideration. The FR zone shows the effect of wind friction
for every structure type.
4.10.15. Generating wind load from a CFD (Computational Fluid Dynamics) file – CFD module
CFD interface The CFD interface allows application of a pressure distribution over domains or load panels. Design codes
provide guidance on pressure coefficient calculation and wind load definition only for structures with a
simple geometric shape (4.10.14 Wind load – SWG module).
In the case of more complex or compound structures these methods are not applicable and pressure
coefficients must be determined individually by a fluid dynamics simulation (CFD modeling) or a wind
tunnel test. The CFD interface helps importing these results to AxisVM.
CFD is a generic interface module to define pressures caused by flowing substances.
Steps of loads generation
Exporting the model An STL (Stereo Lithography) file contains data of a triangular mesh describing the surface of a body.
to an STL file Choose File / Export from the menu then choose the STL file format. Turn on the CFD model radio button.
A 3D model of the structure will be generated including domains and load panels. The resulting file must
be loaded into the CFD software.
User’s Manual X6 Release1 e2 333
The CFD module reads pressure values from an ASCII text file. Each line of the file must contain
• x, y, z global coordinates
• a pressure value
The separator character between these fields can be a space, a comma or a semicolon. The units of
pressure values can be chosen, but coordinates must be in meters.
This method ignores edge loads as surfaces cannot be unambiguously associated to edge points.
4.10.17. Self-weight
Lets you take the self-weight of the elements (that have materials assigned) and domains into account
in the analysis. Self-weight is computed based on the cross-sectional, the mass density of the material,
the gravitational acceleration g, and the length or area of the element. The load is applied as a
distributed load in the direction of the gravitation vector.
A dashed line is drawn along line elements or surface/domain contours. If load intensity labels are
turned on a light blue G appears.
The load has the same effect as the 𝑑𝑇 = = 𝑑𝐿/(𝛼 ⋅ 𝐿) thermal load.
4.10.19. Tension/compression
where,
yG, zG, and Hy, Hz are properties of the cross-section.
=
A positive dT indicates a temperature increase of the beam.
𝑑𝑇 ≠ = 𝑇1 − 𝑇2 is the non-uniform temperature variation that is taken into account in the analysis.
𝑑𝑇 = = 𝑇 − 𝑇𝑟𝑒𝑓𝑒𝑟𝑒𝑛𝑐𝑒 is the uniform temperature variation that is taken into account in the analysis,
where T is the temperature in the center of gravity of the cross-section.
𝑑𝑇 ≠ = 𝑇1 − 𝑇2 is the non-uniform temperature variation that is taken into account in the analysis.
Reference point
336
T1
T2
=
For membranes only dT is taken into account. For plates only dT# is taken into account.
You can define influence line load, only in an influence line type load case.
See 4.10.1 Load cases, load groups
Truss You can specify the value of the relative displacement ex as +1 or –1.
The SE1 module includes several tools that facilitate the execution of Modal Response Spectrum
Analysis (MRSA) in AxisVM. The tools automatically perform seismic load generation in three orthogonal
directions for each vibration mode; calculation of structural response for each vibration mode and
combination of modal responses into a governing seismic effect. The following guide provides
information on the settings and options available in the program and explains how to use them. Due to
the vast literature of MRSA it is not possible to provide a step-by-step theoretical explanation to the
application of such an analysis here.
Design Codes Calculations are performed according to regulations in the general Eurocode standards (particularly
Eurocode 8) complemented by the National Annex that corresponds to the Design Code selected by
the user. Exceptions to this approach are the Swiss and Hungarian National Standards, which are not
based on the Eurocodes. The standards in the program for which MRSA calculation is available are listed
below.
EC Hungarian EN 1998-1:2004
EC Romanian P100-1-2013
EC Czech EN 1998-1:2004
EC Slovakian EN 1998-1:2004
EC Polish EN 1998-1:2004
EC Danish EN 1998-1:2004
EC Austrian EN 1998-1:2004
EC British EN 1998-1:2004
EC Finnish EN 1998-1:2004
EC Norwegian EN 1998-1:2004
NS EN 1998-1:2004/NA:2014
EC Belgian EN 1998-1:2004
EC Swedish EN 1998-1:2004
Romanian STAS
MRSA procedure Seismic loads are generated based on the vibration mode shapes in all design codes. Loads are applied
in linear static analysis and the effects of several mode shapes are combined to get the design internal
forces and displacements of the structure.
Multiple seismic actions can be defined for the structure. These actions cannot be combined.
User’s Manual X6 Release1 e2 339
The following parts provide details on the calculation process. First, the general Eurocode procedure is
presented followed by a summary of country-specific modifications.
The program will automatically create several new load cases for the seismic load and these will be
supplemented by others after the details of seismic load generation are set. The naming convention for
load cases is explained in the static analysis part of this section.
Reference value of
ground acceleration
Behaviour factor Importance factor
Displacement be-
Soil class haviour factor
Spectral function
editor
Design response
spectrum parameters
The first tab of the dialog window lets the user set details of the horizontal response spectrum. The
horizontal response spectrum is used to calculate equivalent static loads for each vibration mode shape
in two orthogonal horizontal directions. The spectrum can either be automatically generated using the
functions given in EC8-1 4.2.4 and the parameters provided by the user, or a custom spectrum can be
defined in the Spectral function editor.
Spectral function editor to create custom spectra
Display time period of calculated modal shapes: Turns on/off the display of time periods provided by the
vibration analysis. Time periods are marked by a vertical pink line at the respective T value.
The automatically generated spectrum uses the following parameters:
agR – Reference peak ground acceleration on rock m/s2
q – Behaviour factor for the reduction of horizontal seismic effects as per EC 8-1 3.2.2.5 (3)P. After
checking Different q factors in X and Y directions, independent qX and qY values can be defined by the
user.
User’s Manual X6 Release1 e2 341
S, TB, TC, TD – Soil-specific parameters that define the spectral shape. If the type of soil is selected from
the list of soil classes, these parameters are automatically loaded according to Tables 3.2 and 3.3 in EC
8-1. Every value can be overridden by the user.
β – Lower bound factor for the horizontal design spectrum. The recommended value in EC 8-1 3.2.2.5
(4)P is 0.2.
γI – Importance factor as per EC 8-1 4.2.5
qd – Displacement behaviour factor as per EC 8-1 4.3.4 (1)P
If the model contains seismic isolators a modified spectrum is used. See 4.9.19 Seismic isolators
Custom spectra are defined in the Spectral function editor after setting the Design spectrum from Para-
metric shape to Custom. Spectra are created in the editor by a piecewise linear approximation of the
spectral shape over a list of pre-defined natural period – spectral acceleration coordinate pairs.
Coordinates can be edited, copied to or pasted from other programs.
The second tab of the Seismic load dialog window corresponds to the parameters of the vertical re-
sponse spectrum. Note that the effect of seismic excitation in the vertical (Z) direction is only taken
into account if a vertical response spectrum is defined. Vertical loads are not generated by default.
Parameters and the layout of this tab are identical to the first one.
A third tab is Direction where the angle of the seismic action can be set. If this
value is not zero seismic loads will be generated not in the global X and Y di-
rections but in X* and Y* directions at the α angle to add the global directions.
The fourth tab is Storeys. It lets the user decide if torsional effects due to accidental eccentricity shall
be taken into account. If the user wishes to consider such effects, then the program calculates additional
torsional moments around the vertical axis for every storey and every mode shape. The magnitude of
torsional moments depends on the horizontal load and eccentricity of each storey.
Horizontal loads are obtained from load cases corresponding to each vibration mode and horizontal
direction.
The size of eccentricity relative to the total size of each storey is defined by the user. An accidental
eccentricity of 5% is recommended in EC 8-1 4.3.2 (1)P.
342
The last tab of the Seismic Load dialog window provides control over the combination of results for
individual modes and individual directions. Results from individual modes can be combined using either
the Square Root of Sum of Squares (SRSS) or the Complete Quadratic Combination (CQC) method. The
latter is considered more appropriate if the vibration modes of the structure are not well separated (i.e.,
vibration frequencies are close to each other).
Selecting the automatic option lets the program automatically decide if application of the CQC method
is warranted by the vibration results. The program considers modes i and j well separated if the following
condition holds: Tj / Ti < 0.9.
Results in the two horizontal and the vertical direction can be combined using either of the two com-
monly used combination methods displayed in the dialog window.
The seismic effects in various directions can also be summed up manually by creating custom load
combination. SMX, SMY and SMZ and load cases connected to these (see below) are also available in
the table of load combinations, see 4.10.2 Load combinations.
Warning! Multiple load cases belonging to the same direction or SMX, SMY, SMZ and SM+/- load
cases must not be taken into account in the same combination collectively.
After the load parameters are set and the Seismic Load dialog window is closed by clicking the OK
button, the program generates a load case for each mode and direction according to the settings de-
fined by the user. The following naming convention is used for these load cases:
Load cases ending in 01X, 02X, ....nX, 01Y, 02Y, ....nY, 01Z, 02Z, ....nZ are the equivalent forces in X,
Y or Z direction corresponding to individual mode shapes.
Load cases ending in 01tX, 02tX, ....ntX, 01tY, 02tY, ....ntY are the torsional moments due to accidental
eccentricity in X or Y direction.
User’s Manual X6 Release1 e2 343
1 = Xa + Ya + Z
2 = Xa + Yb + Z
3 = Xb + Ya + Z
4 = Xb + Yb + Z
MRSA results are absolute values by definition. Therefore, the load cases presented above would contain
only positive values. The least favorable loading scenario is a combination of seismic effects and effects
from other sources such as gravity loading. Creation of this load combination is facilitated in the
program by providing two load cases for each seismic load: seismic effects with only positive and seismic
effects with only negative values are identified by a + and a – sign at the end of the name of their load
cases. Note that the absolute values of corresponding responses (internal forces, displacements, etc.)
are identical in the + and – load cases.
Displacement results shown in the Static tab are automatically scaled by the qd factor specified in the
Seismic Load dialog window.
Second order effects due to geometric nonlinearity shall be taken into consideration when evaluating
the results of MRSA. The program supports such calculations with the data in the Seismic sensitivity of
storeys table in the Table Browser. Independent results are available in the table for each horizontal
direction.
The following quantities are calculated by the program:
θmax – plastic stability index, also known as the inter-storey drift sensitivity coefficient, as per
EC8-1 4.4.2.2 (2)
Ptot – total gravity load above the storey
Vtot – total seismic storey shear
dmax – design inter-storey displacement calculated as the relative displacement of corresponding sto-
rey centroids
S– location of the shear centre
Gm – location of the centroid
M– storey mass
Imz – moment of inertia at the centroid about the Z axis
344
AxisVM does not consider the influence of second order effects automatically through internal force
amplification. The user shall evaluate the sensitivity of the structure to such effects. If the influence of
second order effects can be taken into consideration through force amplification, then the user shall
determine the appropriate value of the amplification factor as per EC8-1 4.4.2.2 (3). That amplification
factor can be specified in the program as the fse factor in the following design tools:
The calculation procedure, dialog windows, assumptions and features are identical to those presented
in the general Eurocode 8 section. The following paragraphs list only the modifications that are
unique to the SIA 261 standard.
3. Set seismic load parameters
The following modifications are applied at the Seismic Load dialog window:
- Horizontal and vertical response spectra are based on SIA261 16.2.4.
- Soil-specific parameters are set according to Table 25 in SIA261.
The calculation procedure, dialog windows, assumptions and features are identical to those presented
in the general Eurocode 8 section. The following paragraphs list only the modifications that are
unique to the DIN EN 1998-1/NA (EC8-1 NA)standard.
3. Set seismic load parameters
The following modifications are applied at the Seismic Load dialog window:
- Horizontal and vertical response spectra are based on EC8-1 NA NDP 3.2.2.5(4)P.
- Soil-specific parameters are set according to Tables NA.4 and NA.5 in EC8-1 NA.
The calculation procedure, dialog windows, assumptions and features are identical to those presented
in the general Eurocode 8 section. The following paragraphs list only the modifications that are
unique to the latest NTC standard.
3. Set seismic load parameters
The following modifications are applied at the Seismic Load dialog window:
- Horizontal and vertical response spectra are based on NTC 2018 3.2.3.2.
- Soil-specific parameters are set according to Tables 3.2.V-VII in NTC 2018 3.2.3.2.
User’s Manual X6 Release1 e2 345
- If the model contains seismic isolators a modified spectrum is used, depending on the limit
state (SLO, SLD, SLV, SLC) set in the seismic parameters dialog according to NTC 2018 3.2.1.
See 4.9.19 Seismic isolators. As multiple seismic actions can be defined for the same model,
seismic results obtained for different limit states can be compared.
- Results of the seismic isolator system are available in a separate table.
The calculation procedure, dialog windows, assumptions and features are identical to those presented
in the general Eurocode 8 section. The following paragraphs list only the modifications that are
unique to the P100-1 standard.
3. Set seismic load parameters
The following modifications are applied at the Seismic Load dialog window:
- Horizontal and vertical response spectra are based on P100-1 3.1(7), 3.1(13), and 3.2(1).
The calculation procedure, dialog windows, assumptions and features are identical to those presented
in the general Eurocode 8 section. The following paragraphs list only the modifications that are
unique to the NPR 9998:2018 standard.
3. Set seismic load parameters
The following modifications are applied at the Seismic Load dialog window:
- Horizontal and vertical response spectra are based on NPR 9998:2018 3.2.2.2.3.
The ag;d acceleration and p, TB, TC, TD parameters shall be specified by the user according to the code or
the code conforming NPR 9998 web tool.
Pushover loads are generated according to the regulations of Eurocode 8 (EN 1998-1:2004) by default.
The load generation uses undamped free vibration frequencies and corresponding mode shapes of the
model, therefore loads can only be generated if a vibration analysis has already been performed.
Steps of pushover The following description shows how to create pushover load cases and set their properties before
load generation performing a nonlinear static analysis.
1. Calculate vibration mode shapes and frequencies.
When running the vibration analysis be sure to use the convert loads to masses option with the
appropriate load case if there are loads defined that need to be considered static. Check the table of
Modal mass factors in the Table Browser. Vibration results will appear only if the Vibration tab is selected.
Although there is no requirement in Eurocode 8 for the minimum value of seismic equivalence coeffi-
cient, it is strongly advised to perform standard pushover analysis only on structures having clearly
dominant mode shapes in each horizontal direction. The coefficients for each mode shape are listed in
the Modal mass factors table. Unlike seismic loads, standard pushover load generation uses a single
vibration mode shape for each load case, therefore the sum of seismic equivalence coefficients is not
important. Thus, there is no need to calculate a large number of modes if the dominant ones are among
the first few.
346
The control node shall be one of the nodes at the top of the structure. It is important to set the direction
of the analysis according to the direction of the parametric load case. The stability of the analysis can
be increased significantly by increasing the number of increments. Following geometric nonlinearity is
recommended for pushover analyses. The analysis is started by clicking the OK button.
Generation of capacity curves and related results are explained in 6.1.4 Pushover capacity curves chapter.
After selecting an imperfection load case, the above icon for imperfection load becomes enabled. After
setting the imperfection parameters the global imperfection is applied to the structure (its displayed
amplitude is magnified to make it more visible).
Imperfection load cases can contribute to load combinations used to perform analysis with geometric
nonlinearity. Nodes will be shifted from their original positions and the other loads in the combination
will be applied to the distorted structure.
Base level It is the Z0 level where the sway begins. Two options are available
• Set it to the Lowest point of the model
• Set it to a custom Z0 level
Structure height The structure height is measured from the Z0 base level. Available options are
from base level • Set it from the highest point of the model
• Set it to a custom h value
Steel/Reinforced Φ = Φ0 𝛼ℎ 𝛼𝑚
concrete
where
2 2
𝛼ℎ is a reduction factor: 𝛼ℎ = with ≤ 𝛼ℎ ≤ 1
√(ℎ[𝑚]) 3
1
𝛼𝑚 is a reduction factor: 𝛼𝑚 = √0.5 (1 + ) where m is the number of columns involved per level
𝑚
Masonry ν=
1
100√ℎ
Other * Arbitrary value can be specified by the user.
348
The only parameter of such load cases is an imperfection based on buckling shapes. Whenever a load
combination contains an imperfection load case, AxisVM calculates with imperfect geometry. Though
imperfection load cases based on buckling shapes can be created, deleted and edited, but load cases
using imperfect shapes are created and deleted automatically according to definition and deletion of
imperfect shapes.
For definition of imperfect geometry see 6.2.1 Creating imperfect geometry from buckling shapes – IMP
module
Tendons can be assigned to a continuous selection of beam, rib, or domain elements. After defining
tendon properties and the tensioning process AxisVM determines the stress distribution in tendons im-
mediately after anchoring and the equivalent loads at the end of tensioning (load case name-T0). After
completing a static analysis, the program determines the time dependent losses of prestress and the
long-term equivalent loads from the result of quasi-permanent combinations (load case name-TI). Ten-
don trajectory tables can be generated with user-defined steps.
Click on this icon of the Loads page to assign tendons to the selected beam or rib elements.
Coordinate systems The geometry of a tendon can be defined in a fixed coordinate system (left icon) or in a moving coordi-
nate system (right icon).
The fixed coordinate system defines x axis on the straight line connecting the first and last node.
Axis z is perpendicular to axis x and the initial cross section’s y axis. While the direction of axis y is deter-
mined in a right-handed coordinate system.
If tensioned elements are not on a straight line but defined by an arc, or a polygon, a moving coordinate
system can be chosen. If the beam is an arc, the coordinate system moves along this arc. If the structure
is defined by a polygon, the geometry is interpolated by a cubic spline. The x axis of this coordinate
system fits to the tangent of the generated curve, while z axis z follows the local z axis of the beam at
each and every point. The y axis is defined by the right-hand rule.
User’s Manual X6 Release1 e2 349
Uncoiled view The position of the tendon is displayed in the working plane. Continuous lines represent
the section of the working plane and the elements. Dashed lines show the center line
and the projection of the beams’ edges on the uncoiled working plane.
3D view The 3D view shows the cable and the elements in 3D.
The base points of the geometry and the inflection points are displayed both in the uncoiled view and
3D view.
Icons on the vertical toolbar beside the tendon list are
Add new tendon
Geometry for the new tendon can be de-
fined using the toolbar beside the dia-
gram. First the tendon properties must
be defined.
Segments (2D) The geometry of the tendon is defined as a series of segments (straight-parabola-arc connections).
Concrete cover If the mouse is near the top or bottom edge of the section
the cursor finds the edge. If Concrete cover is checked the
point entered will not be on the edge but inside the section
as defined by d.
This function works both for segments and splines.
Drawing tendons if Click the icons on the vertical toolbar beside the drawing and enter base points. AxisVM determines the
Spline (2D/3D) trajectory passing through these base points as a cubic spline to minimize curvature. For each basepoint
is selected the angles of tangent can be specified by setting the 𝛼 (top view) and 𝛽 (side view) values in the table.
Enter values between –180° and 180°. Initial values are 0°. Existing base points can be dragged to a new
position using the mouse.
Draw tendon in 2D.
Base points can be created by clicking the diagram or using the coordinate window. Double-click or right
click and choose Complete from the popup menu to end the spline. The tendon position within the cross-
350
section has to be specified only at the first base point. Further base points will be in the local x-z
(uncoiled) plane containing the first base point.
Steps of drawing a tendon in 2D:
1. Select the x position of the cross-section where the tendon basepoint will be defined.
2. Place the tendon to the proper position in the cross-section view.
3. Place the other points of the tendon in the uncoiled view.
Draw tendon in 3D.
The tendon position within the cross-section has to be specified at every basepoint. To end tendon def-
inition, click the right mouse button and select Complete from the popup menu.
Steps of drawing a tendon in 3D:
1. Select the position of the cross-section where you want to define a tendon basepoint.
2. Set the proper tendon position in the cross-section view.
Repeat Step 1 and Step 2 to define all basepoints.
Add new base point
Click the cable to add a new base point. In case of multiple tendons this function works on the active
tendon.
Delete base point
Click on an existing base point to delete it. After deleting the second base point the entire tendon is
deleted. In case of multiple tendons this function works on the active tendon.
Table of base points Base point properties can be edited in the
table. x, y, z are the Cartesian coordinates
of the base points. If Tangent is checked, α
and β (the angle between the x direction
and the tangent on the x-y and x-z plane)
can be edited.
Defining segments Each segment is built from one, two or three basic subsegments. Subsegment ge-
ometry can be chosen from the toolbar icons beside the tendon diagram. The one
letter abbreviations are:
C arc of a circle with a defined radius
P parabolic arc
L straight line
x no intermediate subsegment
Options:
Basic Click on L, C or P symbols (or select L, C or P from the dropdown
list) to append a single line, circular or parabolic arc subsegment
to the existing geometry with tangent continuity.
Composite Click on the three columns of symbols (or select two or three
letter codes from the dropdown list) to set a composite segment
made of two or three subsegments.
Table of segments The properties of the defined segments can be modified in the Table of segments. Editable values and
functions are listed below
Note: not all geometrical parameters are available for different segment types. For instance, the tangent
at the start cannot be modified for straight line-arc segments, as the tangent is already defined by the
coordinates of the start and end points and the radius of the arc.
Tendons selected in the tree can be translated or mirrored. Tendons can be copied or just moved. Copied
tendons inherit the original parameters and the tensioning process assigned to them.
Options. Separate grid and cursor settings for the longitudinal and cross-section tendon diagram.
Concrete parameters Under the concrete parameters the material properties of the concrete can be
checked. Moreover, 𝑒𝑐𝑠 (∞), the long-term value of the concrete shrinkage strain
can be modified.
Tensioning The second tab is to define the tensioning process for tendons by determining the order of certain op-
process erations. Click on the proper symbol to add a new operation to the list.
Anchor on left / right / both side Wedge draw-in of the anchorage device
Note: left is the starting point, right is the end point of the cable.
User’s Manual X6 Release1 e2 353
Include tension loss due to sequential stressing If more than one cable relates to a structural element,
then these cables can be stressed at different time in-
stant. Because of it, stress loss arises in the previously
anchored tensioned cables. With this option the user
can take into account this phenomenon.
Number of steps If the user takes into account the loss from sequential
tensioning, then the number of the tensioning steps
should be set up here.
Results If valid material parameters, geometry and tensioning processes are assigned to every tendon, result
diagrams are displayed on the Results tab. The upper diagram shows the ratio of the actual tension and
the characteristic tensile strength of the selected tendons. The lower diagram shows the chosen equiv-
alent distributed load component for the selected tendons.
Component the component of the equivalent load system can be chosen. Three force compo-
nents and three moment components are available.
Time the stresses in the cable and the equivalent load component can be illustrated
right after the tensioning (choosing T=0) and after long term losses (choosing
T=∞). The components, relating to T=∞ can be chosen only after a static analysis.
1. Tension loss due to friction between tendons and their sleeves at position x measured from the an-
chorage point along the tendon is calculated as
𝜎𝜇 (𝑥) = 𝜎𝑚𝑎𝑥 (1 − 𝑒 −𝜇(Θ+𝑘𝑥) ),
where
𝜎𝑚𝑎𝑥 is the maximum tension in the tendon
Θ is the sum of the absolute angular displacements over a distance x
Long term loss of force due to shrinkage and creep of the concrete and the relaxation of the tendon is
calculated as
𝐸𝑝
𝜖𝑐𝑠 𝐸𝑝 + 0.8Δ𝜎𝑝𝑟 + 𝜑𝜎
𝐸𝑐𝑚 𝑐,𝑄𝑃
Δ𝑃𝑐+𝑠+𝑟 = 𝐴𝑝 Δ𝜎𝑐+𝑠+𝑟 = 𝐴𝑝 ,
𝐸𝑝 𝐴𝑝 𝐴 2
1+ (1 + 𝑐 𝑧𝑐𝑝 )[1 + 0.8𝜑]
𝐸𝑐𝑚 𝐴𝑐 𝐼𝑐
where
Δ𝜎𝑐+𝑠+𝑟 is the tension loss due to the effects above
𝐸𝑐𝑚 is the secant modulus of elasticity of concrete
Δ𝜎𝑝𝑟 is the long-term absolute tension loss due to the relaxation of tendons
in case of 2nd relaxation class:
Δ𝜎𝑝𝑟 = 𝜎𝑚𝑎𝑥 ⋅ 0.6𝜌1000 𝑒 9.1𝜇 ⋅ 5000.75(1−𝜇) ⋅ 10−5 ,
Trajectory table The last tab is to build a trajectory table for the
selected tendons with the desired increment and
optional shift of origin. The trajectory table con-
sists of the local y and z coordinates of the se-
lected tendons at the calculated x positions.
The defined basepoints are always displayed in the
trajectory table.
Copies the drawing on the active tab to the Clipboard as a Windows metafile. This way the diagram can
be pasted to other applications (e.g., Word).
Copy diagram
Ctrl+C
Prints a report of the tensioning using diagrams and tables. Tendons and report items can be selected.
You can choose the position of the drawing (landscape or portrait) and set the scale of it (Print options
Print for drawings).
Ctrl+P
User’s Manual X6 Release1 e2 355
Menu You can reach the following functions via the menu:
File
Edit
Undo/Redo Undoes the effect of the previous command / Executes the command which was undone.
Coordinates Editing of the longitudinal and cross-section diagrams is made easier by a coordinate window.
The display of this window can be turned on and off.
Status On diagrams an information window appears displaying diagram-specific information. Display of this
window can be turned on and off.
On the Loads page by clicking on this icon and selecting domains, tendons can be assigned to the
selected elements.
Domain Tendons are displayed in top view showing geometry of the curve on the middle
plane (baseline).
Uncoiled view Tendon elevation perpendicular to the middle plane as a function of x (the arc length
position on the baseline).
3D view The 3D view shows the cable and the elements in 3D.
Note: The base points of the geometry and the inflection points are shown both in the uncoiled view
and 3D view.
User’s Manual X6 Release1 e2 357
Tendons selected in the tree can be translated or mirrored. Tendons can be copied or just moved. The
Δx, Δy coordinates are defined in the tendon coordinate system (see below). Copied tendons inherit
the original parameters and the tensioning process assigned to them.
Delete tendon. Deletes the selected tendon.
Options. Separate grid and cursor settings for the longitudinal and cross-section tendon diagram.
Concrete parameters Under the concrete parameters the material properties of the concrete can
be checked. Moreover, 𝑒𝑐𝑠 (∞), the long-term value of the concrete shrinkage
strain can be modified.
Tendon geometry The first step is to draw the baseline in top view on the Domain tab using the segment-based or spline-
based geometry tools the same way as in case of post-tensioning of beams. As the baseline must be in
the middle plane of the domain spline definition is limited to 2D. Each tendon has its own coordinate
system (see below).
The second step is to draw the elevation curve along the baseline on the Uncoiled view tab using the
definition method selected in the Tendon properties dialog.
Checking Concrete cover and entering the required value helps tendon definition the same way as in the
case of beams.
358
Trajectory Click on this icon to set the origin (first click) and the local X direction (second click) of the trajectory
coordinate system coordinate sytem (one system for all tendons).
Tensioning The second tab is to define the tensioning process for tendons by determining the order of certain op-
process erations just like for beam tensioning.
Include tension loss During post-tensioning of domains, the elastic loss due to sequential stressing
due to sequential is neglected.
stressing
Results If all tendons have valid material parameters, geometry and tensioning process, result diagrams are
displayed on the Results tab. The upper diagram shows the ratio of the actual tension and the charac-
teristic tensile strength. The lower diagram shows the chosen equivalent distributed load component.
In case of domains only one tendon can be chosen for post-tensioned surfaces at once.
Component The component of the equivalent load system can be chosen. Three force components
(q_x, q_y and q_z) and three moment components (m_x, m_y and m_z) are available.
Time The stresses in the cable and the equivalent load component can be illustrated right
after the tensioning (choosing T=0) and after long term losses (choosing T=∞).
The components, relating to T=∞ can be chosen only after a static analysis.
The calculation of stress loss is done the same way as it was introduced before, for post-tensioned
beams. The calculation of stress losses right after tensioning neglects the elastic shrinking of concrete.
Trajectory table At the last page trajectory table can be created the same way like for beams. The coordinate system for
the trajectories can be defined on the Tendons page.
For the main menu and reporting functions see 4.10.28.1 Post-tensioning of beam and rib elements
User’s Manual X6 Release1 e2 359
Moving loads allow modeling of a drifting load with a constant intensity like a vehicle crossing a bridge
or a crane carriage moving along its runway.
To define a moving load a moving load case must exist. It can be created on the Loads tab clicking the
Load cases and load groups icon. See 4.10.1 Load cases, load groups. Moving load icons will be enabled
only if the current load case is a moving load case. After defining the load new load cases will be created
automatically according to the number of steps specified. Auto-created load cases are maintained and
updated automatically and they cannot be deleted or moved into another load group individually. In-
creasing the number of load steps will create additional load cases. Decreasing this number makes
certain load cases useless. These excess load cases will be removed only before saving the model.
Moving load symbols can be displayed in two ways. The first option is to draw the current phase only.
The second one is to draw other phases in gray.
Open Table Browser to see the load and the load path in tabular format. These tables can be also used
for reporting purposes.
Moving load on line elements is a load pattern moving on a user-defined load path in N steps. The load
pattern can contain any combination of concentrated and distributed loads.
Individual loads in the pattern can be local or global and their position, eccentricity and intensity com-
ponents can be set. This way the vertical load of a crane carriage and the horizontal forces can be
applied together on the runway. Load eccentricity is always parallel to the local y axis. If it is on the left
side when moving along the path its eccentricity is negative. If it is on the right side, its eccentricity is
positive.
Loads can be added to the pattern by clicking the plus icon and filling out the fields in the row. Selected
rows can be deleted by clicking the Delete icon under the plus icon.
Load patterns can be saved under a name and reloaded.
After load pattern definition it is necessary to select the load path. It must be a continuous sequence of
beams or ribs. After selecting the elements constituting the load path the start point and end point has
to be selected. These points must be nodes along the path.
Beside the load path button, the value of N can be set. It determines the number of steps the load
pattern will make evenly along the path.
360
The local z direction of the load pattern will always be the local z direction of the line elements it is
placed on. Lengthening, shortening or breaking a line element of the path will lead to an automatic
recalculation of the load phases.
In the first phase the load with the lowest coordinate in the pattern will be placed over the start point.
In the last phase the load with the highest coordinate in the pattern will be placed over the end point.
Crane runway mode
In the first phase the load with the highest coordinate in the pattern will be placed over the start point.
In the last phase the load with the lowest coordinate in the pattern will be placed over the end point.
Bridge mode
One way: Load moves from start point to end point in N steps.
Round trip: Load moves from start point to end point and back in 2N steps.
This load type is convenient when vehicle loads have to be defined. The load pattern consists of concen-
trated or rectangular surface loads pairs representing the wheels on the axles.
u is the vehicle gauge, a and b refer to the rectangle dimensions. Axle load F will be distributed evenly
on the two wheels. Load patterns can be saved under a name and reloaded.
The load type and direction switches on the left determines the properties of all loads
entered into the table.
Loads can be added to the pattern by clicking the plus icon and filling out the fields in the row. Selected
rows can be deleted by clicking the Delete icon under the plus icon.
After load pattern definition it is necessary to select the load path. It must be a continuous polyline
running through domains or load panels.
The load path does not have to stay in the same plane and can cross holes or empty areas
between domains and load panels.
Path start point and end point is the first and last point of the polyline.
Each phase will contain only the loads actually falling on a domain or load panel. The local z direction of
the load pattern will be the local z direction of the domain or load panel it is placed on. In case of a path
User’s Manual X6 Release1 e2 361
running along the edge of two or more domains in different planes only the domains in the active parts
are taken into account. The local z direction will be chosen finding the domain with the minimum angle
between local z and global Z directions.
Beside the load path button, the value of N can be set. It determines the number of steps the load pattern
will make evenly along the path.
In the first phase the load with the lowest coordinate in the pattern will be placed over the start point. In
the last phase the load with the highest coordinate in the pattern will be placed over the end point.
Changing geometry of the domain/load panel will lead to an automatic recalculation of the load phases.
In the first phase the load with the lowest coordinate in the pattern will be placed over the start point. In
the last phase the load with the highest coordinate in the pattern will be placed over the end point.
Crane runway
mode
In the first phase the load with the highest coordinate in the pattern will be placed over the start point.
In the last phase the load with the lowest coordinate in the pattern will be placed over the end point.
Bridge mode
One way: Load moves from start point to end point in N steps.
Round trip: Load moves from start point to end point and back in 2N steps.
Dynamic nodal loads and acceleration functions can be defined for time-history analysis.
Acceleration functions can be used for seismic analysis. In this case it is recommended to
obtain proper seismic accelerograms and assign these functions to support nodes to analyse the effects
of the earthquake. This method provides more exact results than the response spectrum analysis and
can be used even if nonlinear elements are defined in the model (nonlinear supports, tension-only
trusses, etc.). Its disadvantage is that it cannot be combined with other load types automatically.
To define nodal loads or acceleration functions the current load case must be a dynamic load case.
See 4.10.1 Load cases, load groups
Defining functions
Dynamic loads and accelerations are defined by functions which describe the parameter in time. Function
editor is available from the dynamic load definition dialogs.
Functions must be entered as value pairs in a table. Functions are plotted automatically and can be print-
ed. Functions can be reused. In order to make them available later, save them into the function library.
362
Saved functions can be reloaded, edited and saved under a new name. Functions are saved into separate
*.dfn files in folder c:\ Users \ [user name] \ AppData \ Roaming \ AxisVM \ [version number] \ dfn.
If the total duration of the dynamic analysis exceeds the length of the function in time the parameter values
beyond the end of the function are assumed to be zero.
Formula editing.
The f(t) load function can be entered as a formula. Following operators
and functions are available: +, –, * , /, (, ), sin, cos, tan, exp, ln, log10, log2,
sinh, cosh, tanh, arcsin, arccos, arctan, arcsinh, arccosh, arctanh, int, round,
frac, sqr, sqrt, abs, sgn, random.
random(t) returns a random number between 0 and 1.
A machine rotating about the Y axis has a dynamic load function with the
following X and Z components:
Linear
The original function is considered as piecewise linear and the values of the new function are ob-
tained from the linear segments. In the following picture the original function (blue line) and the
new resampled function (red line) can be seen together.
This method is recommended if the original function is really piecewise linear. If the original function
is a sampled continuous function (e.g., measurement record or a function generated from Fourier
spectral composition), this method cannot approximate the original curved continuous function be-
tween the sample points. Moreover, in the preceding picture several local extreme values of the
original function are lost, because the sampling instances of the new function does not coincide with
the original ones. In such cases the Whittaker-Shannon interpolation method is recommended.
User’s Manual X6 Release1 e2 363
Whittaker-Shannon
The original function is considered as an equidistant sampled continuous function and the values of
the new function are obtained from the reconstructed continuous function.
In the following picture the original function (blue line), the reconstructed continuous function
(green curve) and the new function (red line) can be seen together. It can also be seen that the
originally unsampled extreme values of the continuous function appear in the reconstructed func-
tion.
𝑡 − 𝑡𝑖
1 + cos (𝜋 )
0.5𝑇𝑤
𝑖𝑓 |𝑡 − 𝑡𝑖 | ≤ 0.5𝑇𝑤
𝑊𝑖 (𝑡) = 2
{ 0 𝑖𝑓 |𝑡 − 𝑡𝑖 | > 0.5𝑇𝑤
where 𝑇𝑤 is the length of the cosine wavelet. It is set to 10% of the length of the record.
The following picture shows the original Whittaker-Shannon interpolation function (black curve), the
cosine shape factor (blue curve) and the resultant finite length interpolation function (red curve).
364
In the following picture the wavy ends obtained with the original Whittaker-Shannon interpolation
formula (black curve) and smooth end with the wavelet interpolation (red curve) can be seen to-
gether. The blue circles show the sample points of the original function. It can be seen that in the
middle of the large amplitude part of record the reconstructed continuous functions obtained with
the different interpolation functions almost coincide. But at the ends only the red curve is smoothed.
The Whittaker-Shannon interpolation method is available only for functions obtained from equi-
distant sampling.
Correcting the acceleration record for producing zero final velocity and/or zero final displacement
User’s Manual X6 Release1 e2 365
Measured earthquake acceleration records usually do not produce exactly zero velocity and zero dis-
placement at the end of the record, because of measurement inaccuracies. The nonzero final velocity
appears as an infinite motion of the supports with constant velocity beyond the end of the record. The
nonzero final displacement causes that the supports stop out of their original position at the end of the
record. Any of these inaccuracies can be eliminated by applying a small modification to the acceleration
record. The type of correction can be selected by the checkboxes. The blue graph shows the original
record and the red graph shows the corrected one. The correction rate is defined by the ratio of the RMS
values of the modification and of the original record.
The velocity and the displacement at the end of the record are
𝑇 𝑇
𝑉 = ∫0 𝑎(𝑡)𝑑𝑡 U = ∫0 (𝑇 − 𝑡)𝑎(𝑡)𝑑𝑡
A function previously saved to the library can be loaded by selecting its name from the dropdown list.
The first point of functions must be at t=0. This value pair cannot be changed or deleted. If the
load is applied only at T > 0, the function value must be zero between 0 and T.
It is possible to define a constant (time-independent) load by selecting <Static> from the Dynamic load
functions combo.
The actual value of a load component in t will be calculated as Fi (t ) = Fi f (t ) , i.e., the load
intensity is multiplied by a time-dependent load factor.
If a dynamic load is defined for a support with an existing dynamic load the existing load will be
overwritten.
Modify, delete Dynamic loads can be modified or deleted the same way as static loads.
Dynamic loads are displayed as dashed yellow arrows.
User’s Manual X6 Release1 e2 367
Load position and direction can be set the same way as described in 4.10.5 Point load on domain or load
panel.
A load intensity and a time dependent function can be specified for each component.
Functions can be created and edited in the Function Editor or loaded from the function library.
The actual value of the load at a given time t will be the load intensity multiplied by the time
dependent function: 𝐹𝑖 (𝑡) = 𝐹𝑖 ⋅ 𝑓𝑖 (𝑡)
Dynamic force is displayed as dashed yellow arrow.
The load polygon and load direction can be defined as described in 4.10.10 Domain / Load panel area
load.
A load intensity and a time dependent function can be specified for each component.
Functions can be created and edited in the Function Editor or loaded from the function library.
The actual value of the load at a given time t will be the load intensity multiplied by the time
dependent function: 𝐹𝑖 (𝑡) = 𝐹𝑖 ⋅ 𝑓𝑖 (𝑡)
Dynamic force is displayed as a green area.
Acceleration acts at the bottom of the support string. The acceleration of the supported node can
be different depending on the support stiffness.
If acceleration is defined for a support with an existing acceleration load the existing load will be
overwritten.
If multiple nodal supports are attached to a node, acceleration acts on all supports.
Modify, delete Dynamic support acceleration can be modified or deleted the same way as a static load.
Dynamic support acceleration is displayed as a circle and a yellow arrow.
368
If acceleration is defined for a support with an existing acceleration load the existing load is over-
written.
To specify ground acceleration for seismic analysis nodal support accelerations must be defined.
Modify, delete Dynamic nodal acceleration can be modified or deleted the same way as a static load.
Dynamic nodal acceleration is displayed as a circle and a yellow arrow.
EC DIN EN 1991-1-2:2010-12 NA
German DIN EN 1993-1-2:2010-12 NA
EC MSZ EN 1991-1-2:2005 NA
Hungarian MSZ EN 1993-1-2:2013 NA
EC SR EN 1991-1-2:2004/NA:2006
Romanian SR EN 1993-1-2:2006/NB:2008
EC STN EN 1991-1-2/NA:2006
Slovakian STN EN 1993-1-2/NA:2008
EC PN EN 1991-1-2:2006 NA
Polish PN EN 1993-1-2 NA
EC DS/EN 1991-1-2 DK NA:2011
Danish
EC BS EN 1991-1-2:2002 NA
British BS EN 1993-1-2:2005 NA
EC SFS-EN 1991-1-2:2002 NA
Finnish SFS-EN 1993-1-2:2005 NA
EC NS-EN 1991-1-2:2002/NA:2008
Norwegian NS-EN 1993-1-2:2005/NA:2009
EC NEN EN 1991-1-2:2002/NB:2011
Dutch NEN EN 1993-1-2+C2:2011/NB:2015
User’s Manual X6 Release1 e2 369
Choosing a Fire effect is represented by a fire curve specifying the gas tem-
fire curve perature in the fire compartment as a function of the time.
Standard temperature-time curve (so called ISO fire curve) (EN 1991-1-2):
𝜃𝑔 = 20 + 345 log10 (8𝑡 + 1)
External fire curve (EN 1991-1-2):
𝜃𝑔 = 660(1 − 0.687𝑒 −0.32𝑡 − 0.313𝑒 −3.8𝑡 ) + 20
Hydrocarbon fire curve (EN 1991-1-2):
𝜃𝑔 = 1080(1 − 0.325𝑒 −0.167𝑡 − 0.675𝑒 −2.5𝑡 ) + 20
Parametric fire curve:
∗ ∗ ∗
𝜃𝑔 = 20 + 1325(1 − 0.324𝑒 −0.2𝑡 − 0.204𝑒 −1.7𝑡 − 0.472𝑒 −19𝑡 ) 𝑡 ∗ ≤ 𝑡𝑚𝑎𝑥
∗
Cooling phase:
𝜃𝑔 = 𝜃𝑚𝑎𝑥 − 625(𝑡 ∗ − 𝑡𝑚𝑎𝑥
∗ ∗
∙ 𝑥) 𝑡𝑚𝑎𝑥 ≤ 0.5
∗ )(𝑡 ∗ − 𝑡𝑚𝑎𝑥
∗ ∗
𝜃𝑔 = 𝜃𝑚𝑎𝑥 − 250(3 − 𝑡𝑚𝑎𝑥 ∙ 𝑥) 0.5 ≤ 𝑡𝑚𝑎𝑥 <2
𝜃𝑔 = 𝜃𝑚𝑎𝑥 − 250(𝑡 ∗ − 𝑡𝑚𝑎𝑥
∗ ∗
∙ 𝑥) 𝑡𝑚𝑎𝑥 ≥2
where
𝜃𝑔 [°C] gas temperature
𝑡 [min] time
𝑡 ∗ [h] modified time parameter
370
Parametric Unlike the ISO-, external and hydrocarbon fire Temperature [°C]
fire curve curves, parametric fire curves have a cooling phase.
Using a parametric fire curve, parameters like the
size and ventilation of the fire compartment,
amount of the combustible materials, etc., can be
considered in the calculation. These parameters
considerably influence the maximum gas tempera-
ture and the duration of the fire.
Time [s]
According to EN 1991-1-2, parametric fire curves are valid for fire compartments up to 500 m 2 of floor area,
without openings in the roof and for a maximum compartment height of 4 m.
The following parameters need to be defined for the calculation of the fire curve:
𝐴𝑓 [m2] floor area
𝑞𝑓,𝑑 [MJ/ m2] design value of the fire load density related to the surface area Af of the floor
𝐴𝑣 [m2] total area of vertical openings on all walls
ℎ [m] weighted average of window heights on all walls
𝐴𝑡 [m2] total area of enclosure (walls, ceiling and floor, including openings)
𝑐𝑝𝑤 [J/kg°C] specific heat of boundary of enclosure
𝜌𝑝𝑤 [kg/m3] density of boundary of enclosure
𝜆𝑝𝑤 [W/m°C] thermal conductivity of boundary of enclosure
If the boundary of enclosure consists of different layers, 𝒄𝒑𝒘, 𝝆𝒑𝒘 and 𝝀𝒑𝒘 parameters should be given
𝑨 𝑨
resulting parameter 𝒃 = √𝒄𝒑𝒘 𝝆𝒑𝒘 𝝀𝒑𝒘 equal to 𝒃 = √𝒄𝒑𝒘,𝟏 𝝆𝒑𝒘,𝟏 𝝀𝒑𝒘,𝟏 ∑ 𝟏 + ⋯ + √𝒄𝒑𝒘,𝒏 𝝆𝒑𝒘,𝒏 𝝀𝒑𝒘,𝒏 ∑ 𝒏 .
𝑨 𝑨
In case of NEN EN standard, parametric fire curve cannot be selected, but another prescriptive fire
curve, namely RWS tunnel fire curve is available.
User-defined User-defined fire curve is also available to define the fire effect, thus the program allows to perform fire design
fire curve considering simulation or fire test results.
It is important to note that the calculation of steel temperature based on prescriptive fire curves uses the
assumption that the gas temperature is identical everywhere in the fire compartment and the steel element is
exposed identical heat radiation from every direction. In case of real fires, typically, the gas temperature is
lower compared to e.g., ISO standard fire curve. Accordingly, if we calculate the steel temperature with closed
formula of EN 1993-1-2 standard from gas temperatures recorded in a real fire like it would be gas tempera-
tures from ISO fire curve, we can underestimate the steel temperature because this calculation underestimates
the proportion of radiated heat. The solution may be the use of the so-called adiabatic surface temperature
[U. Wickström, D. Duthinh, K. McGrattan; 2007: Adiabatic surface temperature for calculating heat transfer to
fire exposed structures]) (adiabatic surface temperature may be 15 - 20% higher than gas temperature accord-
ing to the following study [M. Malendowski, A. Glema, W. Szymkuc; 2015: Performance based coupled CFD-
FEM analysis of 3-bay high industrial hall under natural fire in ASFE 2015 proceedings]).
Similarly to custom design seismic spectrum, the points of user-defined fire curve can be defined in function
editor. The maximum number of points is 360.
User’s Manual X6 Release1 e2 371
Steel tempera- The steel temperature is calculated automatically from the selected fire curve and fire duration / required
ture time of fire resistance (R15, R30, R60, etc. – where the number gives the time in minutes). This design
temperature is used then in the course of steel fire design (see 6.6.2 Steel beam fire design according to
Eurocode 3 – SD8 module).
The maximum fire duration is 180 minutes (R180).
The unit of influencing parameters, fire duration and temperatures can be set manually (see 3.3.8 Units and
Formats).
The diagrams can be saved in the Similarly, Drawings Library (see 3.6.11 Save to Drawings Library) by clicking
on the icon situated below the diagram in the right. The saved diagrams may be added later to reports.
The temperature of unprotected steel elements (EN 1993-1-2; ∆t = 5s):
𝐴/𝑉
Δ𝜃𝑎,𝑡 = 𝑘𝑠ℎ ℎ̇ Δ𝑡
𝑐𝑎 𝜌𝑎 𝑛𝑒𝑡,𝑑
where
𝑘𝑠ℎ [-] correction factor for the shadow effect (in case of SIA standard ksh = 1.0)
𝐴/𝑉 [1/m] section factor for unprotected steel members
𝑐𝑎 [J/kgK] specific heat of steel (according to EN 1993-1-2)
𝜌𝑎 [kg/m3] unit mass of steel
ℎ̇𝑛𝑒𝑡,𝑑 [W/m2] design value of the net heat flux per unit area
Calculated steel temperature related to the selected fire curve and fire duration (in case of parametric
fire curve the maximum temperature calculated in the interval) can be seen on the Steel temperature
panel under the combo box. Under the steel temperature reduction factors related to yield strength
and Young’s modulus according to EN 1993-1-2 are shown.
372
Exposition On the Exposition panel it is possible to set how the section is exposed to fire. Exposure type considerably
influences the value of shadow correction factor and section factor. In case of supported section types, the
following exposure types are available for protected / unprotected members:
For unsupported sections the exposure type appears via user specified shadow correction factor and section
factor (ksh and A/V).
Fire protection For steel structures the application of fire protection is often required due
to the slender elements and the fact that steel has a high thermal
conductivity. Check Fire protection on the Exposition panel to take into
account a fire protection. In this case, the Fire protection panel in the
bottom right is enabled. Parameters and their importance has been
already shown above.
Presence or lack of fire protection has an effect on the exposition types as well.
κ1 and κ2 fac- κ1 and κ2 are adaptation factors related to a non-uniform temperature
tors distribution across the cross-section and along the beam, respectively,
according to EN 1993-1-2. These factors can be set on the Temperature
distribution panel.
Temperature For I shapes and box sections, a more accurate temperature distribution may be calculated by clicking on
distribution the Analysis button on the Temperature distribution panel. Temperatures within the cross-section are
analysis calculated by finite difference method. In this case a two-dimensional heat conduction problem is solved.
User’s Manual X6 Release1 e2 373
Depending on the available computational resources and the symmetry of the problem, the calculation may
take a few seconds or a few minutes. The analysis considers the temperature-dependent thermal
conductivity of the steel material according to EN 1993-1-2. The diagram shows the maximum, minimum
and average temperatures as a function of time.
According to the regulations of EN 1993-1-2, in case of non-uniform temperature distribution the maximum
temperature within the cross-section has to be used when verifying the load bearing capacity. So if
temperature distribution results are available, AxisVM uses maximum temperature in further calculations.
κ1 adaptation factor related to non-uniform temperature distribution across the cross-section is
automatically calculated, however it is possible to override the calculated value above Analysis button.
If the section is connected to a reinforced concrete slab on one side, temperature distribution analysis
requires the values of slab thickness and concrete density.
Given tempera- If Given temperature is selected on the Steel temperature panel, a
ture uniform or linear temperature distribution can be specified. Design
temperature may be selected as θ1, θ2 or their average. This design
temperature is used later in the course of steel fire design (see 6.6.2
Steel beam fire design according to Eurocode 3 – SD8 module).
Apply as ther- Both calculated and given temperatures can be applied on the
mal load structural element as thermal load (see 4.10.20 Thermal load on line
elements) to determine internal forces due to restrained thermal de-
formations by a linear static analysis.
A reference temperature must be specified (in fire design it could be set to room temperature).
Using perfect model and linear static analysis significant internal forces may arise from thermal ex-
pansion since possible local stability failures are not considered. These significant forces may be re-
leased by local failures and deformations without considerably risk of global failure. According to the
Eurocode, thermal forces may be neglected in certain cases.
Table Fire effect parameters for each fire load case can be checked in the Table Browser. The listed parameters are:
required time of fire resistance, type of fire curve, fire load density, section factor, shadow factor, exposure
type, properties of fire protection and design steel temperature. The tables can be added to reports.
374
Graphical sym- Fire effects are shown in the main window as red zigzag lines. Symbol properties (thickness, colour) can be
bols in the main set in Preferences / Graphical symbols window.
window
Eurocode EN 1991-1-2
Eurocode 1 Action on Structures
Part 1-2: General actions - Actions on structures exposed to fire
EN 1995-1-2
Eurocode 5: Design of timber structures
Part 1-2: General - Structural fire design
EC DIN EN 1991-1-2:2010-12 NA
German DIN EN 1995-1-2:2010-12 NA
EC MSZ EN 1991-1-2:2005 NA
Hungarian MSZ EN 1995-1-2:2013 NA
CSN EN 1995-1-2/NA 2011
EC
Czech
EC STN EN 1991-1-2/NA:2006
Slovakian STN EN 1995-1-2/NA:2011
EC BS EN 1991-1-2:2002 NA
British BS EN 1995-1-2:2004 NA
EC SFS-EN 1991-1-2:2002 NA
Finnish
SFS-EN 1995-1-2:2004 NA
EC NS-EN 1991-1-2:2002/NA:2008
Norwegian NS-EN 1995-1-2:2004/NA:2010
EC NEN EN 1991-1-2:2002/NB:2011
Dutch NEN EN 1995-1-2 NB
Definition of fire ef- In order to define fire effect on line elements, click on Define fire effect on line elements icon on the Loads
fect toolbar.
Fire curves Fire effect is represented by fire curve specifying the gas temperature in the fire compartment as a func-
tion of the time. In case of timber elements, ISO standard and parametric fire curves can be selected.
These fire curves are presented and discussed in the previous section (see 4.10.31 Fire effect on steel
elements – SD8 module).
In case of NEN EN standard, parametric fire curve cannot be selected.
Calculation of The charring depth is calculated automatically from the selected fire curve and fire duration / required
charring depth time of fire resistance (R15, R30, R60, etc. – where the number gives the time in minutes). This charring
depth is used then in the course of timber fire design (see 6.7.2 Timber beam fire design – TD8 module).
The maximum fire duration is 180 minutes (R180).
The supported cross-sections are the thick rectangle, rounded rectangle and circle sections.
The diagrams can be saved in the exposure, Drawings Library (see 3.6.11 Save to Drawings Library) by
clicking on the icon situated below the diagram in the right. The saved diagrams may be added later to
reports.
The notional charring depth (dchar,n) of exposed timber elements is calculated as follows:
ISO standard fire curve (according to EN 1995-1-2 3.4.2.) – without fire protection:
𝑑𝑐ℎ𝑎𝑟,𝑛 = 𝛽𝑛 𝑡
𝛽𝑝𝑎𝑟 𝑡 𝑡 ≤ 𝑡0
𝑡 2 𝑡0
𝑑𝑐ℎ𝑎𝑟,𝑛 = 𝛽𝑝𝑎𝑟 (1.5𝑡 − − ) 𝑡0 < 𝑡 ≤ 3𝑡0
4𝑡0 4
{ 2𝛽𝑝𝑎𝑟 𝑡0 3𝑡0 < 𝑡 ≤ 5𝑡0
where
𝑞𝑓,𝑑 0.2√Γ − 0.04 (𝑂/𝑏)2 𝐴𝑣
𝑡0 = 0.009 𝛽𝑝𝑎𝑟 = 1.5𝛽𝑛 Γ= 𝑂= √ℎ 𝑏 = √𝜌𝑝𝑤 𝑐𝑝𝑤 𝜆𝑝𝑤
𝑂 (0.04/1160)2 𝐴𝑡
0.16√Γ + 0.08
𝑞𝑓,𝑑 [MJ/ m2] design value of the fire load density related to the surface area Af of the floor
𝐴𝑣 [m2] total area of vertical openings on all walls
ℎ [m] weighted average of window heights on all walls
𝐴𝑡 [m2] total area of enclosure (walls, ceiling and floor, including openings)
𝑐𝑝𝑤 [J/kg°C] specific heat of boundary of enclosure
𝜌𝑝𝑤 [kg/m3] density of boundary of enclosure
𝜆𝑝𝑤 [W/m°C] thermal conductivity of boundary of enclosure
If the boundary of enclosure consists of different layers, 𝒄𝒑𝒘, 𝝆𝒑𝒘 and 𝝀𝒑𝒘 parameters should be
𝑨
given resulting parameter 𝒃 = √𝒄𝒑𝒘 𝝆𝒑𝒘 𝝀𝒑𝒘 equal to 𝒃 = √𝒄𝒑𝒘,𝟏 𝝆𝒑𝒘,𝟏 𝝀𝒑𝒘,𝟏 ∑ 𝟏 + ⋯ +
𝑨
𝑨𝒏
√𝒄𝒑𝒘,𝒏 𝝆𝒑𝒘,𝒏 𝝀𝒑𝒘,𝒏 ∑ 𝑨 .
Notional charring depth of protected timber elements can be calculated only for ISO standard fire curve
according to Section 3.4.3. of EN 1995-1-2 standard. The following parameters are required:
𝑘2 [-] insulation coefficient
𝑡𝑐ℎ [min] time of start of charring of protected members (delay of start of charring due to protec-
tion)
𝑡𝑓 [min] failure time of protection
These parameters need to be specified based on the regulations of the code, the depth and the type of
fire protection material.
Charring rate The software selects the notional charring
rate based on the parameters of timber
material of the selected member. How-
ever, the value can be specified by the
user.
Exposition On the Exposition panel it is possible to set how the member is exposed to fire. Exposure type influences
only the calculation of reduced cross section during fire design (see 6.7.2 Timber beam fire design – TD8
module). In case of supported section types, the following exposure types are available for protected /
unprotected members:
378
Table Fire effect parameters for each fire load case can be checked in the Table Browser. The listed parameters
are: required time of fire resistance, type of fire curve, fire load density, exposure type, properties of fire
protection and calculated notional charring depth. The tables can be added to reports.
Graphical symbols Fire effects are shown in the main window as red zigzag lines similarly to steel members (see 4.10.31 Fire
in the main window effect on steel elements – SD8 module). Symbol properties (thickness, colour) can be set in Preferences /
Graphical symbols window.
Errors If gas temperature or notional charring depth cannot be calculated due to inappropriate parameter
values an error message appears. EN 1991-1-2 impose several restrictions regarding to initial parameters
of parametric fire curves (opening factor, 𝑏 = √𝑐𝑝𝑤 𝜌𝑝𝑤 𝜆𝑝𝑤 , floor area, height, etc.).
By storeys Existing loads placed into separate load cases according to the storey they are applied on.
Cut loads at storey boundaries: Loads extending to multiple storeys are splitted at storey boundaries
only if this option is activated. Otherwise, these loads are placed into the load case of the lowest storey
they cover.
Include storey name in the load case name: If this option is activated generated load case names will
refer to the storey they belong to.
By a structural grid This function splits existing loads along the lines of a
structural grid (see 2.16.8 Structural grids) and place the
resulting loads in separate load cases. The two splitting
options are:
Field by field: Each field gets to a separate load case.
Chessboard: Field loads will be placed in two complimen-
tary load cases following a chessboard pattern.
Split loads inside the structural grid only: If this option is deactivated, loads outside the boundary of the
structural grid are also splitted. If the structural grid was defined for a storey load case names can refer
to the storey the grid belongs to.
To display domain loads splitted into separate load
cases choose the Load group intensities color cod-
ing. This way loads of all load cases within the cur-
rent load group will be displayed.
See 2.16.5 Color coding.
The nodal mass is displayed on the screen as two dark red concentric circles.
4.10.35. Modify
Modify To modify loads:
1. Press the [Shift] key and select loads you want to modify (or the loaded elements).
You can also select by drawing a selection frame or using the Selection toolbar.
2. Click the load type icon on the Toolbar.
3. Check the checkboxes beside the values you want to change.
4. Enter new values.
5. Close the dialog with OK.
Immediate mode If the Loads tab is active click a finite element to modify its loads. If the element has more than one
load only one of them will come up. If you have placed different concentrated and distributive loads on
a beam and click the beam the load nearest to the click position will come up. If more finite elements
380
have been selected their loads can immediately be modified by clicking one of them. If you click an
element which is not selected, selection disappears and you can modify the element load you clicked.
In fact, load modification is similar to the load definition, but does not assign loads to elements
not being loaded and allows access to a specific load property without altering others. You can
switch to the Define radio button to place loads on all the selected elements, lines or surfaces. If
we select elements with loads not matching the load type we choose these loads remain
unchanged.
4.10.36. Delete
[Del] See 3.2.9 Delete
4.11. Mesh
Clicking on the Mesh tab the meshing toolbar becomes available with mesh generation for line elements
and domains, mesh refinement functions and a finite element shape checking.
Automatic detection of overlapping lines and missing intersections during meshing reduces the errors
in model geometry.
Support of multiple core processors can reduce the time of meshing.
Finite element analysis uses linear elements with constant cross-section so arced and variable cross-
section (tapered) line elements must be divided into parts. This is called line element meshing.
The accuracy of the solution depens on the mesh density.
This mesh can be removed or modified just like a domain mesh. Removing a mesh does not delete loads
and properties assigned to the line element.
A mesh can also be defined for linear elements with constant cross-section. It is useful in nonlinear or
vibration analysis when it is required to divide line elements to achieve a higher accuracy.
Maximum element size: Length of the mesh lines cannot exceed the value specified.
Division into N segments: Line elements are divided into N parts.
By angle: Central angle of arced mesh segments cannot exceed the value specified.
User’s Manual X6 Release1 e2 381
Mesh size An average mesh element size can be specified. The actual
mesh can contain smaller and larger elements as well.
Fit mesh to loads Meshes will follow checked loads if load intensity exceeds
the value specified. Point loads will create mesh nodes, line
loads will create mesh lines.
Adjust mesh to col- The mesh must be properly adjusted to column heads to prepare cutting of moment peaks. Turning this
umn heads option on automatically fits the mesh according to the cross-section geometry of connecting columns.
All beams joining to the slab at an angle greater than 45° are identified as columns. This option must be
set to enable the Cut moment peaks over columns option of the Display Parameters dialog. See 6.1.12
Surface element internal forces.
Contour division Uniform mesh size
method Domain boundaries and inner lines will be divided according to the mesh size to ensure the given
element size.
Adaptive mesh size
Adaptive meshing follows domain geometry and refine the mesh by reducing element size wherever
it is necessary.
Smoothing Track bar controls the smoothing of the mesh. Smoothing slows down mesh generation a bit. Moving
the handle to the left end sets minimum smoothing and fast processing while the right end sets maxi-
mum smoothing with slower processing. The result of smoothing depends on domain geometry and
other mesh parameters, so setting higher smoothing does not necessarily result in a better mesh
quality.
If Create mesh only for unmeshed domains is checked no mesh will be created for domains already
meshed.
If Calculation of domain intersections is turned on domain intersections are automatically calculated
before meshing.
The progress of the mesh generation process can be monitored in a window, and can be canceled any
time with the Abort button.
The mesh generator uses only the end points of beam elements that are in the plane of the domain, and
disregards their corresponding line segments. Rib elements are incorporated with their line segments
because they can be defined on surface edges as well.
If there are existing quadrilateral or triangular meshes within the domain, the mesh generator will not
change these meshes, and will integrate them in the new mesh.
382
If a mesh is generated over an existing domain mesh (with a different average element side
length), the new mesh will replace the existing one.
Lets you refine the finite element mesh of surfaces. The elements in the refined mesh have the same
properties (material, cross-section / thickness, references, etc.) as those in the coarse mesh.
You have to manually set the nodal degrees of freedom of the newly generated mesh that were
not set automatically during the process of mesh generation.
The following options are available:
Uniform Lets you refine the entire selected mesh. You must specify the maximum side length of a surface element
in the refined mesh.
before after
User’s Manual X6 Release1 e2 383
Node relative Lets you refine the mesh around the selected nodes (locally around columns, nodal supports). You must
specify a division ratio (0.2-0.8). The command refines the mesh dividing the elements connected to the
respective nodes by the defined ratio.
before after
Edge relative Lets you refine the mesh along the selected edges (locally along edge supports / loads). You must specify
a division ratio (0.2-0.8). The command refines the mesh dividing the elements connected to the
respective edges by the defined ratio.
before after
This command selects all edges connected to one surface element only. This selection helps to find edges
where the meshes of two domains are not connected due some editing problem.
5. Analysis
AxisVM lets you perform linear and nonlinear static, linear and nonlinear dynamic, vibration and
buckling analysis. It implements an object-oriented architecture for the finite element method.
The instructions included in this User’s Manual assume a preliminary knowledge of the finite element
method and experience in modeling. Note that the finite element analysis is only a tool, not a
replacement for engineering judgment.
Details of the analysis can be displayed by expanding one or more category panels. The Messages panel
shows the analysis message log. The Statistics panel shows memory requirements, hardware infor-
mation, model details and calculation times.
Special categories:
Nonlinear analysis: Tracking displays the movements of the tracked node. Convergence shows the
convergence of the iteration process.
Vibration analysis: Frequencies displays how the frequencies converge. Convergence shows the con-
vergence process.
Buckling analysis: Eigenvalues displays how the eigenvalues converge, Convergence shows the conver-
gence process.
Dynamic analysis: Time steps displays the movement of the tracked node, Convergence shows the
convergence process.
Parameters of the latest analysis is saved into the model file and can be studied in the Model Info dialog.
See 2.16.20 Model info
Model optimization To reduce analysis time and memory footprint AxisVM optimizes node order. If the total number of
degrees of freedom is over 1000, it creates an internal three-dimensional graph from the model geom-
etry and begins to partition the system of equations using the substructure method. The system is
stored as a sparse matrix. The parameters of the optimized system of equations appear only at the end
of this process. This process results in the smallest memory footprint and fastest calculation time but it
assumes that the biggest block fits into the available memory. If it doesn’t, AxisVM stores the system as
a band matrix and begins to reduce the bandwidth of the system by iterative node renumbering. If the
two longest rows fit into the available memory the system can be solved. Changes in the memory re-
quirements for the band matrix is displayed real-time. The duration of the optimization process and the
final memory footprint depends on the size of the system and the available memory.
The system of equations can be solved the most efficiently if the whole system fits into the phys-
ical memory. If the system does not fit into the physical memory but its largest block does, the
running time will be moderate.
If the largest block does not fit into the physical memory, the necessary disk operations can slow
down the solution considerably.
Model verification The input data is verified in the first step. If an Error is found a warning message is displayed and you
can then decide whether to cancel or continue the analysis
Performing the AxisVM displays the evolution of the solution process by two progress bars. The bar on the top displays
analysis the current step performed, while the other displays the overall progress of the analysis process.
The equilibrium equations in the direction of constrained degrees of freedom are not included in the
system of equations. Therefore to obtain support reactions you must model the support conditions
using support elements.
Error of the solution Solution error is calculated from the solution of a load case with a known result. It is a good estimation
of the order of errors in displacement results for other load cases.
Info palette shows this error as E(EQ).
If the value of E (Eq) is greater than 1E-06 the reliability of the computed results is questionable. It is
expected, that the Error of the displacements is of the same order.
Result file During the processing of the results the program sorts the results according to the original order of the
generation nodes and prepares them to graphical display.
In the following chapters we ‘ll show the setting of the parameters of the each calculation methods.
386
The term static means that the load does not vary or the variation with the time can be safely ignored.
Linear static Performs a linear static analysis. The term linear means that the computed response (displacement,
internal force) is linearly related to the applied load.
All the load cases are solved in the analysis. Through the geometric linearity, it is assumed that the
displacements remain within the limits of the small displacement theory. Through the material linearity,
it is assumed that all materials and stiffness characteristics are linear-elastic. The materials assigned to
surface elements can be othotropic.
See the description of the truss, gap, support, seismic isolator and spring elements in Chapter 4,
on how to use these elements in a linear analysis.
The relative errors at the end of the iteration process appear in the info window.
E(U): relative error of the displacement convergence
E(P): relative error of the force convergence
E(W): relative error of the work convergence
Values indicating instability appear in red.
Nonlinear static Performs a nonlinear static analysis. The term nonlinear means that the computed response
(displacement, internal force) is nonlinearly related to the applied load. This can be due to the use of
e.g.,gap, link, non-linear support elements, line/surface elements with nonlinear metarial, or taking into
account the geometric nonlinearity.
Select load cases or combinations in the tree view.
AxisVM will perform nonlinear analysis for the selected load cases / combinations and shows a progress
dialog.
User’s Manual X6 Release1 e2 387
Solution control
Force
When Force control is selected, the increments are applied as fractions of the loads (as one
parameter load). It is possible to track the displacement of a node in a given direction. A graph
of this displacement versus increments will be plotted during the analysis.
Displacement
When displacement control is selected, the increments are applied as fractions of the
displacement component of the node specified.
Pushover
Pushover control is a special type of displacement control that allows the use of a constant load
case while having another parametric load case that is increased incrementally. This is essential
for pushover analyses to model P-Δ effects appropriately.
After selecting pushover control, the top of the dialog changes to accommodate the drop-down
boxes for parametric and constant load cases. See 4.10.25 Pushover loads – SE2 module for details
on load definition and recommended analysis settings.
Number of increments
There are two methods to define the number of increments:
1. Equal increments. Specify the number of increments. The default value is 10. When highly
nonlinear behavior is analyzed, you may specify a greater value in order to achieve
convergence.
2. Increment function. Loads are not increasing in a linear way but follow a predefined function.
Using an increment function it is possible to reduce the number of increments where the
behaviour of the structure is linear and increase the number of increments where the
behaviour is nonlinear. Saved functions can be reloaded, edited and saved under a new name
(see 4.10.30 Dynamic loads (for time-history analysis) – DYN module). Functions are saved
into separate *.inc files in folder c:\ Users \ [user name] \ AppData \ Roaming \ AxisVM \
[version number] \ inc.
Convergence criteria
Based on the convergence tolerances you specify, AxisVM will determine if the nonlinear solution
has reached the required accuracy (convergence). Therefore it is important that the convergence
tolerances to be set properly. During the iteration process, the norm of the unequilibrated load
and/or of the iterational displacement increment vector must vanish (to approach zero).
Maximum iterations
You can set the maximum number of the iterations based on the specifics of your model, and of
the incremental solution parameters. By default the value is set to 20. If the convergence is not
achieved within the maximum number of iterations, no results will be obtained.
The iteration control contains the following checks depending on the options selected.
The analysis can be performed either with consideration of creep/shrinkage or without. Further
information related to the consideration of creep and shrinkage can be found in the above referred
sections.
User’s Manual X6 Release1 e2 389
This option is enabled if the model contains elements with nonlinear behaviour (e.g., tension-only
trusses or compression-only supports) or elements with nonlinear material characteristics (trusses,
beams, ribs, membranes, plates, shells). If left unchecked, all elements will respond in a linear way.
Follow geometric nonlinearity of beams, trusses, ribs and shells
The equilibrium is established with respect to the deformed line elements. Depending on the
magnitude of displacement second or third order analysis is performed. The loads are conservative,
they keep their initial direction during the deformation. Geometric nonlinearity can be taken into
account only for truss, beam, rib and shell elements. If there are no elements with nonlinear
characteristic in the model this options is checked by default. If the model contains elements with
nonlinear characteristic this option is left unchecked but can be activated.
Stability warning In geometric nonlinear calculations a stability problem may occur in some increments. It can be a local
messages or global loss of stability or a snap-through phenomenon. Euler’s critical load may also be exceeded in
any beam. In such cases the following investigation procedure is recommended
Allows you to reduce the size of the results file when an incremental nonlinear analysis is performed
with multiple increments (load or displacement) when just the results of the last increment are of
interest to you. You can enable this checkbox when you do not need the results of previous
increments.
You should disable this check-box if you want to trace the load-displacement or other
(nonlinear) response of the structure.
Finite elements with Beam, rib and surface elements made of nonlinear
nonlinear material (elastic or plastic) material are modelled with a dis- peripheral layer
cretized section model. Plates and shells are repre-
sented by layers, cross-section of beams and ribs
are meshed. Stress distribution is determined us- midplane
5.2. Vibration
Lets you determine the lowest natural frequencies and mode shapes corresponding to the free vibration
of an undamped linear structure when no externally applied loads are computed. AxisVM verifies
whether the required number of the lowest eigenvalues has been determined.
The system mass matrix has a diagonal structure and includes only translational mass components.
Select load cases or combinations in the tree view. AxisVM will perform vibration analysis for the se-
lected load cases and shows a progress dialog.
The solution technique applied to the associated generalized eigenvalue problem is designed to
find the lowest real and positive eigenvalues. It is not suitable to find eigenvalues that are zero
or nearly zero.
Solution control
Lets you specify the parameters of the incremental solution process:
First order
The solution does not include the effect of axial forces of truss/beam elements on the system
stiffness.
Second order
The solution include the effect of axial forces of truss/beam elements on the system stiffness.
Tension axial forces have a stiffening effect, while the compression axial forces have a softening
effect. These effects influence the free vibrations of the structure.
Case
Lets you select a case. The loads are converted into masses. If a second-order analysis is
selected, the results of a linear (first-order) static analysis, that precedes the vibration analysis,
will be accounted too.
Number of mode shapes
Lets you specify the number of the vibration mode shapes you want to evaluate.
The default value is 9, maximum number of shapes is not limited. The value specified here can
not be larger than the number of the system’s mass degrees of freedom.
392
The program uses a diagonal mass matrix by default. Due to the lumped mass modeling
technique to achieve the required accuracy the elements must be divided into more elements (by
refining the mesh). Usually at least four finite elements must correspond to each half wave.
A good rule-of-thumb is that beams must be divided into at least eight elements.
The mode shapes are normalized with respect to the mass, {𝑈}𝑇 [𝑀]{𝑈} = 1
The problem of human-induced vibrations on floors is increasing with more efficient design utilizing
stronger materials leading to lighter structures. Walking can cause vibration uncomfortable or unac-
ceptable for occupants of the buildings or make delicate operations impossible (e.g., operating theatres,
precesion laboratories). Footfall analysis considers walking as an external excitation force and
determines the vibration response factor, which is proportinal to the maximal acceleration at the
considered node. Footfall analysis is available only if the configuration includes the FFA module.
In order to analyse the response of a structure to human induced vibration two conditions must be
fulfilled. Model must contain domains and the angle between their normal and the direction of the
gravity should be less then 70°. A domain is considered to be a slab if the angle between its normal and
the direction of gravity is less than 10°, if this angle is between 10° and 70° the surface is considered to
be a staircase. Domains with larger angles between their normal and the gravity are considered to be
walls. AxisVM does not analyse and does not excite nodes of walls as vibration of walls don’t cause
discomfort.
Vibration results The second prerequisite is a vibration analysis with first and/or second order vibration results. Select
to use the vibration results of load cases or combinations that should be taken into account (load cases or
combinations without the appropriate number of mode shapes appear in red).
Choose an excitation method, set the damping ratio and further parameters of the dynamic load. The
result of the analysis is a vibration response factor (R) for the selected excitation method on slabs and
staircases. R must be checked against the recommended values (for details see the FFA Guide (see Help
/ FFA Guide).
It is also important that the boundary and continuity conditions must be set up to reflect the stiffer
behaviour of the structures for dynamic loads (see 5.2 Vibration, Use increased support stiffness).
Furthermore, during the calculation of the modal shapes it is suggested to take into account live loads
with quasi-permanent combination (𝝍𝟐 ), which influences the results favourably due to the significant
increase in the structure’s modal masses.
394
Modal shapes All modal shapes for the load cases/combinations: All shapes of the selected load cases/combinations
to use will be taken into account.
Shapes activated in the table of modal mass factors: Shapes can be made inactive in the table of Modal
mass factors (see 4.10.24.1 Seismic load calculation according to Eurocode 8).
Below frequency limit let the user to set a limit for the eigenfrequencies, which will appear in the
calculation. Only modes with a smaller frequency will be taken into account.
Excitation method Three excitation methods are available. Each one creates a separate R result component.
Full: All nodes of slabs or staircases can excite any other node (R full). This is the most robust method.
Excitation at the extremes of the modal shapes: Excitation takes place only at the two global extrema of
the eigenshapes (R extr.).
Excitation at the node where the response is analysed: The analysed node is only excited by itself (R self).
Excitation until adjacent storeys only: This option is enabled if the model contains at least three storeys.
If checked the analysed node can only be excited with nodes on the same or adjacent storeys.
Damping The critical damping ratio of the analysed structure can be set. Suggestions can be found in [33], [34].
Footfall parameters
Design method Two approaches are available: CCIP-016 [33] and SCI P354 [34]. Their algorithms are described in the
FFA Guide (see Help / FFA Guide). Changing the approach clears the previous results.
Number of steps The CCIP-016 design method allows resonant accelerations to be reduced due to the fact that resonance
(CCIP-016) needs time to develop. So the number of footsteps (the duration of the excitation) has effect on the
results.
Length of the walk- SCI P354 also takes into account the same effect but the relevant parameter is the length of the
ing path walking path. In this approach the most unfavourable length should be given.
(SCI P354)
Mass of the walker Both methods allow deviation from the predefined mass of the walker.
Pace frequency AxisVM analyses the effect of steps at several different pace frequencies within a given range. The lower
and upper bound of the frequency range can be set here.
Weighting curve SCI P354 allows the usage of different weighting curves. These curves take into account the fact that
the human perception of vibration varies with frequency.
Wb – Average areas, medium vibration should be used, if general comfort must be provided in the
building,
Wg – Special protection against vibration should be used if precise hand movements, steady vision
must be ensured.
In CCIP-016 design method only one weighting is applied.
For further details see Help / FFA Guide.
User’s Manual X6 Release1 e2 395
Dynamic analysis determines time-dependent displacements and forces due to dynamic loads or nodal
accelerations. Dynamic analysis can be performed on linear or nonlinear models where a dynamic load
case has been defined and a dynamic load has been applied to the structure.
Solution control Analysis can be performed in equal increments or according to a custom time increment function. Pre-
defined functions can be loaded or a new function can be created using the function editor.
If Equal increments is selected two parameters are required: Time increment and Total time. Analysis uses
the value of Time increment as the increment between time steps and Total time defines the total time
of the analysis.
Tracked node:
The displacement of the selected node in the given direction will be plotted during the analysis.
Rayleigh damping constants (a, b)
Damping matrix is determined from the damping constants, according to the following formu-
las:
𝐌 𝑢̈ + 𝐂𝑢̇ + 𝐊𝑢 = 𝑃(𝑡)
𝐂 = 𝑎𝐌 + 𝑏𝐊
If Consider loads and nodal masses is checked another matrix will be added to M representing loads and
nodal masses.
Save results Due to the considerable result file size result saving options are introduced: Checking Save all steps
means that all result will be saved. Save at regular intervals saves results only at certain model time
coordinates reducing file size.
396
The Whittaker-Shannon interpolation method is available only for functions obtained from equi-
distant sampling.
Convergence crite- If any of the nonlinearity options is checked convergence criteria has to be set and will be taken into
ria account like in a nonlinear static analysis. Otherwise, the actual E(U), E(P) and E(W) values (their final
values appear in the Info window) are compared to the reference values set here.
Iteration control These options are described in section 5.1 Static analysis
User’s Manual X6 Release1 e2 397
Solution method Linear or nonlinear equilibrium equations are solved by the Newmark-beta method. If Δ𝑡 is the time
increment, in 𝑡 + Δ𝑡 we get:
𝐌 𝑢̈𝑡+Δ𝑡 + 𝐂𝑢̇ 𝑡+Δ𝑡 + 𝐊𝑢𝑡+Δ𝑡 = 𝑃(𝑡)
where C is the damping matrix, M is the mass matrix, K is the stiffness matrix.
Δ𝑡 2
𝑈𝑡+Δ𝑡 = 𝑈𝑡 + Δ𝑡 ⋅ 𝑈̇𝑡 + [(1 − 2𝛽)𝑈̈𝑡 + 2𝛽𝑈̈𝑡+Δ𝑡 ]
2
𝑈̇𝑡+Δ𝑡 = 𝑈̇𝑡 + Δ𝑡[(1 − 𝛾)𝑈̈𝑡 + 𝛾𝑈̈𝑡+Δ𝑡 ]
1 1
AxisVM uses 𝛽 = , 𝛾 = .
4 2
The differential equation of the motion is solved by the method of constant mean acceleration. This step-
by-step integration is unconditionally stable and its accuracy is satisfying. AxisVM assumes that no dy-
namic effect is applied in 𝑡 = 0. Time-limited loads appear in 𝑡 > 0. C is calculated from the Rayleigh
damping constants:
𝐂 = 𝑎𝐌 + 𝑏𝐊
Where a and b should be calculated from the damped frequency range (between fi and fj) and the damp-
ing ratio according to the following figure:
2𝜔𝑖 𝜔𝑗
𝑎=𝜉
𝜔𝑖 + 𝜔𝑗
2
𝑏=𝜉
𝜔𝑖 + 𝜔𝑗
where 𝜔𝑖 and 𝜔𝑗 are
angular frequencies
relating to 𝑓𝑖 and 𝑓𝑗 :
𝜔𝑖 = 2𝜋𝑓𝑖
𝜔𝑗 = 2𝜋𝑓𝑗
For details about the behaviour of seismic isolators see 3.1.17.2 Seismic isolators
398
5.4. Buckling
Solution control
Select load cases or combinations in the tree view. AxisVM will run a linear static analysis before the
buckling analysis of the selected load cases.
Lets you specify the parameters of the incremental solution process:
Case
Lets you select a case that will be taken into account. A linear (first-order) static analysis, that
precedes the buckling analysis, will be performed.
Number of mode shapes
Lets you specify the number of the vibration mode shapes you want to evaluate. A maximum
number of 99 can be requested. The default value is 6. The lowest positive eigenvalue is of
main importance.
Convergence criteria
See 5.2 Vibration/Convergence criteria
Shells The buckling of shell elements are fully taken into account in the analysis.
Beams/ribs The buckling of beams/ribs is considered as in-plane buckling (flexural buckling), which means that the
deformed shape of the element remains in a plane and the cross-section does not warp.
For buckling analysis the beam cross-section must be defined by specifying its principal moments of
inertia.
The beam elements must be divided into at least four elements.
Trusses The flexural buckling of truss elements are not considered by the program. You must calculate the
buckling load of each truss manually, or by modeling the trusses by four beam elements with the
corresponding end releases.
If the model contains trusses the critical load parameter of global structural buckling will be
computed only. Buckling of individual trusses is not analysed.
Seismic isolators For details about the behaviour of seismic isolators see 3.1.17.2 Seismic isolators
User’s Manual X6 Release1 e2 399
All finite elements may be used in a linear static, nonlinear static, vibration, buckling and dynamic
analysis. Note that not all elements have geometric stiffness.
The directions in the local coordinate system in which an element has stiffness, and the corresponding
local displacement components are summarized below:
Finite ex ey ez Θx Θy Θz
element u v w
Truss *
Beam * * * * * *
Rib * * * * * *
Membrane * *
Plate * * *
Shell * * * * *
Support
* * * * * *
(only two components are
shown)
Gap *
Rigid
Link * * * * * *
(only two components are
shown for a node-to-node
link)
where:
u, v, w denote the deflections in local x, y, z directions.
Θx, Θy, Θz denote the rotations in local x, y, z directions.
* element has stiffness in the respective direction.
Internal forces The computed internal forces in the local coordinate system are:
1. Define the geometry of the structure, the material and cross-sectional properties of the
members, the support conditions, and the loads.
2. Determine the load transfer path.
3. Determine local discontinuities such as stiffeners, gussets, holes.
4. Determine the type of finite elements that will best model the behavior of the structure. With
this step the properties of structural elements will be concentraded in their neutral axis (point,
axis, or, plane).
5. Determine a mesh type and size for the model. The size of the mesh have to correspond to the
desired accuracy of the results and with the available hardware.
6. Create the model:
a.) Equivalent geometry
b.) Equivalent properties
c.) Topology of the elements
d.) Equivalent support conditions
e.) Equivalent load (static) or masses (vibration, response-spectrum)
7. Check input data (accuracy, compatibility)
8. Run analysis
9. Select important results
10. Evaluate and check the results
a.) Accuracy and convergence of the solution
b.) Compatibility taking into account point 6.d.
c.) Uncommon structures shall be analyzed with other methods and/or software as well.
11. Restart analysis with a correspondingly updated model, if in step 10 a criteria is not satisfied.
12. Evaluate the results by the means of isoline/isosurface plots, animation, tables... Draw
conclusions on the structure’s behavior.
Modelling To build a model of a structure you have to accept many assumptions so you also have to keep the
effects of these assumptions in view when evaluating results.
The finite element method provides an approximative solution for surface models. To make the model
match the real solution you have to use finite element meshes with an appropriate density. Making
finite element meshes you have to take into account the expected stress distribution, the model
geometry and the materials, supports and loads used.
The position af nodes and mesh lines (called the topology of the finite element mesh) depends on the
geometrical discontinuities (irregular contours, line supports) and the discontinuities of loads
(concentrated loads, terraced load values for line loads).
At stress concentration points (sharp corners) you have to refine the mesh. To avoid singularities due
to concentrated effects you can distribute them on a small area around the point of effect.
Arc contours can be approximated as polygons. Using very small tolerance in this approximation leads
to polygons with extreme small sides. The very dense mesh created on this contour may cause the
model exceed the capacity of your computer.
In general if you refine the mesh you get more accurate results.
402
Other (system) er- In some cases, other type of error messages may also appear. Based on the system error code, more
ror messages information is provided about the error on the following link:
http://docwiki.embarcadero.com/RADStudio/Tokyo/en/Input-Output_Errors
404
6. The Postprocessor
Static Lets you display the results of a static analysis. (6.1)
Buckling Lets you display the results of a buckling analysis. (6.2)
Vibration Lets you display the results of a vibration analysis. (6.3)
Dynamic Lets you display the results of a dynamic analysis. (6.4)
R.C. Design Lets you display the results of a reinforced concrete design analysis. (6.5)
Steel design Lets you display the results of a steel design analysis. (6.6)
Timber design Lets you display the results of a timber design analysis. (6.7)
XLAM design Lets you display the results of an XLAM-domain design analysis. (6.8)
Masonry design Lets you display the results of a masonry design analysis. (6.9.1.6)
6.1. Static
The Static menu item allows you to display the tools for displaying and interpreting the static analysis
results.
Display of linear / nonlinear
diagrams
Width of combo boxes and the result scaling factor edit field can be adjusted on each postprocessor
tab separately. Place the mouse between the control and the next control of the toolbar. The cursor will
change to and you can adjust the width of the control. To restore the default sizes click Settings /
Toolbars to default position. See also 3.3.15 Toolbars to default position.
Start a linear static See 5.1 Static analysis
analysis
Start a nonlinear See 5.1 Static analysis
static analysis
Result display Lets you set the options of the graphical display of the results.
parameters You can select the results of a load case/combination or critical load combination.
Display Parameters dialog shows the following options.
User’s Manual X6 Release1 e2 405
Analysis Type Depending on the performed analysis you can select the results of a linear or nonlinear static analysis.
Each analysis type can be further defined:
Case
Lets you display the results of any load case/combination.
Envelope
Lets you display the envelope of the results from the selected load cases and/or load combinations.
The program searches for the minimum and/or maximum values at each location of the selected
result component.
Critical
Lets you generate the critical load combinations, according to the load group definitions, for each
location of the selected result component.
Envelope
AxisVM allows to define and use different envelopes with names. On the left a list of the available
envelopes are listed. Certain basic envelopes are automatically created (envelope of all load cases, all
load combinations or certain combination types (e.g., ULS, SLS Quasipermanent). The composition of
the selected envelope is displayed in the tree of load cases and combinations. Changing the
composition of an envelope results in creating a new, custom envelope. Selecting a custom envelope
and clicking on its name makes the name editable.
If AxisVM main window is divided into sub-windows a different envelope can be chosen for each sub-
window. The name of the selected envelope is also displayed in the status window. Drawings and tables
of the report also contain and display envelope information.
406
Multiple selection is enabled in the tree of load cases and combinations. To check or uncheck a
continuous range of load cases click on the first load case within the range (it will be selected)
then Shift+click on the last load case of the range.
.
Displayed envelopes Select the displayed envelopes from the dropdown list under the list of envelopes. This way you can
control which envelopes are available for result evaluation in the dropdown list of load cases and
combinations.
.
Only the selected envelope Only one envelope will be available which is the currently selected
one.
Only custom envelopes All custom envelopes will be listed
Critical
Investigate all com- By default, this option is off. AxisVM takes into account combinations resulting in an extreme for any
binations resulting result component. In certain design methods however a combination which produces no extremes can
in the same maxi- be more unfavorable.
mum value In this case turn this option on. In design calculations AxisVM will build all possible combinations and
check them according to the design code requirements. As the number of combinations can be ex-
tremely high this option is recommended only if the model size and the number of load cases are small.
In Semi-auto mode AxisVM uses ULS or SLS combination according to the type of calculation but the
formula for creating SLS combinations can be chosen.
Display values If you have selected Envelope or Critical you can choose from the following options:
Min+Max
Displays the minimum and maximum values of the current result component.
Min
Displays the minimum (sign dependent) values of the current result component.
Max
Displays the maximum (sign dependent) values of the current result component.
Display Shape Undeformed
Displays the undeformed shape (original configuration) of the model.
Deformed
Displays the deformed shape of the model.
Automatic scaling of displacements
Scales the real deformation by an automatic scale factor so that the maximum displacement is
represented as 50 pixels. This pre-scaled deformation is the basis of the further scaling by the result
scaling factor. Without this option the basis of the scaling by result scaling factor is the real
deformation.
None
The current result component is not displayed.
Section lines Lets you set the active section lines, planes and segments. If display mode is set to Section line result
diagrams will be drawn only on active (checked) section lines. Symbol of the section planes can be
displayed enabling the Draw section plane contour checkbox. Turning on the Draw diagram in the plane
of elements option changes the appearance of all section diagrams. To change this parameter
individually use the Section lines dialog.
See 2.16.15 Sections
Scale by Lets you set the scale of a diagram drawing. The default value is 1, when the maximum ordinate is
represented as 50 pixels.
If labels overlap the drawing can be made clearer (more readable) by checking Prevent
labels from overlapping (2.16.18 Display options)
After clicking the Miscellaneous Settings... button the following options are available:
User’s Manual X6 Release1 e2 409
Settings for Check Round calculated values and step of rounding to make the calculated values more realistic. Fol-
rebar spacing lowing rebar layout rules for spacing is optional.
Cut moment peaks If the mesh on a domain was created with the option Adjust mesh to column heads moment peaks can
over columns be averaged and cut over column heads by activating this option.
See 6.1.12 Surface element internal forces
410
Case selector to You can select a case from the drop-down list to display:
display Load case, load combination
The k-th increment of a nonlinear analysis
Envelope display
Critical combination
Available result
components
You can select a result component from the drop-down list for display:
Displacement (eX, eY, eZ fX, fY, fZ,eR, fR)
Beam/rib internal force (Nx, Vy, Vz, Tx, My, Mz , MyD *)
Beam/rib stress (Smin, Smax,Tymean, Tzmean)
Relative displacement of beam/rib end releases (ex, ez, ez, fx, fy, fz, eR, fR)
Surface internal force (nx, ny, mx, my, mxy, vxz, vyz, vSz, n1, n2, an, m1, m2, am, nxD, nyD, mxD,
myD)
Surface stress (Sxx, Syy, Sxy, Sxz, Syz, Svm, S1, S2)
Reinforcement design forces (nxD, nyD, mxD+, mxD-, myD+, myD-)
Amounts of reinforcement (axa, aya, axf, ayf)
Crack width (actual reinforcement) (wk(a), wk(f), wk(a,f), wk2(a), wk2(f), wk2(a,f), wR(a), wR(f))
Crack width (required reinforcement) (wk(a)*, wk(f)*, wk(a,f)*, wk2(a)*, wk2(f)*, wk2(a,f)*, wR(a)*,
wR(f)*)
If Min,Max envelope or critical load combination is selected, the Isoline and Isosurface 2D cannot
be selected.
Lets you search for the minimum and maximum value of the current
result component. If you are working on parts, the search will be limited
to the active parts.
AxisVM will mark all occurrences of the minimum / maximum value.
If parts are displayed extreme values are determined from the dis-
played parts only.
6.1.2. Animation
Lets you display the displacements, internal forces, and mode shapes in animated form (frame by frame).
The animation of linear results consists of a sequence of frames that are generated by linear interpolation
between initial values (frame 0) and the actual values of the current result component (frame n), according
to the number of frames (n). Animation of nonlinear or dynamic results generates frames from the results
of individual increments. After choosing Deformed as the Display shape in the Result display options the
animation helps to see the deformation.
The generation process can be cancelled by pressing Esc.
Control buttons
412
Fit in window
Push down the left mouse button to drag the drawing to be animated. Scroll the mouse wheel to zoom
it in/out.
Frames
Lets you set the number of animation frames. You must specify a value between 3 and 99. More
frames produce smoother but slower animation.
This dialog displays linear, nonlinear or dynamic results as diagrams. Two diagrams can be displayed
simultaneously. Each diagram has a result component on its X and Y axis. Points representing consecu-
tive value pairs are connected. Reading coordinates can be changed by dragging the dashed lines or
the black mark of the bottom trackbar. Diagram points can be displayed as a table and exported to
Excel through the Clipboard.
In case of linear analysis all load cases, all load combinations, all steps of a moving load or all members
of an envelope are displayed depending on the chosen case in the main window.
If a result component relating to stress points is selected the cross-section is displayed and a specific
stress point or the cross-sectional min/max value can be selected.
User’s Manual X6 Release1 e2 413
Cross-section for diagram X1-Y1 (the selected component is not a stress point component)
In case of nonlinear analysis all increments of an analysis or all members of an envelope can be dis-
played.
414
In case of dynamic analysis, the bottom trackbar displays time instead of increment numbers.
Toolbar
Print drawing
Prints the diagram (and the table if it is displayed)
Copy to Clipboard
Copies the diagram to the Clipboard.
Table
Turn on/off the table displaying numerical values.
Interval controls
Turns on/off the green interval control rectangles of the bottom trackbar. Dragging them changes the
displayed range of increments or time.
Loads the chosen load case, load combination, increment or time step to the active window.
(available in multi-window mode only)
Loads the chosen load case, load combination, increment or time step to the window.
416
This dialog is only active if results of pushover analysis are available and it helps the user determine
the capacity curve and the target displacement depending on ground motion characteristics.
A combo box on the top of the dialog lets the user select the pushover load case to be analysed.
Results are based on an acceleration-displacement response spectrum with properties specified on
the left side of the dialog. These are identical to the properties of response spectra used for Seismic
loads (See 4.10.24 Seismic loads – SE1 module). Main results of the calculations are shown both on the
bottom left side of the dialog and under the diagrams themselves.
The default dialog displays a capacity curve for both the Multi Degree of Freedom System (MDOF) and
the equivalent Single Degree of Freedom System (SDOF).
The sky-blue curve is the capacity curve of the equivalent single degree of freedom system (SDOF).
It has the same shape as the deeper blue curve for the multi degree of freedom system (MDOF).
Its points are a result of dividing the corresponding force and displacement values of the MDOF curve
by Γ.
Generally, the end point of both capacity curves is the point corresponding to the maximum displace-
ment (divided by Γ for the SDOF curve) set by the user at the beginning of the nonlinear static analysis.
The resulting curve on the figure below shows that the structure is capable of even more displacement,
since the base shear force (vertical axis) is increasing as the displacements are increasing. The maxi-
mum value of the shear force can only be determined if a sufficiently large target displacement is
specified when running the nonlinear pushover analysis.
Toolbar
Print drawing
Prints the current diagram
Copy to clipboard
Copies the current diagram to the Clipboard.
All of the results are based on the N2 method (see 11.32) recommended in Appendix B of Eurocode 8.
The bilinear force-displacement relationship for the SDOF system (green curve) is calculated by taking
the force at the target displacement (𝑑𝑡∗ ) as the force that corresponds to yielding (𝐹𝑦∗ ) and defining
yield displacement (𝑑𝑦∗ ) using the equivalent deformation energy principle.
A vertical red line marks 150% of the target displacement (𝑑𝑡 ) according to Eurocode 8 (4.3.3.4.2.3).
Generally, if the deformation capacity of the structure is above this level (the line style is dash-dot) it
fulfils the deformation capability requirements, otherwise (the lines style is continuous) it fails these re-
quirements.
The Acceleration-Displacement Response Spectrum (ADRS) is shown by switching to the ADRS tab on
the dialog. Both elastic (dashed yellow line) and inelastic (yellow line) ADRS spectra, SDOF (blue line) and
equivalent bilinear capacity curves (green line) are shown here.
A separate line highlights the natural period corresponding to the elastic behavior of the structure. The
intersection of capacity and demand corresponding to the target displacement is marked by a red circle.
Purple dashed line shows the elastic approximation based on the initial stiffness of the bilinear capacity
curve.
Results The variables marked by an asterisk (*) represent the SDOF system’s behavior, while the others corre-
spond to the MDOF system.
Γ transformation factor for computing SDOF characteristics
m* mass of equivalent SDOF system
Fy* base shear force at dm* displacement of the equivalent SDOF system and yield force of the
elasto-perfectly plastic force-displacement relationship
d m* ultimate displacement of the idealized bilinear force-displacement relationship (not neces-
sarily the ultimate displacement of the SDOF system due to the iterating procedure of the N2
method)
dy* yield displacement of the idealized bilinear force-displacement relationship
T* natural period of the equivalent SDOF system
det* target displacement of the equivalent SDOF system with period T* and unlimited elastic be-
havior
dt* target displacement of the equivalent SDOF system considering inelastic behavior
It represents the end of the green bilinear capacity curve.
dt target displacement of the MDOF system considering inelastic behavior
418
6.1.4.3. Drift
Table Browser lets you display the numerical values of the results in a table in customizable form.
If you switched on parts, the table will list the values corresponding to the active parts. If you selected
elements the table will list the selected elements only by default. You can change the range of listed
elements by clicking the property filter button on the Table Browser toolbar.
You can transfer data to other applications via Clipboard. See 2.9 Table Browser.
Setting Display options / Labels / Use finite element numbers controls not only labeling but the
way result tables are compiled.
For example if this option is activated you will find beam internal forces in the Table Browser
under RESULTS / Linear analysis / Internal forces / Finite elements / Beam internal forces. Results
are listed per finite element.
If it is unchecked this path is RESULTS / Linear analysis / Internal forces / Beam internal forces
and the results are listed per structural member.
Displaying results After calling the Table Browser you can set if you need a detailed table
[Ctrl]+[R] and/or the extremes and you can select which components you need the
extremes from. This dialog can be called later from Format / Result Display
Options.
Extremes to find The initial set of extremes to find is determined from the default column visibility of the result table.
The user can set which columns (result components) should be visible by default in the result table. Only
visible result components will be checked automatically.
You can set the components for which you want to find the extreme (maximum and minimum) values.
Among the minimum and maximum values the concomitant values of the different result components
are displayed if the minimum/maximum values occur in a single location or otherwise. If there are
multiple locations the symbol * will appear, and in the Loc (location) column the first occurrence of the
extreme value will be displayed.
When you display the results of critical combinations in addition to the minimum and maximum values,
the load cases that lead to the critical values are included with the following notations:
[ ... ] represents the results of a permanent load case.
{ ... } represents the results of an incidental load case.
( ... ) represents the results of an exceptional load case.
Print Clicking the Print tool button or choosing the File / Print menu item the print dialog appears.
[Ctrl]+[P] See 3.1.12 Print.
6.1.6. Displacements
Nodes
At each node, six nodal displacement components (three translations
and three rotations) are obtained in the global coordinate system.
The resultant values of translations (eR) and of rotations (fR) are also
determined.
In some cases, the displacements and internal forces associated with the same critical or envelope
load combination seem to be inconsistent.
1. In critical or envelop load combinations, the tension side of the beam seems to be contradictory
based on the results of the displacement and internal forces. The components come from
different compilation of loads (SLS and ULS), so the results are not coherent.
2. This special case also shows the effect of different load combinations. The internal force is zero,
but the deformed shape shows (significant) deformation.
3. The last case illustrates the consequences of different local coordinate systems in envelope
load combination. The applied load and the calculated displacement is the same. The internal
force is also identical, but the diagram of the internal force is on the opposite side due to the
various local z axis. This may seems to be contradictory compared to the deformed shape in
the interpretation of the tension or compression side).
This may also occur in seismic combinations.
Displaying the displacements of a cantilever (membrane model):
Diagram with nodal values Section line with nodal values
User’s Manual X6 Release1 e2 421
Isolines Isosurfaces 2D
Beams For each beam element the intermediate displacements are obtained in the local and global coordinate
systems. When displaying the displacements of the structure the beam displacements are related to the
global coordinate system. If you pick the cursor on a beam element the six beam displacement
components related to the element local coordinate system are displayed in a diagram form.
If the model contains a 7 DOF beam (4.9.9.3 7 DOF beam element – 7 DOF module) an additional diagram
appears for the w = dfx/dx specific rotation interpreted in the local system of the element.
You can display displacements of more than one beam element if:
a) The local coordinate system of the elements are almost or entirely identical.
See 2.16.19.3 Drawing/ Contour line angle
b) The local x orientation is the same.
c) The elements have the same material
Save to Drawings Load current load case or combination to the active view
Library
Load current load case or combination to all views
Save to Gallery Load case or
combination Actual displacements
Report Maker
All results Displacements relative to end points
Copy to Clipboard Displacements relative to the left end
Envelopes only
Print Displacements relative to the right end
Load current load case or combination to the active view (in case of multi-window mode the load case
or combination will only be loaded to the active view)
Load current load case or combination to all views (in case of multi-window mode the load case or
combination will be loaded to all views)
Actual displacements
Save diagrams Associative diagrams can be saved to the Drawings Library. Drawings from this library can be inserted
to the Drawings Li- into reports. After changing and recalculating the model diagrams in the library and reports change
brary accordingly.
The following function is only available in Eurocode-NL and Eurocode-B standards and can be
used only for plates.
EN 1990:2002 A1.4.3, To determine the total deflection of a reinforced plate different type of vertical deflections must be
NEN-EN 1990 A1.4.3 calculated by running a series of nonlinear analyses for certain combinations with and without creep.
Use reinforcement in calculation must always be checked.
Calculation of total deflection requires the following values
w1 Initial part of the deflection under permanent loads
Run nonlinear analysis for 1) permanent load cases or 2) characteristic, frequent or quasi-permanent
SLS combinations without variable loads) without creep (φ = 0)
w2 Long-term part of the deflection under permanent loads
The same quasi-permanent SLS combinations must be calculated both with and without creep as w2 is
obtained as the difference between the two results.
Permanent part of the SLS combinations must include the same permanent load cases as the combina-
tion used to determine w1.
w3 Additional part of the deflection due to the variable actions of the relevant combinations
Must be calculated from the same SLS combinations as used to determine w1 but including variable
load cases as well without creep. w3 is obtained as the difference between the two results.
wbij wbij =w2+w3
4. If the envelope contains all the needed combinations, the program finds the set of combinations
that include the same permanent load cases and determine w1, w2, w3, wbij and wtot value for
each set, calculates their envelope and makes the envelope values available as w1, w2, w3, wbij
and wtot result components.
If the model contains domains and a linear analysis has been performed, the Relative displacement
button on the Static tab is also enabled. This feature displays the displacement of domains relative to
their supports.
Click on it to get to the following dialog.
Selection of domains Three methods are available for defining supported nodes:
and supported nodes
1. Automatically: nodes of the selected domain(s) having a nodal, linear or surface support attached
will be considered to be supported if the support has a nonzero stiffness in the local z direction.
The following nodes are also considered to be supported:
a. nodes of the selected domain(s) with columns attached if the angle between the axis of the
column and the local z direction is less than 45°,
b. nodes of the selected domain(s) with walls attached if the angle between their local z directions
is larger than 45°,
2. Automatically (slabs only): The above method but limited to selected slabs (a domain is considered
to be a slab if its local z direction is parallel to the global Z direction).
3. Selection: the user can select both domains and supported nodes. Only nodes within the plane of
the domain will be taken into account.
2. One supported node: the reference plane has the same normal as the domain and goes through
the displaced position of the node.
3. Two supported nodes: the reference plane goes through the two displaced nodes. An in-plane
vector perpendicular to the line going through these nodes is parallel to the similar vector in
the plane of the undisplaced domain.
4. Three or more nodes: in this case the reference plane is fitted to the displaced nodes (it goes
through all supported nodes for three nodes).
The erel relative displacement of a domain node is defined as the dis-
tance of P and P’ where P is the intersection of the reference plane and
a line perpendicular to the original plane of the domain and going
through P’.
Selecting erel from the component list the results of the calculation can
be displayed for any load case, combination, envelope or critical combi-
nation.
424
Reference plane list: Displays a list of reference planes with the domains where erel was calculated from
the given reference plane and a list of supported nodes defining the reference plane. The selected
reference plane can be deleted by clicking on the Delete button or by pressing Del.
Save diagrams Diagrams can be saved to the Drawings Library and inserted into reports the usual way.
to the Drawings Li-
brary
Dynamic analysis provides six nodal velocity components in the global coordinate system (three
translational and three rotational) at each node. The resultant values of translational (vR) and of
rotational ones (vRR) are also determined.
Dynamic analysis provides six nodal acceleration components in the global coordinate system (three
translational and three rotational) at each node. The resultant values of translational (aR) and of
rotational ones (aRR) are also determined.
When displaying the Envelope and Critical results the minimum and maximum values can concomitantly
be displayed.
The internal forces are related to the element local coordinate system, and the positive sign conventions
apply as in the figure above. The moment diagrams are drawn on the tension side of the beam elements.
Displaying the internal forces of a frame:
Nx diagram Vz diagram
Displaying You can display internal forces of more than one beam/truss element if:
beam/truss
internal forces a) the angular deviation of local x and z axes does not exceed the respective maximum values
specified in the Result display parameters / Miscellaneous settings dialog.
b) The elements have the same material.
If you click a beam element all six beam internal force components are displayed in a diagram form.
426
On selecting envelope or critical load combination, the selected beam internal force minimum and
maximum values of the intermediate cross sections will be displayed.
You can display the diagrams corresponding to any load case or combination, as well as envelopes. You
can turn on and off the display of envelope functions and set the position along the member where you
want the results displayed.
Diagrams for Besides the Nx, Vy, Vz, Tx, My, Mz forces and moments a diagram of the B bimoment also appears. In
7 DOF beams load cases and load combinations the Tx diagram also displays the Mw warping torsion.
If the min/max values occur in a single location the concomitant values of the afferent internal force
components are displayed, or the symbol * (if there are multiple locations). An occurrence of such a
location is displayed.
See 6.1.5 Result tables
Three orthogonal internal forces, one axial and two shear forces (Nx,
Vy, Vz). and three internal moments, one torsional and two flexural (Tx,
My, Mz) are calculated at the nodes of each element. The rib can be
used independently (not connected to a surface element), or
connected to a surface element.
The internal forces are related to the element local coordinate system positioned in the center of gravity
of the cross-section, and the positive sign conventions apply as in the figure below. The moment
diagrams are drawn on the tension side of the beam elements.
User’s Manual X6 Release1 e2 427
If the rib is connected eccentrically to a shell element, axial forces will appear in the rib and in the shell.
In this case the design moment can be calculated as follows: 𝑀𝑦𝐷 = 𝑀𝑦 + 𝑒𝑧 ⋅ 𝑁𝑥
Assumptions for the The design moment MyD is always calculated by the program, but the above calculation method gives
calculation of MyD good results only under the following assumptions. In all cases, the user is to check their fulfilment.
component
• there is a fully rigid shear connection between the rib and the shell,
• there is no significant normal force in the rib except from the bending,
• the rib is rigid/large enough compared to the slab so neglecting bending moment of the
slab does not cause considerable error (bending moment of the slab is not included in
MyD).
MyD is also calculated and displayed when elastic shear connection is defined between the slab and rib,
but in this case the program gives a warning message when MyD result component is selected (it appears
only once after selecting this component for the first time after the analysis).
Conditions of the In case of bottom and top ribs, if Elastic type shear connection (see 4.9.9 Line elements /Ribs) is set then
calculation of vxz the specific longitudinal shear force - vxz result component - is also calculated. Using Rigid shear con-
component nection this type of result component is not available).
Displaying rib You can display internal forces of more than one rib element if:
internal forces
a) the angular deviation of local x and z axes does not exceed the respective maximum values
specified in the Result display parameters / Miscellaneous settings dialog.
b) The elements have the same material.
Displaying the internal forces of a ribbed plate:
Tx diagram My min/max envelope
Calculation After defining virtual beams (see 2.16.16 Virtual beams), the program calculates center of gravity in each
section to determine the centre line, on which a finite number of sections are taken, depending on the
geometry, connecting elements and the density of the finite element mesh. In each section, the section
forces are reduced into the intersection of the centre line and the plane of the section. After this proce-
dure, the results of the virtual beam are plotted over the centre line.
Components of in- Three orthogonal internal forces, one axial and two shear
ternal forces forces (Nx , Vy , Vz ) and three internal moments, one tor-
sional and two flexural (Tx, My, Mz) are calculated at the in-
termediate cross-sections of each element.
Internal forces are related to the local coordinate system, and
the positive sign conventions apply as in the figure. Moment
diagrams are drawn on the tension side of the beam ele-
ments.
Result tables The results are listed in all sections, with the global coordinates of the center of gravity.
Surface elements
Membrane nx Plate mx Shell nx
ny my ny
nxy mxy nxy
vxz mx
vyz my
mxy
vxz
vyz
User’s Manual X6 Release1 e2 429
Isoline Isosurface 2D
The x and y index of the plate moments indicates the direction of the normal stresses that occur
due to the corresponding moment, and not the rotation axis.
So, the mx moment rotates about the y local axis, while the my about the x local axis.
The moment diagrams of plate and shell elements are drawn on the tension side. On the top surface
(determined by the local z direction) the sign is always positive, on the bottom surface it is always
negative.
Intensity variation The finite element method is an approximate method. Under normal circumstances the results converge
to the exact values as the mesh is refined.
The refinement of the mesh (the number of the elements used in the mesh), the geometry of the
elements, the loading and the support conditions, and many other parameters influence the results.
Therefore some results will be relatively accurate whereas other results require the user to determine if
they meet the conditions of accuracy that he expects.
The intensity variation values are intended to give you help in identifying the regions in your model
(mesh) where it is possible that the accuracy of the results is not satisfactory, without performing an
additional analysis. This method does not show that the results are good, but will highlight intensity
variations with high magnitudes, where you may want to check and/or refine your mesh.
The allowable values of the intensity variation can be determined based on practice.
Averaging If we model columns connecting to slabs as nodal supports, moment peaks will appear over the supports.
peak moments over If we use a denser mesh these peaks increase due to the nature of the finite element method. A more
columns realistic model takes into account the fact that columns have a nonzero cross-section area. Knowing the
column cross-section moment peaks can be averaged. If we checked the option Adjust mesh to column
heads (4.11.1.2 Meshing of domains), the mesh already follows the column cross-section. After turning on
Cut moment peaks over columns on the Display parameters dialog (6.1 Static), moment diagrams will be
displayed in Isosurfaces 3D mode like the right diagram below.
Diagram without cutting moment peaks Diagram with moment peaks cut
Principal forces The n1, n2 and m1, m2, principal internal forces are computed with their
direction: αn and αm.
The direction of principal forces 1 and 2 can also be queried individually
choosing αn1, αn2 and αm1, αm2 result components.
Above these, the resultant shear force vRz is also calculated with its direction:
αvRz.
The sign conventions are as follows:
𝑚1 ≥ 𝑚2 , 𝑛1 ≥ 𝑛2 , −90° < 𝛼 ≤ +90° (relative to the local x axis)
Shell
Membrane Plate
𝑛𝑥 + 𝑛𝑦 𝑛𝑥 − 𝑛𝑦 2
𝑛1 𝑛1 = + √( 2
) + 𝑛𝑥𝑦 —
2 2
𝑛𝑥 + 𝑛𝑦 𝑛𝑥 − 𝑛𝑦 2
𝑛2 𝑛2 = − √( 2
) + 𝑛𝑥𝑦 —
2 2
2𝑛𝑥𝑦
𝛼𝑛 tg(2𝛼𝑛 ) = —
𝑛𝑥 − 𝑛𝑦
𝑚𝑥 + 𝑚𝑦 𝑚 𝑥 − 𝑚𝑦 2
𝑚1 — 𝑚1 = + √( 2
) + 𝑚𝑥𝑦
2 2
𝑚𝑥 + 𝑚𝑦 𝑚 𝑥 − 𝑚𝑦 2
𝑚2 — 𝑚2 = − √( 2
) + 𝑚𝑥𝑦
2 2
2𝑚𝑥𝑦
𝛼𝑚 — tg(2𝛼𝑚 ) =
𝑚 𝑥 − 𝑚𝑦
𝑣𝑅𝑧 — 2 + 𝑣2
𝑣𝑅𝑧 = √𝑣𝑥𝑧 𝑦𝑧
𝑣𝑦𝑧
𝛼𝑣𝑅𝑧 — 𝛼𝑣𝑅𝑧 = 𝑎𝑟𝑐𝑡𝑔 ( )
𝑣𝑥𝑧
The internal forces can be displayed in diagram, section line, isoline or isosurface style.
The principal directions (αn, αn1, αn2 and αm, αm1, αm2 and αvRz) can be displayed only as
diagrams+++. The color and the size of the direction vector are determined based on the value of the
respective principal internal forces.
If the principal internal force is negative the corresponding direction vector is bounded by two segments
perpendicular to it.
Reinforcement For surface elements nxv, nyv, mxv, myv reinforcement (design) forces and moments are also calculated
forces according to the following rules:
𝑛𝑥𝑣 = 𝑛𝑥 ± |𝑛𝑥𝑦 |, 𝑛𝑦𝑣 = 𝑛𝑦 ± |𝑛𝑥𝑦 |
𝑚𝑥𝑣 = 𝑚𝑥 ± |𝑚𝑥𝑦 |, 𝑚𝑦𝑣 = 𝑚𝑦 ± |𝑚𝑥𝑦 |
The reinforcement design forces can be displayed in diagram, section line and iso-line / surface colored
form.
Intensity variation dnx, dny, dnxy, dmx, dmy, dmxy, dqx, dqy are dimensionless result components showing the intensity
variation within one finite element. The difference between the maximum and minimum value within the
finite element is divided by a reference value. The reference value can be set to any of the following:
• Absolute maximum of the respective force component over the entire model
• Absolute maximum of the respective force component over the active parts
• Custom value
The reference value can be selected in the Result display parameters dialog (see 6.1 Static).
Resultant internal The resultant internal forces 𝑅𝑒𝑅 , 𝑅Θ𝑅 are computed as follows:
forces
𝑅𝑒𝑅 = √𝑅𝑥2 + 𝑅𝑦2 + 𝑅𝑧2 , 2 + 𝑅2 + 𝑅2
𝑅Θ𝑅 = √𝑅x𝑥 y𝑦 z𝑧
Rxyz and Rxxyyzz result components refer to a special display mode where the individual force
or moment components are displayed simultaneously as three arrows pointing in the respective
local direction.
Ratio of horizontal A result component showing the ratio of horizontal and vertical forces is calculated as
and vertical forces 1
𝛼𝑅 = √𝑅𝑥2 + 𝑅𝑦2
𝑅𝑧
Displaying the internal forces of supports in a frame and a shell structure:
Ryy moments ReR resultant forces
Optionally results can be averaged over each structural member separately. The method of averaging
can be selected in the Result display parameters dialog (see 6.1 Static).
Average value For surface support forces a separate display mode is introduced displaying the average values of Rx, Ry
or Rz. Optionally results can be averaged over each domain separately. The method of averaging can be
selected in the Result display parameters dialog (see 6.1 Static). It is also possible to calculate average for
positive and negative regions separately.
6.1.14. Internal forces of line to line link elements and edge hinges
Internal forces AxisVM determines the nx, ny, nz forces and mx, my, mz moments for line-to-line link elements and
edge hinges. If any stiffness component is set to zero the related result component is zero and not
displayed neither in the component combo nor in result tables.
Internal forces are related to the local coordinate system. The posi-
tive sign conventions apply as in the figure.
The strain results are only available in case of materially nonlinear analysis.
Strain components
The following strain components are available for the line elements:
The strain results at stress points are only available in case of an analysis with material nonlinearity. The
following strain values are calculated for truss elements and in each stress point of each cross-section
of the beam/rib elements:
Strain components at
stress points
exx T Axial strain in local x direction, total
(NLP) exx E Axial strain in local x direction, elastic
(NLP) exx P Axial strain in local x direction, plastic
(NLP) eeff Effective plastic strain
(NLP) deeff Effective plastic strain increment
Effective strain
(NLE) eeff
(𝑁𝐿𝐸) 𝑒𝑒𝑓𝑓 = |𝑒𝑥𝑥 𝑇|
The strain results are only available in case of materially nonlinear analysis.
Strain components
The following strain components are available for the surface elements:
The strain results at stress points are only available in case of materially nonlinear analysis. The following
strain components are calculated at each node of the element in the top, center, and bottom fiber:
Deformation
Deformations in local directions
components
ex 𝑒𝑥 = 𝑒𝑥𝑗 − 𝑒𝑥𝑖
ez 𝑒𝑦 = 𝑒𝑦𝑗 − 𝑒𝑦𝑖
ez 𝑒𝑧 = 𝑒𝑧𝑗 − 𝑒𝑧𝑖
Rotational deformations about local axes
fx 𝑓𝑥 = 𝑓𝑥𝑗 − 𝑓𝑥𝑖
fy 𝑓𝑦 = 𝑓𝑦𝑗 − 𝑓𝑦𝑖
fz 𝑓𝑧 = 𝑓𝑧𝑗 − 𝑓𝑧𝑖
The display modes for stress results are the same as for the internal forces. The table of the stress results
are similar to those of internal forces.
Truss The 𝑆𝑥 = 𝑁𝑥 /𝐴𝑥 stress value is calculated for each truss element.
A positive value means tension.
Beams / Ribs The following stress values are calculated in each stress point of each cross-section of the beam/rib
element:
Normal stress from tension/compression and bending is calculated disregarding warping stress:
𝑁𝑥 𝑀𝑦 𝐼𝑧 + 𝑀𝑧 𝐼𝑦𝑧 𝑀𝑧 𝐼𝑦 + 𝑀𝑦 𝐼𝑦𝑧
𝑆𝑥,𝑖 = + 2 𝑧𝑖 − 2 𝑦𝑖
𝐴𝑥 𝐼𝑦 𝐼𝑧 − 𝐼𝑦𝑧 𝐼𝑦 𝐼𝑧 − 𝐼𝑦𝑧
where yi , zi are the stress point coordinates. Positive stress value means tension in the cross-section.
Resultant shear stress is calculated from shear and twisting (Saint-Venant) disregarding warping shear
stress.
User’s Manual X6 Release1 e2 437
where the last two terms are the shear stress from twisting derived from shear flow in closed and open
subsections. 𝑚𝑖 is the distance of the center of gravity from the segment, 𝑡𝑖 is the wall thickness of the
segment. 𝜔, Φ𝑦 and Φ𝑧 are centerline values.
If a cross-section contains two or more separate parts Vi and So,i is not calculated.
𝑉𝑦 𝑉𝑧
Mean shear stresses: 𝑉𝑦,𝑚𝑒𝑎𝑛 = , 𝑉𝑧,𝑚𝑒𝑎𝑛 = , if 𝐴𝑦 = 𝐴𝑧 = 0 then 𝐴𝑦 = 𝐴𝑦 = 𝐴𝑥
𝐴𝑥 𝐴𝑥
Stresses of compo- The modulus-weighted [16, Chapter 1.2] stresses are calculated for each part of the composite cross-
site cross-sections sections. The reference moduli 𝐸𝑟𝑒𝑓 and 𝐺𝑟𝑒𝑓 belong to the material of the outer part of the cross-
section.
Normal stress from tension/compression and bending is calculated disregarding warping stress:
𝐸𝑖 𝑁𝑥 𝑀𝑦 𝐼𝑧 + 𝑀𝑧 𝐼𝑦𝑧 𝑀𝑧 𝐼𝑦 + 𝑀𝑦 𝐼𝑦𝑧
𝑆𝑥,𝑖 = ( + 2 𝑧𝑖 − 2 𝑦𝑖 )
𝐸𝑟𝑒𝑓 𝐴𝑥 𝐼𝑦 𝐼𝑧 − 𝐼𝑦𝑧 𝐼𝑦 𝐼𝑧 − 𝐼𝑦𝑧
where yi , zi are the stress point coordinates and 𝐸𝑖 is the Young’s modulus of the part which the stress
point belongs to.
Resultant shear stress is calculated from shear and twisting (Saint-Venant) disregarding warping shear
stress.
For stress points belonging to thick-walled parts 𝑉𝑖 = √𝑉𝑦,𝑖
2 2
+ 𝑉𝑧,𝑖 ,
where shear stress components are:
𝐺𝑖 𝑉𝑦 𝜕Φ𝑦 𝑉𝑧 𝜕Φ𝑧 𝑀𝑥 𝜕𝜔
𝑉𝑦,𝑖 = ( ( ) + ( ) + [( ) − 𝑧𝑖 ] )
𝐺𝑟𝑒𝑓 𝐴𝑥 𝜕𝑦 𝑖 𝐴𝑥 𝜕𝑦 𝑖 𝐼𝑥 𝜕𝑦 𝑖
𝐺𝑖 𝑉𝑦 𝜕Φ𝑦 𝑉𝑧 𝜕Φ𝑧 𝑀𝑥 𝜕𝜔
𝑉𝑧,𝑖 = ( ( ) + ( ) + [( ) + 𝑦𝑖 ] )
𝐺𝑟𝑒𝑓 𝐴𝑥 𝜕𝑧 𝑖 𝐴𝑥 𝜕𝑧 𝑖 𝐼𝑥 𝜕𝑧 𝑖
where Φ𝑦 and Φ𝑧 are the shear stress functions for shear in y and z direction, is the warping function
and 𝐺𝑖 is the shear modulus of the part which the stress point belongs to.
where the last two terms are the shear stress from twisting derived from shear flow in closed and open
subsections. 𝑚𝑖 is the distance of the centre of gravity from the segment, 𝑡𝑖 is the wall thickness of the
segment. 𝜔, Φ𝑦 and Φ𝑧 are centerline values.
Stress points on the boundary of the two component cross-sections always belong to the thin-
walled cross-section.
where 𝜎𝑦𝑇 and 𝜎𝑦𝐶 are the closest points of the stress-strain diagram of the material (see 3.1.15
Material Library) on the positive and on the negative region and 𝜀𝑒𝑓𝑓,𝑖 is the effective strain
[(NLE) eeff, see 6.1.17 Truss, beam and rib element strains at stress points]
Utilization:
(𝑁𝐿) 𝑆𝑒𝑓𝑓,𝑖
(𝑁𝐿) 𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛𝑖 =
(𝑁𝐿) 𝑓𝑦,𝑖
Plastic / von Mises material model
Effective stress.
Only the normal stress affects the materially nonlinear behaviour of truss, beam and rib elements.
The shear stress has no effect [5]
(𝑁𝐿) 𝑆𝑒𝑓𝑓,𝑖 = |𝑆𝑟,𝑥.𝑖 |
where 𝑆𝑟,𝑥,𝑖 = 𝑆𝑥,𝑖 − 𝐵𝑥,𝑖 is the relative stress with respect to the centre of the yield surface.
See 3.1.15 Material Library
where 𝜀𝑒𝑓𝑓,𝑖 is the effective plastic strain ( (NLP) eeff, see 6.1.17 Truss, beam and rib element strains
at stress points)
Utilization:
(𝑁𝐿) 𝑆𝑒𝑓𝑓,𝑖
(𝑁𝐿) 𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛𝑖 =
(𝑁𝐿) 𝑓𝑦,𝑖
𝜎𝑦𝑇 𝑖𝑓 𝑆𝑥,𝑖 ≥ 0
(𝑁𝐿) 𝑓𝑦,𝑖 = {
𝜎𝑦𝐶 𝑖𝑓 𝑆𝑥,𝑖 < 0
where 𝜎𝑦𝑇 and 𝜎𝑦𝐶 are tensile and compressive yield strength respectively (see 3.1.15 Material Li-
brary)
Utilization:
(𝑁𝐿) 𝑆𝑒𝑓𝑓,𝑖
(𝑁𝐿) 𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛𝑖 =
(𝑁𝐿) 𝑓𝑦,𝑖
Beam stresses Sminmax, Vminmax, Sominmax are minimum / maximum values within the cross-section
and displayed like internal forces.
You can click a beam/rib element to display stress diagrams. On the left the minimum/maximum values
along the line are displayed. Dragging the blue line with the mouse the evaluation position can be
changed. The axonometric diagrams in the middle and the tables on the right show the stress distribu-
tion within the section at the evaluation point.
Select more elements before clicking to display them in one diagram. Continuous beams/ribs can be
displayed in one diagram if conditions described in section 6.1.9 Truss/beam internal forces are satisfied.
440
You can display the diagrams corresponding to any load case or combination, as well as envelopes. You
can turn on and off the display of envelope functions and set the position along the member where you
want the results displayed.
Save diagrams Associative diagrams can be saved to the Drawings Library. Drawings from this library can be inserted
to the Drawings Li- into reports. After changing and recalculating the model diagrams in the library and reports change
brary accordingly.
Selecting envelope or critical combinations only one of the min and max components will appear
depending on the component. If extreme values are located in one cross-section only you will
see values of the other components as well. Otherwise a will appear and the cross-section
location will be the first one.
Result Tables See 6.1.5 Result tables
Result display on a Distribution of normal stress in the beam cross-section can be displayed in the Solid model result display
solid model mode.
User’s Manual X6 Release1 e2 441
The following stress components are calculated at each node of the element in the top, center, and
bottom fiber:
𝑠𝑥𝑥 𝑛𝑥 6 𝑛𝑥 6
𝑠𝑥𝑥 = 𝑠𝑥𝑥 = ± 𝑚 𝑠𝑥𝑥 = ± 2 𝑚𝑥
𝑡 𝑡2 𝑥 𝑡 𝑡
𝑠𝑦𝑦 𝑛𝑦 6 𝑛𝑥 6
𝑠𝑦𝑦 = 𝑠𝑦𝑦 = ± 𝑚 𝑠𝑦𝑦 = ± 2 𝑚𝑦
𝑡 𝑡2 𝑦 𝑡 𝑡
In the case of plane strain membrane elements 𝒔𝒛𝒛 ≠ 𝟎, and is determined as 𝒔𝒛𝒛 = 𝒗 ⋅ (𝒔𝒙𝒙 + 𝒔𝒚𝒚 )
Von Mises stress The Von Mises stress is computed:
2 2 2 2 2
𝑠0 = √0.5 [(𝑠𝑥𝑥 − 𝑠𝑦𝑦 ) + (𝑠𝑦𝑦 − 𝑠𝑧𝑧 ) + (𝑠𝑧𝑧 − 𝑠𝑥𝑥 )2 ] + 3(𝑠𝑥𝑦 + 𝑠𝑦𝑧 + 𝑠𝑧𝑥 )
Only the normal stresses 𝒔𝒙𝒙 , 𝒔𝒚𝒚 , 𝒔𝒛𝒛 and the shear stress 𝒔𝒙𝒚 affect the materially nonlinear
behaviour of surface elements. The shear stresses 𝒔𝒙𝒛 and 𝒔𝒚𝒛 have no effect. Further information
see at References [5]
Actual yield strength
It is the limit of proportionality influenced by the actual multiaxial stress state:
where 𝑟 is the ratio of the tensile and compressive components of the multiaxial stress state:
𝐼1
−1 𝑖𝑓 < −1
𝜎𝑉𝑀
𝐼1 𝐼1
𝑟= 𝑖𝑓 − 1 ≤ ≤1
𝜎𝑉𝑀 𝜎𝑉𝑀
𝐼1
1 𝑖𝑓 >1
{ 𝜎𝑉𝑀
where 𝐼1 is the first invariant of the stress tensor and 𝜎𝑉𝑀 is the von Mises equivalent stress:
2 2 2
𝜎𝑉𝑀 = √0.5 [(𝑠𝑥𝑥 − 𝑠𝑦𝑦 ) + (𝑠𝑦𝑦 − 𝑠𝑧𝑧 ) + (𝑠𝑧𝑧 − 𝑠𝑥𝑥 )2 ] + 3𝑠𝑥𝑦
Utilization:
(𝑁𝐿) 𝑠𝑒𝑓𝑓
(𝑁𝐿) 𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 =
(𝑁𝐿) 𝑓𝑦
where 𝑠𝑟,𝑥𝑥 = 𝑠𝑥𝑥 − 𝑏𝑥𝑥 , 𝑠𝑟,𝑦𝑦 = 𝑠𝑦𝑦 − 𝑏𝑦𝑦 , 𝑠𝑟,𝑧𝑧 = 𝑠𝑧𝑧 − 𝑏𝑧𝑧 and 𝑠𝑟,𝑥𝑦 = 𝑠𝑥𝑦 − 𝑏𝑥𝑦 are the relative stresses
with respect to the centre of the yield surface (see 3.1.15 Material Library)
Only the normal stresses 𝒔𝒙𝒙 , 𝒔𝒚𝒚 , 𝒔𝒛𝒛 and the shear stress 𝒔𝒙𝒚 affect the materially nonlinear
behaviour of surface elements. The shear stresses 𝒔𝒙𝒛 and 𝒔𝒚𝒛 have no effect. [5]
User’s Manual X6 Release1 e2 443
where 𝜀𝑒𝑓𝑓 is the effective plastic strain (NLP) eeff, see 6.1.19 Surface element strains at stress points
Utilization:
(𝑁𝐿) 𝑠𝑒𝑓𝑓
(𝑁𝐿) 𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 =
(𝑁𝐿) 𝑓𝑦
Only the normal stresses 𝒔𝒙𝒙 , 𝒔𝒚𝒚 , 𝒔𝒛𝒛 and the shear stress 𝒔𝒙𝒚 affect the materially nonlinear
behaviour of surface elements. The shear stresses 𝒔𝒙𝒛 and 𝒔𝒚𝒛 have no effect. [5]
Utilization:
(𝑁𝐿) 𝑠𝑒𝑓𝑓
(𝑁𝐿) 𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛 =
(𝑁𝐿) 𝑓𝑦
Stress state for non- It is a result component to identify the tensioned, compressed and sheared regions of the structure. Its
linear materials numerical value is based on the value of the so called stress triaxiality factor that is the ratio of the
hydrostatic and the deviatoric stress component.
𝜎𝑚
(𝑁𝐿) 𝑆𝑡𝑎𝑡𝑒 = 3
𝜎𝑉𝑀
where 𝜎𝑚 is the hydrostatic or mean stress and 𝜎𝑉𝑀 is the von Mises equivalent stress characteristic
for the deviatoric or shear stresses:
2 2 2
𝜎𝑉𝑀 = √0.5 [(𝑠𝑥𝑥 − 𝑠𝑦𝑦 ) + (𝑠𝑦𝑦 − 𝑠𝑧𝑧 ) + (𝑠𝑧𝑧 − 𝑠𝑥𝑥 )2 ] + 3𝑠𝑥𝑦
The following picture illustrates the stress state on the plane of the pricipal stresses S1-S2 with the
belonging values of (NL) State.
(NL) State
Biaxial tension
Pure tension
Mainly tensioned
Mainly sheared
Pure shear
Mainly sheared
Mainly compressed
Pure compression
Biaxial compression
Using (NL) Utilization together with (NL) State the critical regions and the failure mode of them can be
determined.
Result display on a Top and bottom values of surface stresses can be displayed simultaneously in the Solid model result
solid model display mode.
Translational
Description
components
𝐹𝑦𝑇 ℎ𝑎 𝑅𝑥 ≥ 0
(𝑁𝐿) 𝐹𝑦,𝑥 = {
𝐹𝑦𝐶 ℎ𝑎 𝑅𝑥 < 0
where 𝐹𝑦𝑇 and 𝐹𝑦𝐶 are the limit forces defined by the user for tension and compression (see 3.1.17
Spring characteristics library).
Utilization:
𝑅𝑥
(𝑁𝐿) 𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛𝑥 = | |
(𝑁𝐿) 𝐹𝑦,𝑥
446
Utilization:
𝑅𝑥
(𝑁𝐿) 𝑈𝑡𝑖𝑙𝑖𝑧𝑎𝑡𝑖𝑜𝑛𝑥 = | |
(𝑁𝐿) 𝐹𝑦,𝑥
Translational
Description
components
Rotational
Description
components
Displays the internal force influence lines corresponding to the unit applied forces P X, PY, PZ that act in
the positive direction of the global coordinate axes. An ordinate of the influence line represents the
value of the respective internal force that occurs in the respective cross-section caused by an applied
unit force at the position of the ordinate.
User’s Manual X6 Release1 e2 447
Truss Clicking a truss shows the elements’ absolute maximum ordinate value.
Beam Clicking a beam shows the elements’ absolute maximum ordinate value and its location.
Displaying the internal force influence line diagrams of a frame:
Unit force in Z direction Nx influence line
The resultant of all external loads in the global coordinate system (E) is calculated for each load case.
The unbalanced loads for each load case is also displayed (UNB). The unbalanced loads does not appear
on the supports, therefore, if there are non-zero unbalanced load components, it usually means that a
part of the external loads are supported by constrained degrees of freedom and not the supports.
6.2. Buckling
Displays the results of a buckling analysis (buckling mode shapes and critical load parameters).
The module allows imperfections to be taken into account by creating an imperfect geometry by scaling
and superimposing buckling shapes. Second-order effects from imperfections can be considered via ge-
ometrically nonlinear analysis run with imperfect geometry.
The tree on the left lists the buckling mode shapes obtained from buckling analysis. The tree in the middle
displays the shapes of imperfection constructed from the available buckling shapes. The parameters as-
signed to each buckling mode shape can be set on the right side of the window.
Creates a new imperfect shape
Delete imperfect shapes or constituents. In case of multiple selection, all selected items are deleted. De-
leting a load case deletes all associated buckling shapes, deleting an imperfect shape deletes all associ-
ated load cases and buckling shapes.
Use the Add button to add the buckling shapes selected in the left tree to the imperfect shapes selected
in the middle. Multiple buckling shapes can be selected. If a buckling load case is selected, all associated
shapes are copied. Multiple imperfections or their constituents can be selected in the middle tree. Se-
lected buckling shapes are added to each imperfection affected by the selection. Added buckling shapes
are automatically organized into load case folders, and it is not possible to group them or change their
order. If a selected buckling shape is already present in the target imperfection, it is not added again.
Selected buckling shapes can also be added by dragging them over to an imperfect shape. Selected
shapes will be added only to the target of the dragging operation.
User’s Manual X6 Release1 e2 449
Component, To create imperfections, a displacement or rotation component or resultant must be selected. Scaling
Maximum and superposition are performed according to the following formulas: A scaling displacement or rotation
displacement must be specified for each buckling shape. Buckling shapes are scaled so that the absolute value of the
largest deformation in the direction of the selected component is to be equal to the absolute value of
the scaling value. The sign of the scaling value can be used to control whether the direction of the de-
formations of the shape is the same as or opposite to the direction of the buckling shape.
𝑰𝑴𝑷 = ∑ 𝒄𝒊 ∙ 𝜼𝒄𝒓,𝒊
(𝒊)
where IMP is the compiled imperfect shape, 𝜼𝒄𝒓,𝒊 is the ith buckling shape, 𝒄𝒊 is the factor of the ith buckling
shape:
𝒆𝟎,𝒊
𝒄𝒊 =
|𝜼𝒄𝒓,𝒊 |𝒎𝒂𝒙
where 𝒆𝟎,𝒊 is the scaling displacement or rotation, |𝜼𝒄𝒓,𝒊 |𝒎𝒂𝒙 is the absolute value of the maximum defor-
mation in the direction of the selected component in the ith buckling shape.
The entered component and scaling value are applied to all selected shapes.
The constructed imperfect shapes are appended to the dropdown list of the buckling tab, so they can
also be displayed in the main window.
Creating and deleting imperfect shapes automatically creates and deletes the associated load cases, but
they can also be edited among the shape? load cases. See 4.10.27 Geometric imperfections based on
buckling shapes
If a buckling shape is deleted due to changes in the model imperfect shapes that use it as a constituent
are not deleted only invalidated. After running the buckling analysis again these imperfect shapes are
updated automatically from the new shapes.
The updated imperfect shapes need to be reviewed as the order of buckling modes might have
been changed, existing shapes might have disappeared, or new ones might have formed.
6.3. Vibration
Displaying vibration Select Vibration response factor from the end of the list of vibration modes. Vibration response factors
response factor determined with different excitation methods (R full, R extr., R self) can be selected from the list of result
components.
Activated mass The program calculates the activated mass (some article use the term ‘modal mass’) in each X, Y and Z
directions, using the mode shapes ordinates and the masses.
For beams:
Mmod = μ ∫ δ2 (x) ∙ dl
l
For surfaces:
Mmod = μ ∫ δ2 (x, y) ∙ dA
A
User’s Manual X6 Release1 e2 451
where:
μ – the distribution of mass
δ – the normalized mode shape ordinates in the given direction
The calculated ‘activated mass’ values for each mode shapes and for each direction is displayed in the
Frequency Tables/Activated masses table.
The activated masses and frequencies are useful to determine and verify floor designs for vibrations
due to walking. In Eurocode there are few guidelines to the above design. Hence the EU sponsored
research project worked out the OS-RMS90 guideline to vibration design of floors.
Further info can be found in the following PDF file:
http://www.stahlbau.stb.rwth-aachen.de/projekte/2007/HIVOSS/docs/Guideline_Floors_EN02.pdf
6.4. Dynamic
Surface reinforcement can be calculated based on Eurocode 2. The calculation of the reinforcement of
membrane, plate, and shell elements is based on the 3rd stress condition. Reinforcement directions are
the same as the local x and y directions. The nominal moment and corresponding axial strengths are
determined based on the restricted direction optimal design.
Result display on a Top and bottom values can be displayed simultaneously in the Solid model result display mode.
solid model
Reinforcement pa- In the surface reinforcement design, the following parameters must be assigned to the finite elements:
rameters
Nonlinear analysis The value of some parameters considered during nonlinear analysis of surface elements (see 6.5.6 Non-
linear analysis of RC surfaces) can be set here
• taking into account the tensile or flexural tensile strength of concrete;
• final concrete strain due to shrinkage;
• type of dominant nonlinear behaviour (wall, slab or shell)
This feature is available for the following domain types (see 4.9.6 Domain):
• normal domains,
• composite ribbed domains,
• parametric ribbed domains,
• composite trapezoidal plates (if the Trapezoidal plate used only as formwork option is checked).
In case of walls the flexural moments, in case of slabs the membrane forces are calculated elas-
tically.
The purpose of the consideration of nonlinear behavior in nonlinear static analysis is the more
accurate deflection calculation for surface concrete elements by verification in serviceability limit
states considering nonlinear (material and structural) behavior. The verification of reinforced con-
crete elements in ultimate limit states using this fiber integration model is not recommended. It
is also important to note that the present model cannot be used in pushover analysis in order to
take into account nonlinear behaviour of reinforced concrete elements instead of plastic hinges
(see 4.10.24 Seismic loads – SE1 module) because the presented material models do not consider
cyclic degradation in concrete and steel strength / stiffness, buckling of steel reinforcement bars,
etc.
Apply minimum The program determines the minimum top and bottom concrete cover from the environment class ac-
cover cording to the current design code.
Take into account The program determines the required minimum top and bottom reinforcement according to the current
the required mini- design code. If the calculated amount of reinforcement is smaller than these values the required minimum
mum reinforce- is used.
ment
454
Load transfer AxisVM takes into account the load transfer mode when calculating minimum and maximum rebar spac-
ing and minimum steel percentage according to the design code and the respective national annex. Rebar
spacing is adjusted according to design rules only if Follow rebar layout rules for spacing is checked in
Result display parameters / Miscellaneous settings (see 6.1 Static).
If the calculated amount of reinforcement would result in rebar spacings above the maximum the
maximum is used. If it would be under the minimum spacing the slab cannot be reinforced.
Take concrete ten- If this option is activated the program assumes that the stress in concrete remains below the concrete
sile strength into tensile strength and does not calculate crack width
account
Reinforcement cal- In SLS load combinations, the required reinforcement is calculated based on user specified limiting
culation based on cracking width. If the actual reinforcement parameters is adjusted to the reinforcement paramters of the
limiting cracking domain (identical diameters and concrete cover, as defined in this window), the calculated crack width
width will not exceed the limiting value.
In case of a simple load case or an user defined load combination (not categorized as ULS or SLS), the
required reinforcement is calculated considering both ULS and SLS requirements and the greater value
is selected. In case of Auto critical combination formula (see 6.1 Static) required reinforcement is
calculated considering both ULS (in ULS combinations only) and SLS (the type of the SLS combination
applied depends on the standard) requirements and the greater value is shown.
Reinforcement The required reinforcement can be calculated with the consideration of local x/y or custom reinforcement
directions directions. If custom reinforcement directions are selected, ξ/η symbols are used instead of x/y to identify
the reinforcement directions. α is the angle between local x axis of the surface and ξ reinforcement di-
rection, while β refers to the angle between the two (ξ and η) reinforcement directions.
Orthogonal reinforcement mesh with custom α angle (-360°<α<360°) and skew reinforcement (β≠90°)
can be also defined. Lower and upper bounds for angle β are 45° and 135°.
User’s Manual X6 Release1 e2 455
If at least one skewly reinforced surface exists in the model, the labels are replaced on many
drawings and tables. In these cases ξ/η symbols are used instead of x/y. For example, axb is
replaced with aξb (if every surface is skewly reinforced) or with a(x/ξ)b (if there are skewly and
orthogonally reinforced surfaces).
In case of Airdeck and hollow core surface elements, required skew reinforcement cannot be
calculated. If Airdeck or hollow core surface element is included in the selection, Reinforcement
directions panel is not visible.
Required amount of The software creates a summary of the required amounts of reinforcement as Required amount of rein-
reinforcement in forcement in domains (see in the Table Browser) for load cases, load combinations, envelopes and critical
domains combinations. The content of the table can be sorted according to the types of structural elements (slab,
wall or other structure), concrete, rebar steel grades, parameters of the domains and the calculated spe-
cific and total amount of reinforcement. The Comment column displays a warning if the calculation of
required reinforcement failed (e.g., the reinforcement cannot be calculated) or the thickness of the do-
main is different from the thickness found in the reinforcement parameters.
The summary gives information about the minimum and maximum required amount of reinforcement.
The table lists only the results of the domains that are in the active parts or the selected ones (if some
domains are selected).
The calculated amount of reinforcement does not contain the additional reinforcement that comes
from the overlapping of rebars, needed anchor length of rebars, erecting rebars and space bars!
In the interpretation of the total and specific amount of reinforcement, the following shall be taken into
account:
In the case of Ribbed or Composite ribbed slabs (with concrete ribs), the calculated amount of reinforce-
ment includes only the reinforcement of the slabs, not that of the ribs.
The concrete volume of domains includes the volume of the concrete ribs if they are logical and not
actual ribs. In case of actual ribs, only the volume of the slab is taken into account.
In case of a trapezoidal steel deck, using actual domain thickness, the presented concrete volume in-
cludes the volume of the concrete ribs, even if the trapezoidal plate is used only as a formwork.
Setting individual thickness for the domain when defining reinforcement parameters overrides the actual
thickness. The concrete volume and the amount of reinforcement is calculated with that value as a nor-
mal, solid concrete slab. The trapezoidal slab is an exception: the individual thickness means the total
height of the concrete in the trapezoidal slab, the concrete volume is calculated according to this.
456
Plate If 𝑚𝑥 , 𝑚𝑦 , 𝑚𝑥𝑦 are the internal forces at a point, then the nominal moment strengths are as follows:
The moment optimum is: 𝛥𝑚 𝛥𝑚2 =0
=min!
𝑚𝑥 ≥ 𝑚 𝑦
1
m x m xy
Yes No
m txD = m x + m xy m txD = 0 2
m xy
m tyD = m y + m xy m tyD = m y +
mx
m y m xy
Yes No
2
mxy
mbxD = mx + mxy mbxD = mx +
my
mbyD = my + mxy mbyD =0
Results AxisVM calculates the tension and/or compression reinforcements (for doubly reinforced sections).
nx nxy
Yes No
Results AxisVM calculates the tension or compression reinforcements. Compression reinforcement is calculated
only in the points at which the axial compression resistance of the section without reinforcement is
lower than the compressive design axial force.
Shell If 𝑛𝑥 , 𝑛𝑦 , 𝑛𝑥𝑦 , 𝑚𝑥 , 𝑚𝑦 , 𝑚𝑥𝑦 are the internal forces in a point, the design axial forces and moments are
established based on the reserve axial force optimum and reserve moment optimum criterias that were
emphasized, at the membrane reinforcement and plate reinforcement description.
Results The following values are provided as results: axb, axt, ayb, ayt.
These represent the calculated top and bottom reinforcement in x and y directions.
6.5.1.2. Calculation of orthogonal x/y reinforcement according to DIN EN, DIN 1045-1 and SIA 262
Plate, Membrane, Reinforcement of membranes, plates and shells are calculated according to the three-layer method.
Shell The internal forces (𝑛𝑥 , 𝑛𝑦 , 𝑛𝑥𝑦 , 𝑚𝑥 , 𝑚𝑦 , 𝑚𝑥𝑦 ) are calculated in the perpendicular directions of the
reinforcement.
The surface is divided into three layers. Membrane forces for the top and bottom layers are calculated
then design forces and the required amount of reinforcement is determined.
Top layer
Medium layer
Bottom layer
Besides calculating the required reinforcement zones of concrete are checked for shear and
compression according to A, B and C cases. The program considers reduced compressive strength if
tensile stress arises in the concrete and the principal tensile stress exceeds the design tensile strength
of concrete (fctd).
Case A Case B Case C
Error message The error message 'The section cannot be reinforced' appears.
If the compressed zone of the concrete fails due shear forces.
If the compression principal stress is higher than 𝑓𝑐𝑑 .
𝐴𝑥 > 0.04𝐴𝑐 , or 𝐴𝑦 > 0.04𝐴𝑐 , where 𝐴𝑐 is the concrete cross-section area.
Tables The following symbols are used in tables:
(-) compression reinforcement bar
??? the section cannot be reinforced in the corresponding direction
No symbol appears when tension reinforcement is required.
458
Orthogonal reinforcement meshes with reinforcement directions differing from local x and y axes of the
surface and reinforcement meshes with arbitrary reinforcement directions are called skew reinforcement
and calculated as described in this section. ξ/η symbols are used instead of x/y to identify the
reinforcement directions. α is the angle between local x axis of the surface and ξ reinforcement direction,
while β refers to the angle between the two (ξ and η) reinforcement directions.
During the calculation of required reinforcements, a sandwich model with three layers is used.
Calculation methodology described in the paper of Peter Marti and Hans Seelhofer is invoked
(H. Seelhofer & P. Marti: Dimensioning of skewly reinforced concrete plates). The calculation
methodology is identical for membranes, plates and shells.
Calculation Both top and bottom layers of the sandwich model are responsible to resist the bending moments and
membrane forces. The inner layer resists shear forces.
Membrane forces and bending moments are decomposed into equivalent membrane forces in case of
the top and the bottom layer, respectively. Before the decomposition, the internal forces are transferred
from x-y coordinate system into an orthogonal x’-y’ coordinate system where x’ direction is identical
with ξ reinforcement direction.
- bottom layer:
𝑛
𝑏 = 𝑥′ − 𝑥′
𝑚
𝑛𝑥′
2 𝑑𝑣
𝑏
𝑛𝑦′ 𝑚𝑦′
𝑛𝑦′ = −
2 𝑑𝑣
𝑏
𝑛𝑥′𝑦′ 𝑚𝑥′𝑦′
𝑛𝑥′𝑦′ = −
2 𝑑𝑣
- top layer:
𝑛
𝑡 = 𝑥′ + 𝑥′
𝑚
𝑛𝑥′
2 𝑑𝑣
𝑡
𝑛𝑦′ 𝑚𝑦′
𝑛𝑦′ = +
2 𝑑𝑣
𝑡
𝑛𝑥′𝑦′ 𝑚𝑥′𝑦′
𝑛𝑥′𝑦′ = +
2 𝑑𝑣
In order to achieve optimal reinforcement, the thickness of the layers is iteratively changed. The lower
bound for layer thickness is 5 mm, while the upper limit is set to H/2 (in case of membrane-like loading
conditions, where H is the thickness of the surface) or ξc0d (in case of plate-like loading conditions,
where d is the effective depth of the reinforcement mesh). The 𝑑𝑣 level arm is calculated on the fol-
lowing way:
𝑡𝑡 𝑡𝑏
𝑑𝑣 = 𝑚𝑖𝑛 (𝑑𝑏 − , 𝑑𝑡 − )
2 2
Where 𝑡𝑏 , 𝑡𝑡 and 𝑑𝑣 are the thickness of bottom layer, the thickness of top layer and the level arm,
respectively.
The 𝑑𝑏 and 𝑑𝑡 are the effective depths related to the bottom and top reinforcement meshes, respec-
tively. The effective depth is calculated based on the average reinforcement positions of ξ and η rein-
forcements.
The stresses in the top and bottom layers are calculated as follows:
- bottom layer:
𝑛𝑏 𝑛𝑏
𝑏 = 𝑥′ ; 𝜎 𝑏 = 𝑦′ ; 𝜏 𝑏
𝑏
𝑛𝑥′𝑦′
𝜎𝑥′ 𝑦′ 𝑥′𝑦′ =
𝑡𝑏 𝑡𝑏 𝑡𝑏
- top layer:
𝑛𝑡 𝑛𝑡
𝑡 = 𝑥′ ; 𝜎 𝑡 = 𝑦′ ; 𝜏 𝑡
𝑡
𝑛𝑥′𝑦′
𝜎𝑥′ 𝑦′ 𝑥′𝑦′ =
𝑡𝑡 𝑡𝑡 𝑡𝑡
User’s Manual X6 Release1 e2 459
These stresses are transformed into ξ-η coordinate system (𝜎𝜉 , 𝜎𝜂 , 𝜏𝜉𝜂 ) on the following way
(H. Seelhofer & P. Marti: Dimensioning of skewly reinforced concrete plates):
𝜎𝜉 = 𝜎𝑥′ sin 𝛽 + 𝜎𝑦′ cos 𝛽 cot 𝛽 − 2𝜏𝑥′𝑦′ sin 𝛽
𝜎𝑦′
𝜎𝜂 =
sin 𝛽
𝜏𝜉𝜂 = 𝜏𝜂𝜉 = 𝜏𝑥′𝑦′ − 𝜎𝑦′ cot 𝛽
The concrete strut in compression is checked according to the guidelines of Annex F of EN 1992-1-1
and according to the instructions in the paper of Seelhofer and Marti:
(𝜌𝜂 + 𝜌𝜉 )𝑓𝑦 ≤ 𝜈𝑓𝑐𝑑 + 𝜎𝑥′ + 𝜎𝑦′
The program considers reduced compressive strength only if tensile stress arises in the concrete and
the principal tensile stress exceeds the design tensile strength of concrete (fctd).
If the resistance of the concrete strut is not adequate, the layer thickness is increased. Compression
reinforcement is designed in the surface if the upper limit for layer thickness is reached and the utili-
zation ratio of the concrete is still greater than 1.0. If the stresses in the layer and the utilization of
concrete strut are low, the layer thickness is decreased.
Error message The error message 'The section cannot be reinforced' appears in the following cases.
• If the resistance of the concrete strut is not adequate with maximum layer thickness notwith-
standing that compression reinforcement is applied.
• The total cross sectional area of top and bottom reinforcement whether in ξ or η
reinforcement directions is greater than the limit value for maximum reinforcement area. This
limit can be set in the Design codes window under the Reinforced concrete tab
(see 3.3.7 Design codes)
Results The following values are provided as results: aξa, aξf, aηa, aηf
6.5.1.4. Calculation of reinforcement of composite slabs with trapezoidal sheeting according to Eurocode and SIA
standards
The reinforcement of composite slabs with trapezoidal sheeting can be calculated according to the fol-
lowing design codes:
Eurocode 4: EN 1994-1-1:2004
SIA 264: SIA 262:2014
Composite slabs with trapezoidal sheeting can be created as written in the previous sections of the man-
ual (see 4.9.6.5 Trapezoidal steel deck). The geometry of the composite slab needs to satisfy the detailing
rules of the selected design code. The software checks the most important detailing rules (according to
section 4.9.2. of Eurocode 4 and section 7.2. of SIA 264) and it warns the user if the geometry has the be
modified.
The presented calculation method can be used only if the longitudinal shear behaviour is ductile.
The presented method is developed for the calculation of reinforcement in composite slabs with
trapezoidal sheeting. The verification of trapezoidal sheeting without concrete infill is not part of
the module.
460
Longitudinal reinforcement can be calculated with the presented method. Shear reinforcement and shear
resistance are not calculated. Furthermore, if a significant concentrated load is present, the punching
shear resistance of the slab need to be verified. This module does not cover the punching shear verifica-
tion. RC3 module can be used for verification against punching failure if Trapezoidal plate used only as
formwork option is checked.
This calculation does not cover the verification in construction stage! The weight of wet concrete
and temporary supports need to be considered.
Surface reinforce- In case of a composite slab with trapezoidal sheeting,
ment parameters the Surface reinforcement parameters window be-
comes slightly different comparing to the presented
window in previous sections. The main difference is
that parameters need to be defined according to re-
inforcement directions parallel with the ribs and per-
pendicular to the ribs of the slab, respectively, not
parallel with x and y axes. The main direction of the
reinforcement can be specified only for top reinforce-
ment. The concrete cover of the bottom reinforce-
ment perpendicular to the ribs needs to be specified
from the lower edge of concrete plate and not from
the lower edge of the rib. The small figure in the win-
dow helps the understanding of reinforcement direc-
tions and concrete covers.
Due to the fact that the parameters are defined ac-
cording to reinforcement directions parallel with the
ribs and perpendicular to the ribs of the slab, slabs
with different rib directions cannot be selected at the
same time. Similarly, surface reinforcement parame-
ters cannot be defined for a set of domains with dif-
ferent type simultaneously is composite slabs with
trapezoidal sheeting are also selected.
The following features are supported only if the Trap-
ezoidal plate used only as formwork option is checked:
- calculation of shear resistance and rein-
forcement,
- calculation of skew reinforcement,
- nonlinear analysis with the consideration of
cracks and reinforcement.
Single layer If Single layer checkbox is checked, only one layer reinforcement can be given in the concrete plate and
reinforcement bottom reinforcement perpendicular to the ribs cannot be defined.
Consider If Consider trapezoidal plate in calculation checkbox is not checked, the required longitudinal
trapezoidal plate in reinforcement is calculated using the same concrete section but without the consideration of the
calculation trapezoidal sheeting in equilibrium equations.
User’s Manual X6 Release1 e2 461
Calculation of crack In case of composite slabs with trapezoidal sheeting, limiting crack width cannot be given on Cracking
width panel. Independently of the state of Trapezoidal plate used only as formwork checkbox, crack width is
calculated only for the upper edge of the concrete plate. The details of the calculation of crack width can
be found in the following section (see 6.5.7 Cracking).
Crack width values are calculated parallel with the ribs and perpendicular to the ribs. If Trapezoidal plate
used only as formwork checkbox is checked, the software calculated the crack width according to principal
stress directions as for a conventional slab.
Calculation of longi- Composite slabs with trapezoidal sheeting are designed mostly as one-way slabs without the presence
tudinal reinforce- of membrane forces. In the course of calculation required longitudinal reinforcement, the software does
ment not consider the membrane forces in the slab.
Membrane forces are typically caused by modelling issues, and special attention needs to be paid for
the definition of boundary conditions.
Perpendicular to the The calculation of required longitudinal reinforcement perpendicular to the ribs is identical with the cal-
ribs culation of reinforcement for regular slabs (see 6.5.1.1 Calculation of orthogonal x/y reinforcement accor-
ding to Eurocode 2 and 6.5.1.2 Calculation of orthogonal x/y reinforcement according to DIN EN, DIN 1045-
1 and SIA 262. The reinforcement design bending moments are calculated on the same way from mx, my
and mxy bending moments.
Positive and negative design bending moment (mxD, myD) may be calculated for the same position due
to the presence of considerable torsional moment (mxy). If Single layer checkbox is checked, the
longitudinal reinforcement is calculated for the single layer from both positive and negative bending
moments, respectively.
Parallel with the The following model and equations are used for the calculation of longitudinal reinforcement parallel
ribs (lower edge in with the ribs. The width of the reference cross-section is d.
tension)
The stresses in the concrete, in the reinforcement and in the trapezoidal sheeting are calculated accord-
ing to the position of plastic neutral axis. The parts of steel trapezoidal sheeting can be in compression
or in tension based on the position of plastic neutral axis. The effect of local buckling of compressed
parts of the sheeting is taken into account by using effective widths.
If Consider trapezoidal plate in calculation checkbox is not checked, the trapezoidal steel sheeting is
neglected in the equations.
462
0.85fcd
fy,pd N Nc
p,c
As Np,t
Ap
Ns
fy,pd
The equation of equilibrium and the bending resistance are formulated as follows:
𝑥𝑐 𝑥𝑐 𝑥𝑐
𝑀𝑝𝑙,𝑅𝑑 = 𝐴𝑠 𝑓𝑦,𝑑 (𝑧𝑠 − ) + 𝐴𝑝,𝑡 𝑓𝑦𝑝,𝑑 (𝑧𝑝,𝑡 − ) − 𝐴𝑝,𝑐 𝑓𝑦𝑝,𝑑 (𝑧𝑝,𝑐 − )
2 2 2
d
If compression longitudinal reinforcement is required, components representing the forces in the com-
pression reinforcement are added to the equations.
As
Steps of the calculation of required reinforcement: Ns
• The plastic bending resistance of the composite slab is calculated without any additional rein-
d
forcement. If the resistance of the slab is adequate after comparison Nc
with design bending mo-
ment, no0.85fadditional
cd
reinforcement is required and the algorithm quits.
• In cases where 𝑀𝑝𝑙,𝑅𝑑 < 𝑀𝐸𝑑 , additional reinforcement 0.85f is required.
cd
First of all, the position of
fplastic
y,pd Np,c
neutral axis
Nc is compared with 𝑥𝑐0 :
o 𝑥𝑐 < 𝑥𝑐0 : Amount of longitudinal reinforcement and plastic bending resistance are
As Np,t calculated using 𝑥𝑐0 height of compression zone (this is the upper limit of compression
Ap zone where the tension reinforcement is still in plastic state).
Ns
fy,pd ▪ 𝑀𝐸𝑑 < 𝑀𝑝𝑙,0,𝑅𝑑 : No compression reinforcement is required. The exact amount
d
of required tension reinforcement is found with iteration.
▪ 𝑀𝐸𝑑 > 𝑀𝑝𝑙,0,𝑅𝑑 : Additional compression reinforcement is required that is cal-
culated from the difference of bending moments ∆𝑀 = 𝑀𝐸𝑑 − 𝑀𝑝𝑙,0,𝑅𝑑 . The
tension reinforcement is calculated from equilibrium equations 𝑥𝑐 = 𝑥𝑐0 .
o 𝑥𝑐 > 𝑥𝑐0 : Additional compression reinforcement is applied iteratively while one of the
following conditions is met 𝑀𝑝𝑙,𝑅𝑑 > 𝑀𝐸𝑑 or 𝑥𝑐 = 𝑥𝑐0 . In the first case, no additional
A
reinforcement is required, only the calculated compression reinforcement.pIn the latter
d
case, required tension reinforcement is calculated from the difference of bending mo-
ments ∆𝑀 = 𝑀𝐸𝑑 − 𝑀𝑝𝑙,0,𝑅𝑑 . The exact value of compression reinforcement is calcu-
As lated from equilibrium equations 𝑥𝑐 = 𝑥𝑐0
Ns
The effectiveness of shear connection needs to be considered by the calculation of the position of the
plastic neutral axis if η < 1. In this case, we have two neutral axes, one for the concrete and longitudinal
N
reinforcement cand one for the trapezoidal steel sheeting. The equilibrium equation is written twice in
order to obtain the position of these axes. The contribution of trapezoidal sheeting is multiplied with η
0.85fcd
in the first case and the contribution of concrete and reinforcement are multiplied with η in the second
case.
After that, the stresses in concrete and in the reinforcement are calculated using the first neutral axis, but
the stresses in the trapezoidal sheeting are calculated using the second neutral axis.
0.85fcd
Nc
fy,pd
Np,c
Ap Np,t
fy,pd
p
Ns
fy,pd
Parallel with the In this case, the contribution of trapezoidal sheeting is neglected.
ribs (upper edge in
tension)
As
Ns
Nc
0.85fcd
The equation of equilibrium and the bending resistance are formulated as follows:
0.85𝑓𝑐𝑑 𝐴𝑐 = 𝐴𝑠 𝑓𝑦,𝑑
𝑥𝑐 d
𝑀𝑝𝑙,𝑅𝑑 = 𝐴𝑠 𝑓𝑦,𝑑 (𝑧𝑠 − )
2
If compression longitudinal reinforcement is required, components representing the forces in the com-
pression reinforcement are added to the equations.
Note
The software converts the calculated required reinforcement for 1 meter reference width, thus the
unit of presented reinforcement values is mm2/m, etc.
The software calculates and shows the required reinforcement in the ribs for the element under
the mouse's position if the calculated bottom reinforcement result component, that is parallel to
the ribs, is selected.
1.) select surface elements / domains then click the button on the toolbar to specify reinforcement
2.) click the button with no selection, specify reinforcement then draw mesh-independent reinforce-
Rebar statistics The actual surface, beam and column reinforcements can be checked by displaying Rebar statistics in
the Weight report section of the Table Browser. This table lists total length and mass of rebars and the
total reinforced concrete surface and volume per rebar diameter.
464
Min. Thickness Min. Thickness displays the minimum thickness entered as surface reinforcement parameter for the
selected elements, and not the minimum thickness of the elements.
Reinforcement
User’s Manual X6 Release1 e2 465
Add and Delete The applied reinforcement is shown in a tree view on the left. By selecting a reinforcement you can
change its parameters in the right side. Changing the values updates the tree. By selecting a location
(e.g., x Direction / Top Reinforcement) you can set a new reinforcement on the right side and add it.
Use the Delete button (or [Del] key) to delete reinforcement or the Add button (or [Ins] key) to add
reinforcement to a group. If you select a node of the tree view the Delete button (or [Del] key) will delete
all the reinforcements under that node. The Add button (or [Ins] key) will add reinforcement to the
corresponding group.
Calculate rebar positions sets the rebar positions according to the actual concrete cover and primary
directions.
Rebar position is defined as the distance between the
side of the concrete and the axis of the rebar.
Composite slabs with trapezoidal sheeting cannot be selected together with normal or other
types of slabs. If composite slabs with trapezoidal sheeting and other type of slabs are present in
the model, the type of domain has to be selected before the window opens. Reference domain
needs to be selected if more than one composite slabs with trapezoidal sheeting are selected, but
the main geometrical parameters differ. In the case of composite slabs with trapezoidal sheeting,
the actual reinforcements need to be defined according to reinforcement directions parallel with
the ribs and perpendicular to the ribs of the slab.
466
To define mesh-independent reinforcement set the parameters and the reinforcement scheme first then
draw rectangular or polygonal reinforcement domains or apply the reinforcement to entire domains.
If no surfaces or domains are selected clicking the button on the toolbar displays this dialog. Reinforce-
ment can be edited the same way as above.
The bottom toolbar can be used to control the way the actual reinforcement is placed on the structure.
Mesh-independent reinforcement can also be defined.
Displays the selection toolbar to select existing domains or reinforcement domains. The current rein-
forcement is applied when the selection is completed.
Option to draw rectangular reinforcement domains.
Reinforcement is applied only where reinforcement domains fall on surface elements or domains.
The dialog can be reduced to a toolbar. Clicking the triangle icon at the top right corner shrinks or
opens up the dialog. The reinforcement amounts specified are displayed as symbols. The amounts of
top and bottom y reinforcement are written along a vertical line. The amounts of top and bottom x
reinforcement are written along a horizontal line.
Contours of reinforcement domains are identified by the cursor. Clicking reinforcement domains allow
making changes in the reinforcement. [Shift] + clicking selects multiple reinforcement domains. Clicking
on one of the selected domains allow making changes in multiple reinforcement domains. This is the
same method used for elements or mesh-independent loads.
User’s Manual X6 Release1 e2 467
Auto refresh Updates the drawing immediately when the reinforcement changes.
Beam reinforcement parameters and actual beam reinforcement can be assigned to concrete rib and
beam elements without performing design and check of beam reinforcement. Beam reinforcement de-
sign parameters can be different on each finite element. To select only a part of a structural member
check the following option: Settings / Preferences / Editing / Enable selection of finite elements on lines.
Beam reinforcement parameters dialog is suitable to define the reinforcement parameters against uniaxial
bending. This can be done for vertical elements as well.
Design of beam reinforcement (determining the required amount of reinforcement and checking it, see
6.5.11 Beam reinforcement design – RC2 module) uses these parameters.
Cross-section
The list contains all beam design cross-sections found in the selection. The geometry parameters of the
selected list item will be displayed and can be edited. Changes won't lead to recalculation of forces so
cross-section dimensions should not be changed unless it is really necessary.
Concrete grade can also be overridden. Structural class and maximum aggregate size (Dmax) must be
entered.
468
Environment Environment classes and concrete covers must be specified on all four sides of the
classes, beam (+z, -y, +y, -z).
concrete cover Concrete cover is the least distance between the stirrups and the outer surface of
the concrete. The minimum required cover calculated according to the design
code from environment classes and other parameters is displayed in blue at the
end of each row. Checking Apply minimum cover will set the cv edit fields to the
calculated value. Clicking on the link symbol right to the environment classes will
set the same environment class on all sides.
Reinforcement Parameters (i.e., material, diameter, etc.) of stirrups (green) and torsion rebars (yellow) can be entered on
parameters the left side, parameters of corner rebars (red) and other longitudinal rebars (blue) on the right.
Øt is the diameter of the top rebars, Øb is the same for bottom rebars, Øt is the diameter of side rein-
forcement against torsion. Checking Use this rebar and stirrup steel by default sets the default value of
the rebar and stirrup steel grade to the value set in the dialog.
The following parameters are used only in beam reinforcement design. The program distributes
stirrups and rebars according to these parameters before checking the calculated reinforcement.
Step of stirrup spacing: Calculated stirrup spacing values will be integer multiples of the step size entered
here.
Maximum number of applied rebar schemes limits the number of different rebar distribution schemes
applied along the beam (number of top and bottom distributions can be set separately).
Parameters
User’s Manual X6 Release1 e2 469
Cracking Checking Increase reinforcement according to limiting crack width the maximum allowed crack width val-
ues can be entered. In this case the program increases the top / bottom reinforcement (maintaining the
relation 𝐴𝑠 ≤ 0.04𝐴𝑐 ) to reduce the crack width under the specified value. To perform cracking analysis
the load duration must be specified. See 6.5.7.1. Cracking calculation according to Eurocode 2.
If the option Take concrete tensile strength into account is selected no cracking calculations will be per-
formed in points where the tensile stress is below the concrete tensile strength.
Load duration 𝑘𝑡 is a factor depending on load duration.
For short term loads 𝑘𝑡 = 0.6, for long term loads 𝑘𝑡 = 0.8
Check allowed de-
flection The program checks the allowed deflection according to the criteria set for beams and cantilevers. L
represents the beam length. This check will be performed only if the actual concrete grade and cross-
section is set.
For coefficient of seimic forces see 4.10.24 Seismic loads – SE1 module.
Capacity design It can be set for the beams whether they are primary (anti-seismic beams) or secondary seismic elements
(non-anti-seismic beams). Plastic hinges (see 6.5.11 Beam reinforcement design – RC2 module) can be
defined only in case of anti-seismic beams. The ductility class can be selected (M – DCM, H – DCH) that
influences the maximum allowable distance between stirrups in order to provide the expected ductility.
During the calculation of shear resistance, θ=45° is considered for the angle of concrete compression
strut if ductility class H is selected (EN 1998-1-1).
470
Pick up Clicking Pick up allows picking up beam reinforcement design parameters defined for another beam.
Display of actual After actual reinforcement and its prameters have been assigned to the element and the display of cross-
reinforcement section shapes is turned on (Display options / Symbols / Graphics symbols / Cross-section shape,
See 2.16.18 Display options) actual reinforcement can also be seen in the model view.
If a mesh is assigned to the line element (see 4.11.1.1 Meshing of line elements), each finite element can
have different actual reinforcement (Turn on Settings / Preferences / Editing / Enable selection of finite
elements on lines). If line mesh display is disabled, three cross-sections are shown, one at the beginning,
one at the midpoint and one at the end of the element. In case of constant actual reinforcement only
one cross section is shown.
User’s Manual X6 Release1 e2 471
During the definition of beam reinforcement parameters, it is possible to change the cross section. If the
cross section assigned to the line element is not identical to the cross section specified in the beam
reinforcement parameters dialog, the modified cross section is displayed red dashed line in the model
view.
Nonlinear analysis The value of some parameters considered during nonlinear analysis (see 6.5.5 Nonlinear analysis of rein-
forced concrete beam and column elements ) can be set by the user:
• consideration of the tensile or flexural tensile strength of concrete;
• final concrete strain due to shrinkage;
Parameters for reinforcement against biaxial bending can be specified in the dialog below.
Each finite element can have different reinforcement parameters. (Turn on Settings / Preferences / Editing
/ Enable selection of finite elements on lines). Column check based on actual reinforcement can be
performed later (see 6.5.9 Column reinforcement – RC2 module).
This type of reinforcement is used for reinforced concrete elements subjected to biaxial bending.
It can be assigned to horizontal elements as well, though in these cases the use of beam reinforce-
ment parameters (see 6.5.3 Beam reinforcement parameters (uniaxial bending)) may be more practical.
List of existing column reinforcements. You can sort them and delete the marked rows.
Defining The following icons are available on the Defining reinforcement menu:
reinforcement
Parameters Lets you specify the parameters for verification of reinforced concrete columns. The verification covers
bending check with or without axial force, the check against failure caused by shear forces and torsional
effects.
The unfavorable eccentricity increments determined based on the buckling parameters are displayed in
the internal force check table. It can be controlled if eccentricity increments prescribed by the design
code are applied in a certain direction or not. Furthermore, the consideration of second order
eccentricities can also be controlled in each direction. Second order eccentricities can be considered at
the same time or independently in each directions. In checked directions 𝛽𝑦𝑦 and 𝛽𝑧𝑧 buckling length
factors can be specified. (𝛽𝑦𝑦 in x-z plane and 𝛽𝑧𝑧 in x-y plane). It is also possible to change the column
height used in buckling length calculations and select the buckling shape.
You can see coefficient of seimic forces at 4.10.24 Seismic loads – SE1 module.
The considered concrete class in the calculation can be altered compared to concrete class assigned to
the structural element. It allows the user to evaluate structural safety of the column with different con-
crete class without new static analysis.
ϕef is the so-called effective creep ratio according to Section 5.8.4 of EN 1992-1-1 standard.
Related to non-linear analysis of the column (see 6.5.5 Nonlinear analysis of reinforced concrete beam
and column elements) the consideration of tensile strength and shrinkage strain can be specified.
Constant or variable stirrup spacing (three different spacing with different diameters) can be defined to
the column. x/L refers to a relative distance from the bottom of the column related to boundary of zones
with different stirrup parameters. The number of stirupp legs is uniform along the element, however, it
can be different parallel to y and z axes, respectively. The considered shear crack angle may be also
defined by the user. In case of columns with circular cross section, spiral stirrup can be used.
In case of dissipative structures with DCM or DCH ductility class (EN 1998-1-1), the design values of
shear forces are determined in accordance with the capacity design rule (if seismic load case is included
in the selected load combination), on the basis of the equilibrium of the column under end moments
Mi,d (with i=1,2 denoting the end sections of the column), corresponding to plastic hinge formation for
positive and negative directions of seismic loading (see 6.5.9 Column reinforcement – RC2 module). The
following parameters can be specified by the user: the safety factor, the ductility class, the clear length
of the column (clear length: the distance between end sections where plastic hinges can be formed) and
the ratio between the sum of moment capacity of beams and columns connected to the same joint,
respectively.
User’s Manual X6 Release1 e2 473
If global imperfections are taken into account by extra loads acting on the structure,
they must be ignored in column reinforcement calculation. To achieve this, check Local
imperfections only. If a sway buckling shape is selected only a constant local geometric
imperfection is considered. (see 6.5.9 Column reinforcement – RC2 module).
The given buckling shapes and buckling length factors are associated with the local y and z axes
of the section. Accordingly, during the calculation of local geometric imperfections and second
order eccentricities flexural inertias related to y and z axes are taken into account (see 6.5.9 Col-
umn reinforcement – RC2 module).
Reinforcement bars
To position / by Generates a reinforcement bar with a specified diameter to the location of the cursor.
cover If the cursor is on a corner or on the contour line the reinforcement will be generated according to the
concrete cover.
By spacing Inserts evenly N+1 new rebars between two selected points.
On circular arch Inserts evenly N+1 new rebars between a selected starting point and an end point of a circular arch.
If an entire circle is drawn inserts N rebars evenly around the circle.
In this case the concrete cover is the distance from the extreme
fiber to the rebar!
Display of actual After actual reinforcement and its prameters have been assigned to the element and the display of
reinforcement cross-section shapes is turned on (Display options / Symbols / Graphics symbols / Cross-section shape),
see 2.16.18 Display options, actual reinforcement can also be seen in the model view.
See also 6.5.3 Beam reinforcement parameters (uniaxial bending)
Nonlinear analysis The value of some parameters considered during nonlinear analysis (see 6.5.5 Nonlinear analysis of re-
inforced concrete beam and column elements) can be set by the user:
• consideration of the tensile or flexural tensile strength of concrete
• final concrete strain due to shrinkage
In case of nonlinear analysis actual reinforcements, concrete parameters and steel/concrete nonlinear
behaviour can be taken into account (see 5.1 Static analysis) for the following standards:
Eurocode 2: EN 1992-1-1:2010
SIA: SIA 262:2013
Internal forces compatible with strains are calculated through numerical integration of fiber stresses at
Gauss integration points based on ε normal strains, κy and κz curvatures. Concrete cross section is divided
into a number of triangle fibers; for steel reinforcements independent fibers are assigned with circle
shape. Fiber stresses are calculated based on the strain at the center of the fiber and based on con-
crete/steel nonlinear material model. The difference between the material models built in for EC and SIA
standards is only values of strengths, ultimate strains; the shape of the material model is identical.
User’s Manual X6 Release1 e2 475
steel
1.05 f yk
f yk
Es
uk
M kNm
concrete
f ck
Ec ,eff
cu 3
1 / m
In order to avoid convergence problems, concrete material model has been changed on tension side.
Based on the results of verification tests, the calculated displacements are close to displacements calcu-
lated with standardized formulae. The error was lower than 5% in every investigated case.
Tensile stress in the concrete is calculated with the following formula that is based on an approximate
method from section 7.4.3 of EN 1992-1-1:
𝜎 = 𝐸𝜀(1 − 𝜁)
𝑓𝑡 𝛾
𝜁 = min [max [1 − 𝛼𝛽 ( ) ; 0] ; 1]
𝐸𝜀
α and γ factors are selected (as a function of ratio of concrete’s and steel’s Young’s modulus and as a
function of the reinforcement ratio) according to the results of a parametric study in order calculate de-
formations in good agreement with section 7.4.3. of EN 1992-1-1.
β is a coefficient taking account of the influence of the duration of the loading. According to EN
1992-1-1, β may be selected 0.5 and 1.0 in case of sustained loads and short-term loads, respec-
tively. If creep is considered in the nonlinear analysis, β is selected 0.5. If creep is not considered
in the nonlinear analysis, β=0.5 or β=1.0 can also be used. The value for these cases can be speci-
fied in the Design codes window (see 3.3.7 Design codes).
Creep and The consideration of creep and shrinkage in nonlinear static analysis can be set by the initialization of
shrinkage the analysis (see 5.1 Static analysis). The creep is taken into account by the modification of the Young’s
modulus of the concrete. In case of shrinkage, two additional κy and κz curvatures (calculated from the
arrangement and the amount of reinforcement) are added to the calculated strains according to the
given shrinkage strain (see 6.5.3 Beam reinforcement parameters (uniaxial bending) and 6.5.4 Actual rein-
forcement against biaxial bending (column)).
The purpose of consideration of actual reinforcement in nonlinear static analysis is the more ac-
curate deflection calculation for line reinforced concrete elements by the verification in servicea-
bility limit states considering nonlinear material and structural behavior. Using this fiber integra-
tion model for modeling material nonlinearity is not recommended by the verification of rein-
forced concrete elements in ultimate limit states. It is also important to note that the present
model cannot be used in pushover analysis in order to take into account nonlinear behaviour of
reinforced concrete elements instead of plastic hinges (see 4.10.25 Pushover loads – SE2 module)
because the presented material models do not consider cyclic degradation in concrete and steel
strength / stiffness, buckling of steel reinforcement bars, etc.
In case of composite sections, nonlinear behavior is not considered. Elements with composite sec-
tions are treated as linear elements. To consider any creep effect in case of a composite section,
concrete material with reduced Young’s modulus needs to be assigned.
After completing nonlinear analysis with actual reinforcement, the following error messages may appear:
These messages appear to help detecting and correcting errors committed when defining the actual
reinforcement.
In case of the linear static analysis the plate deflection is calculated according to the elastic theory. In
fact the behaviour of RC plates is non-linear due to two opposite effects. The actual reinforcement
increases the bending strength but cracking decreases it.
The non-linear RC plate deflection analysis follows up these two effects taking into account the actual
or calculated reinforcement.
The below presented method, for consideration of material nonlinearity of RC surfaces in non-
linear analysis, can be used for calculation of deflections by verification in serviceability limit
states. Using this fiber integration model for modelling material nonlinearity is not recom-
mended by verification in ultimate limit states. It is also important to note that the present model
cannot be used in pushover analysis in order to take into account nonlinear behaviour of rein-
forced concrete elements instead of plastic hinges (see 4.10.25 Pushover loads – SE2 module) be-
cause the presented material models do not consider cyclic degradation in concrete and steel
strength / stiffness, buckling of steel reinforcement bars, etc.
steel concrete
1.05 f yk
f yk f ck
Es Ec , e
uk Ec ,eff 4 f ctm
Nonlinearity In this case, nonlinear behavior is considered only in the connection of normal strains and membrane
ε-N (Wall) forces. Flexural moments are calculated elastically without consideration of membrane forces.
Nonlinearity In this case, nonlinear behavior is considered only in the connection of curvatures and flexural moments.
κ –M (Slab) Flexural moments are integrated independently in x and y directions, the effect of membrane forces on
the bending behavior is considered. Membrane forces and Mxy are calculated elastically.
User’s Manual X6 Release1 e2 477
In case of slabs with considerable torsion a nonlinear analysis based only on κx-Mx and κy-My
bending behaviour may underestimate the deflection, so it is recommended to use the first ε-
N; κ-M complex model. That can follow nonlinear behaviour related to the six strain and internal
force components.
In case of κ-M nonlinearity model, in order to calculate more realistic deformations, torsion
stiffness in the stiffness matrix is considered on a reduced value if the surface is cracked at a
specific point in both x and y main directions. If the height of cracked zone is considerable
(>0.5H), torsion stiffness is considered on 15% of the original elastic value. Linearly interpolated
value is used between the two states.
The nonlinear analysis of skewly reinforced concrete surfaces is not allowed. These surfaces are
treated as materially linear elements without the consideration of cracking and the effect of
reinforcement. The Young’s modulus is calculated with the consideration of the defined creep
factor.
Amount of rein- Actual or calculated reinforcement can be taken into account in nonlinear static analysis by choosing
forcement Use reinforcement in calculation in the Nonlinear static analysis dialog (see 5.1 Static analysis). The
convergence of such nonlinear analyses is sensitive to the amount of reinforcement. In order to avoid
convergence problems, AxisVM applies the minimum reinforcement prescribed by the design code
where (top or bottom) actual reinforcement is defined only in one direction. The convergence is usually
better in case of actual reinforcement. It is recommended to use the Take into account the minimum
required reinforcement option while defining reinforcement parameters to surfaces (see 6.5.1 Surface
reinforcement parameters and reinforcement calculation - RC1 module)
Elements within the If Adjust mesh to column heads option is selected by domain meshing, finite elements within the con-
contour of columns tour of RC columns can be identified. In order to avoid convergence problems, nonlinear behavior of
these elements is neglected in the calculation.
Creep and shrink- The consideration of creep and shrinkage in nonlinear static analysis can be set by the initialization of
age the analysis (see 5.1 Static analysis). The creep is taken into account by the modification of the concrete’s
Young’s modulus. The shrinkage is considered only in flexural deformations.
Nonlinearity Two calculated (based on the arrangement and the amount of reinforcement and on the state of RC
ε-N; κ -M section) shrinkage induced curvatures (namely κx and κy) are added to the calculated strains according
to the given shrinkage strain (see 6.5.1 Surface reinforcement parameters and reinforcement calculation
- RC1 module).
Nonlinearity The reinforcement in walls is typically symmetrical, thus shrinkage curvatures are not taken into account
ε-N (Wall) in this case.
Nonlinearity Two calculated (based on the arrangement and the amount of reinforcement and on the state of RC
κ –M (Slab) section) shrinkage induced are added to the calculated strains according to the given shrinkage strain.
Warning messages After completing nonlinear analysis with actual reinforcement, the following warning messages may
appear:
• “Normal force exceeds the resistance.”
• ”Actual strain exceeds the limiting strain.”
(Note: In case of the reinforcement or concrete in compression)
6.5.7. Cracking
Design Codes Eurocode 2: EN 1992-1-1:2004
DIN: DIN 1045-1:2001-07
After the assignment of the actual reinforcement the program calculates the
crack width and crack directions in the membrane, plate and shell elements.
The calculation can be perfromed with reinforcement directions relative to the
surface element’s local x and y axes or with arbitrary ξ and η reinforcement
directions. The program displays the crack openings in a color coded mode,
can draw the crack map and the crack angles.
The set of the parameters can be seen in the previous section.
478
Results The calculation of crack width can be performed for required reinforcement (marked with *, e.g., wk*
and wk2*) not only for actual reinforcement. The crack widths calculated for requires reinforcement can
be considered only as approximate values because the final actual reinforcement will be different from
the required reinforcement. In the table of results the following information can be found:
Aa(x/ξ), actual reinforcement in x and y (or in ξ and η in case of skew reinforcement) directions
Aa(y/η)
wk crack width at the axis of the rebar
wk2 crack width at the edge of the slab
xs2 position of the neutral axis relative to the edge on the compressed side
σs2 rebar stress
wR angle of cracking relative to the local x direction
nx, ny, nxy, mx, my, mxy surface forces and moments
𝜖𝑠𝑚 is the strain of the rebar, 𝜖𝑐𝑚 is the strain of the concrete between cracks
𝑘𝑡 𝑓𝑐𝑡𝑚 𝐸
𝜎𝑠2 − (1 + 𝑠 𝜌𝜌,𝑒𝑓𝑓 )
𝜌𝜌,𝑒𝑓𝑓 𝐸𝑐𝑚 𝜎𝑠2
𝜖𝑠𝑚 − 𝜖𝑐𝑚 = ≥ 0.6 ⋅
𝐸𝑠 𝐸𝑠
Φ 𝜎𝑠2 𝜙
𝑠𝑟,𝑚𝑎𝑥 = k 3 ⋅ 𝑐 + k 4 ⋅ 𝑘1 𝑘2 (𝑠𝑟,𝑚𝑎𝑥 = in case of DIN EN 1992-1-1)
𝜌𝜌,𝑒𝑓𝑓 3.6𝑓𝑐𝑡,𝑒𝑓𝑓
where
Φ is the average rebar diameter,
c is the concrete cover,
𝑘1 is a factor depending on rebar surface (ribbed or plain),
𝑘2 is a factor depending on the character of the eccentric tension,
k3 is a factor may be equal to 3.4
k4 is a factor may be equal to 0.425
k3 and k4 factors can be set in the Design codes window under the Reinforced concrete tab
(see 3.3.7 Design codes)
𝑘𝑡 is a load duration factor,
for short-term loads 𝑘𝑡 = 0.6
for long-term (permanent) loads 𝑘𝑡 = 0.4
𝜌𝜌𝑒𝑓𝑓 = 𝐴𝑠 /𝐴𝑐,𝑒𝑓𝑓 is the effective reinforcement ratio
Φ
If plain rebars are used or the spacing of ribbed rebars exceeds 5 ⋅ (𝑐 + ), then
2
𝑠𝑟𝑚𝑎𝑥 = 1.3 ⋅ (ℎ − 𝑥2 ).
The program takes account of the fact that cracking is not perpendicular to any of the reinforcement
directions and calculates its angle relative to the x axis.
User’s Manual X6 Release1 e2 479
𝜖𝑠𝑚 is the strain of the rebar, 𝜖𝑐𝑚 is the strain of the concrete between cracks
𝑓𝑐𝑡𝑚 𝐸
𝜎𝑠2 − 0.4 ⋅ (1 + 𝑠 𝜌𝑒𝑓𝑓 )
𝜌𝑒𝑓𝑓 𝐸𝑐𝑚 𝜎𝑠2
𝜖𝑠𝑚 − 𝜖𝑐𝑚 = ≥ 0.6 ⋅
𝐸𝑠 𝐸𝑠
𝑑 𝜎𝑠2 𝑑
𝑠𝑟𝑚𝑎𝑥 = ≤
3.6 ⋅ 𝜌𝑒𝑓𝑓 3.6𝑓𝑐𝑡𝑚
where
𝑑 is the average rebar diameter
𝜌𝜌𝑒𝑓𝑓 = 𝐴𝑠 /𝐴𝑐,𝑒𝑓𝑓 is the effective reinforcement ratio
The program takes account of the fact that cracking is not perpendicular to any of the reinforcement
directions and calculates its angle relative to the x axis.
6.5.8. Shear resistance and reinforcement calculation for plates and shells
AxisVM calculates the shear resistance of the reinforced plate or shell without shear reinforcement,
shear reinforcement (asw) if required and the maximum shear resistance of plate and shell structures
with shear reinforcement according to the failure of concrete in compression. Furthermore, the resultant
design shear force, the principal direction (𝜙), the difference between the design shear force and the
shear resistance without shear reinforcement and the ratio of the design shear force and maximum
shear resistance with shear reinforcement are also calculated.
This feature is available for the following domain types (see 4.9.6 Domain):
• normal domains,
• composite ribbed domains,
• parametric ribbed domains,
• domains with custom stiffness matrix,
• AIRDECK slabs,
• composite slabs with trapezoidal sheeting (if the Trapezoidal plate used only as formwork op-
tion is checked).
Due to the differences in the regulations related to shear and punching, this calculation and a
verification based on the presented result components cannot substitute punching verification!
The calculation of the shear resistance is based on the actual reinforecement assigned to the
surfaces.
Shear resistance is
𝑉𝑅𝑑,𝑐 = [𝐶𝑅𝑑,𝑐 𝑘(100 ⋅ 𝜌𝑙 𝑓𝑐𝑘 )1/3 + 𝑘1 𝜎𝑐𝑝 ] ⋅ 𝑑 ≥ (𝑣𝑚𝑖𝑛 + 𝑘1 𝜎𝑐𝑝 ) ⋅ 𝑑
where
0.18
𝐶𝑅𝑑,𝑐 = , 𝑘 = 1 + √200/𝑑 ≤ 2.0, 𝑘1 = 0.15
𝛾𝑐
𝑁𝐸𝑑 1/2
𝜎𝑐𝑝 = ≤ 0.2 𝑓𝑐𝑑 , 𝑣𝑚𝑖𝑛 = 0.035 ⋅ 𝑘 3/2 ⋅ 𝑓𝑐𝑘
𝐴𝑐
𝑁𝐸𝑑 is the normal force in the shell perpendicular to the plane of 𝑞𝑅𝑧 .
𝑁𝐸𝑑 is positive in compression.
The reinforcement ratio is 𝜌𝑙 ≤ 0.02
cot 𝜃 + cot 𝛼
𝑉𝑅𝑑,𝑚𝑎𝑥 = 𝛼𝑐𝑤 𝑧𝜈1 𝑓𝑐𝑑
1 + cot 2 𝜃
where
𝑓𝑐𝑘
𝜈1 = 𝜈 = 0.6 [1 − ]
250
𝛼𝑐𝑤 = 1.0 ← 𝜎𝑐𝑝 = 0
𝜎𝑐𝑝
𝛼𝑐𝑤 = (1 + ) ← 0 < 𝜎𝑐𝑝 ≤ 0.25𝑓𝑐𝑑
𝑓𝑐𝑑
𝛼𝑐𝑤 = 1.25 ← 0.25𝑓𝑐𝑑 < 𝜎𝑐𝑝 ≤ 0.5𝑓𝑐𝑑
𝜎𝑐𝑝
𝛼𝑐𝑤 = 2.5 (1 − ) ← 0.5𝑓𝑐𝑑 < 𝜎𝑐𝑝 ≤ 𝑓𝑐𝑑
𝑓𝑐𝑑
𝑧 is the level arm of internal forces
Calculation of level The level arm of internal forces is calculated according to one of the two following ways based on the
arm of internal settings specified by the user (See 6.5.1 Surface reinforcement parameters and reinforcement calculation
forces - RC1 module):
• calculation from internal forces based on the equilibrium of a three-layered sandwich model;
• approximately: 𝑧 = 0.9𝑑
Calculation of re- If 𝑣𝐸𝑑 is greater than 𝑉𝑅𝑑,𝑐 , shear reinforcement is required to resist shear forces. The required reinforce-
quired shear rein- ment (e.g., [mm2/m2]) is calculated with the help of the following formula. If 𝑣𝐸𝑑 is greater than 𝑉𝑅𝑑,𝑚𝑎𝑥 ,
forcement the section cannot be reinforced against shear force. In these cases, the software warns the user with a
warning message.
𝑣𝐸𝑑
𝑎𝑠𝑤 =
𝑧𝑓𝑦𝑤𝑑 1𝑚[(cot 𝜃 + cot 𝛼) sin 𝛼]
𝑉𝑅𝑑,𝑐 , 𝑉𝑅𝑑,𝑚𝑎𝑥 shear resistances, furthermore the difference between design shear force and the shear
resistance 𝑉𝐸𝑑 − 𝑉𝑅𝑑,𝑐 and 𝑉𝐸𝑑 /𝑉𝑅𝑑,𝑚𝑎𝑥 can also be displayed with isolines and isosurfaces.
Shear resistance is
𝑉𝑅𝑑𝑐 = 𝑘𝑑 𝜏𝑐𝑑 𝑑
𝑓𝑠𝑑
where 𝜏𝑐𝑑 = 0.3√𝑓𝑐𝑘 /𝛾𝑐 ; 𝑘𝑑 = 1/(1 + 𝜀𝑣 𝑑𝑘𝑔 ); 𝑘𝑔 = 48⁄(16 + 𝐷𝑚𝑎𝑥 ); 𝜀𝑣 = 1,5 .
𝐸𝑠
If direction of 𝑉𝐸𝑑 resultant design share force is not parallel to either x or y reinforcement, 𝜀𝑣 is increased
by 1/(cos4 𝜙 + sin4 𝜙).
Calculation of level The level arm of internal forces is calculated according to one of the two following ways based on the
arm of internal settings specified by the user (See 6.5.1 Surface reinforcement parameters and reinforcement calculation
forces - RC1 module):
• calculation from internal forces based on the equilibrium of a three-layered sandwich model;
• approximately: 𝑧 = 0.9𝑑
Calculation of re- If 𝑣𝐸𝑑 is greater than 𝑉𝑅𝑑,𝑐 , shear reinforcement is required to resist shear forces. The required reinforce-
quired shear rein- ment (e.g., [mm2/m2]) is calculated with the help of the following formula. If 𝑣𝐸𝑑 is greater than 𝑉𝑅𝑑,𝑚𝑎𝑥 ,
forcement the section cannot be reinforced against shear force. In these cases, the software warns the user with a
warning message.
𝑣𝐸𝑑
𝑎𝑠𝑤 =
𝑧𝑓𝑦𝑤𝑑 1𝑚[(cot 𝜃 + cot 𝛼) sin 𝛼]
𝑉𝑅𝑑,𝑐 , 𝑉𝑅𝑑,𝑚𝑎𝑥 shear resistances, furthermore the difference between design shear force and the shear
resistance 𝑉𝐸𝑑 − 𝑉𝑅𝑑,𝑐 and 𝑉𝐸𝑑 /𝑉𝑅𝑑,𝑚𝑎𝑥 can also be displayed with isolines and isosurfaces.
The reinforced column check can be performed based on the following design codes:
Design Codes Eurocode 2: EN 1992-1-1:2004 + (MSZ, DIN, SR, DS, NBN, BS, NEN, SFS, NS, SS, NTC, CSN, PN,
ELOT NAs)
DIN: DIN 1045-1:2001-07
SIA: SIA 262:2013
Reinforcement bars Actual reinforcement can be entered the same way as described in 6.5.4 Actual reinforcement against
biaxial bending (column)
Column Check The program performs the verification against bending with or without axial force, shear forces and
torsional moments of the column based on the cross-section properties and reinforcement
parameters and determines the eccentricity increments for the forces in the selected columns (or any
𝑁𝑥 , 𝑀𝑦𝑏 , 𝑀𝑧𝑏 , 𝑀𝑦𝑡 , 𝑀𝑧𝑡 values) based on the given buckling parameters and according to the
requirements of the current design code.
Calculates 𝑁𝑥,𝐸𝑑 , 𝑀𝑦,𝐸𝑑 , 𝑀𝑧,𝐸𝑑 design forces using the eccentricity increments and checks if these points
are within the interaction diagram.
The display of the diagram can be set in the Display Parameters window.
In case of shear verification, the software determines the shear resistance of the concrete and the
stirrups and the upper bound of shear resistance related to the failure of concrete strut in compression.
In case of verification against torsional effects, the software determines the torsion resistance of the
concrete and the upper bound of torsion resistance related to the failure of concrete strut in
compression. During the calculation of utilization, the combined effect of shear and torsion is also
accounted.
482
N-M diagram Displays the Nx-My, or Nx-Mz load-moment strength interaction diagram.
This display mode can be used with cross-sections that are symmetric. You can display the design
values of the internal forces, by enabling the Write values to check-box.
The design values of the internal forces are displayed as follows:
Load eccentricity
limit curves
𝑀𝑦,𝑅𝑑,𝑖 𝑀𝑧,𝑅𝑑,𝑖
Displays the load eccentricity limit curves based on or .
𝑁𝑖 𝑁𝑖
Shear/torsion check
In order to increase the performance of the calculation, the column is divided into several sections.
New section boundary is created where any of internal forces reaches min./max. value, where stirrup’s
parameters change and where Vy, Vz or Tx internal force diagram changes.
kT, η and Asl,T result components are calculated at the bottom and at the top of every sections.
Maximum values are shown only for every section in the window.
See 6.5.9.5 Shear and torsion check of reinforced concrete columns
Detailed verification results are available for each section, if the cursor is moved over the black
lines representing the column axis.
Special The Window / Special window layouts menu item offers different predefined result window layouts
window layouts that can be customized by the user later.
Design calculations The details of calculations according to the current design code are displayed as a multi-page docu-
ment. References to sections and formulas of the design code appear in blue. It is only available if
column check is performed and column design results are available.
Clicking on the Settings icon beside the Design calculations button allows setting the basic units for
force and length used in the design calculations. More detailed design calculation can be selected as
well.
Detailing rules The software checks the most critical detailing rules according to the selected design code. If there
are not fulfilled detailing rules, the software warns the user in warning messages and in the design
calculations.
Utilization values The calculated utilizations are available in table format. If N-M surface, N-My diagram, N-Mz diagram,
in table My-Mz diagram or Critical eccentricity curves display mode is active, the presented utilizations are
related to the column verification against bending with or without axial force. If Shear/torsion check
display mode is active, the results of shear and torsion verification are shown.
The Column internal force check table contains the maximum normal forces and moments at the top
and bottom end of the selected columns and different eccentricity values. Additional columns dis-
playing 𝑀𝑦𝑅𝑑,𝑚𝑖𝑛 , 𝑀𝑦𝑅𝑑,𝑚𝑎𝑥 , 𝑀𝑧𝑅𝑑,𝑚𝑖𝑛 , 𝑀𝑧𝑅𝑑,𝑚𝑎𝑥 moment resistance maximum values at the given Nx
are also available.
Utilization – bending The program calculates two types of utilization. The first one is ε (N = const.), the moment utilization:
with or without axial this is defined on the My-Mz diagram as the ratio of the distance of design force point from the origin
force to the distance of the intersection point of the curve and the half line drawn from the origin through
the same point from the origin. The second one is ε (e = const.) the utilization for constant eccentricity.
486
It is defined in the N-My-Mz space as the ratio of the distance of design force point from the origin
to the distance of the intersection point of the N-M surface and the half line drawn from the origin
through the same point from the origin.
Besides the utilizations, the program lists the design shear resistances related to local y and z axes,
the shear resistant reduction factor of stirrups due to torsion and the additional longitudinal rein-
forcement required for torsion.
see 6.5.9.5 Shear and torsion check of reinforced concrete columns
Show failed columns Check this option to display the failed columns only.
only
Custom internal forces Adds new custom internal forces to the table: By clicking this button, new custom internal
in result table forces can be assigned to the column. The software saves these internal forces and it takes into
account these values in column design check together with internal forces coming from the
analysis.
Deletes selected rows from the table.
Pasting custom Custom internal forces can be pasted in the table from the clipboard as well. If Paste lines at the end
internal forces option is selected, the table previously copied to the clipboard needs to consist of exactly 11 columns
from the clipboard according to the number and order of internal force columns in the result table (Nx, Mya, Mza, Myf, Mzf,
Txa, Vya, Vza, Txf, Vyf, Vzf). The same number of columns and rows need to be selected in the result table
as the copied table has if Overwrite table beginning from current cell option is selected.
User’s Manual X6 Release1 e2 487
Result components If there are reinforced concrete columns with actual column reinforcement in the model, new results
components become available under R. C. Design tab. These new result components show the calcu-
lated utilization values and required additional longitudinal reinforcement due to torsion:
• N-M utilization,
• V-T utilization,
• Asl,T (required additional longitudinal reinforcement due to torsion),
• Maximum utilization (greater of N-M utilization and V-T utilization).
Results in the Design results are also available in the Table Browser (see 2.9 Table Browser), where the results of
Table Browser every reinforced concrete column are available in summary tables.
Eccentricity calculation in AxisVM is based on a method assuming that the value of bending moments
along the column are the sum of first order bending moments, bending moments due to geometrical
imperfections and second order bending moments (EN 1992-1-1 5.8.8.2). Any of the sections along the
column may be critical.
𝑀𝐸𝑑 = 𝑀0 + 𝑀𝑖 + 𝑀2
where
𝑀0 first order bending moment
𝑀𝑖 bending moment due to geometrical imperfections
𝑀2 second order bending moment
(A. J. Bond et al., How to Design Concrete Structures using Eurocode 2. UK, 2006 )
According to this method the verification of a column with arbitrary boundary conditions can be per-
formed if the bending moments are available in every section.
First order bending AxisVM analyses the first order bending moment diagram along the column and it divides the column
moments based on the number of moment extremes. This way the effect of intermediate forces on the column
can be taken into account. Bending moments between extreme values are determined with linear in-
terpolation.
Bending moments In case of braced elements (non-sway buckling shapes) bending moments due to geometrical imper-
due to imperfections fections are taken to be constant along the column because 1) the axis of columns on different storeys
are not necessarily collinear, 2) if the column has rigid connections, a local curvature causes the same
bending moment (N·ei) at the ends than at the middle but of opposite sign.
Buckling shapes of braced elements:
In case of sway buckling shapes, linear bending moment diagram is assumed along the beam if it is
specified in the code.
Buckling shapes of sway elements:
488
If global imperfections are taken into account by extra loads acting on the structure, they must be
ignored in column reinforcement calculation. To achieve this, check Local imperfections only. If a sway
buckling shape is selected only a constant local geometric imperfection is considered (calculated with
a β ≤ 1.0 buckling factor).
Second order mo- Based on the selected design code, second order moments are calculated from the first order bending
ments moments (extended with bending moments due to imperfections) or based on the buckling shape. The
first calculation method is called nominal stiffness method, while the second one is the nominal curva-
ture method.
Nominal curvature Second order moments are calculated with trigonometric functions assuming that the bending mo-
method ment at the maximum amplitude is equal to N·e2. The maximum amplitude is obtained according to
the selected buckling shape (see 6.5.4 Actual reinforcement against biaxial bending (column)).
𝛽 − 0.5 2𝜋𝑥
𝑒2 = [1 − ] cos ( ) 𝑒2,0
𝜋𝑥 0.5 𝐿
𝑒2 = sin ( )𝑒 −
𝐿 2,0
1) 6) 1−𝛽 𝜋𝑥
[1 − ] sin ( ) 𝑒2,0
0.5 𝐿
𝜋𝑥 𝜋𝑥
𝑒2 = cos ( )𝑒 𝑒2 = cos ( )𝑒
2𝐿 2,0 𝐿 2,0
2) 7)
3𝜋𝑥 3𝜋𝑥 𝜋
𝑒2 = cos ( )𝑒 𝑒2 = cos ( + ) 𝑒2,0
2𝐿 2,0 2𝐿 2
3) 8)
2𝜋𝑥 𝜋𝑥 𝜋
𝑒2 = cos ( ) 𝑒2,0 𝑒2 = cos ( + )𝑒
𝐿 2𝐿 2 2,0
4) 9)
𝜋𝑥
𝑒2 = cos ( )𝑒
𝐿 2,0
5)
e2,0 is the maximum amplitude of second order eccentricity.
Nominal stiffness This method ignores the buckling shape. The sum of first order bending moments and bending mo-
method ments due to imperfections are multiplied based on the ratio of design normal force and the buckling
force. For further details see EN 1992-1-1. Dutch and Danish NA-s require application of this method.
Minimal eccentricity Design codes usually require the consideration of a minimum eccentricity if the calculated eccentricity
is lower than this limit. In AxisVM, if the sum of first order eccentricity, eccentricity due to imperfections
and second order eccentricity does not reach the minimum, eccentricity due to imperfections is in-
creased in order to ensure that the sum of eccentricities is not less than the minimum
b)
Note: Figures and illustrations in this section have been saved considering case (b). If the second
order eccentricities were considered simultaneously (case(a)), the figures and illustrations would
show different representation of eccentricities.
User’s Manual X6 Release1 e2 489
6.5.9.2. Check of reinforced columns according to Eurocode 2 (bending with axial force)
Nominal stiffness 𝛽
𝑒2 = (𝑒𝑒 + 𝑒𝑖 ) [1 + ],
method 𝑁𝐵 ⁄𝑁𝑑 − 1
𝜋2
where 𝛽 =
𝑐
𝑁𝐵 is the buckling load based on nominal stiffness
𝐸𝐼 = 𝐾𝑐 𝐸𝑐𝑑 𝐼𝑐 + 𝐾𝑠 𝐸𝑠 𝐼𝑠 is the nominal stiffness, where
𝐸𝑐𝑑 is the design value of the modulus of elasticity of concrete, see 5.8.6 (3)
𝐼𝑐 is the moment of inertia of concrete cross-section
𝐸𝑠 is the design value of the modulus of elasticity of reinforcement, 5.8.6 (3)
𝐼𝑠 is the second moment of area of reinforcement, about the centre of area of the concrete
𝐾𝑐 is a factor for effects of cracking, creep etc, see 5.8.7.2. (2), (3)
𝐾𝑠 is a factor for contribution of reinforcement, see 5.8.7.2. (2), (3)
Eccentricities are determined in both bending planes. The program checks the following design situa-
tions:
𝜆𝑦 ⁄ 𝜆𝑧 ≤ 2 and 𝜆𝑧 ⁄𝜆𝑦 ≤ 2, otherwise
𝑒𝑑𝑦 ⁄𝑏𝑒𝑞 𝑒𝑑𝑧⁄ℎ𝑒𝑞
≤ 0,2 or ≤ 0,2 *
𝑒𝑑𝑧 ⁄ℎ𝑒𝑞 𝑒𝑑𝑦 ⁄𝑏𝑒𝑞
if 𝑒𝑑𝑦 ⁄𝑏𝑒𝑞 > 𝑒𝑑𝑧 ⁄ℎ𝑒𝑞 𝑀𝑑𝑦,1 = 𝑁𝑑 (𝑒𝑒𝑧 + 𝑒𝑖𝑧 )
𝑀𝑑𝑦,1 = 0 𝑀𝑑𝑧,1 = −𝑁𝑑 (𝑒𝑒𝑦 + 𝑒𝑖𝑦 ± 𝑒2𝑦 )
𝑀𝑑𝑧,1 = −𝑁𝑑 (𝑒𝑒𝑦 + 𝑒𝑖𝑦 ± 𝑒2𝑦 ) 𝑀𝑑𝑦,2 = 𝑁𝑑 (𝑒𝑒𝑧 + 𝑒𝑖𝑧 ± 𝑒2𝑧 )
otherwise 𝑀𝑑𝑧,2 = −𝑁𝑑 (𝑒𝑒𝑦 + 𝑒𝑖𝑦 )
𝑀𝑑𝑦,2 = 𝑁𝑑 (𝑒𝑒𝑧 + 𝑒𝑖𝑧 ± 𝑒2𝑧 )
𝑀𝑑𝑧,2 = 0
AxisVM checks whether the calculated design loads (Mdy, Mdz, Nd) are inside the N-M strength
interaction diagram. If it is not satisfied in any of the design situations, the column with the given cross-
section and reinforcement fails.
User’s Manual X6 Release1 e2 491
Possible reasons:
Possible reasons:
, diagrams:
Longitudinal reinforcement bars thinner than 1/15 of the stirrup distance will be ignored for compression.
6.5.9.3. Check of reinforced columns according to DIN1045-1 (bending with axial force)
16
𝜆𝑚𝑎𝑥 = max {25; 𝑁𝑑
},
√𝐴
𝑐 𝑓𝑐𝑑
AxisVM checks whether the calculated design loads (Mdy, Mdz, Nd) are inside the N-M strength inter-
action diagram. If it is not satisfied in any of the design situations, the column with the given cross-
section and reinforcement fails.
Longitudinal reinforcement bars thinner than 1/15 of the stirrup distance will be ignored for compression.
6.5.9.4. Check of reinforced columns according to SIA 262 (bending with axial force)
AxisVM checks whether the calculated design loads (Mdy, Mdz, Nd) are inside the N-M strength
interaction diagram. If it is not satisfied in any of the design situations, the column with the given
cross-section and reinforcement fails.
The calculation takes the following assumptions: , diagrams
Longitudinal rebars will not be taken into account for compression if any of the following criteria is met (s
is the stirrup distance):
< 8, s > 15 , s > amin, s > 300 mm
Shear resistant According to design codes and design manuals, the transverse reinforcement cannot be fully utilized
reduction factor of for shear and torsion, respectively. For this reason, the program calculates the utilization of the stirrups
stirrups due to considered only torsional effects (kT) and it calculates a reduced shear resistance for stirrups against
torsion (kT) shear forces, the considered reduction factor is 1-kT (only used for the external stirrups). This reduced
shear resistance is used by the calculation of the overall utilization with shear and torsion interaction
(ηVyVzT). The equilibrium forces in the stirrups against shear and torsional effects, respectively:
If the utilization of stirrups considered only torsion (kT) exceeds 1.0, the resistance of stirrups cannot
be considered against shear forces. In these cases, overall utilization with shear and torsion interaction
is ηVyVzT cannot be calculated that is marked with ??? symbol.
Torsion is considered by the verification only if torsional moment considerably influences the shear
resistance (kT > 0.01), the torsion moment exceeds 1% of torsional cracking moment, or the following
condition is satisfied:
𝑉𝑦,𝐸 𝑉𝑧,𝐸 𝑇𝑥,𝐸
+ + >1
𝑉𝑦,𝑅,𝑐 𝑉𝑧,𝑅,𝑐 𝑇𝑥,𝑅,𝑐
where Vy,R,c, Vz,R,c and Tx,R,c are the shear resistance of the concrete related to local y and z axes and the
torsional cracking moment, respectively. Otherwise, the shear resistant reduction of stirrups due to
torsion is not considered (kT = 0).
494
Additional In case of a reinforced concrete beam/column, longitudinal reinforcement is also required against tor-
longitudinal rein- sional moment if condition above is satisfied. This longitudinal reinforcement should be distributed
forcement required over the length of side of the section. Bending reinforcement may not be considered against torsion.
for torsion (Asl,T) Torsional effects are typically not considerable in case of columns, torsional longitudinal reinforcement
cannot be defined. The program calculates an additional longitudinal reinforcement may be required
due to torsion.
If additional longitudinal reinforcement is
necessary a warning message appears. If
Do not display this message option is
checked, the message is not displayed
while the Checking concrete columns win-
dow is open.
Overall shear utiliza- Rectangular column section:
tion without torsion 𝑉𝑦,𝐸 𝑉𝑧,𝐸
𝜂𝑉𝑦𝑉𝑧 = + ≤1
(ηVyVz) 𝑉𝑦,𝑅 𝑉𝑧,𝑅
where Vy,E and Vz,E are calculated design shear forces related to local y and z axes, while Vy,R and Vz,R
are the design shear resistances related to local y and z axes, respectively.
In case of circular sections, the program calculates the resultant shear force from Vy,E and Vz,E. The
resultant shear force is compared to shear resistance of the section calculated in resultant’s direction:
2 2
√𝑉𝑦,𝐸 + 𝑉𝑧,𝐸
𝜂𝑉𝑦𝑉𝑧 = ≤1
𝑉𝑅
Overall utilization ηVyVzT utilization is evaluated similarly to ηVyVz, however, in this case the resistance of external stirrup
with shear and tor- legs is reduced with a factor (1 – kT) based on the torsional utilization.
sion interaction Torsion is considered only by the verification if torsional moment considerably influences the shear
(ηVyVzT) resistance (kT > 0.01), the torsion moment exceeds 1% of torsional cracking moment, or the following
condition is satisfied:
𝑉𝑦,𝐸 𝑉𝑧,𝐸 𝑇𝑥,𝐸
+ + >1
𝑉𝑦,𝑅,𝑐 𝑉𝑧,𝑅,𝑐 𝑇𝑥,𝑅,𝑐
Utilization of the con- In case of both shear and torsion verification, the resistance cannot exceed a shear force and torsional
crete compression moment value, respectively, related to the failure of concrete compression strut. If shear forces and
strut (ηVyVzT,max) torsional moments are both present, the effect of shear and torsion interaction need to be considered
by the calculation of the utilization of concrete compression strut.
In case of circular sections, the program calculates the resultant shear force from Vy,E and Vz,E. The
resultant shear force is compared to shear resistance of the section calculated in resultant’s direction.
The utilization is calculated as follows:
2 2
√𝑉𝑦,𝐸 + 𝑉𝑧,𝐸
𝑇𝑥,𝐸
𝜂𝑉𝑦𝑉𝑧𝑇,𝑚𝑎𝑥 = + ≤1
𝑉𝑅,𝑚𝑎𝑥 𝑇𝑥,𝑅,𝑚𝑎𝑥
If ηVyVzT,max exceeds 1.0, the resistance of the section can be increase only by changing concrete class
or increasing sections dimensions.
Shear resistance of The calculation shear resistance of reinforced concrete columns with circular cross section is mainly
circular cross section identical with the calculation performed in case of rectangular cross sections, there are only few mod-
ifications. For further information, we recommend the following research: John Orr - Shear Capacity of
Circular Concrete Sections.
User’s Manual X6 Release1 e2 495
a) By the calculation of shear resistance of the concrete (Vy/z,R,c) and the maxmin design shear
resistance (Vy/z,R,max), 0.8D is used as the width of cross section, where D is the diameter.
b) In case of circular stirrups, a reduction factor (λ1) is used by the calculation of shear resistance
due to the fact that the stirrups are not parallel to the shear force along the height of the
section. λ1 factor is calculated on the following way [Turmo et. al (2008): Shear truss analogy
for concrete members of solid and hollow circular cross sections] [J. Orr (2009): Shear Capacity
of Circular Concrete Sections], where z0 is the distance between the center of tension rein-
forcement and the center of gravity of the section, z is the level arm and rs is the distance
between the center stirrup reinforcement and the center of gravity of the section:
1
𝜆1 = ∫ √1 − ((𝑧0 − 𝑧𝑋)⁄𝑟𝑠 )2 𝑑𝑋
0
c) In case of spiral stirrups, another reduction factor is also used to consider that the vertical
component of resisting force is not parallel to shear force. λ2 factor is calculated on the fol-
lowing way [I. Feltham (2004): Shear in reinforced concrete piles and circular columns], where
p is the pitch of the spiral stirrup:
−0.5
𝑝 2
𝜆2 = (( ) + 1)
2𝜋𝑟𝑠
Level arm During the calculation of shear resistance of a reinforced concrete section, the level arm (z) of internal
forces plays an important role. The level arm is interpreted as the distance between the center of ten-
sion reinforcement and the center of compression zone. In case of beams, z is approximately selected
0.9d, where d is the effective depth. However, if axial force is present, this approximation cannot be
used. The program automatically calculates the level arm of internal forces with the consideration of
the axial force in the section.
The program checks the detailing rules in the standard regarding minimum/maximum longitu-
dinal reinforcement ratios, maximum distance of stirrups and minimum diameter of stirrups.
Warning message calls the attention of the user in the information window and instantly after
the design calculations if any of these detailing rules is not fulfilled. During the definition of the
parameters, the user is informed of the fulfilment of detailing rules regarding maximum dis-
tance of stirrups and minimum diameter of stirrups by clicking button.
See 6.5.4 Actual reinforcement against biaxial bending (column)
6.5.9.6. Shear and torsion check according to Eurocode 2 and DIN 1045-1
6.5.9.8. Capacity design: calculation the design value of shear force according to Eurocode and SIA
In case of dissipative structures with DCM or DCH ductility class, in order to avoid shear failure in
columns, the design values of shear forces are determined in accordance with the capacity design rule
(if seismic load case is included in the selected load combination), on the basis of the equilibrium of
the column under end moments Mi,d.
Calculation of Mi,d ∑ 𝑀𝑅𝐵
𝑀𝑖,𝑑 = 𝛾𝑅𝑑 𝑀𝑅𝑐,𝑖 𝑚𝑖𝑛 {1; }
∑ 𝑀𝑅𝐶
where:
Plastic hinge If plastic hinge may be formed only at the top of the column (e.g., due to hinged support, etc.), it can
at the top be set in the parameter window that plastic hinge is expected only at the top. In this case, the design
value of seismic shear force is calculated on the following way:
If seismic shear force from seismic analysis is greater than the shear force calculated based on
capacity design rules, shear force from seismic analysis is used as V Ed,EQ.
Results
The above-described calculation of design seismic shear force is performed only if there is seismic load
case in the selected load combination. A seismic shear force calculated in accordance with capacity
design rules is added to non-seismic shear forces in y and z directions respectively. The design value
of shear force including seismic shear force coming from capacity design is represented by dashed and
solid red lines. The seismic shear force needs to be considered with both positive and negative sign
considering positive and negative directions of seismic loading. Solid red line represents greater ab-
solute values.
The program checks the detailing rules in the standard regarding minimum/maximum longitudinal re-
inforcement ratio, maximum distance of stirrups and minimum diameter of stirrups. Warning message
calls the attention of the user in the information window and instantly after the design calculations if any
of these detailing rules is not fulfilled. During the definition of the parameters, the user is informed of the
fulfilment of detailing rules regarding maximum distance of stirrups and minimum diameter of stirrups
by clicking button. (See 6.5.4 Actual reinforcement against biaxial bending (column)) Checking the
fulfilment of further standard specific detailing rules is the responsibility of the user.
User’s Manual X6 Release1 e2 499
Design check of composite columns subjected to compression/tension force and biaxial bending can
be performed according to the following design codes:
Eurocode 4: EN 1994-1-1:2010
SIA : SIA 264:2014
The chapter 3.1.16.1 Cross-Section Editor shows how to create a composite section in the software.
Columns fulfil the following criteria can be designed only in RC2 module according to the design codes:
• steel grade needs to between S235 to S460;
• strength class of concrete need to be between C20/25 to C50/60;
• column is isolated and part of a framed structure where the other structural members are
either composite or steel members;
• the column section is doubly symmetrical.
It has to be noted that both Eurocode and SIA standard prescribes further criteria that need to be
checked by the user. The program shows warning messages in the following cases:
• if the steel grade or the strength class of concrete is out of the acceptable range;
• if the effect of local buckling cannot be neglected (the design check performed by the software
does not take into account the effect of local buckling);
• if the steel contribution ratio (δ) is out of the acceptable range.
The steps of design check of composite columns are very similar or identical with the steps need to be
done in course of the design of conventional reinforced concrete columns (i.e., how to define reinforce-
ment, which button need to be clicked). For this reason, the steps described in 6.5.4 Actual reinforcement
against biaxial bending (column) and 6.5.9 Column reinforcement – RC2 module sections are not dupli-
cated, however, the differences are underlined.
Parameters
This window is used to define the design parameters similarly to conventional reinforced concrete col-
umns. Most of the parameters are described in previous section see 6.5.4 Actual reinforcement against
biaxial bending (column). Additionally, the steel grade of steel sections and reduction factor (ρ) consid-
ering the effect of shear force on bending and compression resistance can be specified. ρ factor is used
if the shear force Va,Ed on the steel section exceeds 50% of the design shear resistance Vpl,a,Rd of the steel
section. This factor is used to reduce the considered yield strength of the steel material int the web of
the steel section:
(1 − 𝜌) ∙ 𝑓𝑦𝑑
500
Definition of The reinforcement of the section can be defined in the following window. It is identical compared to
reinforcement definition of reinforcement in case of conventional reinforced concrete column see 6.5.4 Actual rein-
forcement against biaxial bending (column). In case of composite columns, the concrete cover on the
steel section (Δ) can be also specified.
Column check The program performs the verification against bending with or without axial force of the column based
on the cross-section and given design properties and reinforcement parameters. The details of
evaluation of the bending utilization are explained in the next sections. The graphical interpretation of
the results and the result tables are discussed in the aforementioned section see 6.5.9 Column rein-
forcement – RC2 module
The utilization of the column subjected to uniaxial y-y and z-z bending moments are evaluated on the
following way:
𝑀𝑦,𝐸𝑑
𝜂𝑦 = ≤ 1.0
𝛼𝑀,𝑦 𝑀𝑦,𝑁,𝑅𝑑
𝑀𝑧,𝐸𝑑
𝜂𝑧 = ≤ 1.0
𝛼𝑀,𝑧 𝑀𝑧,𝑁,𝑅𝑑
where
• My,N,Rd and Mz,N,Rd are the design value of the bending resistance of composite section taking
into account the compressive normal force,
• αM,y and αM,z are coefficients related to bending of a composite column about the y-y axis and
the z-z axis respectively (S235 – S355 αM=0.9; S420 – S440 αM=0.8).
Interaction curves for combined compression and uniaxial bending related y-y and z-z axes are
calculated separately. The utilization of the column subjected to biaxial bending is calculated on the
following way:
𝑀𝑦,𝐸𝑑 𝑀𝑧,𝐸𝑑
𝜂= + ≤ 1.0
𝑀𝑦,𝑁,𝑅𝑑 𝑀𝑧,𝑁,𝑅𝑑
It may happen that in the info window or in the result table ??? symbol can be seen instead of
utilization of the column. This symbol means that the utilization of the column cannot be
evaluated. In most of the cases, the reason of this error is that the design axial force is higher
than the compression/tension resistance of the section, thus, the bending moment based
utilization of the column cannot be calculated.
User’s Manual X6 Release1 e2 501
Design bending My,Ed and Mz,Ed bending moments contain first order bending moment, the bending moment coming
moment from the imperfection and second order effects. The second order effects are taken into account with
the following multiplier coefficient (k) that is used to multiply the sum of first order bending moments
and bending moments from the imperfection:
𝛽
𝑘= ≥ 1.0
𝑁𝐸𝑑
1−
𝑁𝑐𝑟,𝑒𝑓𝑓
where
• β is an equivalent moment factor,
• Ncr,eff is the critical normal force for the relevant axis and corresponding to the effective flexural
stiffness calculated as follows.
𝜋 2 (𝐸𝐼)𝑒𝑓𝑓,𝐼𝐼
𝑁𝑐𝑟,𝑒𝑓𝑓 =
𝐿2𝑐𝑟
1
(𝐸𝐼)𝑒𝑓𝑓,𝐼𝐼 = 0.9 ∙ (𝐸𝑎 𝐼𝑎 + 𝐸𝑠 𝐼𝑠 + 0.5 ∙ 𝐸𝑐𝑚 𝐼)
1 + 𝜑𝑡 𝑐
where
• EaIa is the bending stiffness of steel section,
• EsIs is the bending stiffness of reinforcement,
• Ecm, φ and Ic are the secant modulus of elasticity of concrete, the creep coefficient and second
moment of area of the un-cracked concrete section.
The beam reinforcement design module can be applied to beam structural elements modeled by beam
or rib finite elements, that have the same material and constant or variable rectangular or T cross
sections, assuming that the load is applied in the symmetry plane of the cross section.
The computed longitudinal top and bottom reinforcement are of the same steel grade, while the stirrups
could have steel grade different from the longitudinal ones.
Beam reinforce- Beam design (calculating the required amount of reinforcement, placing stirrups and rebars, and check-
ment parameters ing the reinforced beam) uses parameters entered in the Beam reinforcement parameters dialog (see
6.5.3 Beam reinforcement parameters).
The program performs design calculations described below. Every other analysis, if prescribed by
the design code, have to be done by the user.
The present version of the module does not deal with out of plane bending, the effect of complex
internal force states, lateral-torsional buckling or the effect of peak stresses perpendicular to the
axis due to the action of concentrated forces and is not suitable for the reinforcement design of
short cantilevers.
Design calculations Click the Design calculation button on the Beam reiforcement dialog to display the details of calculations
according to the current design code. References to sections and formulas of the design code appear
in blue. If the Design or Calculated beam reinforcement tab is active, the design calculations are based
on the calculated reinforcement. If the Actual beam reinforcement tab is active, the short report is based
on actual reinforcement.
Clicking on the Settings icon beside the Design calculations button allows to set the level of elaboration
and the basic units for force and length used in the design calculations. Important results also appear
converted to standard AxisVM units (see 3.3.8 Units and Formats).
For further description about Design calculations window see 6.6 Steel design
and 6.6.1.3 Diagrams and design calculations
User’s Manual X6 Release1 e2 503
Longitudinal The longitudinal reinforcement against bending is displayed in blue, compression reinforcement in red,
reinforcement the minimum reinforcement required by the design code in gray.
against bending
Stirrup spacing The allowable maximum stirrup spacing is displayed in black, the calculated spacing in blue, and the
minimal spacing according to the design code in gray.
Checking Auto sets the detected width values. Uncheck it to set the Actual width and the Theoretical
width reduction manually. It lets you specify the a1 and a2 segments on the side of the support that will
be ignored in the calculations. The internal forces are linearly interpolated within the segments.
Checking Shear force reduction activates a method described in the design code to reduce shear force
above the supports. Uncheck Enabled to ignore the support in the design calculation.
Display results The display of diagrams and labels on each tab can be customized.
Plastic hinges In case of dissipative structures with DCM or DCH ductility class (EN 1998-1-1), in order to avoid shear
failure of beams, the design values of shear forces are determined in accordance with the capacity design
rule (if seismic load case is included in the selected load combination), on the basis of the equilibrium of
the beam under: a) the transverse load acting on it in the seismic design situation and b) end moments
Mi,d (with i=1,2 denoting the end sections of the beam), corresponding to plastic hinge formation for
positive and negative directions of seismic loading.
The following parameters can be given by the user: the safety factor, the distance between plastic hinges
and the reinforcement of plastic hinges. If Actual reinforcement checkbox is checked, the reinforcement
parameters related to plastic hinges are calculated automatically based on the defined actual beam re-
inforcement (see 6.5.11.3 Checking actual beam reinforcement). The length below the plastic hinges in
the window is the length of critical region where the program considers stricter requirements regarding
detailing rules of shear reinforcement in order achieve the expected ductility.
User’s Manual X6 Release1 e2 505
Considering capacity design rules, the design shear force is calculated on the following way (if seismic
load case is included in the selected combination):
where:
• VEd,0 denotes the shear force from non-seismic actions (from dead loads, etc.),
• VEd,EQ is the design value of seismic shear force.
𝑀1,𝑑 + 𝑀2,𝑑
𝑉𝐸𝑑,𝐸𝑄 = 𝛾𝑅𝑑
𝑙
where:
• l is the distance between plastic hinges (xcr2 – xcr1).
The seismic shear force is considered with both positive and negative sign considering positive and
negative directions of seismic loading.
If seismic shear force from seismic analysis is greater than the shear force calculated based on
capacity design rules, seismic shear force from seismic analysis is used as VEd,EQ.
The program considers detailing rules regarding max. distance of stirrups by the calculation of
stirrup arrangement. Checking the fulfilment of further standard specific detailing rules of beams
with plastic hinges is the responsibility of the user.
After clicking on the Calculated beam reinforcement tab, the following diagrams can be displayed:
Envelope of reinforcement (As), side reinforcement against torsion (ASL), cracking (wk), stirrup spacing
(sw), moment resistance (MyRd), maximum shear force (VzRd), torsion resistance (TzRd), bending moment
utilization (MyEd/MyRd), shear utilization (VzEd/VzRd), torsion utilization (TzEd/TzRd), relative deflection
(ez,rel), absolute deflection (ez,abs).
Shear and torsion The program calculates the shear and torsion resistance:
- the shear resistance without shear / torsion reinforcement (𝑉𝑅𝑑,𝑐 and 𝑇𝑅𝑑,𝑐 ),
- the maximum shear resistance / torsion moment limited by the concrete compression strut (𝑉𝑅𝑑,𝑚𝑎𝑥
and 𝑇𝑅𝑑,𝑚𝑎𝑥 ),
- shear resistance of shear reinforcement (𝑉𝑅𝑑,𝑠 ) and the maximum torsion moment (𝑇𝑅𝑑,𝑠 )
The amount of stirrup area taken into account for torsion by the following:
𝑉𝐸𝑑,𝑖
𝐴𝑠𝑤,𝑇 = ⋅ 𝐴𝑠𝑤 ,
𝑉𝐸𝑑,𝑖+𝑉𝐸𝑑
where 𝐴𝑠𝑤 is the area of one stirrup leg,
𝑇
𝑉𝐸𝑑,𝑖 = 𝐸𝑑 ⋅ 𝑧 is the shear force from torsion,
2⋅𝐴𝑘
𝑉𝐸𝑑 is the design shear force.
During the calculation of torsional resistance, only the external stirrup legs are considered.
Check table Two rows of data are displayed for each section, one for the top reinforcement and one for the bottom.
(ULS) Each row contains ULS internal forces and values of different intermediate result components.
A shorter summary is also available displaying values only at certain important sections.
The rebar scheme displays the number of rebars in the flange extension (outside of the web) in [square
brackets]. Rebars in the web is displayed row by row from the outside in (round brackets).
Check table Similar tables are generated for SLS internal forces, crack width and deflection values.
(SLS) Top cracking is calculated from the (max) moment causing tension within the upper part. Bottom crack-
ing is calculated from the (min) moment causing tension within the lower part. If no tension appears on
a side (max is negative or min is positive), calculations are performed with zero moment. In this case the
table shows zero and shows the actual moment in brackets.
Internal forces that appear in these two tables are different only if the reinforcement was calcu-
lated for an envelope or critical combination. If a load case or an individual load combination was
selected the internal forces will be the same.
Changing actual On the Actual beam reinforcement tab, it is possible to change the actual longitudinal reinforcement or
beam reinforcement stirrup spacing assigned to finite elements.
After changing the reinforcement or stirrup spacing, the results are recalculated automatically, if Display
/ Recalculate automatically is checked. If this feature slows down the definition (e.g., the beam has to be
checked for a huge number of critical combinations) deactivate this option. In this case click on the
Calculate button at the bottom to recalculate the results.
508
Longitudinal reinforcement
If this button is down the longitudinal reinforcement can be defined.
Select finite elements or remove the selection by clicking or dragging the mouse. To add non-adjacent
elements to the selection press Shift during the operation. Diameter of corner rebars and other longitu-
dinal rebars are set among the beam reinforcement parameters. Edit boxes allow changing the number
of top and bottom rebars in the selected elements. The – and + buttons decrease/increase these values.
The diagram of the actual reinforcement is filled with sky blue, the As reinforcement calculated for the
current load case or combination can be seen behind it as a red diagram filled with pink.
Checking the actual The program displays the selected results the same way as on the previous tab but for the actual rein-
reinforcement forcement. The tables are the same but display results for the actual reinforcement.
User’s Manual X6 Release1 e2 509
Stirrup spacing
If this button is down the actual stirrup spacing can be set.
A diagram of applied stirrup spacing (s*) and the specific area of reinforcement (as*) appears at the top.
A grid appears on the right displaying the stirrup spacing as a step function constant in continuous
sections along the beam. The x coordinate of the start point of section, the L length of the section, and
the s* spacing over that section can be edited. If the results have been calculated the utilization of each
section is also displayed (if this value is greater than 1 the background is red otherwise it is green).
Editing After clocking on the Apply calculated reinforcement / stirrup spacing button the table shows the calcu-
stirrup spacing lated values. Use the buttons below to edit the table describing the step function
510
Shear & torsion re- The design is based on the following values of design shear resistance:
inforcement design
of stirrups
𝑉𝑅𝑑,𝑐 Design shear resistance of the cross-section without shear reinforcement.
𝑉𝑅𝑑,𝑚𝑎𝑥 Maximum shear force that can be transmitted without the failure of the inclined
compression bars.
𝑉𝑅𝑑,𝑠 Design shear resistance of the cross-section with shear reinforcement.
𝑇𝑅𝑑,𝑐 Design torsional resistance of the cross-section without shear reinforcement.
𝑇𝑅𝑑,𝑚𝑎𝑥 Maximum torsional moment that can be transmitted without the failure of the inclined
compression bars.
AxisVM calculates the shear & torsion reinforcement assuming that shear crack inclination angle is 45.
The relation between the capacity of inclined compression concrete bars and the design values is
checked.
𝑉𝐸𝑑 𝑇𝐸𝑑
+ ≤ 1,
𝑉𝑅𝑑,𝑚𝑎𝑥 𝑇𝑅𝑑,𝑚𝑎𝑥
where
If the cross-section does not fail it is checked if shear & torsion reinforcement is required according to
the formula
𝑉𝐸𝑑 𝑇𝐸𝑑
+ ≤ 1,
𝑉𝑅𝑑,𝑐 𝑇𝑅𝑑,𝑐
1⁄
where 𝑉𝑅𝑑,𝑐 = [𝐶𝑅𝑑,𝑐 𝑘(100𝜚𝑙 𝑓𝑐𝑘 ) 3 + 𝑘1 𝜎𝑐𝑝 ] 𝑏𝑤 𝑑 and 𝑇𝑅𝑑,𝑐 = 2𝑓𝑐𝑡𝑑 𝑡𝑒𝑓𝑖 𝐴𝑘
User’s Manual X6 Release1 e2 511
The cross-section fails if critical shear force is higher than the shear resistance of the compressed
concrete beams, i.e.:
𝑉𝐸𝑑 𝑇𝐸𝑑
+ >1
𝑉𝑅𝑑,𝑚𝑎𝑥 𝑇𝑅𝑑,𝑚𝑎𝑥
Design rules applied in calculation:
On the basis of equation 9.2.2 (9.5N) 𝜌𝑤;𝑚𝑖𝑛 = 0.08 √𝑓𝑐𝑘 /𝑓𝑦𝑘 and of equation 9.2.2 (9.4) 𝜌𝑤 = 𝐴𝑠𝑤 /𝑠𝑏𝑤 ,
so the ratio of shear reinforcement is 𝑠𝑚𝑎𝑥,1 = 𝐴𝑠𝑤 /𝜌𝑤𝑚𝑖𝑛 𝑏𝑤 9.2.2 (9.6N) states that: 𝑠𝑚𝑎𝑥,2 = 0.75 𝑑.
The software designs compressed longitudinal rebars if the height of the compressed zone exceeds a
limit height. Beyond this limit, the stress does not reach yield strength in the tensioned reinforcement
and yielding cannot be assumed.
The required top and bottom reinforcement along the beam and the moment diagram shift is calculated
for each load case.
Due to inclined cracks tension reinforcement is designed for a force greater than calculated from M/z.
This is taken into account by different design codes by shifting the moment diagram.
Minimum (Mmin 0) and maximum (Mmax 0) values of the moment diagram and the corresponding
reinforcement on tension and compression side is determined. Tension reinforcement is displayed in
blue, compression reinforcement in red, the minimal tension reinforcement required by the design code
appears in grey.
In case of designed compressed longitudinal bars, the stirrup spacing may not be greater than 20 times
of the diameter of the designed compressed main bars.
Cracking is calculated according to 6.5.7.1 Cracking calculation according to Eurocode 2
512
Shear & torsion re- The design is based on the following three values of design shear resistance:
inforcement design
of stirrups
𝑉𝑅𝑑,𝑐𝑡 Design shear resistance of the cross-section without shear reinforcement.
𝑉𝑅𝑑,𝑚𝑎𝑥 Maximum shear force that can be transmitted without the failure of the inclined
compression bars.
𝑉𝑅𝑑,𝑠𝑦 Design shear resistance of the cross-section with shear reinforcement.
For cross sections with shear reinforcement we can choose between the regular method (45° cracking)
and Variable Angle Truss (VAT) method.
If the assumed compression trusses have reserve (𝑉𝑅𝑑,𝑚𝑎𝑥 > 𝑉𝐸𝑑 ) according to the regular method, the
VAT method will lead to considerable savings in shear reinforcement.
By changing the shear crack inclination angle the compressed concrete beams gets more load while
shear reinforcement gets less.
The program is calculating the value
𝜎𝑐𝑑
1.2 − 1.4
𝐴
𝑓𝑐𝑑 𝑠𝑤
cot Θ =
𝑉
1 − 𝑅𝑑,𝑐
𝑉𝐸𝑑
In case of regular concrete: 0.58 ≤ cot Θ ≤ 3.0
In case of light concrete: 0.58 ≤ cot Θ ≤ 2.0 DIN 1045-1 10.3.4 (3)
The regular method assumes the angle of shear cracks to be 45, so cot Θ = 1
𝐴
𝑉𝑅𝑑,𝑠𝑦 = 𝑠𝑤 𝑓𝑦𝑑 𝑧 cot Θ DIN 1045-1 10.3.4 (7)
𝑠𝑤
is the shear resistance due to the shear reinforcement.
Stirrup reinforce- Resistant twisting moment on the basis of the failure of the compressed concrete bar:
ment from twisting
2𝐴𝑠𝑤
moment 𝑇𝑅𝑑,𝑠𝑦 = 𝑓 𝐴 cot Θ
𝑠𝑤 𝑦𝑑 𝑘
The stirrup distance:
𝟐𝑨𝒔𝒘
𝒔𝒘 = 𝒇 𝑨 𝐜𝐨𝐭 𝚯
𝑻𝑬𝒅 𝒚𝒅 𝒌
𝟐𝑨𝒔𝒍
Longitudinal reinforcement is calculated from twisting moment 𝑻𝑹𝒅,𝒔𝒚 = 𝒇𝒚𝒅 𝑨𝒌 𝐭𝐚𝐧 𝚯 ,
𝒖𝒌
so 𝑨𝒔𝒍 = 𝑻𝑬𝒅 𝒖𝒌 /𝟐𝒇𝒚𝒅 𝑨𝒌 𝐭𝐚𝐧 𝚯, which should be placed evenly along the cross-section contour.
User’s Manual X6 Release1 e2 513
The software designs compressed longitudinal rebars if the height of the compressed zone exceeds a
limit height. Beyond this limit, the stress does not reach yield strength in the tensioned reinforcement
and yielding cannot be assumed.
If from the calculation a greater height than x0 is obtained, compressive steel cross section is applied,
but the sum of the compressive and tensile steel cross section cannot exceed 8% of the concrete cross
section.
The software calculates for each load case and cross section the lower and upper reinforcement, and
the value of the moment shifting.
Due to oblique cracks the tension reinforcement is designed for a tension force greater than calculated
from M / z.
This is taken into account by design codes by shifting the moment diagram (DIN 1045-1 13.2.2)
Minimum (Mmin 0) and maximum (Mmax 0) values of the moment diagram, and the corresponding
tension and compression reinforcements are determined. On the reinforcement diagram the tension
reinforcement is displayed in blue, the compressive in red, and the minimal tension reinforcement
according to the design code in grey.
In case of designed compressed longitudinal bars, the stirrup spacing may not be greater than 12 times
of the diameter of the designed compressed main bars.
𝐴𝑠𝑤
Construction rules Ratio of stirrup reinforcement: 𝜌𝑤 = 𝑏𝑤
𝑠
considered in the
From the above expression: 𝑠𝑚𝑎𝑥,1 = 𝐴𝑠𝑤 /𝜌𝑤 𝑏𝑤 , where 𝜌𝑤 = 0.16𝑓𝑐𝑡𝑚 /𝑓𝑦𝑘
program
Minimal value of w is may calculated from Table 29. in DIN 1045-1 13.1.3
The smax stirrup distance is taking into account Table 31. in DIN 1045-1 13.2.1
The maximum stirrup distance from twisting moments is uk / 8.
The software sends warning message and does not draw any reinforcement diagram in the
following cases:
Message The cross section is not acceptable for shear/torsion
Event Any of the following conditions is not satisfied:
2 2
𝑇𝐸𝑑 𝑉𝐸𝑑
𝑉𝑅𝑑,𝑚𝑎𝑥 > 𝑉𝐸𝑑 or [ ] +[ ] ≤1
𝑇𝑅𝑑,𝑚𝑎𝑥 𝑉𝑅𝑑,𝑚𝑎𝑥
Solution Increase the cross section of the concrete, or/and the concrete grade.
Message The cross section is not acceptable for bending (As + As2 > 0.08 * Ac)
Event The cross sectional area of the longitudinal reinforcement is greater than 8% of the concrete cross
section
Solution Increase the cross section of the concrete, or/and the concrete grade, or/and the steel grade.
Shear & torsion re- The shear reinforcement design is based on three values of the shear resistance:
inforcement design
of stirrups
𝑉𝑅𝑑 The shear resistance of the cross section without shear reinforcement.
𝑉𝑅𝑑,𝑐 The maximum shear force that can be transmitted without the failure of the assumed
compression bars.
𝑉𝑅𝑑,𝑠 The shear resistance of the cross section with the shear reinforcement.
No shear reinforcement is required if 𝑉𝑑 ≤ 𝑉𝑅𝑑
1
𝑉𝑅𝑑 = 𝑘𝑑 𝜏𝑐𝑑 𝑑𝑏𝑤 , where 𝑘𝑑 = , d in m, 𝑘𝑣 = 2.5
1+𝑘𝑣 𝑑
The software designs compressed longitudinal rebars if the height of the compressed zone exceeds a
limit height. Beyond this limit, the stress does not reach yield strength in the tensioned reinforcement
and yielding cannot be assumed.
Due to oblique cracks the tension reinforcement is designed for a tension force greater than calculated
from M / z.
This is taken into account by shifting the moment diagram.
Minimum (Mmin 0) and maximum (Mmax 0) values of the moment diagram, and the corresponding
tension and compression reinforcements are determined. On the reinforcement diagram the tension
reinforcement is displayed in blue, the compressive in red, and the minimal tension reinforcement
according to the design code in grey.
In case of designed compressed longitudinal bars, the stirrup spacing may not be greater than 15 times
of the diameter of the designed compressed main bars.
User’s Manual X6 Release1 e2 515
AxisVM sends a warning message and does not draw any reinforcement diagram in the following
cases:
Message The cross section is not acceptable for shear/torsion
Event The utilization of concrete cross-section is greater than 1.
Solution Increase the cross section of the concrete, or/and the concrete grade.
The check of resistance of compression struts and the calculation of required shear reinforcement are
incorporated.
This feature is available for the following domain types (see 4.9.6 Domain):
• normal domains,
• composite ribbed domains,
• parametric ribbed domains,
• domains with custom stiffness matrix,
• AIRDECK slabs,
• composite slabs with trapezoidal sheeting (if the Trapezoidal plate used only as formwork op-
tion is checked).
Openings and slab The RC3 module automatically recognizes domains being closer than 6d to the selected column/wall
contour end. When finding the punching perimeter, it is important to decide if an opening must be considered
as an opening or its size make it necessary to handle it as an edge of the slab. Edges of openings larger
than 2 m2 at the edge of the slab and edges of inner openings larger than 5 m2 is considered to be an
edge of the slab.
516
Punching shear control perimeters are determined based on the column cross-section and the effective
plate thickness. Plate edges and holes are taken into account if they are closer to the column than six
times the effective plate thickness. If column cross-section is concave, a convex section is used instead.
If a rib element is connected to the column within the plane of the plate, analysis cannot be
performed!
Materials
Concrete, Concrete and steel grade used in calculation. First, these parameters are automatically selected based
rebar steel grade on model settings, but they can be changed in the window.
Plate Plate thickness is automatically selected but it can be changed in this window if By reinforcement param-
thickness (h) eter is turned off. In the info window the minimum mushroom head thickness is displayed as H1. The
minimum mushroom head without punching shear reinforcement is displayed as H2.
Reinforcement ratio Reinforcement ratios (𝜌𝑥 , 𝜌𝑦 ) can be taken from the actual or calculated reinforcement or set Custom
ratio and specify manually.
Parameters
Shear reinforce- Angle between the plate and the punching shear rebars (45°-90°).
ment angle
Radial rebar Radial rebar spacing is the difference between the radii of two neighboring perimeters of shear rein-
spacing forcement.
The OK button is not available until basic design criterion is met: 𝑠𝑟 ≤ 0.75 ⋅ 𝑑. (section 9.4.3. (1))
Distance of the first Distance of the first punching perimeter of shear reinforcement from the convex edge of the column.
perimeter of shear
reinforcement
User’s Manual X6 Release1 e2 517
factor 𝑀𝐸𝑑 𝑢1
Calculated (6.4.3. (3) (6.39)) 1+𝑘 ⋅
𝑉𝐸𝑑 𝑊1
Internal column 1,15
Approximate value by
column position* (6.4.3. (6)) Edge column 1,4
Corner column 1,5
Custom user-specified value
*For structures where the lateral stability does not depend on frame action between the slabs and the
columns, and where the adjacent spans do not differ in length by more than 25%.
Take soil reaction If this option is checked, soil reaction within the rebar circle is considered when calculating the punching
into account force. This effect increases with the radius and can reduce the size of the necessary reinforcement area.
Its values per rebar circles are listed in the Punching analysis results dialog.
The program verifies the punching shear resistance according to EN 1992-1-1 6.4.4. (2) along control
perimeters within 2d distance from the column edge.
You can see coefficient of seimic forces at 4.10.24 Seismic loads – SE1 module.
Open... Loads a saved parameter set.
After entering all parameters control perimeters will appear and the required number of punching re-
bars is displayed in the info window.
AxisVM calculates the effective parts of the control perimeter based on plate edges and holes. Contin-
uous lines show that reinforcement is needed. AxisVM displays the required amount of reinforcement
for each line. The info window shows the amount of critical punching reinforcement. When calculating
the length of the critical perimeter it is assumed that rebar spacing on the perimeter is not above 2d
but the fulfilment of this requirement is not checked. If this requirement is not met, the user should
choose a smaller diameter or place additional rebars.
Calculate If this option is checked the reinforcement for control perimeters beyond 2d will be calculated from the
reinforcement actual length of the perimeter. For control perimeters closer than 2d the reinforcement calculated for
for each control the critical perimeter is applied.
perimeter If this option is not checked the reinforcement calculated for the critical perimeter will be used for all
perimeters.
Calculation of By default the punching force is calculated from the differ-
punching force ence in axial forces of connected columns.
Check Calculation of punching force by integration to find
the punching force by integrating shear forces of the slab.
The line of integration is at 2d from the convex outline of
the column.
In case of base plates, the punching force is integrated
at 0.5d distance from the convex contour of the column
(VEd,0). This value is used to check the resistance of con-
crete compression struts (compared to the design value
of the maximum punching shear resistance).
Calculation process The required punching reinforcement is calculated based on the following principles:
The column-slab connection does not fail if the shear stress is less than or equal to the design value of
the maximum punching shear resistance along the control section and the design value of the punching
shear resistance of the plate with punching shear reinforcement:
𝑣𝐸𝑑 ≤ 𝑣𝑅𝑑,𝑚𝑎𝑥 and 𝑣𝐸𝑑 ≤ 𝑣𝑅𝑑,𝑐𝑠
𝑣𝐸𝑑 design value of the shear stress
𝑣𝑅𝑑,𝑚𝑎𝑥 the design value of the maximum punching shear resistance along the control section
𝑣𝑅𝑑,𝑐𝑠 the design value of the punching shear resistance of the plate with punching shear reinforce-
ment
𝑉𝐸𝑑
𝑣𝐸𝑑 = 𝛽
𝑢𝑖 ⋅ 𝑑
where 𝑢𝑖 is the length of the control perimeter, d is the mean effective thickness of the plate.
𝛽 is a factor expressing additional stress due to eccentric forces:
𝑀𝐸𝑑 𝑢𝑖
𝛽 = 1+𝑘 ⋅
𝑉𝐸𝑑 𝑊1
518
Eurocode assumes that the critical section is at a distance of 2d from the edge of the cross-section. The
length of the critical perimeter and the static moment is calculated considering plate edges and holes
of the actual geometry.
Design value of the punching resistance of the connection without punching shear reinforcement is:
𝑣𝑅𝑑,𝑐 = 𝐶𝑅𝑑,𝑐 𝑘(100𝜌1 𝑓𝑐𝑘 )1/3 + 𝑘1 𝜎𝑐𝑝 ≥ 𝑣𝑚𝑖𝑛 + 𝑘1 𝜎𝑐𝑝
𝜌1 = √𝜌𝑥 𝜌𝑦 (𝜌1 = √𝜌𝜉 𝜌𝜂 in case of skew reinforcement)
If 𝑣𝐸𝑑 > 𝑣𝑅𝑑,𝑐 then the required punching reinforcement is determined along the critical perimeter
𝑑 𝐴𝑠𝑤 𝑓𝑦𝑤𝑑,𝑒𝑓
𝑣𝑅𝑑,𝑐𝑠 = 0.75 ⋅ 𝑣𝑅𝑑,𝑐 + 1.5 ⋅ ⋅ sin 𝛼 ; 𝑣𝐸𝑑 ≤ 𝑣𝑅𝑑,𝑐𝑠
𝑠𝑟 𝑢1 𝑑
The reinforcement for each perimeter and the perimeter where no punching reinforcement is needed
is calculated based on the formula:
𝑉𝐸𝑑
𝑣𝐸𝑑 = 𝛽 ≤ 𝑣𝑅𝑑,𝑐
𝑢𝑖 𝑑
Results for the critical perimeter are calculated first (these are displayed in the Punching analysis results
dialog). Then the required amount of reinforcement is determined for reinforcement circles defined in
the parameters dialog. The critical perimeter is red, reinforcement circles are black. Dashed line shows
the perimeter where the distance of points from the column is six times the effective plate thickness.
A thin blue line shows the perimeter where no punching reinforcement is needed.
This is also the outline of the mushroom head which can be designed with thickness hmin,2 and without
punching reinforcement.
A thick blue line shows the perimeter where the critical punching force exceeds the compressing
strength of the concrete so the plate with the original thickness cannot be properly reinforced. This is
the outline of the mushroom head which can be designed with thickness hmin,1 and with punching rein-
forcement. Punching capacity can be increased by setting the plate thicker, using a better concrete
grade or columns with bigger cross-section area.
Saves the drawing to the Drawing Library.
Saves the current punching parameters under a name. You can load back the saved parameters with
the button Loading... on Punching parameters dialog.
Inflates the plate boundary so that the entire column cross section is within the boundary.
Clicking on the Settings icon beside the Design calculations button allows setting the units for force and
length used in the design calculations.
520
Info window Under the design code, element identifier and materials the following pa-
rameters are displayed.
h plate thickness
d effective plate thickness
𝜌𝑥 , 𝜌𝑦 ratios of longitudinal reinforcement in x and y directions
α angle between the plate and the punching reinforcement
a1 distance of the first perimeter of shear reinforcement from the convex edge
of the column
sr distance between perimeters of shear reinforcement
fse factor scales only the internal forces from seismic load
VEd design value of punching force
VEd,0 design value of punching force integrated at 0.5d distance from the convex
contour of the column (only for base plates)
MEdy, MEdz design value of bending moments
u0 control perimeter at the column perimeter
u1 critical control perimeter at 2d
ϐ* calculated eccentricity factor
vEd0 shear stress along the u0 perimeter
vEd shear stress along the u1 perimeter
vRd,max the design value of the maximum punching shear resistance
vRd,c the design value of the punching shear resistance of the slab without punch-
ing shear reinforcement
vEd0/vRd,max utilization on the u0 perimeter
vEd/vRd,c utilization (tension in concrete)
fywd,ef tension in the punching reinforcement
Asw punching reinforcement area on the critical control perimeter
vRd,cs the design value of the punching shear resistance of the slab with punching shear reinforcement
Nsr number of reinforcement circles
hmin,1 minimum plate thickness required with punching reinforcement
hmin,2 minimum plate thickness required without punching reinforcement
User’s Manual X6 Release1 e2 521
Punching analysis of wall ends and wall corners can be performed using RC3 module including the
calculation of required reinforcement if necessary. The software generates the punching control perim-
eters according to the relative position of the walls. Plate edges and holes are taken into account if they
are closer than six times the effective plate thickness.
After clicking on the tool button shown above, a node needs to be selected that identifies the wall end
or the wall corner. The program identifies automatically the walls connecting to the node. Walls can be
represented by vertical domains or line supports with section (with stiffnesses calculated from wall pa-
rameters) (see 4.9.12 Line support). The software is able to identify both elements.
If a rib element is connected to the wall end/corner within the plane of the plate, analysis cannot
be performed!
Most of the parameters can be set in the dialog has been already presented in the previous sections
(see 6.5.12.1.1 Punching analysis of columns according to Eurocode).
The punching force is generally calculated according to one of the following ways in case of wall ends
and corners:
• by integration of section force along the wall(s)
• by integration of shear forces along a segment around the wall end/corner
acp
lcp lcp
acp
The second method is implemented in the software. acp and lcp parameters (used to generate section
line) can be specified by the user. The punching force is calculated from the shear forces along the line
on the following way presented below.
The acp and lcp parameters can be given in terms of d (effective depth) in the Punching parameters
window. In the case of acp=2 and lcp=2, the position of the integration line will be at 2d distance from
the edge of the wall and the ends of the integration line will be at also 2d distance from the wall
end/corner.
522
In case of base plates, the punching force is integrated at 0.5d distance from the contour of the
wall end/corner (VEd,0). This value is used to check the resistance of concrete compressed strut
(compared to the design value of the maximum punching shear resistance).
The distribution of shear forces along the integration line is shown by clicking on this tool button:
It has to be noted that the line given by acp and lcp parameters is a tool used to calculate the
punching force and it is not considered as punching perimeter. The punching perimeter is se-
lected according to the design code (see 6.5.12.1.1 Punching analysis of columns according to Euro-
code and 6.5.12.2.1 Punching analysis of columns according to SIA 262).
Calculation of VEd Calculated from maximum shear force:
punching force
𝑉𝐸𝑑 = 𝑣𝐸𝑑,𝑚𝑎𝑥 ∙ 𝐿
where
• vEd,max is the maximum value of shear force along the integration line;
• L is the length of the integration line.
2 + 𝑣2
𝑣𝐸𝑑 = √𝑣𝑥𝑧 𝑦𝑧
Soil reaction In case of wall end/wall corner punching analysis the soil reaction force is ignored as the punching
perimeters are not closed polygons and the punching force is typically integrated at 2d from the con-
tour of the wall thus the most of the soil interaction is already neglected.
Graphical interface The graphical interface and buttons are described and presented in detail in the previous section (see
and results 6.5.12.1.1 Punching analysis of columns according to Eurocode). Design results are listed in an info win-
dow and shown graphically in Plate punching analysis window. If punching reinforcement is required,
calculated reinforcement is written on punching perimeters. After clicking on the Design calculations
button, a report of the calculation details is displayed. Design calculations and drawings can be saved
into the documentation.
User’s Manual X6 Release1 e2 523
Punching analysis is enabled if reinforcement parameters and actual reinforcement are defined and as-
signed to the reinforced concrete slab.
After clicking the tool button, please select a column or a support with stiffnesses calculated from column
parameters for analysis.
Some details of design process are discussed in 6.5.12.1.1 Punching analysis of columns according to
Eurocode section and not repeated here.
If a rib element is connected to the column within the plane of the plate, analysis cannot be per-
formed!
Parameters related to punching design check can be specified in the following window similarly to Euro-
code conforming design.
Materials
Concrete, Concrete and steel grade used in calculation. First, these parameters are automatically selected based on
rebar steel grade model settings, but they can be changed in the window.
Plate Plate thickness is automatically selected but it can be changed in this window if By reinforcement param-
thickness (h) eter is turned off.
Reinforcement ratio Reinforcement ratios (𝜌𝑥 , 𝜌𝑦 ) can be taken from the actual or calculated reinforcement or set Custom
ratio and specify manually.
Parameters
Approximation step Approximation level can be selected according to SIA 262:2013 (section 4.3.6.4.2). Approximation steps
1 and 2 can be selected.
𝛽 Angle between the plate and the punching shear rebars (45°-90°).
Radial rebar Radial rebar spacing is the difference between the radii of two neighboring perimeters of shear rein-
spacing forcement.
𝑠1
Distance of the first Distance of the first punching perimeter of shear reinforcement from the convex edge of the column.
perimeter of shear
reinforcement
𝑠0
bsr Limit value for width of column strips for the determination of bending resistances.
Calculation of The punching force can be calculated from the difference in axial forces of connected columns or by the
punching force integration of shear forces of the slab (see 6.5.12.1.1 Punching analysis of columns according to Eurocode).
Design principles The software determines the punching perimeter at 0,5𝑑𝑣 distance (section 4.3.6.2.2) from the convex
edge of the column where punching resistance need to be checked. The resistance of compressive struts
is calculated according to section 4.3.6.5.7:
𝑉𝑅𝑑,𝑚𝑎𝑥 = 2 ∙ 𝑘𝑟 ⋅ 𝜏𝑐𝑑 ⋅ 𝑑𝑣 ⋅ 𝑢
If 𝑉𝑑 > 𝑉𝑅𝑑,𝑚𝑎𝑥 , the resistance of compressive struts is not adequate. In this case, the plate thickness or
concrete strength need to be changed.
The punching shear resistance without shear reinforcement is calculated on the following way (section
4.3.6.3.1):
𝑉𝑅𝑑,𝑐 = 𝑘𝑟 ⋅ 𝜏𝑐𝑑 ⋅ 𝑑𝑣 ⋅ 𝑢
If 𝑉𝑑 ≤ 𝑉𝑅𝑑,𝑐 , punching shear reinforcement is not necessary, otherwise the required shear reinforcement
is calculated as follows (sections 4.3.6.5.2 (65) and 4.3.6.5.3 (66)):
𝑉𝑑
𝛴𝐴𝑠𝑤 = ,
𝑘𝑒 ⋅ 𝜎𝑠𝑑 ⋅ 𝑠𝑖𝑛𝛽
where
𝐸𝑠 ⋅ 𝜓 𝑓𝑏𝑑 𝑑
𝜎𝑠𝑑 = ⋅ (1 + ⋅ )
6 𝑓𝑠𝑑 𝜙𝑠𝑤
𝑟𝑠 𝑓𝑠𝑑 𝑚𝑠𝑑 3⁄2 1,4⋅𝑓𝑐𝑡𝑚
𝜓 = 1,5 ⋅ ⋅ ⋅( ) and 𝑓𝑏𝑑 =
𝑑 𝐸𝑠 𝑚𝑅𝑑 𝛾𝑐
This required reinforcement needs to be applied between 0,35 ⋅ 𝑑𝑣 and 𝑑𝑣 according to the code. Based
on the given design parameters, the number perimeters of shear reinforcement are calculated can be
placed between 0,35 ⋅ 𝑑𝑣 and 𝑑𝑣 . Required reinforcement for a perimeter is calculated by dividing 𝛴𝐴𝑠𝑤
with the number of perimeters. This 𝐴𝑠𝑤 required reinforcement is placed (outside 𝑑𝑣 distance) on each
perimeter until it is needed.
User’s Manual X6 Release1 e2 525
Info window Under the design code, element identifier and materials the following
parameters are displayed.
ℎ plate thickness
𝑑 effective plate thickness
fse factor scales only the internal forces from seismic load
𝑉𝑑 design value of punching force
𝑉𝑑,0 design value of punching force integrated at 0.5d distance from the
convex contour of the column (only for base plates)
𝑀𝑑𝑦 ,𝑀𝑑𝑧 design value of bending moments
𝑘𝑒 reduction factor for punching shear perimeter
𝑢 length of punching shear perimeter
𝑢’ 𝑢′ = 𝑘𝑒 𝑢 ∙ 𝑢
𝑉𝑅𝑑,𝑚𝑎𝑥 the design value of the maximum punching shear resistance
𝑉𝑅𝑑,𝑐 the design value of the punching shear resistance of the slab without
punching shear reinforcement
𝑉𝐸𝑑 ⁄𝑉𝑅𝑑,𝑚𝑎𝑥 utilization for 𝑉𝑅𝑑,𝑚𝑎𝑥
𝑉𝐸𝑑 ⁄𝑉𝑅𝑑,𝑐 utilization for 𝑉𝑅𝑑,c
𝑉𝑑,𝑠 punching force for calculation of shear reinforcement
𝜎𝑠𝑑 design value of tensile stress in shear reinforcement
𝛴𝐴𝑠𝑤 required shear reinforcement
𝐴𝑠𝑤 required shear reinforcement for one perimeter
𝑉𝑅𝑑,𝑠 design value of punching shear force for calculation of reinforcement
𝑁𝑠𝑟 number of perimeters of shear reinforcement
526
6.5.12.2.2. Punching analysis of wall ends and wall corners according to SIA 262
The software determines the punching shear force as it is described in 6.5.12.1.2 Punching analysis of wall
ends and corners according to Eurocode. In case of SIA conforming design, β factor is not used. Reduction
factor for perimeter length (ke) can be specified by the user.
After clicking on tool button shown above, a node needs to be selected that identifies the wall end or the
wall corner. The program identifies automatically the walls connecting to the node. Walls can be repre-
sented by vertical domains or line supports with section (with stiffnesses calculated from wall parameters)
(see 4.9.12 Line support). The software is able to identify both elements.
Some details of design process are discussed in 6.5.12.1.2 Punching analysis of wall ends and corners ac-
cording to Eurocode section and not repeated here.
If a rib element is connected to the wall end/corner within the plane of the plate, analysis cannot
be performed!
Details of design principles and equations are discussed in the previous section. see Punching analysis of
columns according to SIA 262. In case of wall ends and corners, the approximation level cannot be selected
(section 4.3.6.4.2. of SIA 262:2013 chapter), the software uses approximation level 1, thus it is assumed
that msd/mRd 1.0 and the rs distance can be specified by the user.
User’s Manual X6 Release1 e2 527
AxisVM can determine the necessary size and reinforcement of rectangular or circular pad foundations
(with or without pedestal), and can check the footing against sliding and punching according to Euro-
code7 and MSz. It determines the settlement of the foundation as well.
Footing size The size of the foundation can be entered or let AxisVM calculate it. If AxisVM calculates the size a max-
imum value must be specified.
Using the soil profile and the internal forces this module determines the necessary size of the foundation
in an iterative process. Then it calculates the effective area of the foundation for load cases and combi-
nations, the design forces, moments and resistances, determines the settlement (for load cases and Ser-
vice Limit State [SLS] combinations), efficiencies and the shear reinforcement if necessary. The module
also checks the stability of the footing.
Step sides must not be bigger than the respective side of the foundation.
The coordinate system used in footing calculations is the coordinate system of the support. How-
ever, eccentricities are calculated relative to the center of the footing.
Footing design Click the Footing design icon and select one or more nodal supports with a vertical or slanted column. (If
parameters supports have been already selected, the dialog is displayed at the first click).
Footing design parameters have to be specified in a dialog.
Footing At the Footing tab select the footing type (simple plate / stepped / sloped, rectangular or circular) and
set the geometry parameters and the friction coefficient between the footing and the blind concrete
and the partial factor for the friction coefficient.
528
Rectangular footing
bx and by are the sides,
the column is concentric,
value or upper limit of bx and by must be entered
You can see coefficient of seimic forces at 4.10.24 Seismic loads – SE1 module.
Under the edit fields the footing and the column is displayed in top view. Given sizes
are drawn as continuous lines, upper limits as dashed lines. The forces appear as red
crosses placed according to their eccentricities. This diagram is for orientation pur-
poses only because the actual eccentricities are calculated taking into account the
self -weight of the footing and the backfill reducing the eccentricity.
If the button Show all support forces is down, the view is scaled to show all force crosses. If the
button is up only crosses within the bounding rectangle of the footing are displayed. These
forces act at the top of the footing and does not include the weight of the backfill, footing and
blind concrete.
Reinforcement On the Reinforcement tab reinforcement calculations can be activated.
Rebar steel grade, cT and cB top and bottom concrete covers can be entered (i.e., the least distance
between the surface of embedded reinforcement and the outer surface of the concrete). The top two
rows under Diameter and Direction represent the top rebars (the 1st row is the outer one, the 2nd row
is the inner one), the bottom two rows represent the bottom rebars (the 3rd row is the inner one, the
4th row is the outer one). The actual scheme in local x-z view is displayed accordingly.
Punching reinforcement calculations can be activated only for simple plate footings.
For the parameters see 6.5.12 Punching analysis – RC3 module
Soil At the Soil tab you can specify the soil profile and the properties of the backfill. Soil profiles can be saved
under a name and can be reloaded.
Properties of the selected layer is displayed in the Soil group box. Properties of the backfill is displayed
in the Backfill group box.
Soil layer properties can be changed. These changes can be applied to the soil layer clicking the Modify
layer button. Layer name and description can be modified. Layer color can be changed clicking the small
color rectangle beside the name. Soil library icon is placed beside the color rectangle. Clicking this icon,
a soil library is displayed with predefined layer properties.
530
Saves the soil profile under a name. This way you can reload the same soil profile for other
footings in the model.
If Save a copy to the soil profile library is checked the soil profile is also saved to a library. This
way you can reload the same soil profile in different models.
Opens the soil profile library.
The function available on the Soil toolbox are: Add new soil layer,
Move up, Move down, Delete.
Add new soil layer Adds a new soil layer with the properties and layer thickness set in
the group box. The new layer always gets to the bottom of the soil
profile.
Move up Moves the selected soil layer up within the soil profile.
Move down Moves the selected soil layer down within the soil profile.
Delete Deletes the selected soil layer from the soil profile.
User’s Manual X6 Release1 e2 531
Soil database Clicking the Soil database icon two tables are dis-
played. After selecting a soil and clicking the OK
button (or double clicking the soil) properties of
the selected soil are copied to the Soil or Backfill
group box.
Modify soil layer Name, color, description and physical properties of the selected layer can be edited. Click on this button
to apply changes to the selected layer.
Undrained loading Under undrained loading there is no volume change, since water cannot escape. The soil is fully satu-
rated, shear strength is a constant value that can be determined by experiments. In this case user must
enter the cuk shear strength.
Passive If this option is turned on the sliding resistance is increased taking into account the passive soil pressure.
earth pressure Active soil pressure increases the horizontal forces. These effects are usually neglected to be on the safe
side. Activating this option requires extra watchfulness.
γm,EP is the mobilization factor of passive earth pressure.
Checks Required checks and allowed maximum efficiencies can be selected on the Checks tab.
If Settlement evaluation is activated the program checks whether the settlement is greater than the limit
specified here but does not increase footing size.
Stability check allows setting the maximum allowed eccentricity ratio to ensure that the soil will be
compressed under the footing to avoid overturning.
Bearing resistance and Eccentricity check cannot be deactivated.
532
For rectangular footings 𝛾𝑒𝑐𝑐 = 0.5 means that the the bounding rectangle of the eccentricity limit
curve has the dimensions 𝑏𝑥 /2 and 𝑏𝑦 /2, where 𝑏𝑥 and 𝑏𝑦 are the footing dimensions. The eccentrity
factor is calculated according to the Type of eccentricity limit:
Linear (diamond), Quadratic (ellipse) Axis of rotation (rectangle)
𝑒𝑥 𝑒𝑦 2 |𝑒𝑥 | |𝑒𝑦 |
𝛾𝑒𝑐𝑐 = | | + | | 𝑒𝑥 2 𝑒𝑦 𝛾𝑒𝑐𝑐 = max ( ; )
𝑏𝑥 𝑏𝑦 𝛾𝑒𝑐𝑐 = √( ) + ( ) 𝑏𝑥 𝑏𝑦
𝑏𝑥 𝑏𝑦
For circular footings the eccentricity limit curve is a circle, 𝛾𝑒𝑐𝑐 = 𝑒/𝑟 , where 𝑟 = 𝑑𝑝 /2
Stability AxisVM calculates the moment of actions around the axes of overturning, sums up the stabilizing and
destabilizing moments then checks the following:
𝑀𝑑𝑠𝑡
Δ𝐸𝑄𝑈 = | | ≤ Δ𝐸𝑄𝑈,𝑙𝑖𝑚
𝑀𝑠𝑡𝑏
Sliding of the It is checked whether the horizontal force applied to the footing causes sliding failure between the
footing (or the footing and the soil using the following formula:
blind concrete) 𝐻𝑑 ≤ 𝑅𝑑,𝐻𝑠 + 𝑅𝑝,𝑑
on the soil Where the sliding resistance is
Drained condition: 𝑅𝑑,𝐻𝑠 = 𝑉𝑑 𝑡𝑎𝑛𝛿𝑑
For cast-in-situ concrete the design value of the structure-ground interface friction angle can be con-
sidered equal to the critical state angle of the shearing resistance: 𝛿𝑑 = 𝜑𝑐𝑣,𝑑
Undrained condition: 𝑅𝑑,𝐻𝑠 = 𝐴′ 𝑐𝑢,𝑑 and 𝑅𝑑,𝐻𝑠 ≤ 0,4𝑉𝑑 *
* If water or air can get between the footing and the soil EN 1997-1 6.5.3. (12)P requires an additional
check. It can be activated on the Soil tab by checking the following option
Sliding of the It is checked whether the footing slides on the blind concrete using the following formula:
footing on the 𝐻𝑑 ≤ 𝑅𝑑,𝐻𝑏 + 𝑅𝑝,𝑑
𝜇
blind concrete The sliding resistance being 𝑅𝑑,𝐻𝑏 = 𝑉𝑑 𝑐𝑐
𝛾𝜇
where 𝜇𝑐𝑐 is the friction coefficient between the footing and the blind concrete, 𝛾𝜇 is the partial factor
for the friction coefficient.
Reinforcement of If rebar positions and diameter are specified the module determines the necessary amount of top and
the foundation bottom reinforcement in x and y direction according to the following diagram.
base plate The minimum requirement is always taken into account.
User’s Manual X6 Release1 e2 533
Calculating Eurocode 7 allows different design approaches (DA). These are certain combinations of
according to partial factors for actions, material properties and resistances. Partial factor sets applied to actions are
Eurocode 7 referred to as A1, A2, sets applied to material properties are M1, M2, sets applied to resistances are R1,
R2, R3. (See EN 1997-1:2004, Annex A) Each design approach combine these partial factor sets.
The program checks A1+M1+R1 (DA1 / 1) and A1+M1+R2 (DA 2) for critical ULS combinations,
A2+M2+R1 (DA1 / 2) and A2+M2+R3 (DA3) for critical SLS combinations.
So for each critical combination two results are calculated.
If design was performed for a user-defined load combination set this combination to ULS or SLS oth-
erwise the footing may be overdesigned.
Bearing resistance is 𝑞𝑅𝑑 = 𝑠𝛾 ⋅ 𝛾 ′ ⋅ 𝐵′ ⋅ 𝑁𝛾 ⋅ 𝑖𝛾 ⋅ 𝑏𝛾 ⋅ 0.5 + 𝑠𝑞 ⋅ 𝑞 ⋅ 𝑁𝑞 ⋅ 𝑖𝑞 ⋅ 𝑏𝑞 + 𝑠𝑐 ⋅ 𝑐 ⋅ 𝑁𝑐 ⋅ 𝑖𝑐 ⋅ 𝑏𝑐
Sliding check calculates if the footing meets the following criterion between the footing and the blind
concrete and between the blind concrete and the soil: 𝐻𝑑 ≤ 𝑅𝑑 + 𝑅𝑝,𝑑
where 𝐻𝑑 is the design value of the horizontal force, 𝑅𝑑 is the design shear resistance, 𝑅𝑝,𝑑 is the pas-
sive soil resistance at the side of the footing.
Design shear resistance is obtained from the formula 𝑅𝑑 = 𝑉𝑑 tan 𝛿𝑑 , where 𝑉𝑑 is the design vertical
tan 𝜑
action, 𝛿𝑑 is the design angle of friction: 𝛿𝑑 = arctan ( ), where 𝜑 is the angle of interface friction,
𝛾𝜑
𝛾𝜑 is the partial factor of shearing resistance, prescribed by the design approach.
Punching check The module checks the shear resistance of the foundation (𝑣𝑅𝑑,𝑚𝑎𝑥 ), at the perimeter of the column
and determines the necessary amount of shear reinforcement.
The calculation reduces the punching force by the soil reaction on the effective area
(and within the critical punching line).
The punching check is passed if 𝑣𝐸𝑑 ≤ 𝑣𝑅𝑑
Without shear reinforcement 𝑣𝑅𝑑 = min {𝑣 𝑣𝑅𝑑,𝑐 }, with shear reinforcement 𝑣𝑅𝑑 = min {𝑣𝑣𝑅𝑑,𝑐𝑠 }.
𝑅𝑑,𝑚𝑎𝑥 𝑅𝑑,𝑚𝑎𝑥
Predicting the AxisVM calculates the elastic settlement caused by additional stress in soil layers.
settlement of Loads cause the following stress at depth of z under the center of the centrally loaded rectangle of
footing the footing (after Boussinesq-Steinbrenner):
4𝜎0 𝑏 𝑎(𝑎2 + 𝑏 2 ) − 2𝑎𝑧(𝑅 − 𝑧) 𝑏𝑧 𝑎(𝑅2 + 𝑧 2 )
𝜎𝑧 = {arctan [ ⋅ 2 2 2
]+ 2 ⋅ 2 }
2𝜋 𝑧 (𝑎 + 𝑏 )(𝑅 − 𝑧) − 𝑧(𝑅 − 𝑧) 𝑏 + 𝑧 (𝑎 + 𝑧 2 )𝑅
2
534
where the distance between the characteristic point and the central axes are 0.37B' and 0.37L'.
a and b are the dimensions of the four parts of the loaded rectangle according to the following table:
𝑎 𝑏
AxisVM breaks up the user defined soil layers into 10 cm sublayers and calculates the stress due to
soil weight and the stress caused by loading at the bottom of the sublayer. The change in sublayer
thickness is calculated according to the following formulas:
𝜎𝑎𝑖 𝜎𝑖−1 + 𝜎𝑖
Δℎ𝑖 = ℎ𝑖 ; 𝜎𝑎𝑖 =
𝐸𝑠𝑖 2
𝜎𝑎𝑖 is the average stress caused by loading in sublayer i
𝜎𝑖−1 is the average stress caused by loading at the top of sublayer i
𝜎𝑖 is the average stress caused by loading at the bottom of sublayer i
𝐸𝑠𝑖 : the Young modulus of the sublayer i
The predicted settlement at a given depth is calculated as the sum of the changes in sublayer thick-
nesses for the sublayers above the level:
𝑚
𝑠𝑚 = ∑ Δℎ𝑖
𝑖=0
AxisVM calculates the limit depth, where 𝜎 = 0.1 ⋅ 𝜎𝑜𝑏 (i.e., the extra stress caused by loading falls un-
der the 10% of the stress due to soil self-weight.
If this condition is not met at the bottom of the layer structure a settlement estimation is made based
on the settlement at this point and the stress ratio (> 0.1) is calculated.
If the stress caused by loading at the footing base plane is smaller than the stress due to the
original soil layers settlement is not calculated.
AxisVM calculates the settlement for all load cases and SLS combinations. Stress and settlement func-
tions are displayed for the selected load case. Settlement function s(z) is the total settlement of layers
above z.
Seismic check If vibration analysis is completed and the seismic loads are generated seismic check of footings can be
performed. Enter the following parameters:
User’s Manual X6 Release1 e2 535
The seismic resistance of the footing is verified according to EN 1998-5 Annex F. Especially the value of
the Cyclic undrained shear strength for cohesionless soil (τcy,u) must be determined carefully.
SIA Geotechnical calculations are performed according to EN 1997-1 using the safety factors according to
SIA. Structural calculations follow SIA 262, punching analysis is performed according to 6.5.12 Punching
analysis – RC3 module.
Results The designed foundation will be displayed in top view with soil layers, punching circles and places
dimension lines automatically. The 3D model can be zoomed in and out, shifted and rotated just like
the main model.
536
If stress caused by loading at the bottom of the layer structure is still more than 10% of the stress due
to soil self-weight the limit depth cannot be determined as the further structure of the soil is unknown.
In this case the info window displays the value of the settlement function at the bottom of the layer
structure as >value.
To improve the estimation further soil layer information must be added.
Footing This table displays the forces of the selected supports and the most important results including calcu-
internal forces lated geometry.
As support forces are calculated in the local system of the support the x and y directions are the local
x and y directions of the support. If the supports are global, these are the global X and Y directions.
Rx, Ry, Rz, Rxx, support forces
Ryy, Rzz
qEd design bearing pressure
qRd design bearing resistance
qEd /qRd soil utilization factor
Axb local x direction bottom reinforcement (if calculated)
ayb local y direction bottom reinforcement (if calculated)
axt local x direction top reinforcement (if calculated)
ayt local y direction top reinforcement (if calculated)
τEd /τRd utilization based on footing displacement relative to the blind concrete
τEd2 /τRd2 utilization based on blind concrete displacement relative to the soil
User’s Manual X6 Release1 e2 537
Detailed Displays the data in the table of Footing internal forces and the following results:
internal forces
Design approach design approach used to calculate the results of the line
cx, c y x and y size of the effective rectangle
e x, e y eccentricity of action in x and y direction
Rebars xb rebar scheme in bottom x direction (if calculated)
Rebars yb rebar scheme in bottom y direction (if calculated)
Rebars xt rebar scheme in top x direction (if calculated)
Rebars yt rebar scheme in top y direction (if calculated)
τEd design shear stress between the footing and the blind concrete
τRd design shear resistance between the footing and the blind concrete
τEd2 design shear stress between the soil and the blind concrete
τRd2 design shear resistance between the soil and the blind concrete
VRdc minimum shear design resistance without punching reinforcement
VRdmax maximum shear design resistance without punching reinforcement
VRdcs shear design resistance with punching reinforcement
u1 length of the critical line
Asw shear reinforcement along the punching line
Stress ratio ratio of stress caused by loading and the stress due to self-weight of the soil (if
limit depth is below the bottom of the layer structure its value is determined at
that point and is greater than 0.1, otherwise it is 0.1)
Limit depth the depth where stress ratio is 0.1 (if limit depth is greater than the bottom of
the layer structure a ? is displayed)
Copies image to the Clipboard
Display parameters
Turns on and off symbols of the drawing.
538
Using RC5 module, reinforcement can be assigned to reinforced cores and walls and the design check
of cores/walls subjected to bending and axial force can be performed.
Reinforcement can be assigned to virtual beams or virtual strips (see 2.16.16 Virtual beams) after
clicking the icon on the left or clicking on the axis of a virtual beam/strip. The icon is visible if there is
reinforced concrete domain, there is at least on virtual beam/strip and statical or dynamical results are
available in the model.
After clicking on the icon, one virtual beam or strip can be selected in the course of selection procedure
(see 2.16.1 Selection). If the selection is invalid, the software warns with a warning message. The
selection may be invalid in the following cases:
After a valid selection, the Reinforced concrete cores and walls window is opened where storeys and
reinforcement can be assigned to cores and walls and the design check can be performed. Altogether,
there are three tabs on the window: 1) Virtual beams; 2) Virtual strips; 3) Overall utilization.
Virtual beams can be used to design reinforced concrete cores with the assumption that sections
remain approximately plane before and after loading. This assumption is likely not valid in case of large
and short cores and in case of the presence of considerable warping torsional moment and a section
that is sensitive to warping effects.
The module can check concrete wall sections (virtual strips) or cores (virtual beams) subjected to
bending with axial force (N-M interaction). The shear check is available only for virtual strips, but the
shear utilization cannot be summed up with the N-M (bending with axial force) utilization. In the case
of complex cores, the shear check can only be performed for the separate virtual strips.
For eccentric internal forces (N-M) the design check of reinforced concrete cores is performed similarly
to the design check of reinforced concrete columns. The Nx-My-Mz strength interaction diagram of the
section is calculated and it is checked whether the point representing Nx,Ed, My,Ed, Mz,Ed design forces is
within the interaction diagram. During design check of core reinforcement, buckling length coefficient
cannot be defined, since modern design codes (e.g., EN 1992-1-1) includes methods wherewith the
effect of imperfections and second order effects can be taken into account on structural level. In
case of reinforced concrete cores, it is assumed that the integrated design internal forces include
the effect of imperfections and second order effects.
Virtual strips can be used by the design check of wall ends or wall segments considering possible
buckling failures of the wall between floors. The program is able to calculate the additional eccentricities
from imperfections and second order effects regarding a storey, based on the given buckling length
coefficient. After sum with the initial eccentricities (calculated from the integrated internal forces), the
program calculates the Nx,Ed, My,Ed, Mz,Ed design forces. It is assumed that the integrated design
internal forces include the effect of imperfections and second order effects on structural level.
The design check is performed similarly to the case of reinforced concrete cores that the Nx-My-Mz
strength interaction diagram of the section is generated and it is checked whether the point
representing Nx,Ed, My,Ed, Mz,Ed design forces is within the interaction diagram.
On the overall utilization tab, design results of reinforced concrete cores calculated with virtual beams
and design results of wall ends/segments calculated with virtual strips can be summed. It may be
necessary because the Nx-My-Mz strength interaction diagram of cores is generated without
consideration the loss of stability of compressed wall ends and inner wall segments. The program
supports the use of various summation rules as it can be seen in the following figures.
540
In case of warping-sensitive cores and the presence of considerable warping torsional moments,
the section warps and it does not remain plane after loading. It is recommended to Ms perform
the design check for wall segments of the core separately in such cases because the section can-
not be designed as a whole.
Virtual beams
Virtual beams of Virtual beams in the model are
the model listed on the left. If at least
storeys are assigned to the
virtual beam, the name of the
virtual beam is written bold
letters. After clicking on a
virtual beam written with bold
letters, the storeys of the
selected virtual beam are listed
in the list-box on the right of
the window.
If actual reinforcement is
assigned to the storey, the
storey is shown with bold
letters.
If a virtual beam is selected which has no storeys, the list on the right remains empty and only the
middle button is visible below the list. After clicking on this button, a new window opens where new
storeys can be added, storeys can be deleted and modified.
The storeys can be deleted (delete icon), modified (pencil) and new storeys can be added (+ and arrow).
The storeys are saved to the virtual beam/strip. If storeys are assigned to the virtual beam/strip, the
cross section will be displayed together with circles on the axis of virtual beam/strip indicating the
boundaries of the defined storeys.
Reinforcement can be deleted from the selected storey with this button. It is visible only if
reinforcement is assigned to the storey.
User’s Manual X6 Release1 e2 541
By clicking on table button, the design results of the selected storey of the selected virtual beam can
be seen summarized in a table (see below). This button is visible only if reinforcement is assigned to
the storey.
Define and check After clicking on the Define and check reinforcement button (or double clicking on a storey), a new
reinforcement dialog opens (see 6.5.14.1 Defining reinforcement) where new reinforcement can be assigned to the
storey or an existing reinforcement can be modified.
Instead of clicking on Define and check reinforcement button, alternatively the window for
define and check reinforcement opens if you double-click on the selected storey.
Virtual beams and Virtual beams and virtual strips in the model
virtual strips in the that have storeys and reinforcement are listed
model on the left. If actual reinforcement is missing
from at least one storey, the name of the
virtual beam/strip is written with red letters. In
this case, the overall utilization cannot be
calculated for every storey.
The list supports the multiple selection which
is available while Ctrl button is holding. To
create a virtual beam – virtual strip pair, a
virtual beam and a compatible virtual strip
need to be selected. If a virtual beam is
selected, the incompatible virtual strips will be
displayed with light gray color.
A virtual beam and a virtual strip are compatible if they have the same number of storeys and
their storeys are identical.
Calculation of The created virtual beam – virtual strip pairs are listed on the right, below the Calculate overall
overall utilization utilization label.
This button is visible only if a virtual beam and a compatible virtual strip are selected in the list on the
left. After clicking, a new virtual beam – virtual strip pair will be added to the list. The name is composed
from the names of virtual beam and the virtual strip.
This button can be used to delete virtual beam – virtual strip pair from the list.
The combination rule of utilizations for the selected virtual beam – virtual strip pair can be defined
below the list. The different combination rules are available.
1. η = max(η1, η2)
2. η = (η1)a + (η2)b
When creating a new virtual beam – virtual strip pair, the software sets the first combination rule
automatically. To modify the combination rule of the selected virtual beam – virtual strip pair, after
setting the new rule save button need to be clicked.
By clicking on table button, the design results of the selected virtual beam – virtual strip pair can be
seen summarized in a table (see below).
User’s Manual X6 Release1 e2 543
Display If storeys and actual reinforcement are assigned to the virtual beam/strip, the cross section with
reinforcement will be displayed together with circles on the axis of virtual beam/strip indicating the
boundaries of the defined storeys. The name of the actual reinforcement assigned to the storey is
displayed close to the middle of the storey on the axis of the virtual beam/strip.
With the association of storeys and actual reinforcement to virtual beams and virtual strips,
practically, we assign reinforcement to non-existing elements (virtual beam and virtual strip) in
order to integrate the stresses in domains and surfaces and to perform design check for the
integrated internal forces. The given width of virtual strips or the defined storeys do not
necessarily match to boundaries of domains or the finite element mesh. For this reason,
reinforcement parameters and actual reinforcement are independent of reinforcement
parameters (see 6.5.1 Surface reinforcement parameters and reinforcement calculation - RC1 module)
and actual reinforcement (see 6.5.2 Actual reinforcement) assigned to surfaces. Contradiction
between reinforcement parameters and actual reinforcements defined two different ways is not
investigated. Rebar statistics in weight report (see 3.2.13 Weight Report) does not incorporate
actual reinforcements assigned to virtual beams and strips.
544
Define and check Parameters for reinforcement and actual reinforcement can be specified in the dialog (Design check of
reinforced concrete core/wall) below. This dialog can be opened after clicking on Define and check
reinforcement button on Virtual beams and Virtual strips tabs, respectively. The functionalities, dialogs
and buttons are very similar to functionalities, dialogs and buttons described and presented by column
reinforcement (see 6.5.4 Actual reinforcement against biaxial bending (column) and 6.5.9 Column
reinforcement – RC2 module). On the page of Reinfrocement bars, the steel material, the concrete
material, the out-of-plane buckling parameters (only in case of virtual strips) can be specified and rebars
with certain diameter can be placed in the cross section.
Cross-section In this dialog window the design cross-section can be seen. This design cross-section is inherited from
the virtual beam/strip. By the definition of virtual beams and strips, the cross-section can be
automatically created from the intersected surface elements or a custom cross-section can be defined
by the user. The cross-section cannot be modified in this window, it is a property of the virtual
beam/strip. The cross-section can be changed only on Virtual beams window (see 2.16.16 Virtual
beams).
Local coordinate The local coordinate system of the virtual beam/strip plays an important role in the design check pro-
system cedure due to the fact that the internal forces are integrated from the stresses according to this coor-
dinate system. The orientation of the cross-section displayed in Design check of reinforced concrete
core/wall window is also consistent with this local coordinate system.
Parameters Lets you specify the parameters used in the course of design check.
In case of virtual strips, 𝛽𝑦𝑦 buckling length factor can be specified. (𝛽𝑦𝑦 related to buckling in x-z plane).
It is also possible to select the buckling shape. The buckling length is calculates as the product of
buckling length factor and the height of the wall in the selected storey.
The considered concrete class in the calculation can be altered compared to concrete class assigned to
the structural element. It allows the user to evaluate structural safety with different concrete class with-
out new static analysis. ϕef is the so-called effective creep ratio according to Section 5.8.4 of EN 1992-
1-1 standard.
You can see coefficient of seimic forces at 4.10.24 Seismic loads – SE1 module.
In addition to the above settings, in the case of virtual strips, the diameter and spacing of the secondary
reinforcement can be set, and the shear check can be requested. In addition to the shear check of the
cross-section, the check of sliding shear can also be selected. The angle of the shear crack can be
specified. Selecting sliding shear check, two parameters depending on the roughness of the interface
can be set: f and c.
Virtual beam
Virtual strip
546
If global imperfections are taken into account by extra loads acting on the structure, they
must be ignored in reinforcement calculation. To achieve this, check Local imperfections
only. If a sway buckling shape is selected only a constant local geometric imperfection is
considered (see 6.5.14.3 Calculation of eccentricity).
To position / by Generates a reinforcement bar with a specified diameter to the location of the cursor.
cover If the cursor is on a corner or on the contour line, the reinforcement will be placed considering the
concrete cover.
Spacing Inserts evenly N+1 new rebars between two selected points.
Spacing with Inserts new rebars between two selected points considering constant s rebar distance. If the distance
constant rebar between to end points is not integral multiple of s rebar distance, the distance between the last and
distance the last but one rebar will be less than s.
Spacing on circular Inserts evenly N+1 new rebars between a selected starting point and an end point of a circular arch.
arch If an entire circle is drawn, the program inserts N rebars evenly around the circle.
Placing rebars at Places rebars in positive and negative corners (> 5°) if there is no rebar in the corner.
the corners
N (division number) The division number which defines the number of rebars as N+1.
s (rebar distance) The rebar distance that is considered in spacing with constant rebar distance.
The design check can be performed based on the following design codes:
Eurocode 2: EN 1992-1-1:2010 + (MSZ, DIN, SR, DS, NBN, BS, NEN, SFS, SS, NS, NTC, CSN, PN, ELOT
NAs)
SIA : SIA 262:2013
Wall check After selecting Core/wall check tab, the program performs the verification against bending with or
without axial force or the shear check based on the cross-section properties and reinforcement
parameters. In case of N-M interaction, the module determines the eccentricities coming from
imperfection and second order effects based on the given buckling parameters and according to
requirements of the current design code.
The program calculates 𝑁𝑥,𝐸𝑑 , 𝑀𝑦,𝐸𝑑 , 𝑀𝑧,𝐸𝑑 design internal forces (see 6.5.14.3 Calculation of eccentricity)
and it checks if these points are within the interaction diagram.
The design check of reinforced concrete cores and walls is performed very similarly to the design check
of reinforced concrete columns, thus diagrams, functionalities and buttons are very similar or identical
that are introduced in previous sections (for further details see 6.5.9 Column reinforcement – RC2 mod-
ule).
The displayed diagram and its properties can be set in the Display Parameters window.
The shear check is only available for virtual strips, if the check of shear is requested among the param-
eters of the reinforcement parameters of the wall. The walls are only verified against the shear forces
parallel to the plane of the wall.
In the result window, the following figures and diagrams are presented:
a) the side view of the wall strip and its reinforcement (in case of wider walls, only the end seg-
ments of the wall are shown)
b) the design shear force (Vy) and the shear resistance (blue line)
c) in the case of sliding shear, the design shear force at the bottom of the wall and the sliding
shear resistance (blue line)
d) the maximum utilization (η) along the height of the wall.
In the case of ribbed walls, calculating the shear resistance, only the cross-section of the wall domain
(without ribs) and its reinforcement is considered.
548
Internal forces The Design results table contains the maximum normal forces and bending moments at the top and
bottom of the selected storey of the invetigated core/wall. The internal forces are integrated from the
stresses and they are associated with the local coordinate system of the virtual beam/strip.
Detailing rules The software checks the most critical detailing rules according to the current design code. If any of
these rules are violated a warning message appears.
Eccentricities due to imperfections and second order effects are calculated only in the course of design
check of reinforcements assigned to virtual strips. These eccentricities are summed with the initial
eccentricities calculates from the internal forces. In case of reinforcements assigned to virtual beams,
the software does not calculate eccentricities due to imperfections and second order effects and the
initial eccentricities are design eccentricities because modern design codes (e.g., EN 1992-1-1) includes
methods wherewith the effect of imperfections and second order effects can be taken into account
on structural level. It is assumed that the internal forces already represent these effects.
The calculation of eccentricities due to imperfections and second order effects in case of reinforcements
assigned to virtual strips is performed similarly to the calculation presented by reinforced concrete
columns (see 6.5.9 Column reinforcement – RC2 module). The major difference is that eccentricities are
calculated only in direction perpendicular to the plane of the wall (it is assumed that this direction is
identical with local z axis).
In case of reinforcements assigned to virtual strips, it is assumed that the local z axis of the virtual
strip is perpendicular to the wall. For this reason, it is assumed that the weak axis is the local y
axis. Among the reinforcement parameters, the buckling length coefficient related to y weak
axis can be specified by the user. Review of the orientation of the local coordinate system is
recommended before performing design check.
User’s Manual X6 Release1 e2 549
𝑒𝑒 is the initial eccentricity calculated from the first order axial force and bending moment.
Initial eccentricities at the ends of the investigated section:
𝑒𝑒 = 𝑀𝐼 /𝑁𝑑
Initial eccentricities at intermediate cross-section of the investigated section (equivalent eccentricity):
550
0.6 ⋅ 𝑒𝑎 + 0.4 ⋅ 𝑒𝑏
𝑒𝑒 = max { } and |𝑒𝑎 | ≥ |𝑒𝑏 |,
0.4 ⋅ 𝑒𝑎
where 𝑒𝑎 and 𝑒𝑏 are the initial eccentricities at the ends of the investigated section.
𝑁𝐸𝑑 1
,
𝐴𝑠 𝑓𝑦𝑑
𝑛= 𝐴=
1+0.2𝜑𝑒𝑓
𝐵 = √1 + 2𝜔, 𝜔= ,
𝐴𝑐 𝑓𝑐𝑑 𝐴𝑐 𝑓𝑐𝑑
The prescribed calculation method of 𝒆𝟐 can be nominal curvature method or nominal stiffness method
according to NA of the selected code. In some NAs, the application of both methods is allowed. In
these cases, nominal curvature method is applied.
The prescribed calculation method of 𝒆𝟐 can be nominal curvature method or nominal stiffness
method according to NA of the selected code. In some NAs, the application of both methods is
allowed. In these cases, nominal curvature method is applied.
Nominal curvature 1 𝑙02
method 𝑒2 =
𝑟𝑐
1 𝑓𝑦𝑑
where = 𝐾𝑟 𝐾𝜑
𝑟 𝐸𝑠 ⋅0.45⋅𝑑′
8 ≤ 𝑐 ≤ 𝜋 2 based on the first order bending moment diagram
𝑁𝑢′ −𝑁𝐸𝑑
𝐾𝑟 = min { ; 1.0} , 𝐾𝜑 = max{1 + 𝛽𝜑𝑒𝑓 ; 1.0},
𝑁𝑢′ −𝑁𝑏𝑎𝑙
𝑓𝑐𝑘 𝜆
𝛽 = 0.35 + − , where 𝑓𝑐𝑘 is in N/mm2 unit
200 150
ℎ
𝑑′ = + 𝑖𝑠 , where 𝑖𝑠 is the radius of inertia of the longitudinal reinforcement
2
Nominal stiffness 𝛽
𝑒2 = (𝑒𝑒 + 𝑒𝑖 ) [1 + ],
method 𝑁𝐵 ⁄𝑁𝑑 − 1
𝜋2
where 𝛽 =
𝑐
𝑁𝐵 is the buckling load based on nominal stiffness
𝐸𝐼 = 𝐾𝑐 𝐸𝑐𝑑 𝐼𝑐 + 𝐾𝑠 𝐸𝑠 𝐼𝑠 is the nominal stiffness, where
𝐸𝑐𝑑 is the design value of the modulus of elasticity of concrete, see 5.8.6 (3)
𝐼𝑐 is the moment of inertia of concrete cross section
𝐸𝑠 is the design value of the modulus of elasticity of reinforcement, 5.8.6 (3)
𝐼𝑠 is the second moment of area of reinforcement, about the centre of area of the concrete
𝐾𝑐 is a factor for effects of cracking, creep etc, see 5.8.7.2. (2), (3)
𝐾𝑠 is a factor for contribution of reinforcement, see 5.8.7.2. (2), (3)
User’s Manual X6 Release1 e2 551
Eccentricities are determined in both bending planes. The program checks the following design situ-
ations:
Virtual beams Virtual strips
𝑀𝑑𝑦 = 𝑁𝑑 𝑒𝑒𝑧 𝑀𝑑𝑦 = 𝑁𝑑 (𝑒𝑒𝑧 + 𝑒𝑖𝑧 ± 𝑒2𝑧 )
𝑀𝑑𝑧 = −𝑁𝑑 𝑒𝑒𝑦 𝑀𝑑𝑧 = −𝑁𝑑 𝑒𝑒𝑦
AxisVM checks whether the calculated design loads (Mdy, Mdz, Nd) are inside the N-M strength
interaction diagram. If it is not satisfied in any of the design situations, the core/wall with the given
cross-section and reinforcement fails.
6.5.14.3.2. Check reinforcement according to SIA 262 (bending with axial force)
Eccentricities are determined in both bending planes. The program checks the following design
situations:
Virtual beams Virtual strips
𝑀𝑑𝑦 = 𝑁𝑑 𝑒1𝑧 𝑀𝑑𝑦 = 𝑁𝑑 (𝑒1𝑧 + 𝑒0𝑧 ± 𝑒2𝑧 )
𝑀𝑑𝑧 = −𝑁𝑑 𝑒1𝑦 𝑀𝑑𝑧 = −𝑁𝑑 𝑒1𝑦
AxisVM checks whether the calculated design loads (Mdy, Mdz, Nd) are inside the N-M strength
interaction diagram. If it is not satisfied in any of the design situations, the core/wall with the given
cross-section and reinforcement fails.
General principles, In case of virtual strips, the module provides the ability to calculate the shear resistance and to deter-
limitations mine the shear utilization of the wall strips. The shear utilization cannot be summed up by the utiliza-
tion coming from N-M interaction. In the case of cores consisting of several wall strips, the strips can
only be checked separately.
In the calculation, it is supposed that the vertical and horizontal reinforcement has an adequate an-
chorage.
The shear check consists of two parts:
1) The module calculates the shear strength of the concrete cross-section and the applied rein-
forcement, considering the angle of shear crack set among the reinforcement parameters.
The cross-sectional area of the ribs connected to the wall domain and their reinforcements
are not considered in the calculation.
The analysis can only be applied to walls for which the angle of the shear crack is considered
to be valid.
2) The module calculates the sliding shear resistance at the bottom of the walls, if this type of
check is requested.
The maximum shear utilization of a storey is calculated based on the result of the above two check.
Lever arm During the calculation of shear resistance of a reinforced concrete section, the level arm (z) of internal
forces plays an important role. The level arm is interpreted as the distance between the center of ten-
sion reinforcement and the center of compression zone. In case of beams, z is approximately selected
0.9d, where d is the effective depth. However, if axial force is present, this approximation cannot be
used. The program automatically calculates the level arm of internal forces with the consideration of
the axial force in the section.
User’s Manual X6 Release1 e2 553
−|𝑉𝑑 | cot 𝛼 𝑁𝑑
𝜎𝑑 = +
𝑧 ⋅ 𝑏𝑤 𝑙𝑤 𝑏𝑤
where:
• 𝑁𝑑 is the dimensioning value of normal force acting on the cross-section of the virtual strip.
It is negative in case of compression.
• 𝑉𝑑 is the dimensioning value of shear force,
• 𝛼 is the compression field inclination,
• 𝑙𝑤 is the length of the shear wall (lengtt of the virtual strip).
User’s Manual X6 Release1 e2 555
6.5.14.5. Results
The design results of reinforced concrete cores and walls are available in many ways in the software.
Drawings
Design check drawings (Nx-My-Mz, Nx-My, Nx-Mz and My-Mz strength interaction diagrams, critical
eccentricity curves, shear check) can be saved into the Drawings Library by clicking Save to Drawings
Library button (see 3.6.10 Drawings Library).
Tables
Design results table is available via Design check of reinforced concrete core/wall and Reinforced concrete
cores and walls windows, as well.
If the table is opened from the Design check dialog, the presented results depend on the active design
check drawing. (If the active result view is Nx-My-Mz, Nx-My, Nx-Mz and My-Mz strength interaction
diagrams or critical eccentricity curves, then only the result of N-M interaction is summarized. If the
active result view is Shear check (available only for virtual strips), only the shear utilization is displayed.)
The table is editable if it is opened through Design check of reinforced concrete core/wall window. In
this case, columns can be turned on/off and new custom force and moment values can also be added
to the table. If the table is opened from Reinforced concrete cores and walls window, the table is not
editable.
In this case, the load case/combination, envelope or automatic critical load combination considered in
the design check can be selected from the dropdown list. After changing the load situation, the design
checks are performed automatically, that can take a few seconds.
556
Design results table for overall utilizations (available only for N-M interaction) can be opened only
through Reinforced concrete cores and walls window. The table summarizes the design results for the
virtual beam and the virtual strip and it shows the overall utilization as well.
If the utilization cannot be calculated due to any reason, that is marked with ??? symbol. If the calculated
utilization exceeds 1.0, the utilization is displayed with red color.
The overall utilization is the utilization of N-M interaction. Shear check results of strips are not taken
into account.
Result components If there are storeys and actual reinforcements assigned to virtual beams/strips in the model and static
(linear/non-linear) or dynamic results are available, new result components become available on R. C.
Design tab related to design check results of cores and walls:
• Utilization (beam) – if there are storeys and actual reinforcements assigned to virtual beams
in the model,
• Utilization (strip) – if there are storeys and actual reinforcements assigned to virtual strips in
the model,
• Overall utilization – if at least virtual beam – virtual strip pair is defined in Reinforced concrete
cores and walls window.
• N-M utilization (Virtual strip) - if there are storeys and actual reinforcements assigned to vir-
tual strips in the model,
• V utilization (Virtual strip) - if there are storeys and actual reinforcements assigned to virtual
strips in the model and shear check is requested.
(Overall utilization results are displayed on virtual strip of virtual beam – virtual strip pairs.). The overall
utilization is the utilization of N-M interaction. Shear check results of strips are not taken into account.
Results in the Design results are also available in the Table Browser (see 2.9 Table Browser), where the results of every
Table Browser reinforced concrete wall are available in summary tables.
User’s Manual X6 Release1 e2 557
Stresses and strains within reinforced concrete elements are important in many cases (e.g., serviceability
checks, fatigue analysis). The RC6 module allows stress-strain analysis of columns and beams with ac-
tual reinforcement for Nx-My-Mz internal forces, taking into account the effect of actual reinforcement,
cracks in concrete and nonlinear behaviour of the materials. RC2 module is a prerequisite to use RC6
module because an actual column or beam reinforcement is a requirement (see 6.5.3 Beam reinforce-
ment parameters (uniaxial bending) and 6.5.4 Actual reinforcement against biaxial bending (column)).
The aim of the analysis is to obtain the strain and the curvatures of the section for given internal forces.
Due to the fact that this problem is not continuous and not linear, an iteration is required to find the
equilibrium.
Click on the icon, then select one (and only one) reinforced concrete structural member
(see 2.16.1 Selection). The icon is active if there is at least one reinforced concrete structural member
(beam, rib, truss) in the model with actual reinforcement. The following selection problems trigger a
warning message
If the selection is accepted, the Stress-strain analysis for reinforced concrete sections dialog appears
where the stress-strain analysis can be performed for any section along the member.
558
Load case, load You can select a case from the drop-down list to
combination, display:
envelope or critical • Load case, load combination
combination • The k-th increment of a nonlinear analysis
• Envelope display
• Critical combination
Position of the Stress-strain analysis is performed for internal forces in a given cross-section. The location of the cross-
cross-section section can be set with the slider at the bottom of the window.
Prints image.
Saves the drawing into the Drawings Library (see 3.6.10 Drawings Library and 3.6.11 Save to Drawings
Library)
Stress-strain analysis parameters - see 6.5.15.1 Stress-strain analysis parameters
Recalculation It may happen that the heuristic algorithm does not find the equilibrium or its precision is not adequate.
In these cases, stress-strain analysis can be repeated by clicking on the Recalculation button.
User’s Manual X6 Release1 e2 559
The parameters considered in stress-strain analysis can be entered in the following dialog.
Materials
Concrete and rebar steel grade can be
specified.
Internal forces
Internal force components can be
turned on/off. Only the checked com-
ponents are considered in stress-
strain analysis. At least one compo-
nent must be selected.
Seismic
For coefficient of seimic forces see
4.10.24 Seismic loads – SE1 module.
Concrete material The nonlinear concrete material model can be selected from a dropdown list. The available material
model models are different for Eurocode and SIA standards, respectively.
Eurocode
Custom Parameters of the material model are specified by the user. It can be used to consider tension strength
or creep effect on concrete stiffness.
560
SIA
Design codes specify bilinear, parabola-rectangle and nonlinear material models especially for
the design of cross-sections. For this reason, the stiffness of concrete is not necessarily equal to
the stiffness used by hand calculations. If a custom material model is selected for concrete, the
Young’s modulus of the concrete can be entered. This material model can be used if tension
strength or creep effect is considered.
The behaviour of steel is described by bilinear material model.
Filtering the results If envelope or critical combination results are selected in the Stress-strain analysis for reinforced concrete
sections dialog, the analysis is performed for multiple load cases and combinations. The results of these
analyses can be seen in the result tables, however, only the results of the critical case/combination are
presented graphically and listed in the info window. The presented case/combination can be critical in
terms of stress in concrete or in reinforcement. Set the required option under Filtering.
In the course of the analysis, the strain and the curvatures of the section are calculated for given internal
forces considering nonlinear material properties and cracking of concrete. Supported design codes are
Eurocode 2: EN 1992-1-1:2010
SIA : SIA 262:2013
RC6 module can use parallel computing if envelope or critical combination is selected, thus it is
recommended to turn on the option of multiple threads (Settings / Preferences / Analysis).
Iteráció Due to the fact that the problem is nonlinear and not continuous, an iteration need to be performed
to find the equilibrium. The RC6 module uses a heuristic algorithm based on a genetic algorithm.
User’s Manual X6 Release1 e2 561
In some cases, it may happen that the heuristic algorithm does not find the equilibrium or its
precision is not adequate (the accepted maximal difference in axial force is 1% or 1 kN and in
bending moment is 1 kNm). Stress-strain analysis can be repeated by clicking on Recalculation
button.
Integration Internal forces compatible with strains are calculated with the help of so-called fiber model through
numerical integration based on ε normal strains, κy and κz curvatures. The considered concrete material
model and rebar steel grade can be specified by the user.
6.5.15.3. Results
Results Results of the stress-strain analysis are presented in result tables and as a diagram in the main window.
Possible errors in the stress-strain analysis:
• the algorithm does not find the equilibrium, the stress-strain analysis should be repeated
• the internal forces exceed the resistance of the cross-section
Envelope and The results of the critical case/combination are presented as a diagram and listed in the info window if
critical combination envelope or critical combination is selected. The name of the load case/combination is shown above
the drawing.
Info window
fse – coefficient of seismic forces
Nx – axial force
My – bending moment
Mz – bending moment
Nx – axial force (integrated)
My – bending moment (integrated)
Mz – bending moment (integrated)
δNx – difference between input and integrated axial force
δMy – difference between input and integrated bending
moment
δMz – difference between input and integrated bending
moment
ε – strain at the centroid
κy – curvature (y axis)
κz – curvature (z axis)
x – height of compression zone
d – effective height
562
Result tables
By clicking on tables button, a result table opens with the detailed results of stress-strain analysis.
Custom internal forces can be given in the result table opened by clicking on this icon on the left (see
2.9 Table Browser). Stress strain analysis results for custom internal forces are available only in this result
table.
New row
Delete rows
Recalculation
User’s Manual X6 Release1 e2 563
It is assumed that the cross-sections do not have holes in them and are made of plates with a thickness
less than or equal to 40 mm.
The cross section should be constant or tapered. It is also assumed that the loads on single-symmetric
cross-sections act in the plane of symmetry, that is the plane of bending. For general shapes with no
plane of symmetry only Axial Force-Bending-Shear (N-M-V) and Compression-Bending-Buckling (N-M-
Buckling) is checked.
AxisVM performs only the checks listed below. All the other checks specified in the design code
like constrained torsion, strutting forces, joints, etc. has to be completed by the user.
The principal axes of an arbitrary cross section have to be coincident with the local y and z axes.
564
Classification of The program performs the classification of the cross-section based on EN 1993-1-1, Table 5.2. Whenever
cross-section possible, it determins the class based on the actual stress distribution of combined compression and
bending. If there is no compression in any point of the cross-section, it is automatically classified as
class 1.
This information is given by the program as auxiliary results. The checks are mostly
defined by interaction formulae. The definition and the detailed conditions of the application of the
variables contained by the equations can be found in the design code.
In the following, 𝑁𝑅𝑘 = 𝑓𝑦 𝐴; 𝑀𝑦,𝑅𝑘 = 𝑓𝑦 𝑊𝑦 and 𝑀𝑧,𝑅𝑘 = 𝑓𝑦 𝑊𝑧
where 𝑊𝑦 = 𝑊𝑝𝑙,𝑦 and 𝑊𝑧 = 𝑊𝑝𝑙,𝑧 for class 1 or 2 cross sections,
𝑊𝑦 = 𝑊𝑒𝑙,𝑦 and 𝑊𝑧 = 𝑊𝑒𝑙,𝑧 for class 3 cross sections and
𝑊𝑦 = 𝑊𝑒𝑓𝑓,𝑦 and 𝑊𝑧 = 𝑊𝑒𝑓𝑓,𝑧 for class 4 cross sections.
User’s Manual X6 Release1 e2 565
Axial Force- The member can be in tension or in compression. The check is performed on the basis of
Bending-Shear EN 1993-1-1, 6.2.1 (7).
𝑁𝐸𝑑 𝑀𝑦,𝐸𝑑 + Δ𝑀𝑦,𝐸𝑑 𝑀𝑧,𝐸𝑑
+ + ≤1
𝑁𝑅𝑘 /𝛾𝑀0 𝑀𝑦,𝑅𝑘 /𝛾𝑀0 𝑀𝑧,𝑅𝑘 /𝛾𝑀0
Δ𝑀𝑦,𝐸𝑑 = 𝑁𝐸𝑑 ⋅ 𝑒𝑁,𝑦 : it differs from zero only when the cross section is in class 4 and the original cross
section is asymmetric to axis y.
High shear
If the shear force is greater than 50% of the shear resistance, the effect of shear force is considered as
detailed below.
For section class 1. and 2. allowance is made on the resistance moment accoding to EN 1993-1-1, 6.2.8.
For section class 3. and 4. stresses are calculated and the general and conservative formula in EN 1993-
1-1, 6.2.1 (5) is applied. This is done for section types: I, T, C, box and pipe. For other section types (L
shape, rectangular and round sold shapes, and user defined shapes) the effect of hight shear has to be
calculated by the user.
where 𝑀𝑁,𝑦,𝑅𝑑 , 𝑀𝑁,𝑧,𝑅𝑑 are reduced moment resistances based on the effect of shear force and axial force
(EN 1993-1-1 6.2.8. and 6.2.9.1). For pipe sections, the reduced moment is calculated as follows:
2
𝑛1.7 𝑁𝐸𝑑 𝑉𝐸𝑑
𝑀𝑁,𝑦,𝑅𝑑 = 1.04 ⋅ (1 − 𝜌 − 0.7 ); 𝑛 = ; 𝜌 = (2 − 1)
(1 − 𝜌) 𝑁𝑝𝑙,𝑅𝑑 𝑉𝑝𝑙,𝑧,𝑅𝑑
2
𝑛1.7 𝑁𝐸𝑑 𝑉𝐸𝑑
𝑀𝑁,𝑧,𝑅𝑑 = 1.04 ⋅ (1 − 𝜌 − 0.7
); 𝑛 = ; 𝜌 = (2 − 1)
(1 − 𝜌) 𝑁𝑝𝑙,𝑅𝑑 𝑉𝑝𝑙,𝑦,𝑅𝑑
For bi-axial bending the criterion in EN 1993-1-1 6.2.9.1. (6) should be satisfied:
𝛼 𝛽
𝑀𝑦,𝐸𝑑 𝑀𝑧,𝐸𝑑
[ ] +[ ] ≤1
𝑀𝑁,𝑦,𝑅𝑑 𝑀𝑁,𝑧,𝑅𝑑
The check is based on the form of equations (6.61) and (6.62) of EN 1993-1-1, 6.3.3 :
𝑁𝐸𝑑 𝑀𝑦,𝐸𝑑 + Δ𝑀𝑦,𝐸𝑑 𝑀𝑧,𝐸𝑑
+ 𝑘𝑦𝑦 + 𝑘𝑦𝑧 ≤1
𝜒𝑦 𝑁 /𝛾𝑀 𝜒𝐿𝑇 𝑀𝑦,𝑅𝑘 /𝛾𝑀 𝑀𝑧,𝑅𝑘 /𝛾𝑀
𝑅𝑘 1 1 1
𝑁𝐸𝑑 𝑀𝑦,𝐸𝑑 + Δ𝑀𝑦,𝐸𝑑 𝑀𝑧,𝐸𝑑
+ 𝑘𝑧𝑦 + 𝑘𝑧𝑧 ≤1
𝜒𝑧 𝑁 /𝛾𝑀 𝜒𝐿𝑇 𝑀 /𝛾𝑀 𝑀𝑧,𝑅𝑘 /𝛾𝑀
𝑅𝑘 1 𝑦,𝑅𝑘 1 1
𝛥𝑀𝑦,𝐸𝑑 = 𝑁𝐸𝑑 ⋅ 𝑒𝑁,𝑦 : it differs from zero only when the cross section is in class 4 and the original cross
section is asymmetric to axis y.
566
The stability interaction formula is valid for beams that have an axis of symmetry in their plane of
bending. For C and U cross sections, the results are only informative.
Compared to earlier versions of the software, the lateral torsional buckling of beams with C or U
cross section is no longer checked in the plane of symmetry, but it is checked for an eccentric
load that is parallel to the web, and that goes through the center of the web.
Shear /y The check is performed on the basis of EN 1993-1-1, 6.2.6.
𝑉𝑦,𝐸𝑑
≤1
𝑉𝑐,𝑦,𝑅𝑑
Elastic sections:
𝐼 ∙ 𝑡 𝑓𝑦
𝑉𝑐,𝑦,𝑅𝑑 =
𝑆 √3𝛾𝑀
0
Supported sections:
Plastic or non-supported sections:
𝑓𝑦
𝑉𝑐,𝑦,𝑅𝑑 = 𝐴𝑣
√3𝛾𝑀0
Supported sections:
Plastic or non-supported sections:
𝑓𝑦
𝑉𝑐,𝑧,𝑅𝑑 = 𝐴𝑣
√3𝛾𝑀0
Web Shear- The check is performed for cross-sections with web (I and box sections) based on EN 1993-1-5 7.1, 6.2.8,
Bending-Axial Force 6.2.9 assuming that the web is parallel to the local z axis.
2
𝑀𝐸𝑑 𝑀𝑓𝑅𝑑 𝑉𝐸𝑑
+ (1 − ) ⋅ (2 − 1) ≤ 1
𝑀𝑝𝑙,𝑅𝑑 𝑀𝑝𝑙,𝑅𝑑 𝑉𝑏𝑤,𝑅𝑑
In case of high shear force or high axial force formulas in EN 1993-1-1 6.2.8, 6.2.9 are applied.
User’s Manual X6 Release1 e2 567
L ✓ ✓ — ✓ ✓ — —
U ✓ ✓ approximative ✓ ✓ — —
C ✓ ✓ approximative ✓ ✓ — —
Round ✓ ✓ ✓ ✓ ✓ — —
Rectangular ✓ ✓ ✓ ✓ ✓ — —
(*)
Double-sections
2I
if a=0 → welded box ✓ ✓ ✓ ✓ ✓ ✓ ✓
2L
✓ ✓ — ✓ ✓ — —
2L
if a=0 → T ✓ ✓ ✓ ✓ ✓ — ✓
2U opened ][
✓ ✓ — ✓ ✓ — —
2U opened ][
if a=0 → I ✓ ✓ ✓ ✓ ✓ ✓ ✓
2U closed []
✓ ✓ — ✓ ✓ — —
2U closed []
if a=0 → Box ✓ ✓ ✓ ✓ ✓ ✓ ✓
2LX
✓ ✓ — ✓ ✓ — —
4X
✓ ✓ — ✓ ✓ — —
4X
if a=0 → X
✓ ✓ — ✓ ✓ — —
568
(*) For double-section types, it is assumed that the connection between the two elements is continuous
(EN 1993-1-1 6.4.4: Closely spaced built-up members), and the buckling check is performed accordingly.
If the distance between the two sections is zero, the program will assume that the connection between
the elements is continuous and will replace the two with one section (I, T, X or box). The connection
needs to be calculated by the user.
(**) These sections are designed only if local coordinates are the same as principal directions.
(***) Calculate with the total height of the cross-section, stiffening effect of the middle flanges is neglected.
If the manufacturing process of the section is cold-formed or other, the member is not designed.
User’s Manual X6 Release1 e2 569
For the design based on Eurocode 3, the following design parameters should be defined and assigned
to the structural members to perform ULS checks:
ULS
(Ultimate Limit
State)
Cross-section
5
For the steel design, a cross-section size can be defined, that is different from the original cross-section
used for the static calculation. If the cross-section size is modified, the steel design checks will be per-
formed for the original internal forces and deflections, but with the modified cross-section size. Similarly,
in case of designing for fire, the original fire reduction factors are used (see 6.6.2 Steel beam fire design
according to Eurocode 3 – SD8 module). When viewing the results, a modified steel design member is
indicated by a * after the name of the cross-section, to distinguish it from the original cross-section. In
the drop-down menu, we can select from the cross-section sizes, that were loaded earlier into the model.
Clicking on this icon will allow us to load further cross-sections from the Cross-section library or by
defining a parametric shape. On the right side of this icon, the original cross-section sizes can be seen.
Design approach By section class (elastic/plastic): both elastic and plastic design methods are allowed, depending on the
section class of the structural member.
Elastic design: all checks use elastic design methods. Resistances are calculated from elastic cross-section
properties; in Class 4 effective cross-section properties are used.
Section class Automatic classification classifies the cross-section by the actual stress values.
Design member If the design member has a sway frame buckling mode in local x-y or x-z plane the respective bracing
bracing must be turned off. These settings affect the automatic flexural buckling calculation (AutoNcr), and in
case of lateral torsional buckling the lateral supports for the AutoMcr method. Furthermore, they affect
equivalent uniform moment factors Cmy and Cmz of the stability interaction check (EN 1993-1-1: 2005
Annex B: Method 2: Table B.3).
Assemble design The program assembles design members from the selected elements before performing design
members calculations. Design members consist of finite elements with the same material, same type (truss, beam
or rib) and local system orientation. Design members must have constant or linearly varying cross-
section and excentricity along the whole member. Finite elements must be on the same line.
Steel design members are not the same as the structural members (see 3.2.14 Assemble structural
members)
570
Factor for seismic forces see 4.10.24 Seismic loads – SE1 module
Stability AxisVM performs checks against buckling, lateral-torsional buckling and web shear buckling. Each check
parameters can be activated separately by clicking on the checkbox before its name. For example, if it is sure that
there is no need to check lateral-torsional buckling that part of the check can be deactivated and no
parameters has to be specified.
Buckling To determine the Ky and Kz buckling resistance around y
(flexural) and z axes (in the local x-z and y-z plane) end support con-
ditions must be defined.
There are three ways to specify buckling behaviour.
Ky, Kz buckling factors (effective length factors). The buckling factor will be multiplied by the length of the
design members and not the total length of the selected elements. Note that connecting elements might
lead to separation of a selected member into multiple design members (see Assemble design members
above)!
Ly, Lz buckling lengths. The entered buckling length will be used regardless of the length of the design
member. This often leads to more straightforward design in case of complex structures.
Auto. The buckling length of design members is calculated automatically. The so-called AutoNcr method
determines buckling length based on the model geometry and the distribution of internal forces in the
model. The buckling length of each design members is calculated after considering the stabilizing effect
of connected other members. This method is based on the rules recommended by the European Con-
vention for Constructional Steelwork (ECCS TC8: Rules for Member Stability in EN 1993-1-1: Background
documentation and design guidelines).
The original method was developed for vertical columns of simple frames. The improved algo-
rithm of AxisVM can handle any three-dimensional structures but for special geometries the error can be
considerable. In such cases it is recommended to check if the calculated buckling length values are within
the expected range. For complex structures it is also advisable to determine the critical load parameter
solving an eigenvalue problem (see 6.2 Buckling) or to perform a nonlinear analysis with material non-
linearity and geometric imperfection.
The buckling length heavily depends on whether the design member has a sway buckling mode.
Check the Braced in local x-y / x-z plane checkboxes according to the sensitivity of the member to sec-
ond-order effects. It is important to recognize for instance that columns of a portal frame typically buckle
in sway mode in-plane, while its beam does not (because both end points are supported by the columns).
AxisVM takes into account the effect of all beam elements (including steel beams with no steel
design parameters defined or non-steel beams). As a conservative assumption, all columns are consid-
ered to be in sway mode while all beams are in non-sway mode by default.
User’s Manual X6 Release1 e2 571
The algorithm takes into account hinges, rigid and semi-rigid connections. Nonlinear connections
are represented by their initial stiffness. Nodal supports and constrained nodal degrees of freedom are
also taken into account. Due to constrained nodal DOFs planar structures can show very small buckling
length for out of plane buckling. This is not an error but the consequence of the constrained motion at
the end points of finite elements. For planar structures it is recommended to specify the buckling length
for out of plane buckling or to model the actual supports instead of constraining nodal DOFs.
The buckling length of design members is heavily affected by the distribution of internal forces.
As these forces are different in each load case and combination, the calculated buckling length also
depends on the selected load case or combination! Calculation speed can be increased in the conceptual
design phase by neglecting the influence of internal force distribution; uncheck the Take N into account
checkbox in the Design Parameters dialog to do so. When the internal forces are not taken into account,
the design parameter under consideration is assumed to have uniform normal force distribution, while
other members are assumed unloaded.
Limitations
This buckling length calculation method can determine the critical load parameter only for structures
made of truss, beam or rib elements. Design members are considered to have piecewise constant cross-
section. The calculation method is not available for tapered beams yet. Effects from other types of struc-
tural components (e.g., plates, springs, rigid bodies, line or surface supports) are ignored. Only nodal
supports are considered; the influence of line and surface supports is not taken into account.
Additional information on the automatic flexural buckling coefficient calculation tool is available from
the menu Help / AutoNcr Guide.
If the Steel design tab is selected the calculated Ky or Kz buckling length factors can be displayed by
selecting Ky or Kz as a result component from the Resistances group.
Buckling of design
members
If a buckling analysis has been completed for the structure a summary of buckling of design members is
displayed in the Table Browser (e.g., RESULTS / Linear analysis / Steel design / Buckling / Load case or
combination / Mode 4). Nonlinear results are available only if geometrical nonlinearity was ignored and
the final load factor is 1.0. Another table (…Steel design / Buckling / Minimum Ncr) displays the smallest
critical force (Ncr) for each design member with other buckling data.
If design calculations failed for a certain member the cause of the error (wrong dimensions, wrong type
etc.) is also displayed in the Design member column (just like in the table of Utilization).
The critical force is calculated from the Euler formula
𝜋𝐸𝐼
𝑁𝑐𝑟 = = 𝑁𝑚𝑖𝑛 𝛼𝑐𝑟
𝐿𝑐𝑟 2
572
The buckling length of the equivalent Euler beam is 𝐿𝑐𝑟 = 𝐾𝑐𝑟 𝐿, where L is the physical length of the
design member and Kcr is the buckling factor.
αcr Critical load parameter
Nmin The smallest normal force along the beam
Ncr The critical force
Ly,cr Buckling length about the y axis (in z direction)
Lz,cr Buckling length about the z axis (in y direction)
Ky,cr Buckling factor about the y axis
Kz,cr Buckling factor about the z axis
λ Slenderness
Lateral-torsional K is a factor related to the constraints against warping. Its value must be between 0.5 and 1.
buckling - if warping is not constrained it is 1.0.
- if warping is constrained at both ends of the beam, it is 0.5.
- if warping is constrained at one of the ends of the beam, it is 0.7.
See in detail: Appendix F1 of ENV 1993-1-1.
Za is the load position relative to the center of gravity (a signed value).
Load position can be set to As defined or Custom in the dropdown list. As defined means that the load
position is determined from the eccentricity of a concentrated or distributed beam load. Nodal loads are
applied in the center of gravity of the cross-section. If Custom is chosen, load position must be selected
from the following options by radio buttons: Top, Center of gravity, Bottom or Custom (entering the Za
value). The analysis assumes that the local z axis is within the plane of symmetry and that the forces
causing lateral torsional buckling act parallel to this axis.
Calculation of Mcr Two options are available to calculate the critical moment of the lateral-torsional buckling (Mcr).
(critical moment)
1.) By formula
𝜋 2 𝐸𝐼𝑧 𝑘 2 𝐼𝑤 (𝑘𝐿)2 𝐺𝐼𝑡 2
𝑀𝑐𝑟 = 𝐶1 2
[√( ) + 2 + (𝐶2 𝑧𝑔 − 𝐶3 𝑧𝑗 ) − (𝐶2 𝑧𝑔 − 𝐶3 𝑧𝑗 )]
(𝑘𝐿) 𝑘𝑤 𝐼𝑧 𝜋 𝐸𝐼𝑧
Meaning of the parameters can be found in the literature or in the Appendix F1.2 of ENV 1993-1-1. The
value of 𝐶1 , 𝐶2 , 𝐶3 parameters depend on the shape of the moment curve and the k factors.
To enter parameter values, choose Custom C1, C2, C3.
In certain cases, 𝐶1 can be calculated automatically. Choose C1 Lopez formula from the Calculation
method for Mcr combo. This option is not available if the steel structural member is a cantilever or
𝐾𝑧 > 1.
𝐶2 must be entered if external loads are applied to the structural member and the point of application is
not coincident with the shear center of the cross section. In case of a single-symmetric cross-section 𝐶3
shall also be entered. 𝐶 parameter values can be set using ENV 1993-1-1, F1.2.
Calculating Mcr is also possible according to the method given in the national appendix of Dutch stand-
ard (NEN-EN 1993-1-1 + C2: 2011 / NB: 2011 Bijlage D). Based on the beam's moment diagram and its
internal load the C1 and C2 coefficients are automatically calculated in a conservative way according to
the given procedures and tables in the appendix (Calculated C1,C2 (NL NAD)). In some load cases and
support conditions, it is hard to automatically choose between any of the categories covered in the ap-
pendix, therefore we let the user define C1 and C2 (User-defined C1, C2 (NL NAD)).
2.) The AutoMcr method
This method makes a separate finite element model for each design member and calculates Mcr directly
for each load combination making 𝐶1 , 𝐶2 , 𝐶3 unnecessary but increasing calculation time. This method
handles variable cross-sections and cantilevers as well. The finite element sub-model of a beam contains
at least 30 finite elements where each node has four degrees of freedom essential to determine lateral
torsional buckling: 1) 𝜈, lateral shift in local y direction, 2) Θ𝑥 torsion, 3) Θ𝑧 lateral rotation, 4) w, warping.
This method builds the beam stiffness form two parts: the first one is linear, the second one has geo-
metric nonlinearity.
It applies loads with their eccentricity then reduces the calculation to an eigenvalue problem. The
method is developed for bending constant cross-sections in their plane of symmetry, so for variable
cross-sections the program creates the appropriate number of finite elements. [See Yvan Galea: Moment
critique de deversement elastique de poutres flechies presentation du logiciel ltbeam, CTICM, 2003]
User’s Manual X6 Release1 e2 573
Lateral supports AxisVM determines the support conditions for the sub-model
automatically by default.
Alternatively, the following methods may be chosen:
Estimated from kz, kw: the position of the supports is estimated
similarly a sin version 12 of AxisVM;
Fork supports at both ends;
User defined: the user may edit/define the supports.
The four stiffness components of a lateral supports are 𝑅𝑦 , 𝑅𝑥𝑥 , 𝑅𝑧𝑧 , 𝑅𝑤 .
It the mentioned option above was set to Automatic, the program determines the support conditions as
follows:
Model support: based on the supports along the beam defined earlier in the Elements tab; the 𝑅𝑦 , 𝑅𝑥𝑥
and 𝑅𝑧𝑧 values are directly taken from the support stiffness values, while 𝑅𝑤 = 0.
Connecting elements: line and shell elements directly connected to the designed member provide some
support against lateral torsional buckling. The estimated support stiffness values are summarized in the
AutoMcr Guide. See Help / AutoMcr Guide
Limitations
The sub-model does not take into account the effect of nodal DOFs.
Without having proper supports the sub-model has no stability against lateral torsional buckling. To
avoid this kind of instability the following conditions must be met: 1) 𝑅𝑦 is nonzero in at least one point
and 𝑅𝑦 or 𝑅𝑧𝑧 is nonzero in another one. 2) 𝑅𝑥𝑥 is nonzero in at least one point.
If the first condition is not met, the default settings are applied: 𝑅𝑦 and 𝑅𝑥𝑥 is rigid at both ends. It is an
approximation of the 𝑘𝑧 = 𝑘𝑤 = 1 in the ENV formula. For a cantilever beam, the default setting is a
support on one end with rigid 𝑅𝑦 , 𝑅𝑥𝑥 and 𝑅𝑧𝑧 .
If the second condition is not met, i.e., no torsional support was defined, the default setting is to make
𝑅𝑥𝑥 rigid at one end.
Lateral support conditions can be edited by clicking the … button.
A dialog appears with a table of lateral supports of the design members created from the selection.
Design members are listed on the left. The table shows the lateral supports of the selected item.
The last item of the list is Same supports on selected elements. Selecting this item, the table shows only
the common supports on the selected design members. Adding or deleting supports will change the
support configuration of all selected design members.
Add new support.
Adds a new line to the table and allows entering support properties.
Delete selected supports.
Deletes selected rows of the table.
The following properties can be edited: support position, eccentricity of the support in z direction relative
to the center of gravity of the cross-section, stiffness components. The last column shows the support
type:
The table displays connecting elements even if the angle between their axis and the design member axis
is greater than 15°, but only if the direction vector of the connecting element has a nonzero component
in the local y axis of the design member (lateral support). So, if the design member is a horizontal beam,
vertical columns and other horizontal beams in line with the design member will not appear in the table.
Table fields allow entering numbers and certain parametric values as well:
Pos. L is the length of the design member. So relative position can be entered like L/2 or 2*L/3.
Exc. h is the cross-section height. So, eccentricity can be entered like h/2 or 2*h/3.
R… Stiffness components: m will be interpreted as 1E+10.
In case of a connecting element with a length of a and an E*I stiffness 6*EI/a can be entered.
Right click on a list item on the left for additional operations Right click on a table row for additional operations
Changing the Assemble design members option will redefine design members so lateral supports
will be reset to default values and all modifications will be lost.
User’s Manual X6 Release1 e2 575
AxisVM version 12 estimated support conditions from kz and kw values. Models created with this version
will appear with Estimated from kz, kw setting.
In order to help using and understanding the AutoMcr method and defining the lateral supports cor-
rectly, guidelines and examples can be found in the AutoMcr Guide. See Help / AutoMcr Guide
Load position Za is the z coordinate of the point of application of the transversal load (relative to the center of gravity
of the cross-section), based on ENV 1993-1-1, Figure F1.1. It is a signed value and must be defined as
the ratio of this distance to the height of the cross-section. The positions of the center of gravity and the
top or bottom of the cross section can also be chosen by radio-buttons.
Web Shear For shapes with webs, the web can be supported or not with stiffeners:
Buckling No stiffeners: assumes no transversal stiffeners along the structural member.
Transversal stiffeners: In any cases the program assumes that there are transversal stiffeners (non-rigid
end post) at the ends of the structural members (e.g., at the supports).
You can see coefficient of seimic forces at 4.10.24 Seismic loads – SE1 module.
The parameters for the serviceability limit state (SLS) checks can be set on the SLS tab in the Design
parameters window.
SLS
(Serviceability Limit
State)
The program checks the deflections of beams and the horizontal displacements of columns. Inclined steel
design members are categorized by the angle between their centerline and the horizontal plane. If this
angle is above α = 45°, then the member is considered a column, and if it is below α = 45°, then it is
considered a beam. This α value can be modified at Checking inclined members.
If the critical load combinations are generated automatically, the program performs the SLS checks
in the characteristic load combinations.
576
Vertical deflections Deflections can be checked according to the z or y local axis directions of the steel design member. The
check in each direction can be activated/deactivated by checking/unchecking the box before the limit
values. The deflection limit values shall be set according to the length (L) detailed below.
Deflections can be considered in several ways: based on the actual displacements or based on the
displacements relative only to the left, only to the right or to both end points of the design member.
Length (L) Deflections of a design member and the deflection limits (e.g., L/300) are based on one of the chosen
length values below:
Design member length: L is equal to the length of the design member;
Based on connecting members and supports: If the Assemble design members parameter in the ULS tab is
set to the first option (members are broken at connecting elements and supports), then this length is
equal to the design member’s length, and thus this is the same as the length above. If it is set to the
second option (the selected beams are checked as one design member), then the program checks if there
are any supports or connecting members along the design member in the direction of the deflection
check that prevents the deflection of the design member. Between these, the SLS check is performed for
each section of the member. For example, checking a 15-meter-long beam, if there are supporting
columns in the end points and at one-third of the length of the beam, then the program performs SLS
checks separately for the first 5-meter-long section and the remaining 10-meter-long section of the beam.
Custom length: any value can be entered.
Pre-camber Pre-camber can be defined for the design members, where the pre-camber amounts (uz and uy) are
interpreted in the local coordinate-system. The shape of the pre-camber can be set with three parameters:
Quadratic or Linear curves may be chosen, for which the maximum value (u) and its location along the
member (xmax) shall also be defined by the user. As an exception, if the actual displacements are set by
the user (e.g., only a section of a beam is selected as a design member), then the pre-camber values are
constant along the length. If the length L is set to the design member length or to a custom length, the
pre-camber is zero in the end points of the design members. If L is Based on connecting members and
supports, the pre-camber will be zero in the end points and at the supports found by the program in the
relevant direction.
It is important to note, that the pre-camber value set here has an effect in the SLS checks only, by
reducing the previously calculated deflections that is based on the original shape.
Horizontal displace- Horizontal displacements can be checked in two directions at a time, according to the global or local
ments coordinate system. If the global coordinate system is chosen, the program checks the displacements in
horizontal X and Y axis directions. If the local coordinate system is chosen, then the horizontal components
of the deflections (wx and wy) are checked in the local z and y directions of the design member.
The height (H), along which the program performs displacement checks, can be the whole length of the
structure, or only the height of the design member or the level. In the first case, the actual deflections are
considered, while in the latter cases, only the relative displacements along the considered height are
checked. In case of inclined members, the height is defined as the vertical component of the design
member length. The height can be set by the user:
• Design member length (relative horizontal displacements);
• Height of the entire structure (actual horizontal displacements);
• Measured from height h (actual horizontal displacements).
User’s Manual X6 Release1 e2 577
Diagrams You can display the diagrams corresponding to all individual checks and their envelope by clicking on the
steel design member. Results for any position of any steel design member in any load case or combination
can be obtained by setting the combo boxes and dragging the tracking line. If a check cannot be per-
formed with the current cross-section the respective diagram is replaced by a cross.
Design calculations Clicking on the Design calculations button a report of the calculation details can be displayed. All strength
and stability checks appear as formulas completed with substituted actual values and references to the
design code.
The report consists of the six basic interaction checks listed above and several partial results which make
it easier to follow the calculations and provide useful details for cross-section optimization.
The partial results are:
1. Axial Plastic Resistance
2. Plastic Moment Resistance about the y axis
3. Plastic Moment Resistance about the z axis
4. Plastic Shear Resistance in z direction
5. Bending-shear interaction check
6. Bending-axial force interaction check
7. Flexural Buckling Resistance
8. Lateral-Torsional Buckling Resistance
Clicking on the Settings icon beside the Design calculations button allows setting the basic units for force
and length used in the design calculations. Important results also appear converted to standard AxisVM
units (see 3.3.8 Units and Formats).
578
Details of The details of calculations according to the current design code are displayed as a multi-page document.
calculations References to sections and formulas of the design code appear in blue.
Substitution Substitution into formulae can be turned on / off. Eliminating substitution makes the report somewhat
shorter.
Select the font size of the report.
Clicking on this icon adds the design calculation to the current report.
User’s Manual X6 Release1 e2 579
Optimization checks the design members for the same internal forces ignoring stiffness changes
due to changing dimensions. In certain structures recalculation of the model may show consider-
able changes in internal force patterns. In these cases, several consecutive optimizations may find
the more efficient structure.
Optimization uses the steel design parameters previously assigned to the design members.
Cross-section types suitable for optimization are: I, asymmetric I, rectangular, T, C, 2U shapes and pipes.
Variable cross-sections cannot be optimized.
Optimization groups The first step of the optimization is to create optimization groups from the existing steel design members.
Each member of an optimization group must have the same cross-section and optimization will assign
the same cross-section to the group members. The actual optimization can be started from the second
tab (Optimization).
The list of optimization groups (see it on the left side of the Design optimization groups tab) shows the
common cross-section and the number of design members within the group (<n>). Select a group and
set the optimization parameters on the right (see below).
Objective Objective of optimization can be (1) minimum weight, (2) minimum height or (3) minimum width.
of optimization
This defines the objective function. The process will search for the cross-section with a utilization <1 for
all group members and closest to the objective. This cross-section is called optimized cross-section. The
objective is reached separately for each group.
A maximum of utilization can also be set. Limiting the utilization can help when optimizing statically
indeterminate structures where cross-section modifications can cause big changes in internal forces.
Optimization It is possible to ignore certain checks during the optimization process. All strength checks are always
checks performed but checks for flexural buckling, lateral torsional buckling and web buckling can be deac-
tivated.
Optimization types There are two ways to define the range of cross-sections to be checked. Optimization from predefined
shapes works on a given number of cross-sections while Parametric optimization finds the optimal shape
within different geometry parameter ranges.
Optimization from This method finds the optimal cross-section from a given number of predefined shapes. Candidates can
predefined shapes be selected from model cross-sections and from the library. Candidates must have the same cross-section
type as the original cross-section of the group.
The range of candidates can be reduced by setting Constraints. Only cross-sections between the limits
for height and width will be used as candidates (other sections will appear greyed).
580
If a group contains more than one design member all members will be checked. Members are checked
along their entire length. Not all candidates will be checked. The program analyses only those necessary
for finding the global optimum.
Parametric This method finds the optimal cross-section within different geometry parameter ranges.
optimization Many different optimum search algorithms are known and used successfully for optimizing frame struc-
tures. Due to the nonlinearity of the problem and the large number of local optimums it is hard to find a
global optimum with pure mathematics. It is even harder if the optimization has to perform not only
strength checks but also stability analysis.
AxisVM uses the so-called Particle Swarm Optimization (PSO), a stochastic computational method for
finding optimum. It is an evolutionary algorithm developed in the 1990s.
The PSO process runs for a given number of iterations and due to its stochastic nature, it can find multiple
local optimums. The number of iterations is determined by the program trying to balance running time
and the fullest possible mapping of the search-space. Moreover, if the algorithm finds no changes in the
result after a long period it assumes that it is the global optimum and quits. In multi-threaded
mode the search space is partitioned among the threads.
The algorithm estimates the necessary number of iterations and selects the optimization method. It can
be either a simple linear search or a PSO optimization. The decision is based on the following parameters:
(1) size of the search space, (2) iteration step size, (3) number of fixed parameters, (4) the objective of the
optimization, (5) single- or multi-threaded mode (see Settings / Preferences / Analysis) (6) range of beams
taken into account (all or x% of the most efficient beams). Reducing the number of beams taken into
account make the calculation faster but reduces the precision. This is useful if optimization groups consist
of many members with different utilization and the utilization comes from a check which is also per-
formed in the optimization.
The calculation can start parallel search processes using all processor cores going through the search
space faster. A typical graph for the optimization process is like this:
The blue line shows the decreasing population diversity on a reversed scale, the filled graph displays the
fitness value in the percentage of the initial (the lower the better). If fitness value does not change through
many iterations and the population diversity is small we can accept the current result by jumping to the
optimization of the next group or pressing Esc to quit optimization.
Skip to the next group. Stops the optimization of the current group and jumps to the next one.
The current status is considered to be the result.
Stop the optimization. Stops the entire optimization. The current status of each group is considered to
be the result. Same as pressing the Esc key.
User’s Manual X6 Release1 e2 581
The advantage and drawback of this algorithm is its stochastic nature. Running the optimization multiple
times for the same problem can lead to slightly different results. This tendency is stronger in large search-
spaces. For small search spaces like pipes within a narrow size range a simple linear search is made (an-
alysing all candidates for finding the best one).
Too big intervals and/or too small increments makes the search space extremely large and as a result,
the calculation time increases and/or the convergence slows down. So, it is important to set the ranges
around the estimated optimum.
If an optimization group contains multiple design members, the overall utilization will be the highest
utilization of the members. Therefore, it is not recommended to place members with very different length
or internal forces into the same optimization group.
Optimization After setting the parameters go to the Optimization tab to run the optimization for the selected load
case, combination, envelope or critical combination.
The table displays the group parameters and the results of the optimization (weight per length unit,
weight reduction, width and height. The Opt. column can be used to control which group is to be opti-
mized.
If optimization was based on predefined shapes a dropdown list can be opened with all checked and
usable cross-sections sorted from higher utilization to lower.
The Replace column controls which cross-sections are to be replaced. Clicking on the Replace cross-sec-
tions button will actually replace the cross-sections in the selected groups.
EN 1993-1-2
Eurocode
Eurocode 3: Design of steel structures
Part 1-2: Structural fire design
EC DIN EN 1993-1-2:2010-12 NA
German
EC SZ EN 1993-1-2:2013 NA
Hungarian
EC SR EN 1993-1-2:2006/NB:2008
Romanian
EC STN EN 1993-1-2/NA:2008
Slovakian
EC PN EN 1993-1-2 NA
Polish
EC BS EN 1993-1-2:2005 NA
British
EC SFS-EN 1993-1-2:2005 NA
Finnish
Fire design of steel Steel beam fire design (SD8 module) is based on the ”ordinary” steel beam design (SD1 module;
structures according see 6.6.1 Steel beam design according to Eurocode 3 – SD1 module), so SD1 module is a prerequisite to
to EN 1993-1-2 use SD8 module.
EN 1993-1-2 (EC3-1-2) discusses fire design of steel structures. EC3-1-2 is very close to EN 1993-1-1, so
we note only the difference.
Steel beam fire design can be performed in AxisVM if 1) the selected load combination includes a fire
load case (4.10.1 Load cases, load groups) 2) if a fire effect has been assigned to any of the selected
elements (4.10.31 Fire effect on steel elements – SD8 module) in that fire load case. If different fire effects
have been assigned to elements of a design member, an error message appears and fire design cannot
be performed.
If the selection contains elements without fire effect the ordinary steel design will be performed on them.
Certain steel fire design parameters also have to be assigned to the elements (see below).
Member analysis EC3-1-2 discusses different methods: member analysis, analysis of a part of the structure and analysis of
the entire structure. Member analysis is the most widespread verification method due to its simplicity.
Analysis of a part or whole of the structure usually requires complex nonlinear numerical models and
nonlinear statical analysis.
Guiding principles of member analysis according to EC3-1-2:
• Design member internal forces can be calculated with linear static analysis;
• „Only the effects of thermal deformations resulting from thermal gradients across the cross-
section need to be considered. The effects of axial or in-plain thermal expansions may be ne-
glected.” (EC3-1-2, 2.4.2 (4));
• „The boundary conditions at supports and ends of member may be assumed to remain un-
changed throughout the fire exposure.” (EC3-1-2, 2.4.2 (5)).
SD8 performs member analysis. It has to be checked if that structure/structural part can be verified with
member analysis according to the guidelines of EC3-1-2.
In certain cases, analysis of a part of the structure can be also executed in AxisVM if 1) boundary condi-
tions are set carefully, 2) fire effects are taken into account as thermal load and 3) we consider the guide-
lines in section 2.4.3 of EC3-1-2.
User’s Manual X6 Release1 e2 583
ky,θ, kE, ky,θ,4 According to the EC3-1-2, the following reduction factors for mechanical properties of carbon steels are
considered in steel fire design in AxisVM:
• ky,θ – reduction factor for yield strength (Class 1, 2 and 3 cross-sections), relative to yield strength
at 20 °C;
• ky,θ,4 – reduction factor for yield strength (Class 4 cross-section), relative to yield strength at 20
°C;
• kE – reduction factor for Young’s modulus, relative to Young’s modulus at 20 °C.
It has to be noted that in case of the design of stainless-steel different reduction factors need to be used.
θs [°C] ky,θ ky,θ,4 kE
20 1.000 1.000 1.000
100 1.000 1.000 1.000
200 1.000 0.890 0.900
300 1.000 0.780 0.800
400 1.000 0.650 0.700
500 0.780 0.530 0.600
600 0.470 0.300 0.310
700 0.230 0.130 0.130
800 0.110 0.070 0.090
900 0.060 0.050 0.0675
1000 0.040 0.030 0.045
1100 0.020 0.020 0.0225
1200 0.000 0.000 0.000
Steel fire design For the fire design based on EC3-1-2, fire design parameters should be defined and assigned to the
parameters structural members. Cick the button displayed below in red rectangle.
This button is visible only if
1) SD8 module is part of the configuration and
2) steel fire design according to the current design code is supported.
The button adds an additional panel on the right side. If it remains hidden, fire design of the element will
not be performed.
584
Flexural buckling It is possible to define different buckling length or buckling length coefficient for the calculation of buck-
ling resistance about y and z axes in fire, as deflection of the element is often prevented by thin-walled
elements (purlins, etc.) that lose their stiffness very quickly on elevated temperature. The definition/cal-
culation of buckling length is the same as in case of normal temperature design (blue rectangle). If the
automatic buckling length calculation is selected, the fire design module uses the same buckling length
as in normal temperature design.
If Flexural buckling is checked, buckling resistance calculation and verification are performed on both
normal and elevated temperature.
Lateral-torsional 𝐾_𝜔: factor related to the constraints against warping can be different on elevated temperature.
buckling If Lateral-torsional buckling is checked, resistance calculation and verification are performed on both nor-
mal and elevated temperature.
Mcr – critical Different parameters can be defined for the consideration and calculation of critical bending moment.
bending moment The way of calculation of critical bending moment and load position are the same as in normal tempera-
ture design (blue rectangle).
If AutoMcr is selected, automatically calculated or user defined lateral supports are taken into account by
the calculation of Mcr.
Flexural buckling + Verification against flexural torsional buckling is performed if Flexural buckling and Lateral-torsional buck-
Lateral-torsional ling are checked and the stability parameters are defined. The verification method is explained below.
bucklings Equivalent uniform moment factors (βM,y, βM,z and βM,LT of EC3-1-2; Fig. 4.2) need to be defined. It is pos-
sible to calculate these factors automatically based on the bending moment diagram.
Check only EN 1993-1-1 and EN 1993-1-2 standards are similar, but EN 1993-1-2 does not discuss verification in
prescribed failure several failure modes (e.g., strength failure caused by internal forces in interaction with each other or
modes shear web buckling failure). According to the relevant literature, EN 1993-1-1 verification formulae can
be used in these cases but with reduced yield strength and Young’s modulus. Nevertheless, these verifi-
cations are not prescribed by the code.
If Check only prescribed failure modes is checked, only EC3-1-2 design checks are performed.
Classification of AxisVM classifies cross-sections according to Table 5.2 of EN 1993-1-1. In case of fire design, the ε pa-
cross-sections rameter is calculated according to EC3-1-2:
235
𝜀 = 0.85√
𝑓𝑦
Check only Axial force-Bending-Shear [N-M-V] (EN 1993-1-1, 6.2.1, 6.2.8, 6.2.9.3)
prescribed failure (EN 1993-1-2, 4.2.1-4.2.3, Annex E)
modes Compression-Bending-Buckling (flexural in plane or torsional) [N-M-Buck] (EN 1993-1-2, 4.2.3.5, Annex E)
Axial force-Bending-Lateral Tors. Buckling [N-M-LTBuckl] (EN 1993-1-2, 4.2.3.5, Annex E)
Shear /y [Vy] (EN 1993-1-2, 4.2.3.3, 4.2.3.4, Annex E)
Shear /z [Vz] (EN 1993-1-5, 5.1-5.3)
(EN 1993-1-2, 4.2.3.3, 4.2.3.4, Annex E)
Web shear-Bending-Axial force [Vw-M-N]
(EN 1993-1-1, 6.2.1, 6.2.8, 6.2.9)
(EN 1993-1-5, 5.1-5.3, 7.1)
(EN 1993-1-2, 4.2.1-4.2.3, Annex E)
User’s Manual X6 Release1 e2 585
Resistances
Plastic resistance (axial) [Npl,fi,Rd] (EN 1993-1-2, 4.2.3.1, Annex E)
Effective resistance (when subjected to uniform compression) [Neff, fi,Rd] (EN 1993-1-2, 4.2.3.1, Annex E)
Elastic Shear Resistance /y axis [Vel,fi,y,Rd]
(EN 1993-1-1, 6.2.6, EN 1993-1-2, 4.2.3.3, 4.2.3.4, Annex E)
Elastic Shear Resistance /z axis [Vel,fi,z,Rd]
(EN 1993-1-1, 6.2.6, EN 1993-1-2, 4.2.3.3, 4.2.3.4, Annex E)
Plastic Shear Resistance /y axis [Vpl,fi,y,Rd]
(EN 1993-1-1, 6.2.6, EN 1993-1-2, 4.2.3.3, 4.2.3.4, Annex E)
Plastic Shear Resistance /z axis [Vpl,fi,z,Rd]
(EN 1993-1-1, 6.2.6, EN 1993-1-2, 4.2.3.3, 4.2.3.4, Annex E)
Shear Web Buckling [Vb,fi,Rd] (EN 1993-1-5, 5.1-5.3, EN 1993-1-2, 4.2.3.3, 4.2.3.4, Annex E)
Axial force-Bending-
Shear
Check only Elastic design check (Class 3 cross-sections)
prescribed failure General formula:
modes
𝑁𝐸𝑑 𝑀𝑦,𝐸𝑑 𝑀𝑧,𝐸𝑑
+ + ≤1
𝑓𝑦 𝑘𝑦,𝜃 𝑓𝑦 𝑘𝑦,𝜃 𝑓𝑦 𝑘𝑦,𝜃
𝐴 𝑊𝑒𝑙,𝑦 𝑊𝑒𝑙,𝑧
𝛾𝑀𝑓𝑖 𝛾𝑀𝑓𝑖 𝛾𝑀𝑓𝑖
If shear force is greater than 50% of the shear resistance or torsional moment is present, the following
yield condition is checked (only for I, T, C, box and pie sections):
2 2 2
𝜎𝑥,𝐸𝑑 𝜎𝑧,𝐸𝑑 𝜎𝑥,𝐸𝑑 𝜎𝑧,𝐸𝑑 𝜏𝐸𝑑
( ) +( ) −( )( )+ 3( ) ≤1
𝑓𝑦 𝑘𝑦,𝜃 ⁄𝛾𝑀𝑓𝑖 𝑓𝑦 𝑘𝑦,𝜃 ⁄𝛾𝑀𝑓𝑖 𝑓𝑦 𝑘𝑦,𝜃 ⁄𝛾𝑀𝑓𝑖 𝑓𝑦 𝑘𝑦,𝜃 ⁄𝛾𝑀𝑓𝑖 𝑓𝑦 𝑘𝑦,𝜃 ⁄𝛾𝑀𝑓𝑖
∆𝑀𝑦 ,𝐸𝑑 = 𝑁𝐸𝑑 ∙ 𝑒𝑁 ,𝑦 : it differs from zero only when the cross section is in Class 4 and the original cross-
section is asymmetric to axis y.
Plastic design check
For I, pipe and box sections in section Class 1 and 2, the resistance check is performed according to EN
1993-1-1 6.2.10 and EC3-1-2. Allowance is made for the effect of both shear force and axial force on the
resistance moment. Beyond the resistance check of pure axial force and pure shear force, the following
criteria should also be satisfied:
𝑀𝑦,𝐸𝑑 𝑀𝑧,𝐸𝑑
≤ 1; ≤1
𝑀𝑁,𝑓𝑖,𝑦,𝑅𝑑 𝑀𝑁,𝑓𝑖,𝑧,𝑅𝑑
586
where MN,fi,y,Rd, MN,fi,z,Rd are reduced moment resistances based on the effect of shear force and axial force
(EN 1993-1-1 6.2.8. and 6.2.9.1). For pipe sections, the reduced moment is calculated as follows:
2
𝑛1.7 𝑁𝐸𝑑 𝑉𝐸𝑑
𝑀𝑁,𝑓𝑖,𝑦,𝑅𝑑 = 1.04 ⋅ (1 − 𝜌 − ) ; 𝑛 = ; 𝜌 = (2 − 1)
(1 − 𝜌)0.7 𝑁𝑝𝑙,𝑓𝑖,𝑅𝑑 𝑉𝑝𝑙,𝑓𝑖,𝑧,𝑅𝑑
2
𝑛1.7 𝑁𝐸𝑑 𝑉𝐸𝑑
𝑀𝑁,𝑓𝑖,𝑧,𝑅𝑑 = 1.04 ⋅ (1 − 𝜌 − ); 𝑛 = ; 𝜌 = (2 − 1)
(1 − 𝜌)0.7 𝑁𝑝𝑙,𝑓𝑖,𝑅𝑑 𝑉𝑝𝑙,𝑓𝑖,𝑦,𝑅𝑑
For bi-axial bending the criterion in EN 1993-1-1 6.2.9.1. (6) should be satisfied (I, box and pipe shaped
cross-sections):
𝛼 𝛽
𝑀𝑦,𝐸𝑑 𝑀𝑧,𝐸𝑑
[ ] +[ ] ≤1
𝑀𝑁,𝑓𝑖,𝑦,𝑅𝑑 𝑀𝑁,𝑓𝑖,𝑧,𝑅𝑑
Other cross-section types:
𝐴𝑓𝑦 𝑘𝑦,𝜃 𝑊𝑝𝑙,𝑦 𝑓𝑦 𝑘𝑦,𝜃 𝑊𝑝𝑙,𝑧 𝑓𝑦 𝑘𝑦,𝜃
𝑁𝑝𝑙,𝑓𝑖,𝑅𝑑 = ; 𝑀𝑝𝑙,𝑓𝑖,𝑦,𝑅𝑑 = ; 𝑀𝑝𝑙,𝑓𝑖,𝑧,𝑅𝑑 =
𝛾𝑀𝑓𝑖 𝛾𝑀𝑓𝑖 𝛾𝑀𝑓𝑖
𝑁𝐸𝑑 𝑀𝑦,𝐸𝑑 𝑀𝑧,𝐸𝑑
+ + ≤1
𝑁𝑝𝑙,𝑓𝑖,𝑅𝑑 𝑀𝑝𝑙,𝑓𝑖,𝑦,𝑅𝑑 𝑀𝑝𝑙,𝑓𝑖,𝑧,𝑅𝑑
If the shear force is greater than 50% of the shear resistance, the bending moment resistance is reduced
considering the effect of shear force according to EN 1993-1-1 6.2.8. and EC3-1-2 (MV,fi,y,Rd, MV,fi,z,Rd).
Supported sections:
Unsupported cross-section types or plastic design:
𝑉𝑦,𝐸𝑑 𝑓𝑦 𝑘𝑦,𝜃
≤ 1; 𝑉𝑝𝑙,𝑓𝑖,𝑦,𝑅𝑑 = 𝐴𝑣
𝑉𝑝𝑙,𝑓𝑖,𝑦,𝑅𝑑 √3𝛾𝑀 𝑓𝑖
588
Shear /z [Vz]
Check only AxisVM performs the following design check:
prescribed failure
𝑉𝑧𝐸𝑑
modes ≤1
min(𝑉𝑧,𝑓𝑖,𝑅𝑑 , 𝑉𝑏,𝑓𝑖,𝑅𝑑 )
𝑉𝑏,𝑓𝑖,𝑅𝑑 shear web buckling resistance (for I and box cross-sections) is based on the shear resistance of
webs. The calculation of 𝑉𝑧,𝑓𝑖,𝑅𝑑 is shown below.
No guidelines may be found in EC3-1-2 about verification against shear web buckling in fire. In this
case, the verification is performed according to sections 5.1-5.3 of EN 1993-1-5 with reduced yield
strength and Young’s modulus due to the elevated steel temperature (EC3-1-2). Furthermore, research
results of André Reis, Nuno Lopes, Paulo Vila Real are also considered by this verification (A. Reis, N.
Lopes and P. Vila Real: Numerical study of steel plate girders under shear loading at elevated tempera-
tures, Journal of Constructional Steel Research, 117 (2006) 1-12).
Class 1, 2 and 3 cross-sections:
𝑓𝑦 𝑘𝑦,𝜃
𝑉𝑏,𝑓𝑖,𝑅𝑑 = 𝜒𝑤,𝜃 ℎ𝑤 𝑡𝑤
√3𝛾𝑀𝑓𝑖
Class 4 cross-sections:
𝑓𝑦 𝑘𝑦,𝜃,4
𝑉𝑏,𝑓𝑖,𝑅𝑑 = 𝜒𝑤,𝜃 ℎ𝑤 𝑡𝑤
√3𝛾𝑀𝑓𝑖
where
0.61
𝜒𝑤,𝜃 = 1.2 𝜆̅𝑤,𝜃 <
1.2
0.55 0.61
𝜒𝑤,𝜃 = 0.1 + 𝜆̅𝑤,𝜃 ≥
̅
𝜆𝑤,𝜃 1.2
ℎ𝑤 235
𝜆̅𝑤,𝜃 = ; 𝜀 = 0.85√
37.4𝑡𝑤 𝜀√𝑘𝜏 𝑓𝑦
Check only pre- According to EN 1993-1-1, 6.2.6, EN 1993-1-2, 4.2.3.3, 4.2.3.4, Annex E:
scribed failure
modes Elastic design check (class 3 cross-sections)
𝑉𝑧,𝐸𝑑 𝐼𝑦 ∙ 𝑡 𝑓𝑦 𝑘𝑦,𝜃
≤ 1; 𝑉𝑒𝑙,𝑓𝑖,𝑧,𝑅𝑑 =
𝑉𝑒𝑙,𝑓𝑖,𝑧,𝑅𝑑 𝑆𝑦 √3𝛾𝑀
𝑓𝑖
Supported sections:
Unsupported cross-section types or plastic design:
𝑉𝑧,𝐸𝑑 𝑓𝑦 𝑘𝑦,𝜃
≤ 1; 𝑉𝑝𝑙,𝑓𝑖,𝑧,𝑅𝑑 = 𝐴𝑣
𝑉𝑝𝑙,𝑓𝑖,𝑧,𝑅𝑑 √3𝛾𝑀 𝑓𝑖
Web shear-Bending- In case of I and box cross-sections, this interaction design check is also performed according to section
Axial force 7.1 of EN 1993-1-5 and sections 6.2.8, 6.2.9 of EN 1993-1-1 with reduced yield strength and Young’s
modulus due to the elevated steel temperature (EC3-1-2).
Check only
2
prescribed failure 𝑀𝐸𝑑 𝑀𝑓,𝑓𝑖,𝑅𝑑 𝑉𝐸𝑑
modes + (1 − ) ⋅ (2 − 1) ≤ 1
𝑀𝑝𝑙,𝑓𝑖,𝑅𝑑 𝑀𝑝𝑙,𝑓𝑖,𝑅𝑑 𝑉𝑏,𝑓𝑖,,𝑅𝑑
it means that the utilization is high and the safety of the element against failure cannot be confirmed
even at room temperature in accidental design situation based on the regulations of EC3-1-2. In this
case, it is very likely that the resistance is not adequate in persistent design situation neither. If AxisVM
displays 1001 °C as critical temperature in case of an element, it means that the utilization is very low
and the critical temperature is above 1000 °C.
Diagrams Click on the steel design member to display the diagrams all individual checks and their envelope. Re-
sults for any position of any steel design member in any load case or combination can be obtained by
setting x or dragging the tracking line. If a check cannot be performed with the current cross-section
the respective diagram is replaced by a cross.
If the selected or the critical combination contains fire load case and fire design was performed for the
selected steel design member, buckling coefficients in fire and critical temperature are shown.
Design calculations Click on Design calculations button to see the design details. The calculation can be saved into the
documentation. In case of fire design, the type of fire curve, the design temperature and the critical
temperature are also shown.
590
Result table Among steel design result tables, Utilization in fire summarizes the fire design results in order to help
the communication with fire safety engineers.
Besides the basic initial parameters, the following parameters are listed: required time of fire resistance
(e.g., R30), type of selected fire curve, design steel temperature, critical failure mode, utilization and
critical temperature.
Cross-section opti- If SD9 module is installed, steel cross-section optimization can be performed for load combinations that
mization contain fire load case with respect to steel fire design rules. During the optimization, design steel tem-
perature is not recalculated automatically. For this reason, the utilization may be slightly different after
analysis of the optimized structural configuration.
See 6.6.1.4 Steel cross-section optimization - SD9 module
From AxisVM X4 version this module, earlier called SD2, is built-in into the basic packages. AxisVM
calculates the moment-curvature diagram, the resistance moment and initial strength of steel column-
beam bolted joints based on Eurocode3 (Part 1.8 Design of Joints).
Type of joints Beam to column or beam to beam joints.
Assumptions:
- The beam and column cross-sections are rolled or welded I shapes.
- The beam end plate connect to the flange of the column.
- The pitch range of the beam is beetwen ± 30°.
- The cross-section class should be 1, 2 or 3.
- The normal force in the beam should be less than 0.05* Npl,Rd
The program checks if these requirements are met.
The steps of the Select the beam and one of its end nodes.
design (We can select several beams in one process if the selected beams have the same material and cross-
section properties and connected columns also have the same material and cross-section properties.)
User’s Manual X6 Release1 e2 591
This dialog lets you assign the parameters of the joint in three steps.
Bracings We can assign horizontal, diagonal bracing plates and web thickening plates to increase the strength
of the connection.
Horizontal bracings
Diagonal bracing
End plate
Bolt rows can be assigned to the tensile part of the end plate.
Bolts
User’s Manual X6 Release1 e2 593
The program places bolts in two columns symmetrical to the beam web. The same type of bolts is used
in the connection.
Bolt parameters: Size, Quality, Material safety factor, Number of bolt rows, distance of bolt columns (d)
In case of automatic positioning of bolts the program places bolt rows in equal distances.
The program checks the required minimal distances between bolts and from the edge of plates.
Turn off the option Use default positions to place the bolt rows individually.
An error message will appear if the distances does not meet the requirements.
Minimal bolt distances are checked based on EC2: 2.2 d between bolts, 1.2 d from edge of plate and in
the direction perpendicular to the force.
Results When we click on the Results tab AxisVM calculates the Moment-curvature diagram, the design resistant
moment (𝑀𝑟𝐷 ) and the initial strength of the connection (𝑆𝑗,𝑖𝑛𝑖𝑡 ).
A warning message will appear if the resistant moment is less than the design moment. The
calculation method considers shear forces and normal forces together with the moments. As a
consequence we can get different resistant moments (MrD) for the same connection depending
on the load cases (or combinations). Therefore AxisVM checks the MrD-MsD condition in all load
cases.
594
Icon Bar
Save the connection parameters. Saved parameters can be loaded and assigned to other beam-end
joints later.
List of the existing joints
Additional parameters (𝑓𝑠𝑒 Coefficient for seismic forces, see 4.10.24 Seismic loads – SE1 module)
Verification of steel The SC1 module is to verify bolted steel connections by checking if they meet the require-
connections ments for structural joints specified in the design code. The module
The module uses the “component method” defined and supported by the relevant design
codes for the verification.
Design codes Steel connection verification is supported according to
EN 1993-1-1 (2005): Eurocode 3: Design of steel structures - Part 1-1: General rules and rules for buildings
EN 1993-1-8 (2012): Eurocode 3: Design of steel structures - Part 1-8: Design of joints
The range of supported steel joint types, restrictions and instructions to define joints are included in a
PDF document: Steel connections – SC1 application (sc_manualX6_en.pdf) available on the installation
pendrive.
User’s Manual X6 Release1 e2 595
Rendered view Parameters of joints are saved into the AxisVM model. Joint details are displayed in the rendered view.
Further information Further details are available in the SC1 manual. Download >>
596
EUROCODE 5 The timber beam design module can be applied to the following cross-sections and timber materials:
(EN 1995-1-
1:2004)
a) Rectangle for solid timber, Glued laminated timber (Glulam) and for Laminated veneer lumber (LVL)
b) Round for solid timber
Material properties The material database contains the solid, Glulam and LVL timber material properties according to the
related EN standard. (Solid timber: EN338, Glulam: EN 1194)
Density Notation
Characteristic value of density ρk
Mean value of density ρmean
Timber classes Timber elements must have a service class. Service class can be set in the line elements
definition dialog, at Service Class field. See 4.9.9 Line elements
Service classes (EN 1995-1-1, 2.3.1.3):
Service class 1 – where the average moisture content in most softwoods will not exceed 12%. This cor-
responds to a temperature of 20°C and a relative humidity of the surrounding air only exceeding 65%
for a few weeks per year.
Service class 2 – where the average moisture content in most softwoods will not exceed 20%. This cor-
responds to a temperature of 20°C and a relative humidity of the surrounding air only exceeding 85%
for a few weeks per year.
User’s Manual X6 Release1 e2 597
Service class 3 – where the average moisture content in most softwoods exceeds 20%.
Design strength and other design properties of the timber materials depend on the service class.
Load-duration clas- Timber design module requires information on the load duration. So, if a timber material has
ses been defined in the model load case duration class can be entered.
See 4.10.1 Load cases, load groups
Design strength The design values of strength are calculated from the characteristic values of strength according to the
components following formulas:
In case of 𝑓𝑡,90,𝑑 , 𝑓𝑐,0,𝑑 , 𝑓𝑐,90,𝑑 , 𝑓𝑣,𝑑 (Solid, Glulam, LVL timbers): 𝑓𝑑 = 𝑘𝑚𝑜𝑑 𝑓𝑘 ⁄𝛾𝑀
In case of 𝑓𝑚,𝑑 (Solid, Glulam, LVL timbers): 𝑓𝑑 = 𝑘𝑚𝑜𝑑 𝑘ℎ 𝑓𝑘 ⁄𝛾𝑀
In case of 𝑓𝑡,0,𝑑 (Solid and Glulam timbers): 𝑓𝑑 = 𝑘𝑚𝑜𝑑 𝑘ℎ 𝑓𝑘 /𝛾𝑀
In case of 𝑓𝑡,0,𝑑 (LVL timber): 𝑓𝑑 = 𝑘𝑚𝑜𝑑 𝑘𝑙 𝑓𝑘 /𝛾𝑀
where
𝑘𝑚𝑜𝑑 is a modification factor (EN 1995-1-1, 3.1.3)
𝑘ℎ is the depth factor (EN 1995-1-1, 3.2, 3.3, 3.4)
𝑘𝑙 is the length factor for LVL timber (EN 1995-1-1, 3.4)
𝑓𝑘 is the characteristic strength
𝛾𝑀 is the partial factor of material (EN 1995-1-1, Table 2.3)
The values of kmod factor can be set in the Design codes window under the Timber tab (see 3.3.7
Design codes).
kh factor The 𝑓𝑚,𝑘 and 𝑓𝑡,0,𝑘 characteristic strength values are determined for a reference depth of beam. In case
of solid and Glulam timber if the depth (h) of the cross-section less than the reference value, the design
strength is multiplied with the following factor.
150 0.2
Solid timber: 𝑘ℎ = min {( ) ; 1.3} (if 𝜌𝑘 ≤ 700 kg/m3)
ℎ
600 0.1
Glulam: 𝑘ℎ = min {( ) ; 1.1}
ℎ
In case of LVL timber if the depth (h) of the cross-section not equal to the reference value, the design
strength is multiplied with the following factor.
300 𝑠
LVL: 𝑘ℎ = min {( ) ; 1.2}
ℎ
(where s is the size effect exponent, h is the cross-section depth in mm.
Reference depths for solid timber is 150 mm, for Glulam it is 600 mm, for LVL it is 300 mm.
kl factor The 𝑓𝑡,0,𝑘 characteristic strength value of LVL timber is determined for a reference length of beam. If the
length of the beam is not equal to the reference length, the design strength is multiplied by the follow-
ing factor.
3000 𝑠/2
𝑘𝑙 = min {( ) ; 1.1}
𝑙
(where s is the size effect exponent).
𝑙 is the beam length in mm. Reference length: 3000 mm.
Stiffness values for Analysis type Modulus (SLS) Modulus (ULS)
analysis First order, 𝐸𝑚𝑒𝑎𝑛 𝐸𝑚𝑒𝑎𝑛
𝐸𝑚𝑒𝑎𝑛,𝑓𝑖𝑛 = 𝐸𝑚𝑒𝑎𝑛,𝑓𝑖𝑛 =
linear elastic 1 + 𝑘𝑑𝑒𝑓 1 + 𝜓2 𝑘𝑑𝑒𝑓
𝐺𝑚𝑒𝑎𝑛 𝐺𝑚𝑒𝑎𝑛
𝐺𝑚𝑒𝑎𝑛,𝑓𝑖𝑛 = 𝐺𝑚𝑒𝑎𝑛,𝑓𝑖𝑛 =
1 + 𝑘𝑑𝑒𝑓 1 + 𝜓2 𝑘𝑑𝑒𝑓
Second order, 𝐸𝑑 = 𝐸𝑚𝑒𝑎𝑛 /𝛾𝑀 𝐸𝑑 = 𝐸𝑚𝑒𝑎𝑛 /𝛾𝑀
linear elastic
𝐺𝑑 = 𝐺𝑚𝑒𝑎𝑛 /𝛾𝑀 𝐺𝑑 = 𝐺𝑚𝑒𝑎𝑛 /𝛾𝑀
Frequency 𝐸𝑚𝑒𝑎𝑛 , 𝐺𝑚𝑒𝑎𝑛 𝐸𝑚𝑒𝑎𝑛 , 𝐺𝑚𝑒𝑎𝑛
• In case of Glulam the laminates are parallel to the y axis of the cross-section.
• in case of LVL the laminates are parallel to the z axis of the cross-section.
AxisVM performs the following checks only. All the other checks specified in the design code like
supports, joints, etc. has to be completed by the user.
Normal force-Bend- The design value of normal force can be tension or compression.
ing Tension and moment (EN 1995-1-1, 6.2.3)
𝜎𝑡,0,𝑑 𝜎𝑚,𝑦,𝑑 𝜎𝑚,𝑧,𝑑
+ + 𝑘𝑚 ≤1
𝑓𝑡,0,𝑑 𝑓𝑚,𝑦,𝑑 𝑓𝑚,𝑧,𝑑
𝜎𝑡,0,𝑑 𝜎𝑚,𝑦,𝑑 𝜎𝑚,𝑧,𝑑
+ 𝑘𝑚 + ≤1
𝑓𝑡,0,𝑑 𝑓𝑚,𝑦,𝑑 𝑓𝑚,𝑧,𝑑
Normal force-Bend- For lateral torsional buckling check the program assumptions that the beam is bending in z-x plane
ing- (about y axis). If there is simultaneous 𝑀𝑧 moment on the beam and the compression stress from 𝑀𝑧
LT buckling reach the 3% of the 𝑓𝑐,0,𝑑 a warning message appears.
Bending only (EN 1995-1-1, 6.3.3)
𝜎𝑚,𝑑
≤1
𝑘𝑐𝑟𝑖𝑡 𝑓𝑚,𝑑
Shear-Torsion There is no rule in EC5 for case of simultaneous shear force and torsional moment.
In this case the program uses the interaction formula according to DIN EN 1995-1-1/NA:2010.
Shear(y) , Shear(z) and torsion
2 2
𝜏𝑣,𝑦,𝑑 𝜏𝑣,𝑧,𝑑 𝜏𝑡𝑜𝑟,𝑑 𝜏𝑣,𝑦,𝑑 𝜏𝑣,𝑧,𝑑
max [ ; ; +( ) +( ) ]≤1
𝑓𝑣,𝑑 𝑓𝑣,𝑑 𝑘𝑠ℎ𝑎𝑝𝑒 𝑓𝑣,𝑑 𝑓𝑣,𝑑 𝑓𝑣,𝑑
where,
𝑘𝑠ℎ𝑎𝑝𝑒 is a factor for the shape of cross-section.
For round cross-section 𝑘𝑠ℎ𝑎𝑝𝑒 = 1.2,
for rectangular cross-section 𝑘𝑠ℎ𝑎𝑝𝑒 = min{1 + 0.15 ⋅ ℎ⁄𝑏 ; 2.0}
Moment-Shear In case of curved beams, the program checks the tensile stress perpendicular to the grain from My and
Vz forces. (EN 1995-1-1, 6.4.3.)
Moment(y)-Shear(z)
𝜏𝑑 𝜎𝑡,90,𝑑
+ ≤1
𝑓𝑣,𝑑 𝑘𝑑𝑖𝑠 𝑘𝑣𝑜𝑙 𝑓𝑡,90,𝑑
𝑘𝑑𝑖𝑠 is a factor which takes into account the effect of the stress distribution in the apex zone (𝑘𝑑𝑖𝑠 =
1.4 for curved beams)
𝑘𝑣𝑜𝑙 is a volume factor 𝑘𝑣𝑜𝑙 = [𝑉0 ⁄𝑉]0.2
600
Design Parameters For the design based on Eurocode 5, the following design parameters should be defined and assigned
to the design members:
Layer thickness In case of Glued laminated timber (Glulam) arcs thickness of one layer has to be defined.
Grain direction Set of grain direction in case of tapered beam. The grain direction can be parallel with the top edge or
with the bottom edge. The top edge lays in the +z direction of the cross-section.
Stability parame-
ters
Buckling Ky, Kz: buckling length factors corresponding to the y and z axis, respectively.
𝐾𝑦 = 𝑙𝑒𝑓,𝑦 ⁄𝑙 ; 𝐾𝑧 = 𝑙𝑒𝑓,𝑧 /𝑙
where 𝑙 is the member length
𝑙𝑒𝑓,𝑦 is the buckling length in x-z plane of the member
𝑙𝑒𝑓,𝑧 is the buckling length in x-y plane of the member
Lateral torsional KLT: lateral buckling length factors corresponding to the z axis.
buckling 𝐾𝐿𝑇 = 𝑙𝑒𝑓 /𝑙
where 𝑙 is the member length
𝑙𝑒𝑓 is the lateral buckling length of the member corresponding to the z axis.
When the load not applied to the centre of gravity, the program modifies the lateral buckling length
according to the following:
- if the load is applied to the compression edge of the member the 𝑙𝑒𝑓 is increased by 2ℎ
- if the load is applied to the tension edge of the member the 𝑙𝑒𝑓 is decreased by 0.5ℎ
pz 0.9
Fz
0.8
Fz Fz
0.96
¼ ½ ¼
pz
0.42
Fz
0.64
M M
1.0
M ½M
0.76
M 0
0.53
M ½M
0.37
M M
0.36
pz
0.5
Fz 0.8
Design members The design is performed on design members that can consist of one or more finite elements (beams
and/or ribs). A group of finite elements can become a design member only if the finite elements in the
group satisfy some requirements checked by the program: to be located on the same straight line or
arc, to have the same material, cross-section and to have joining local coordinate systems.
The program allows two methods to define design members as follows:
602
SLS The parameters for the serviceability limit state (SLS) checks can be set on the SLS tab in the Design
(serviceability limit parameters window. The program checks the deflections of beams.
state)
The program checks the final deflections (ufin). When determining the deformations, the effect
of creep is considered throughout the whole loading (in each SLS load combination) by using the
kdef parameter of each beam.
If the critical load combinations are generated automatically, the program performs the SLS
checks in the quasi-permanent load combinations.
Deflections can be checked according to the z or y local axis directions of the timber design member.
The check in each direction can be activated or deactivated by checking the box in front of the limit
values. The deflection limit values shall be set according to the length (L) detailed below.
User’s Manual X6 Release1 e2 603
Deflections can be considered in several ways: based on the actual displacements or based on the dis-
placements relative only to the left, only to the right or to both end points of the design member.
Length (L) Deflections of a design member and deflection limits (e.g., L/300) are based on one of the following
length values:
It is important to note, that the pre-camber value set here has an effect in the SLS checks only,
reducing the previously calculated deflection based on the original shape.
Result diagrams By clicking on a design member the program displays the diagrams corresponding to all the checks.
Design calculations Clicking on the Design calculations button a report of the calculation details can be displayed. All strength
and stability checks appear as formulas completed with substituted actual values and design code refer-
ences.
604
Clicking on the Settings icon beside the Design calculations button allows setting the basic units for force
and length used in the design calculations. Important results also appear converted to standard AxisVM
units (see 3.3.8 Units and Formats). For further details see 6.6.1 Steel beam design according to Eurocode
3 – SD1 module
EC DIN EN 1995-1-2:2010-12 NA
German
EC MSZ EN 1995-1-2:2013 NA
Hungarian
EC SR EN 1995-1-2:2004/Ac:2006
Romanian
EC STN EN 1995-1-2:2008/NA:2011
Slovakian
EC BS EN 1995-1-2:2004 NA
British
EC NEN EN 1995-1-2/NB:2016
Dutch
EC SFS-EN 1995-1-2:2004 NA
Finnish
Fire design of Timber beam fire design (TD8 module) is based on methods of general timber beam design (TD1 module;
timber structures see 6.7.1 Timber beam design – TD1 module), thus TD1 module is a prerequisite to use TD8.
according to EN EN 1995-1-2 (EC5-1-2) discusses fire design of timber structures. This section describes the differences
1995-1-2 between the general design and fire design.
Timber beam fire design can be performed in AxisVM if
1) the selected load combination includes a fire load case (see 4.10.1 Load cases, load groups)
2) if a fire effect has been assigned to any of the selected elements (see 4.10.32 Fire effect on timber
elements – TD8 module) in that fire load case.
If different fire effects have been assigned to elements of a design member fire design cannot be per-
formed and an error message appears.
For selected elements without fire effect the general timber design will be performed. Timber fire design
parameters also has to be assigned to the elements (see below).
Member analysis EC5-1-2 discusses different methods: member analysis, analysis of a part of the structure and analysis of
the entire structure. Member analysis is the most widespread verification method due to its simplicity.
Analysis of a part or whole of the structure usually requires complex nonlinear numerical models and
nonlinear statical analysis.
Guiding principles of member analysis according to EC5-1-2:
• Design member internal forces can be calculated with linear static analysis;
• The boundary conditions at supports may be assumed to be constant with time. (EC5-1-2, 2.4.2
(4)).
TD8 module performs member analysis. It has to be checked whether the structure/structural part can be
verified with member analysis according to the guidelines of EC5-1-2.
Timber fire design Fire design based on EC5-1-2 requires fire design parameters defined and assigned to structural
parameters members. This button is available only if 1) TD8 module is part of the configuration and 2) the current
User’s Manual X6 Release1 e2 605
design code supports timber fire design. The button adds an additional panel on the right side. If it
remains hidden, fire design of the element will not be performed.
Buckling coeffi- It is possible to define different buckling length coefficients in fire because deflection of the element is
cients in fire often prevented by other elements that may lose their stiffness or resistance in shorter period of time.
The definition/calculation of buckling length is the same as in case of normal temperature design.
Reduced cross-sec- EC5-1-2 provides two different methodology to calculate the mechanical resistance in fire:
tion method
• reduced cross-section method,
• reduced properties method.
In the software, the reduced cross-section method is implemented. As a first step, the so-called effective
charring depth is calculated on the following way using the notional charring depth (𝑑𝑐ℎ𝑎𝑟,𝑛 ; see 4.10.32
Fire effect on timber elements – TD8 module)
𝑑𝑒𝑓 = 𝑑𝑐ℎ𝑎𝑟,𝑛 + 𝑘0 𝑑0
where 𝑑0 = 7𝑚𝑚, 𝑘0 is a coefficient
The cross-section is reduced with this effective charring depth on the exposed sides. This reduced cross-
section is called effective cross-section.
Design checks in The TD8 module, similarly to the TD1 module, performs the following design checks:
fire
Axial force-Bending [N-M] (EN 1995-1-1, 6.2.3, 6.2.4)
Compression-Bending-Buckling (in plane) [N-M-Buckling] (EN 1995-1-1, 6.3.2)
Axial force-Bending-Lateral tors. buckling [N-M-LT buckling] (EN 1995-1-1, 6.3.3)
Shear (y) - Shear(z) - Torsion (x) [Vy- Vz-Tx] (EN 1995-1-1, 6.1.7, 6.1.8)
Moment /y - Shear /z (tensile stress perp. to the grain) [My-V z] (EN 1995-1-1, 6.4.3)
The details of the calculation are presented in the previous section (see 6.7.1 Timber beam design – TD1
module), thus we note only the differences in this section:
1. different strength needs to be considered,
2. different stability parameters are considered.
Calculation of According to EC5-1-2, the timber strength is calculated on the following way:
strength of timber
𝑓20
𝑓𝑑,𝑓𝑖 = 𝑘𝑚𝑜𝑑,𝑓𝑖
𝛾𝑀,𝑓𝑖
𝑓20 = 𝑘𝑓𝑖 𝑓𝑘
where
𝑓20 is the 20% fractile strength at normal temperature
𝑘𝑚𝑜𝑑,𝑓𝑖 = 1 is the modification factor for fire
𝑘𝑓𝑖 is a coefficient, 𝛾𝑀,𝑓𝑖 is a partial factor
606
Stability checks Buckling length (Ky,fi and Kz,fi) and lateral torsional buckling length (KLT,fi) coefficient given in Design
parameters window are used to evaluate the flexural buckling and lateral torsional buckling resistances.
Result diagrams Click on a timber design member to display diagrams for all individual checks and their envelope. Results
for any position of any timber design member in any load case or combination can be obtained by setting
x or dragging the tracking line. If design check cannot be performed with the current cross-section, the
diagram is replaced by a cross.
If the selected or the critical combination contains fire load case and fire design was performed for the
selected design member, buckling coefficients in fire and charring depth are also shown.
Design calculations Click on Design calculations button to see the design details. The calculation can be saved into the doc-
umentation. In case of fire design, the type of fire curve and the calculated effective charring depth are
also shown.
User’s Manual X6 Release1 e2 607
Result table Among timber design result tables, Utilization in fire summarizes the fire design results in order to help
the communication with fire safety engineers.
Cross-section opti- If TD9 module is present, timber cross-section optimization can be performed for load combinations that
mization contain fire load case with respect to fire design rules.
Design codes There is no currently valid, overall design regulation for XLAM (CLT) panels. For further details on the
detailed calculation procedure and the references see the “XLAM Theory and Design Guide”.
Material properties There is a dedicated XLAM material in the material library with predefined stiffness and strength proper-
ties. If the user wishes, the material properties can be arbitrarily modified in the material library associated
with the model.
*it is assumed that GR,mean / Gmean = 0,1 holds, independently of the strength class
Density Denotement
Apparent density ρk
Mean value of the density ρmean
Service classes See in chapter 6.7.1 Timber beam design – TD1 module
Load duration See in chapter 6.7.1 Timber beam design – TD1 module
classes
Characteristic value The stress values are to be determined independently due to bending, normal and shear actions.
of strength
𝜎𝑚𝑥,𝑡 – relevant value of the normal stress in x direction due to bending, on the upper half of the re-
gion (the side of te region associated with the positive direction of the local z axis).
𝜎𝑚𝑥,𝑏 – relevant value of the normal stress in x direction due to bending, on the lower half of the region
(the side of the region associated with the negative direction of the local z axis).
𝜎𝑚𝑦,𝑡 – relevant value of the normal stress in y direction due to bending, on the upper half of the re-
gion (the side of the region associated with the positive direction of the local z axis).
𝜎𝑚𝑦,𝑏 – relevant value of the normal stress in y direction due to bending, on the lower half of the region
(the side of te region associated with the negative direction of the local z axis).
𝜎𝑛𝑥 – relevant value of the normal stress in x direction due to normal forces.
𝜎𝑛𝑦 – relevant value of the normal stress in y direction due to normal forces.
𝜏𝑥𝑧,𝑚𝑎𝑥 – relevant shear stress acting in direction z on the plane whose normal is in direction x.
𝜏𝑦𝑧,𝑚𝑎𝑥 – relevant shear stress acting in direction z on the plane whose normal is in direction y.
𝜏𝑟𝑥,𝑚𝑎𝑥 – relevant rolling shear stress acting in direction z on the plane whose normal is in direction x.
𝜏𝑟𝑦,𝑚𝑎𝑥 – relevant rolling shear stress acting in direction z on the plane whose normal is in direction y.
User’s Manual X6 Release1 e2 609
When clicking on a node, one can see the distribution of each calculated stress component through
the thickness:
Design value of The design values of strength can be calculated according to the formula:
strength 𝑓𝑐0𝑑 , 𝑓𝑐90𝑑 , 𝑓𝑣𝑑 , 𝑓𝑟𝑑 : 𝑓𝑑 = 𝑘𝑚𝑜𝑑 𝑓𝑘 ⁄𝛾𝑀
𝑓𝑚90𝑑 , 𝑓𝑡0𝑑 , 𝑓𝑡90𝑑 : 𝑓𝑑 = 𝑘𝑠𝑦𝑠 𝑘𝑚𝑜𝑑 𝑓𝑘 /𝛾𝑀
𝑓𝑚0𝑑 : 𝑓𝑑 = 𝑘𝑠𝑦𝑠 𝑘𝑚𝑜𝑑 𝑘𝑓𝑖𝑛 𝑓𝑘 /𝛾𝑀
where,
𝑘𝑚𝑜𝑑 modification factor for duration of load and moisture content (EN 1995-1-1, 3.1.3)
𝑘𝑠𝑦𝑠 system strength factor
𝑘𝑓𝑖𝑛 bending strength reduction factor
𝑓𝑘 charasteristic value of strength
𝛾𝑀 partial safety factor of the material (EN 1995-1-1, Table 2.3)
ksys factor 𝑘𝑠𝑦𝑠 = min {1 + 0.025 ⋅ 𝑛; 1.2} (𝑛 denotes the number of layers, but 𝑛 > 1)
In case of compression
2
𝜎𝑐,0,𝑑 |𝜎𝑚,0,𝑑 |
( ) + ≤1
𝑓𝑐,0,𝑑 𝑓𝑚,0,𝑑
610
Shear – Torsion 2 2
|𝜏 𝑇,𝑑 | 𝜏𝑑 𝜏𝑟,𝑑
(V-T) +( ) +( ) ≤1
𝑓𝑇,𝑑 𝑓𝑣,𝑑 𝑓𝑟,𝑑
When clicking on a node, one can see the distribution of the calculated utilization component through
the thickness:
User’s Manual X6 Release1 e2 611
MD1 module allows performing the typical analysis of unreinforced masonry walls and to specify their
design parameters required for the design check. An overall and complex verification process is imple-
mented for masonry walls subjected to mainly vertical and in-plane shear loading.
It provides a solution for the global analysis of one or multi-storey masonry walls with connecting joints.
Joints usually has considerable influence on the load-bearing capacity of the walls.
The verification is available for Eurocode, NTC and SIA standards.
General principles The following types of analysis are available in the module:
A) Stability analysis of walls subjected to mainly vertical loads. This check takes into account only
the eccentricities perpendicular to the plane of the wall, irrespective of the width of the wall
strip. Two-directional eccentricity cannot be analysed.
B) Complex stability check handling two-directional normal force eccentricity coupled with in-plane
shear check.
The A) and B) checks use different user-defined design sections (see 6.9.1.3). Check type B is only
performed if shear check is requested in the Masonry wall strip parameters dialog (see 6.9.1.2 Pa-
rameters of a masonry wall strip)
Only unreinforced masonry walls can be checked assuming that the wall has constant wall thickness along
a storey, only single leaf wall can be specified.
The masonry walls can be checked according to EN 1996-1-1 (chapter 6.1), NTC (chapter 4.5.6.2) and SIA
266 (chapter 4.3.1 and 4.4.) considering the following assumptions and application limits.
The above design codes assume that the walls are horizontally supported by the slabs, so standalone or
cantilevered walls cannot be analysed.
Various parameters can be added to the model (see Joint settings) which can result additional bending
moments on the structural model (multi-storey walls with various wall thickness, joint settings, the effect
of eccentric slab reaction force).
These effects are not negligible in the calculation of the load-bearing capacity of the elements, but accu-
rate modelling may require significantly complicated models in some cases. To facilitate the design check,
two approximative models are provided (see Calculation of additional bending moments). In all cases, the
user should decide if these solutions are applicable.
The MD1 module works with virtual strips (see 2.16.16 Virtual beams). Design parameters are assigned to
virtual strips and the check can only be performed on the integrated results of the strips.
Virtual strips can include multiple storeys, so the whole wall segment can be analysed as a complete
structural system, but storeys can have different design parameters.
Chapters 6.9.1.7…6.9.1.10 serves information about standard specific limitations.
Design parameters Click on the Design parameters icon, to specify the storeys of the masonry wall strip and check the wall
in the Masonry walls dialog.
Click on the icon or on the axis of a virtual strip to assign design parameters (see 2.16.16 Virtual beams).
The icon is enabled only if there is at least one virtual strip in the model with static or dynamic results.
First the virtual strip must be selected (see 2.16.1 Selection). Only one virtual strip can be selected, its
domains cannot have materials other than masonry, the strip must be continuous, the domains cannot
be tapered and must be vertical with a maximum deviation of 15°. If any of these criteria is not met a
warning message appears otherwise the Masonry walls dialog is displayed and the selected virtual strip
will be selected from the list of masonry virtual strips.
Masonry walls dialog opens automatically when clicking on a (valid) virtual strip in the model.
612
Virtual strips The Masonry walls dialog shows a list of Virtual strips of the model. These are the existing masonry virtual
strips. Selecting a virtual strip displays the list of storeys assigned to the selected strip (Storeys). Selecting
a storey displays the design parameters assigned to the selected storey.
As design parameters are assigned to storeys the first step is define storeys for the selected strip.
If a virtual strip already contains storey(s), then its title is shown in bold. The number in the parentheses
indicates the number of storeys assigned to the virtual strip.
Storeys The first step is to assign storeys to the strips. Design checks are performed on storeys. The set levels are
identified as the levels of the slabs which support the wall laterally.
Click on the Edit storeys icon to define storeys for the virtual strip.
Automatic design After setting the storeys, the following parameters are taken from the model based on the properties of
parameters the domains associated with the virtual strip.
The type of the slab and its thickness can be modified later independently of the model in the
Masonry joint window.
• Design cross-sections (4 or 6 different positions, see later) based on the connecting structures.
If properties of slabs connecting to the storeys are overridden changes in the domain parameters
no longer update the masonry parameters. These shall be updated by the user by deleting the
masonry parameters and redefining it again, or the parameters has to be changed manually.
Restore automatic Some or all of the automatic design parameters of the selected storey of the actual virtual strip can be
design parameters restored with this button. These are
• Masonry material (and its parameters),
• Wall thickness,
• Masonry joints (top and bottom joint, with all the reference cross-sections),
• Design parameters (all the data displayed in the Design parameters dialog except the material).
For details, see 6.9.1.3.
Table of global Clicking on the Table of global design parameters icon, the design parameters of the virtual strips can be
design parameters seen summarized in a table.
For details, see 6.9.1.6 Results.
Check Click on the Check button to perform design check of the storeys.
For details, see 6.9.1.4 Design check of masonry walls, definition of design- and joint parameters
The name of a storey can be changed by right-clicking on the name and choosing Rename.
Other design pa- Other design parameters are set according to the design code recommendations or the default setting
rameters predefined in AxisVM.
Design parameters If design parameters have not been changed, Auto is displayed in the first line of the list, otherwise Custom
of masonry wall is displayed.
Side view of a storey The Masonry walls dialog shows the side of view of the selected masonry wall strip with the connecting
slabs.
Coordinate system The side view of the wall and the internal forces of the strip are displayed in the local coordinate system
of the virtual strip. The same system is used in the Masonry walls and the Design check of masonry walls
dialogs.
614
Closing the Masonry walls dialog, the main window displays the vertical section of the wall with its joints
(see 6.9.1.5).
Settings The following settings are assigned to the entire virtual strip:
Settings of global The global settings of the masonry wall strips are available by clicking on the Settings of global design
design parameters parameters icon.
Alignment If the selected virtual strip contains storeys with different wall thickness values, the user can adjust the
reference plane of the walls. By default, the program matches the mid-planes (Axis) of the walls (like in
case of internal walls).
In case of facade walls or in other situations, the walls can be arranged to the Left or Right (based on the
local coordinate system of the virtual strip), matching the left or right sides of the wall. In these cases, the
effect of an additional bending moment has to be taken into account (see next section).
The side view displays the axis of the virtual strip and the axis of the wall with the largest wall thickness
as a dashed green line. The dashed yellow line shows the axis of the wall on the selected storey.
If all walls in the strip have the same thickness the Alignment option is disabled.
Design parameters Clicking on the Design parameters of masonry walls icon, the design parameters of the storeys can be seen
of masonry wall summarized in a table. The icon is only active if the model already contains at least one (masonry) virtual
strip that has storey(s).
For details, see 6.9.1.6 Results.
Design calculations Clicking on the Design calculations icon, the software generates detailed report of the results based on
the active virtual strip and the load combination/load case selected in the model. The icon is only active
if the model already contains at least one (masonry) virtual strip that has storey(s).
For details, see 6.9.1.6 Results.
Design results Clicking on the Design results icon, the design results of the virtual strips can be seen summarized in a
table. The icon is only active if the model already contains at least one (masonry) virtual strip that has
storey(s).
For details, see 6.9.1.6 Results.
Out-of-plane settings
Calculation of Settings of the masonry wall strip (alignment of walls, masonry joint settings, stress distribution and the
additional slab reaction force) may introduce additional eccentricities so bending moments has to be taken into
bending moments account. If the model does not include these effects the masonry module offers two different models to
approximate and calculate them: Hinged model or Continuous model. For details see 6.9.1.7 Principles of
checking. The design bending moment is the sum of the moment of the virtual strip and the results of
the above models.
If Calculation of additional bending moments is unchecked, no additional bending moments are calcu-
lated.
In-plane settings These settings are related to the in-plane checks:
Shear check Selecting Shear check the module performs both type A and B checks as detailed in chapter 6.9.1. If this
feature is not activated, only checking type A will be performed.
Calculating design The initial and second-order eccentricities in the plane of the wall are calculated using the following
eccentricities parameters
htot Total height of the wall measured from the top of the foundation
α Ratio of the vertical loads stabilized by the wall strip (0< α ≤1).
Check domain as a Check domain as a masonry wall is a simplified check of a single storey wall without existing virtual strips.
masonry wall The virtual strip is automatically created during the process. A simple masonry region has to be selected
and a reference point for the virtual strip.
The created virtual strip is always 1 m wide, but this value can be changed any time editing the settings
of virtual strips (see 2.16.16 Virtual beams). If the entered starting point is near to the edge of the domain,
the axis of the virtual strip is automatically shifted to ensure that the entire strip falls within the domain.
Finishing the selection, the Design check of masonry walls window opens, where the design parameters
and joints can be set and the results can be checked.
The masonry wall strip defined by the function of Check domain as a masonry wall can also be managed
by the Design parameters function like the other virtual strips.
Note: using this function the software assumes that calculation of additional bending moment (see later)
is not needed and the shear check is not marked. To change these settings, click on the automatically
created virtual strip and make changes through the Masonry walls dialog.
6.9.1.4. Design check of masonry walls, definition of design- and joint parameters
Design check of Click on the Check button of the Masonry walls dialog to display the Design check of masonry wall dialog
masonry wall where the design parameters of the wall storeys can be specified and design check is performed.
Set the first dropdown list on the top toolbar to the desired storey of the wall and set the load case or
combination.
The left side of the window displays the Vertical section or the Side view of the selected masonry storey.
(If shear check is not active, only the section of the wall can be seen.) Dimensions and joints are shown in
detail including the connecting slab, thermal insulation, etc. In the Side view of the wall, the connecting
wall domains are also visible in a pale colour, which is a help to specify the design sections (openings) for
shear design.
The coordinate system in the right/left corner (depending on the selected view) shows the coordinate
system of the virtual strip.
The green axis shows the axis of the virtual strip. If the wall thickness is not the same on all storeys, this
axis is identical with the axis of the thickest wall. The yellow axis is the axis of the storey. If wall thickness
is not the same on all storeys the yellow axis is not necessarily coincide with the green axis.
The material colour used is the same as the colour of masonry elements. Filling pattern cannot be
changed and it does not follow the type of masonry elements.
User’s Manual X6 Release1 e2 617
Vertical section
Side view
Active storey
If the mouse is moved over the masonry wall, the cursor changes to a hand symbol. Double-click
on the wall to get to the Design parameters of masonry wall dialog.
If the mouse is moved over the top or bottom joint of the masonry wall, the cursor changes to a
hand symbol. Double click to open the Masonry joints dialog.
Reference cross- Position of reference cross-sections can be entered at the Top / Center / Bottom. Internal
section forces of the virtual strip are read in these positions for design check.
Reference If shear check is not selected, only 3 reference cross-sections can be set (with labels 1, m, 2). They can be
cross-sections specified when the Vertical section of the wall is active. These sections belong to the check type A (see
1, m, 2 before).
618
Thick, blue lines show the positions of reference cross-sections. These lines can be dragged with the
mouse between the limits (black dashed lines).
The position of reference cross-sections is automatically adjusted to the edge of concrete slabs at the
top and the bottom if concrete slabs are present according to the joint settings. Click on the small button
right to the edit field to restore the automatic position.
In case of SIA 266 standard, the position of the middle cross-section cannot be set.
Reference If shear check is activated three more reference cross-sections can be set in the Side view of the wall.
cross-sections These are the positions where design check B) are performed (complex stability and shear check). The
1v, mv, 2v effects of openings and connecting domains can be considered with these positions (shear design is
(position of the performed on the wall section between the lower and upper sections). These positions are independent
openings) of sections 1, m, 2 and are shown in green with signs 1v, mv, 2v. Only the top and bottom positions can
be set, the middle reference line is always halfway between them.
Setting the position Default positions of sections 1v, mv, 2v are determined based on the connecting domains. These can be
of openings modified by dragging the reference lines or they can be set directly in the model after clicking on the
Position of opening icon. The icon is only available on the Internal forces panel (if Side view is selected).
Further information Using the three edit boxes on Internal forces tab, the positions can be given by exact values. These values
to the reference are accepted only if they are between the limits (dashed lines).
cross-sections
The position of arrows above the thick blue lines represents the eccentricity of the compression force in
that specific cross-section. The colour of arrows is blue but if the resistance of the cross-section is not
adequate according to the design check the colour changes to red. If the resistance cannot be calculated
due to an error, the arrow is not shown.
If positions of reference cross-sections change both considered internal forces and design results
will change.
Internal forces In the Design check of masonry wall window, the following internal force diagrams are shown:
Vertical section
Nx Axial force (design axial force)
My Bending moment about local y axis
DMy Additional bending moment due to eccentric effects (calculated only if Calculation of additional
bending moments option is checked on Masonry walls window)
SMy Design bending moment (the sum of bending moment of the virtual strip and additional bending
moment)
Vz Shear force of the virtual strip perpendicular to the wall plane
Side view
Design parameters Design parameters of the selected storey can be specified in the Design parameters of masonry wall dia-
of masonry wall log:
Masonry material After the definition of storeys, the module automatically sets the material of the masonry wall based on
the material of the involved domains. This default material can be changed here.
Editable material The following parameters can be edited. (Changes in this window do not change the parameters of the
parameters domain material. These values are used only in the design check of this particular storey of the selected
masonry wall.):
As it is explained in the previous section, the masonry material is selected automatically after the
definition of storeys at the first time. In case of orthotropic materials, the orientation of local axes
needs to be revised.
Any changes in this window related to the material of the masonry wall does not imply the change
of the material of the domains connected to the virtual strip. The material of the domains needs
to be changed by the user if necessary.
620
SIA 266 • h0 is the sum of the height of masonry element and the
height of mortar layer,
• rmax is the maximum acceptable crack width,
• vmax/hw is the allowed relative displacement considering ref-
erence sections 1v and 2v (where hw is the distance between
the two reference sections). For maximum values see Table
8 in the standard.
• εx,max maximum permissible value according to Table 8 in the
standard,
• boundary conditions (E1 / E2 / E3).
Stress distribution In case of one-side slabs, constant and linear stress distribution
can be considered under the slab reaction force. The stress dis-
tribution set has a significant effect on the calculated additional
bending moment.
Additional bending moment is calculated as the product of slab reaction force and eccentricity based on
the position of resultant of stress distribution. The slab reaction calculated from the difference between
the axial forces of subsequent storeys.
Stress distribution can be selected only if Calculation of additional bending moments option is
checked on Masonry walls window.
In case of the topmost storey, the total axial force at the top is considered to be slab reaction force.
For multi-storey structures, it is highly recommended to use virtual strips that span all of the sto-
reys if the wall is continuous, otherwise, the software cannot calculate the additional bending mo-
ments properly.
If SIA 266 design code is selected, only constant stress distribution can be considered.
Seismic For further details about the coefficient of seismic forces see 4.10.24 Seismic loads – SE1 module.
User’s Manual X6 Release1 e2 621
Joint settings The parameters of joints can be specified in Masonry joint window:
The two drawings show the top and bottom masonry joints. Click on the top or bottom one to change
the active joint. The dashed purple frame shows which joint is active. Parameters of the active joint can
be edited on the right.
If the mouse is moved over the top or bottom joint of masonry wall in the Design check of masonry
wall window, the cursor changes to a hand symbol. Double click to open the Masonry joints dialog.
Type of the slab Possible cases are: Slab is not specified, Continuous slab, One-side slab. Further parameters depend on the
selected case. The lower indices indicate the joint position (1 = top; 2 = bottom).
Slab is not specified Slab is not specified option can be used if the user does
not want to define the parameters of the slab.
Optional parameters:
e1,N is the eccentricity of the slab reaction (it is originated from the ’jump’ in the axial forces or the differ-
ence between the axial forces of subsequent storeys.). (This parameter is only available for top joints).
The sign of the eccentricity is selected according to the coordinate system of the virtual strip.
erel,1 is the relative eccentricity based on the wall thickness (calculated).
The calculated relative eccentricity is revised relative to the limit set in Design parameters without
the consideration of the (other) internal forces that reduces the eccentricity of the slab reaction.
In the verification, at the top section, the effective wall thickness is identical with the wall thickness of the
storey. This rule is also valid for the bottom section, but if the thickness of the wall below the investigated
storey is lower than the wall thickness of the actual storey, then only the lower wall thickness is taken into
account in the calculation.
Continuous slab In case of Continuous slab, it is assumed that the slab re-
action loads are centric. The storey height is calculated
with the consideration of slab thickness.
Optional parameters:
• v is the thickness of the slab.
622
Optional parameters:
Further parameters The Further parameters list shows relevant parameters that cannot be read from the drawings.
Toolbar
Save to Saves the drawing to the Drawing Library.
Drawings Library
Switch between The investigated storey can be selected from the list below. The arrows beside the list can be used to
storeys switch one storey up/down.
Load case / In the drop-down list of load combinations, the user can select the load case to be displayed.
combination
User’s Manual X6 Release1 e2 623
Interpreting of the Selecting Critical combination or Envelope, the results displayed belong to
results in case of the load case or combination resulting in the maximum utilization based
Critical combina- on the results of all cross-sections. This is the default mode of operation,
tions or Envelope called Auto.
There may be cases when the maximum utilization of each reference cross-
section comes from different combination which can cause misunderstand-
ing if the results are not analysed in depth.
To display the maximum utilization of a certain reference section, select a
reference cross-section from the drop-down list.
The combination resulting in the maximum utilization is displayed above the internal force diagrams.
Note: if considering different load combinations, a continuous internal force diagram cannot be
plotted and only the results of related load combinations can be compared and interpreted
together.
The overall results of the sections can be compared and checked simultaneously in the Result table.
The design calculation report uses the Reference position or Auto function selected here. Auto
function is used if the design calculation report was created from the Masonry walls dialog.
6.9.1.6. Results
Without shear check When shear check is not requested, the result panel appear as follows:
Eurocode NTC SIA 266
User’s Manual X6 Release1 e2 625
Results In case of shear check, the user can choose between the result types. Summary or detailed results of
in case of different verifications can be displayed.
shear check The horizontal symbol list at the top summarize the design results in reference cross-sections 1, m, 2 and
1v, mv, 2v (see 6.9.1.4 Design check of masonry walls, definition of design- and joint parameters). The
background of verified sections is grey, failed sections are red. The selected position is blue if the section
is verified otherwise it remains red. The font color of the selected position is always white.
In case of Eurocode, the following options are available:
Results displayed on In case of each reference cross-section, on its reference line, the program indicates the point of action
the Vertical section of the vertical force based on the internal forces (the eccentricity does not contain the initial eccen-
or on the Side view tricity/ minimum eccentricity).
If the resistance of the wall section is adequate then the colour of the arrow is blue, otherwise, it is red.
If the vertical resistance cannot be calculated then the arrow is not displayed.
626
Tables
Table of global de- Global design parameters table is available from the Design check of masonry wall dialog and displays the
sign parameters design parameters assigned to masonry wall strips.
Design results Design results table is available both from the Design check of masonry wall and Masonry walls dialogs. If
the table is opened from Masonry walls window, load case/combination, envelope or automatic critical
load combinations can be selected from a list-box which is considered in the design check. After modi-
fying selection in this list-box, the window automatically performs the design checks that can take only a
few seconds. Without shear check the table appears as follows:
Design calculations Click on Design calculations button to see the design details. The calculation can be saved into the report.
Design calculations is available via Design check of masonry wall and Masonry walls windows, as well.
If it is opened from Masonry walls window, it summarizes the design details for the selected storeys of
the actual masonry wall strip.
If it is opened from Design check of masonry wall window, only the design results of the investigated
storey are summarized.
If shear check is requested, the design calculation contains more calculation units and it appears in more
complex form. Clicking on the Design calculation icon, the user may specify the elements to be included
in the documentation:
If the position of top (1 - 1v), middle (m - mv), or bottom (2 - 2v) reference section coincide, only the
results of the critical sections are detailed.
628
Masonry wall If masonry wall parameters are assigned to virtual strips in the model and static (linear/non-linear) or
utilization result dynamic results are available, a new result component becomes available on the Masonry wall design tab
component related to design check results.
If the utilization cannot be calculated at any storey, the program displays a red dashed line on the nega-
tive side of the diagram indicating the error.
Results in the Design results are also available in the Table Browser (see 2.9 Table Browser), where the results of every
Table Browser reinforced concrete wall are available in summary tables.
User’s Manual X6 Release1 e2 629
Checking The software can perform two types of analysis with the following restrictions:
A. Design check of masonry walls subjected to mainly vertical loads.
In this checking only the out-of-plane eccentricities can be taken into account, shear check is not
included. The verification is based on the reference cross-sections signed 1, m, 2. (Only the ec-
centricities perpendicular to the plane of the wall are taken into account, irrespective of the width
of the wall strip, two-directional eccentricity cannot be analysed.)
B. Design check of masonry bracing walls, which contains a complex stability analysis consid-
ering out-of-plane and in-plane eccentricity of the axial force and an in-plane shear check
as well.
This verification is concerned to the reference cross-sections signed 1v, mv, 2v, but executed only
if shear check is requested.
Naming conventions During the design check 6 different reference cross-section are distinguished:
• the reference sections 1, m, 2 are identical to the design sections explained in chapter 6.1 in
Eurocode (top, middle and bottom sections);
• sections 1v, mv, 2v are the top, middle and bottom sections and used for shear check.
Stability check In case of reference sections signed with 1, m and 2, only the out-of-plane eccentricities of the axial force
are taken into account in the analysis. At sections 1v, mv, 2v both eccentricities (out-of-plane and in-plane)
are considered. The two cases are discussed together below.
630
General principles Considering the general principles of the standard (chapter 6.1.) the program executes the following
verifications:
The resistance of the wall section is adequate if the next criterion
𝑁𝐸𝑑 ≤ 𝑁𝑅𝑑
and the detailed rules are satisfied.
The following formula is applied to determine the vertical resistance of a wall:
N𝑅𝑑 = Φ𝑦 ∙ Φ𝑧 ∙ 𝑡 ∙ 𝑏 ∙ 𝑓𝑑
where y and z are the reduction factors related to y and z direction. (In case of reference cross-sections
1, m and 2, the capacity reduction factor z = 1 because the in-plane eccentricity is not taken into ac-
count).
At the top and bottom sections, the capacity reduction factor is calculated by the effective wall thickness
- teff (see masonry joint settings).
In case of shear check, if the top 1v or bottom 2v cross-section is adjusted to a corner of internal opening
and the distance between the reference cross-section (1v or 2v) and the top/bottom of the wall is greater
than the width of the wall, the section is considered as ‘middle’ section during the verification. Reference
cross-section mv is always considered as ‘middle’ section.
The design compressive strength is: 𝑓𝑑 = 𝑓𝑘 /𝛾𝑀
If the cross-sectional area of a wall strip is less than 0,01 m2, the design compressive strength should be
lowered: f𝑑,𝑟𝑒𝑑 = 𝑓𝑑 ∙ (0,7 + 3𝐴), where A is the loaded horizontal gross cross-sectional area of the wall,
expressed in square metres.
In the module, only the cross-sectional area of the virtual strip can be considered, irrespective the total
width of the concerned domain.
Checking at The design eccentricities, the reduction factors and the vertical resistance of the sections are calculated
the top or bottom of as follows:
the wall
Out-of-plane The design eccentricity is calculated by the next equation:
eccentricity 𝑀𝑖𝑑,𝑦
𝑒𝑖,𝑧 = + 𝑒ℎ𝑒,𝑧 + 𝑒𝑖𝑛𝑖𝑡,𝑧 ≥ 0,05 ∙ 𝑡
𝑁𝑖𝑑
where ei,z is the eccentricity at the top (i=1) or the bottom (i=2) of the wall.
The program does not separate the eccentricities depending on the load type (vertical/horizontal) and
indicates the sum of them in the calculation as follows. The eccentricity based on the internal forces
calculated to the axis of the wall:
𝑀𝑖𝑑,𝑦
𝑒0,𝑧 = + 𝑒ℎ𝑒,𝑧
𝑁𝑖𝑑
In the verification, the program applies lowered wall thickness (teff), if not the whole section is working
(due to the masonry joint settings). In this case, the eccentricity is calculated to the axis of the effective
wall thickness:
𝑒0,𝑖,𝑧 = 𝑒𝑖,𝑧 − Δ𝑧
where z is the difference between the axis of the wall and effective wall thickness.
(If the wall thickness t is not reduced due to the joint settings, then 𝑡𝑒𝑓𝑓 = 𝑡 and 𝑒0,𝑖,𝑧 = 𝑒0,𝑧 .)
Considering the rules of the standard and the markings above, the design eccentricity comes from the
following expression:
𝑒𝑖,𝑧 = 𝑒0,𝑖,𝑧 + 𝑒𝑖𝑛𝑖𝑡,𝑧 ≥ 0,05 ∙ 𝑡𝑒𝑓𝑓
The initial eccentricity is measured in the unfavourable direction relative to the axis of the effective thick-
ness, the minimum eccentricity is calculated by teff.
The i,y capacity reduction factor is calculated as follows:
2 ∙ 𝑒𝑖,𝑧
Φ𝑖,𝑦 = 1 −
𝑡𝑒𝑓𝑓
User’s Manual X6 Release1 e2 631
fvd is the design value of the shear strength of masonry based on the average of the vertical stresses over
the compressed part of the wall.
When calculating the resistance by the above methods, if the reference cross-section 1v or 2v are
considered as ‘top’ or ‘bottom’ section in the verification (the distance between the top/bottom
level and the position of the reference cross-section lower than the width of the wall), then lowered
wall thickness teff is taken into account in the calculation.
Notes to the Usually, there is only slight difference between the results of above methods but it depends on many
result of the shear parameters. In case of the interaction check, the area of the compressed zone is lower, but the design
check shear strength may be greater due to the greater average vertical stresses. However, the design shear
strength is limited by the maximum shear strength according to the National Annexes.
Typically, the interaction check gives the critical (lower) capacity.
Error messages The following error messages appears in the results:
General principles Following the general principles of the standard (NTC 2018) a wall has to be checked at three sections
(top, mid-height of the wall, bottom) similar to Eurocode 6.
In the standard, it is supposed that the wall has hinges at the top and bottom. The user has to build such
a model with the right settings that follows the guidelines of the standard.
Signs During the design check 6 different reference cross-section are distinguished:
• the reference sections signed as 1, m, 2-es are identical to the design sections explained in
chapter 6.1 in Eurocode (top, middle and bottom sections);
• sections signed as 1v, mv, 2v are the top, middle and bottom sections and used for shear check.
Stability check In case of reference sections signed with 1, m and 2, only the out-of-plane eccentricities of the axial force
are taken into account in the analysis. At sections 1v, mv, 2v both eccentricities (out-of-plane and in-plane)
are considered. The two cases are discussed together below.
The resistance of the wall section is adequate if the following criterion
𝑁𝑑 ≤ 𝑁𝑟
and the detailed rules are satisfied.
The program uses the following formula to determine the vertical resistance of a wall:
N𝑟 = 𝑓𝑑,𝑟𝑖𝑑 ∙ 𝑡 ∙ 𝑏
and
634
f𝑑,𝑟𝑖𝑑 = Φ𝑦 ∙ Φ𝑧 ∙ 𝑓𝑑
where y and z are the reduction factors related to y and z direction. (In case of reference cross-sections
1, m and 2, the capacity reduction factor z=1 because the in-plane eccentricity is not taken into account).
At the top and bottom sections, the reduction factor is calculated by the effective wall thickness (see
masonry joint settings).
The design compressive strength is f𝑑 = 𝑓𝑘 /𝛾𝑀
Checking at The design eccentricities, the capacity reduction factors and the vertical resistance of the sections are
the top or bottom of calculated as follows:
the wall
Out-of-plane In NTC, the design eccentricity is calculated by the following equation:
eccentricity 𝑒1,𝑧 = |𝑒𝑠,𝑧 | + 𝑒𝑎,𝑧
where es,z is the eccentricity of vertical loads in the local z direction.
The program does not separate the eccentricities depending on the load type (vertical/horizontal) and
indicates the sum of them as e0,z.
In the verification, the program applies lowered wall thickness (teff), if not the whole section is working
(due to the masonry joint settings). In this case, the eccentricity is calculated to the axis of the effective
wall thickness:
𝑒0,1,𝑧 = 𝑒0,𝑧 − Δ𝑧
where z is the difference between the axes of the wall and effective wall thickness.
(If the wall thickness t is not reduced due to the joint settings, then 𝑡𝑒𝑓𝑓 = 𝑡 and 𝑒0,𝑖,𝑧 = 𝑒0,𝑧 .)
Finally, the design eccentricity comes from the following expression:
𝑒1,𝑧 = 𝑒0,1,𝑧 + 𝑒𝑎,𝑧
Where 𝑒𝑎,𝑧 = ℎ/200 (h is the storey height).
The initial eccentricity ea,z is measured in the unfavourable direction relative to the axis of the effective
thickness (teff).
Based on the initial parameters, the factor of i,z is calculated by linear interpolation using the table given
in the standard (see below).
The relative eccentricity: e𝑟𝑒𝑙,1,𝑧 = 𝑒1,𝑧 /𝑡𝑒𝑓𝑓 , where teff is the effective wall thickness.
The eccentricity coefficient m is calculated by the following: 𝑚 = 6 ∙ 𝑒𝑟𝑒𝑙,1,𝑧 ,
and the slenderness: 𝜆 = ℎ0 /𝑡, where ℎ0 = 𝜚 ∙ ℎ.
In-plane The in-plane eccentricity and the capacity reduction factor are calculated according to the principles of
eccentricity EN 1996-1-1 standard (see details above).
Vertical Considering the total wall width b, the vertical resistance is the following:
resistance N𝑟 = Φ1,𝑦 ∙ Φ1,𝑧 ∙ 𝑓𝑑 ∙ 𝑡𝑒𝑓𝑓 ∙ 𝑏
Checking in the The design eccentricities, the capacity reduction factors and the vertical resistance of the middle sections
middle of the wall are calculated as follows:
height
Out-of-plane The design eccentricity is calculated by the following expression:
eccentricity 𝑒1,𝑧
𝑒2,𝑧 = + 𝑒𝑣,𝑧
2
where e2,z is the design eccentricity at the top in the local z direction, ev,z eccentricity is calculated from
horizontal loads.
The program does not separate the eccentricities depending on the load type (vertical/horizontal) and
indicates the sum of them as e0,z.
The design eccentricity comes from the following expression:
𝑒𝑎,𝑧
𝑒2,𝑧 = 𝑒0,𝑧 + ≥ 𝑒𝑎,𝑧
2
Where 𝑒𝑎,𝑧 = ℎ/200
User’s Manual X6 Release1 e2 635
In the above formula, half of the ea,z is taken in the calculation considering the hinged model recom-
mended in the standard.
If the user uses hinges for the domains and also applies the model called Hinged model to determine the
additional bending moments, then this formula gives the same result as the standard. The applied models
have to be considered by the users.
Based on the initial parameters, the factor of 2,z is calculated by linear interpolation using the table given
in the standard (see below).
𝑒2,𝑧
The relative eccentricity: e𝑟𝑒𝑙,2,𝑧 = , where t is the total wall thickness.
𝑡
In-plane The in-plane eccentricity and the capacity reduction factor are calculated according to the principles of
eccentricity EN 1996-1-1 standard (see details above).
Vertical Considering the total wall width b, the vertical resistance is the following:
resistance N𝑟 = Φ2,𝑦 ∙ Φ2,𝑧 ∙ 𝑓𝑑 ∙ 𝑡𝑒𝑓𝑓 ∙ 𝑏
Interpolation table The following table (given in the standard, Tab. 4.5.III) is used to determine the factor of y.
for the determina-
tion of y factor
m = 6e/t
Slenderness -
0 0,5 1,0 1,5 2,0
Shear check Shear resistance is calculated according to the principles of EN 1996-1-1 standard (see details above).
Design shear strength is determined according to NTC standard.
Design and detailing The following rules are checked:
rules
• slenderness of the wall: ≤ 20.
• minimum wall thickness: t ≥ 150 mm (only this value is considered, the minimum value depend-
ing on the type of the unit is not checked).
• minimum area of wall: 0,04 m2 (applying the rule given Eurocode 6).
• maximum relative eccentricity (the default value is 0.33, but it can be modified).
Checking
- at the top or bottom:
𝑒1
𝑒𝑟𝑒𝑙,1 = ≤ 0,33
𝑡𝑒𝑓𝑓
- in the middle of the wall height:
𝑒2
𝑒𝑟𝑒𝑙,2 =
≤ 0,33
𝑡
The relative eccentricity is checked in both direction (y and z).
636
General principles The principles of SIA 266: 2015 design code calculation method differ significantly from the procedure
described in Eurocode 6. Based on the standard, the wall has to be checked in ultimate and serviceability
limit state (ULS and SLS) as well.
The program performs ULS or SLS verification depending on the type of combination. If the type is not
known, both checking is performed, but only the result of the critical state is displayed.
The design code contains constraints on the static model, so the user should take care of the correct
choice of the static model. For the assumptions, see details below.
Calculating the additional bending moment only the Hinged model is available, and only constant stress
distribution can be applied.
The applied static model (E1, E2 and E3) has to be selected as the program cannot determine it from the
model.
Signs of design Four different design cross-sections are distinguished in the verification
cross-sections
• cross-sections with a sign of 1 (top) or 2 (bottom),
• cross-sections with a sign of 1v (top) or 2v (bottom) for shear check.
User’s Manual X6 Release1 e2 637
ULS verification In ULS verification, the wall is checked in two sections of the wall: at the top and the bottom of the wall.
Four different analysis are performed:
• stress limitation at cross-sections 1 and 1v,
• stability check considering the internal forces of the virtual strip at the reference position of
sections 1 and 2v,
• stress limitation at cross-sections 2 and 2v,
• shear check at the top cross-section (1v).
The verifications of cross-sections 1v and 2v are performed only if shear check is requested.
At design cross-sections 1 and 2 only the out-of-plane eccentricity of the axial force is taken into account,
but at sections 1v and 2v the in-plane eccentricity is also considered.
Stress limitation Firstly, the relative eccentricity is checked according to the standard: (ez/tw) is limited by the formula (6)
and Figure 4 (ez/tw ≤ 0,375). On the other hand, the program calculates the load-bearing capacity con-
sidering the compressed zone of the wall and it is compared to the design axial force.
The compressed width of the section considering out-of-plane eccentricity:
t 𝑛𝑜𝑚 = 𝑡 − 2 ∙ 𝑒0,𝑧
In the verification, the program applies lowered wall thickness (teff), if not the whole section is working
(due to the masonry joint settings).
e0,z is the eccentricity is calculated to the axis of the effective wall thickness:
𝑁𝑅𝑑,𝑛𝑜𝑚 = 𝑙𝑐 ∙ 𝑡𝑛𝑜𝑚 ∙ 𝑓𝑥𝑑
where lc is equal to lw when checking cross-sections 1 and 2. In case of cross-sections 1v and 2v the
compressed length of the wall is (l=l2) lowered due to the in-plane bending moment:
2 ∙ 𝑀𝑧,𝑖,𝑑
𝑙2 = 𝑙𝑤 −
𝑁𝑥𝑑
Stability check In case of cross-section 1, the program follows the regulations of the standard (SIA 266:2015) described
in chapter 4.3.1.3 (11):
𝑁𝑥𝑑 ≤ 𝑘𝑁 ∙ 𝑙𝑤 ∙ 𝑡𝑤 ∙ 𝑓𝑥𝑑
In case of cross-section 2v, formula (17, in chapter 4.3.3.1) is used considering in-plane bending moment:
𝑁𝑥𝑑 ≤ 𝑘𝑁 ∙ 𝑙2 ∙ 𝑡𝑤 ∙ 𝑓𝑥𝑑
where
• Nxd is the design vertical force at the top of the wall,
• lw width of the wall (in the program it is equal to the width of the virtual strip)
• tw wall thickness,
• fxd is the design compressive strength, f𝑥𝑑 = 𝑓𝑥𝑘 /𝛾𝑀
• l2 length of the compressed part of the wall assuming constant stress distribution:
2 ∙ 𝑀𝑧2𝑑
𝑙2 = 𝑙𝑤 −
𝑁𝑥𝑑
The factor of kN is calculated by linear interpolation based on the set static model (E1, E2 and E3) using
the design curves given in the standard (SIA 266:2015, Figure 5).
The interpolation is determined by the relative slenderness of the wall (hw / tw or hcr / tw) and the relative
eccentricity (ez / tw) calculated at the top of the wall.
638
tnom is the width of the compressed zone due to the out-of-plane bending:
t 𝑛𝑜𝑚 = 𝑡 − 2 ∙ 𝑒0,𝑦
(In the verification, the program applies lowered wall thickness - teff, if not the whole section is
working due to the masonry joint settings).
fyd is the design compressive strength in the direction parallel to the bed joints:
𝑓𝑦𝑘
f𝑦𝑑 =
𝛾𝑀
The factor of kV is calculated by linear interpolation based on the design curves given in the standard (SIA
266:2015, Figure 9). hw is the height difference of cross-sections 1v and 2v. The software makes also inter-
polation between the 3 given curves depending on the ratio of fxd/fyd in the range of 0.30…0.75.
Error messages The following error messages appears in the results:
SLS verification The program follows the regulations of the standard (SIA 266:2015) described in chapter 4.4.1.5 (21):
The maximum crack width can be calculated by the next formula:
𝑁𝑥 ∙ ℎ0
𝑟 = 𝑘𝑟 ∙
𝐸𝑥𝑘 ∙ 𝑙𝑤 ∙ 𝑡𝑤
where
Nx is the vertical force on the top of the wall,
h0 is the total height of the masonry unit and mortar,
Exk moduli of elasticity,
lw width of the wall (in the program it is equal to the width of the virtual strip),
tw wall thickness,
The factor of kr is calculated by linear interpolation based on the set static model (E1, E2 and E3) using
the design curves given in the standard (SIA 266:2015, Figure 12).
The interpolation is determined by the relative slenderness of the wall (hw/hE or hcr/hE) and the relative
eccentricity (ey/tw) calculated at the top of the wall.
The parameter of hE is calculated as follows:
3
𝐸𝑥𝑘 ∙ 𝑙𝑤 ∙ 𝑡𝑤
ℎ𝐸 = 𝜋 ∙ √
12 ∙ 𝑁𝑥
The wall is adequate for the SLS verifications, if 𝑟 ≤ 𝑟𝑚𝑎𝑥 , where rmax is the maximum crack width accord-
ing to 4.2.2.2. The default value of the maximum crack width is 0,2 mm, but it can be set among the
design parameters.
In case of If shear check is requested, then the SLS verification is extended by the followings according to 4.4.2.1 in
shear check SIA 266:2015.
The relative displacement – the value of ’v’ according to formula (22) - between the top (1v) and bottom
(2v) sections is calculated by the relative deformation of the virtual strip.
The maximum strain εx,max is determined by the formula (23):
6 ∙ (𝑀𝑧1 + 𝑉 ∙ ℎ𝑤 ) 𝑁𝑥
𝜀𝑥.𝑚𝑎𝑥 = 2 ∙𝑡
−
𝐸𝑥𝑘 ∙ 𝑙𝑤 𝑤 𝐸𝑥𝑘 𝑙𝑤 ∙ 𝑡𝑤
∙
The relative displacement and the maximum strain are always determined by the top (2v) and bot-
tom (2v) sections defined for the shear check.
The above results are compared to the limits set among the design parameters of a storey.
Error messages The following error messages appears in the results:
If you do not want others to use your work as a basis for their models, but you would like to let them
see, the model can be saved in an AxisVM Viewer (*.AXV) file format (see File/Export). The market version
cannot read AXV files but the Viewer can. This format guarantees that your work will be protected.
AxisVM Viewer Ex- Owners of the AxisVM market version can buy the Viewer Expert version which lets the user print dia-
pert grams, tables and reports or place temporary dimension lines and text boxes. No changes can be saved.
Any external program can get these descriptions, read object properties or call the functions provided
through the interface. A program can launch AxisVM, build models, run calculations and get the results
through the AxisVM COM server. This is the best way to
• build and analyse parametric models,
• finding solutions with iterative methods or
• build specific design extension modules.
DLL modules placed in the Plugins folder of AxisVM are automatically included in the Plugins menu
imitating the subfolder structure of the Plugins folder. The AxisVM COM server specification and pro-
gramming examples are downloadable from the AxisVM website, www.axisvm.eu
User’s Manual X6 Release1 e2 641
9. Examples
Geometry:
Material: Steel
Cross-section: I 240
Loads:
Results AK-ST-I.axe
Geometry:
Material: Steel
Cross-section: I 240
Loads:
Results AK-ST-II.axe
Verify The equilibrium must be verified taking into account the deflections.
User’s Manual X6 Release1 e2 643
Material: Steel
Cross-section: I 240
Results AK-KI.axe
Buckling mode:
Geometry:
Material: Steel
Cross-section: I 240
Results AK-RZ-I.axe
Frequency [Hz]
Mode Cosmos/M AxisVM
1 6.957 6.957
2 27.353 27.353
3 44.692 44.692
4 48.094 48.094
5 95.714 95.714
6 118.544 118.544
User’s Manual X6 Release1 e2 645
Material: Steel
Cross-section: I 240
Results AK-RZ-II.axe
Frequency [Hz]
Mode Cosmos/M AxisVM
1 0.514 0.514
2 11.427 11.426
3 12.768 12.766
4 17.146 17.145
5 27.112 27.109
6 39.461 39.456
646
E=880 kN/cm2
=0
t=0.10 m
p=100 kN/m
Mesh: 4x16
Results VT1-ST-I.axe
ez( B)
[mm] 15.09 15.09
n(xA )
[kN/m] 1800.00 1799.86
User’s Manual X6 Release1 e2 647
E=880 kN/cm2
=0
t=0.15 m
p=50 kN/m2
Mesh: 8x8
Results
Convergence
analysis
Meshes:
648
Results VL2-ST-I.axe
Convergence
analysis
Meshes:
User’s Manual X6 Release1 e2 649
10. References
1. Bathe, K. J., Wilson, E. L., Numerical Methods in Finite Element Analysis, Prentice Hall, New Jersey, 1976
2. Bojtár I., Vörös G., A végeselem-módszer alkalmazása lemez- és héjszerkezetekre, Műszaki Könyvkiadó, Budapest,
1986
3. Chen, W. F., Lui, E. M., Structural Stability, Elsevier Science Publishing Co., Inc., New York, 1987
4. Hughes, T. J. R., The Finite Element Method, Prentice-Hall, Inc., Englewood Cliffs, New Jersey, 1987
5. Owen D. R. J., Hinton E., Finite Elements in Plasticity, Pineridge Press Limited, Swansea, 1980
6. Popper Gy., Csizmás F., Numerikus módszerek mérnököknek, Akadémiai Kiadó Typotex, Budapest, 1993
7. Przemieniecki, J. S., Theory of Matrix Structural Analysis, McGraw Hill Book Co., New York, 1968
8. Weaver Jr., W., Johnston, P. R., Finite Elements for Structural Analysis, Prentice-Hall, Inc., Englewood Cliffs, New
Jersey, 1984
9. Dr. Szalai Kálmán, Vasbetonszerkezetek, vasbeton-szilárdságtan, Tankönyvkiadó, Budapest, 1990. 1998
11. Dr. Kollár László: Vasbetonszerkezetek I., Vasbeton-szilárdságtan az Eurocode 2 szerint, Műegyetemi Kiadó, 1997
12. Dr. Bölcskei E., Dr. Dulácska E.: Statikusok könyve, Műszaki Könyvkiadó, 1974
13. Dr. Dulácska Endre: Kisokos, Segédlet tartószerkezetek tervezéséhez, BME Építészmérnöki Kar, 1993
14. Porteous, J., Kermani, A., Structural Timber Design to Eurocode 5, Blackwell Publishing, 2007
15. Dulácska Endre, Joó Attila, Kollár László: Tartószerkezetek tervezése földrengési hatásokra, Akadémiai Kiadó, 2008
16. Pilkey, W. D., Analysis and Design of Elastic Beams - Computational methods, John Wiley & sons, Inc., 2002
17. Navrátil, J., Prestressed Concrete Structures, Akademické Nakladatelství Cerm®, 2006
18. Szepesházi Róbert: Geotechnikai tervezés (Tervezés Eurocode 7 és a kapcsolódó európai geotechnikai szabványok
alapján), Business Media Magyarország Kft., 2008
27. Paz,M., Leigh, W., Structural Dynamics - Theory and Computation, Fifth Edition, Springer, 2004
28. Chopra, A. K., Dynamics of Structures - Theory and Applications to Earthquake Engineering, Third Edition, Pearson
Prentice Hill, 2007
30. Weaver, W., Jr., P. R. Johnston, Structural Dynamics by Finite Elements, Prentice-Hall, 1987
31. Bathe, K. J., Finite Element Procedures, Prentice-Hall, 1996
32. Borst, R., Crisfield, M. A., Remmers, J. J. C., Verhoosel, C. V., Non-Linear Finite Element Analysis of Solids and
Structures, Second Edition, John Wiley & Sons Ltd., 2012
33. Willford, M.R., Young, P. A Design Guide for Footfall Induced Vibration of Structures, Concrete Society, 2006
34. Smith, A. L., Hicks, S. J., Devine, P. J. Design of Floors for Vibration: A New Approach, The Steel Construction Institute, Ascot, 2009
35. https://en.wikipedia.org/wiki/Von_Mises_yield_criterion
650
36. https://en.wikipedia.org/wiki/Bresler_Pister_yield_criterion
37. https://en.wikipedia.org/wiki/Whittaker%E2%80%93Shannon_interpolation_formula
38. https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem
39. Wilson, E. L., Three-Dimensional Static and Dynamic Analysis of Structures – A Physical Approach With Emphasis on
Earthquake Engineering, Computers and Structures Inc., Berkeley, 2002