Flac2d (Getting Started)
Flac2d (Getting Started)
2-1
2 GETTING STARTED
This section provides the first-time user with an introduction to FLAC. Getting Started contains instructions for program installation and start-up on your computer. It also outlines the recommended
procedure for applying FLAC to problems in geo-engineering and includes simple examples that
demonstrate each step of this procedure.
If you are familiar with the program but only use it occasionally, you may find this section (in
particular, Section 2.6) helpful in refreshing your memory on the mechanics of running FLAC.
More-complete information on problem solving is provided in Section 3.
FLAC can be operated in command-driven or graphical, menu-driven mode. For most of the
examples in this manual, input is entered and results are viewed using the command-driven mode.
We believe this is the clearest way for you to understand the operating procedures for FLAC.
As explained previously in Section 1.1, the command-driven structure allows FLAC to be a very
versatile tool for use in engineering analysis. However, this structure can present difficulties for
new, or occasional, users. Command lines must be entered as input to FLAC, either interactively
via the keyboard or from a remote data file, in order for the code to operate. There are over 40 main
commands and nearly 400 command modifiers (called keywords) which are recognized by FLAC.
The menu-driven mode is an easy-to-use alternative to the command-driven procedure. All the
commands in FLAC can be accessed by point-and-click operation from the graphical mode. We
call this mode the GIIC for Graphical Interface for Itasca Codes; eventually the GIIC will operate
with all Itasca software.
Getting Started contains the following information.
1. A step-by-step procedure to install and start up FLAC on your computer is given in
Section 2.1.
2. This is followed in Section 2.2 by instructions on running FLAC. Section 2.2.1 describes
the procedure for running FLAC in the command-driven mode, and includes installation
tests (Section 2.2.1.1) and a tutorial (Section 2.2.1.2) to help you become familiar with
common input commands. Section 2.2.2 introduces the GIIC and provides a tutorial on
running FLAC in menu-driven mode (Section 2.2.2.4).
3. There are a few things that you will need to know before creating and running your own
FLAC model i.e., you need to know the FLAC terminology. The nomenclature used
for this program is described in Section 2.3. The definition of a FLAC finite difference
grid is given in Section 2.4. You should also know the syntax for the FLAC input language
when running in command-driven mode; an overview is provided in Section 2.5.
4. The mechanics of running a FLAC model are described in separate steps; in Section 2.6,
each step is discussed separately and simple examples are provided.
5. The sign conventions, systems of units and precision limits used in the program appear
in Sections 2.7, 2.8 and 2.9, respectively.
6. The different types of files used and created by FLAC are described in Section 2.10.
2-2
Users Guide
GETTING STARTED
2-3
* Note that the installation program works only under 32-bit Windows. The term Windows in this
section means Windows 95 and upward. The files are in uncompressed form on the CD, so you can
also copy them directly. However, this procedure will not install the hardware key drivers that are
required for execution.
2-4
Users Guide
Table 2.1
Directory
Section Files
Fishtank
FLAC
Grid Generation
Intface
Interfaces
Models
Theory
FISH
Program Guide
Library
Tutorial
Fluid
GUI
Options
Dynamic
Dynamic Analysis
Thermal
Thermal Option
Problems
Example Applications
Verify
Verification Problems
Struct
Tutorial
Getting Started
FISH
Solving
Problem Solving
JRE
Manuals
Models
System
Utility
GETTING STARTED
2-5
from the command line in text mode, and by pressing the File / Exit button in graphics mode to
return to text mode.
The Windows-console version of FLAC is compiled with the Watcom Fortran 11.0 compiler. The
GIIC is written in JAVA using JAVA(TM) Runtime Environment, standard edition version 1.2.2.
2.1.4 Memory Allocation
Automatic memory allocation logic has been implemented in FLAC for Intel-based computers.
When loaded, FLAC will, by default, adjust the size of the main array to take up 8 MB RAM, or the
maximum amount available, if it is less than 8 MB. This means that if other programs are resident
when FLAC is executed, the size of the main array may be decreased and smaller allowable problem
sizes will result.
You can change the amount of memory used by FLAC by modifying the shortcut to FLAC from the
FLAC v4.0 icon in the Itasca Codes group. In the shortcut properties, add the amount of memory (in
MB) to the end of the target string. If the amount of memory requested is more than that available,
FLAC will still load, but with the maximum available memory. The amount of memory allocated
for FLAC is printed in the start-up screen.
As a guide, Table 2.2 summarizes the approximate maximum numbers of zones (of Mohr-Coulomb
material) that can be created for different sizes of available RAM, in the single-precision version
of FLAC.
2-6
Users Guide
Table 2.2
Available RAM
(MB)
8
16
32
64
30,000
60,000
120,000
240,000
GETTING STARTED
2-7
time can be affected by other programs running in the background. If you notice a significant delay
in the initialization of the graphics mode, it may be necessary to close other Windows applications.
2.1.7 Program Initialization
On start-up, FLAC will look for the file named FLAC.INI in the current directory and then, if
not found, in the directory pointed to by the ITASCA environment variable. (By default, this is the
\ITASCA\System directory.) The FLAC.INI file can contain any FLAC commands that preset
attributes of the program that you may wish to apply every time FLAC is used. When FLAC is installed from the CD-ROM, the FLAC.INI file is automatically installed in the \ITASCA\FLAC
directory and contains the command GIIC. This will cause the GIIC to start up each time FLAC is
loaded.
If the file FLAC.INI does not exist, FLAC simply continues in command-driven mode without
error. Note that some commands in a FLAC.INI file may result in an error message. For example,
if you attempt to give properties to a grid prior to defining the grid, the normal error message will
arise.
2.1.8 Version Identification
The version number of FLAC follows a simple numbering system that identifies the level of updates
in the program. There are three numerical identifiers in the version number that is,
Version I.JK
where
I
is an integer that is incremented when minor modifications are officially released as an update to the current version.
In addition to the version number, sub-version numbers are also used to identify minor changes
to FLAC that have been made since the official version was released. Users may access the latest
sub-version of the current version of FLAC via the Internet. (Contact Itasca for further information.)
However, FLAC with a sub-version number greater than that of the officially-released version should
be used with caution, because not all features have been fully tested.
By typing the command
print version
at the flac:command-line prompt, the complete version number, including the sub-version number, can be obtained.
2-8
Users Guide
file.dat
on the command line, in which FILE.DAT is the user-assigned name for the data file. You will
see the data entries scroll up the screen as FLAC reads each line (if SET echo is on).
2.2.1.1 Installation Tests
Three simple data files, TEST1.DAT, TEST2.DAT and TEST3.DAT, are included in the
\Itasca\FLAC directory so that you can verify whether FLAC is properly installed on your
computer. These files test the calculation kernel, the graphics screen plotting, and the hardcopy
plotting facilities for your computer. In order to run the third test, your current default Windows
printer must be connected to the LPT1 port.*
To run these tests, first start up FLAC following the procedure in Section 2.1.6. When you start
up FLAC after installation from the CD-ROM, the code will be operating in the GIIC. To change
to command-driven mode, press the Ok button to close the FLAC Options dialog, and then press
File / Exit GIIC; FLAC will switch to text mode.
* The FLAC hardware key is transparent to other uses of the LPT1 port. The printer can be attached
directly to the key.
GETTING STARTED
2-9
and press <Enter>. Several data entries should scroll up the screen, and a simple model will be
executed for 100 calculation steps. Example 2.1 contains the results of a successful TEST1.DAT
run.
Example 2.1 FLAC output from TEST1.DAT
major stress
I
J
1 -9.553-9.493-9.308-8.834
A screen plot of this model should appear. The plot is a contour plot of displacements. The plot
window may be moved by dragging the title bar, and it may be resized by dragging a border or
corner. It may also be maximized, restored or closed. Only one plot screen can be created at a
time, and the plot screen must be closed in order to continue processing FLAC commands. Press
the <Enter> key to close the plot window and return to the flac: prompt.
If your current default Windows printer is connected to the LPT1 port, type
call test3.dat
and the plot shown in Figure 2.1 should be sent to your printer. If you do not have a printer
connected, type
quit
2 - 10
Users Guide
JOB TITLE :
LEGEND
12-Apr- 0 15:04
step
100
-8.333E-01 <x< 5.833E+00
-8.333E-01 <y< 5.833E+00
4.000
Boundary plot
0
3.000
2E 0
Y-displacement contours
-2.50E-03
-2.00E-03
-1.50E-03
-1.00E-03
-5.00E-04
0.00E+00
2.000
1.000
0.000
Figure 2.1
1.000
2.000
3.000
4.000
5.000
GETTING STARTED
2 - 11
To set up the initial finite difference grid, use the GRID command:*
grid 5,5
This command will create an initial grid (or mesh, if you prefer) that is 5 zones (or elements) wide
by 5 zones high. Now, give the zones a material model and properties. For this example, we use
the Mohr-Coulomb elasto-plastic model. Type in the following commands:
model mohr
prop bulk=1e8 shear=.3e8 fric=35
prop dens=1000 coh=1e10 ten=1e10
Here, we have specified the Mohr-Coulomb model. Every zone in the grid could conceivably have a
different material model and property. However, by not specifying a range of zones directly behind
the MODEL command, FLAC assumes that all zones are to be Mohr-Coulomb. The properties are
given next including the bulk modulus (in Pa), shear modulus, the angle of internal friction,
the mass density, the cohesion and the tensile strength. Any consistent set of engineering units
can be used when assigning properties in a FLAC model (see Section 2.8). Note that very high
cohesion and tensile strength values are given. These are only initial values that are used during the
development of gravitational stresses within the body. In effect, we are forcing the body to behave
elastically during the initial development of the gravitational stresses. This avoids any plastic yield
during this initial phase of the analysis. The reasons for this will become obvious once you gain
experience with the explicit-simulation procedure.
Now that a grid and model properties have been defined, data pertaining to the simulation can be
plotted or printed. Issue the following command:
print
The x- and y-coordinates will appear in tabular form in the physical positions of the gridpoints.
You will note that the table has i (column) and j (row) going from 1 to 6 along the top and left-hand
edge of the table. Therefore, each gridpoint and zone has an i (column) and j (row) associated with
it. In this example, the gridpoint range is i from 1 to 6 and j from 1 to 6, whereas the zones range is
from 1 to 5 for i and 1 to 5 for j . If you require greater clarification on this point, see Figure 2.29
in Section 2.4. To see a plot of the grid, give the following command:
plot
grid
This will create a plot of the grid on the screen. After viewing, press <Enter> to get back to the
flac: prompt.
* See the command reference list in Section 1.3 in the Command Reference for further details. Note
that command words can be abbreviated (see Section 2.5).
Alternatively, an elastic model could initially be used to set up the virgin stresses, followed by
changing the model to Mohr-Coulomb prior to any excavation, applied loads, or other simulations.
The plotting window will be set automatically unless otherwise specified by using the WINDOW
command.
2 - 12
Users Guide
In order to make a hard copy of a plot, enter the command COPY and the plot will be sent (by
default) to the current Windows printer connected to the LPT1: port.*
Alternatively, we can send the plot to a file for printing at some later time. For example, the
commands
set plot emf
copy grid.emf
will create a Windows-enhanced metafile plot GRID.EMF of the last-viewed plot. The file can
then be directly imported to a word processor program such as Microsoft Word.
If a PCX file is desired instead, the SET pcx command will allow PCX files to be generated by
pressing <F2> when the plot is displayed on screen. See Section 1.3 in the Command Reference
for a full description of this command.
Note that if we do not assign coordinates to the grid (by using the GENERATE or INITIAL command),
then the x- and y-coordinates are assigned equal to the number of the gridpoint minus 1. For
example, in the previous grid plot, the lower left-hand gridpoint is assumed to be the origin and
is given the coordinate (0,0). The bottom right-hand corner gridpoint (6,1) is given the coordinate
(5,0). The user is completely free to assign any chosen coordinates by using the GENERATE and
INITIAL commands. To keep this example simple, we leave the grid at 5 m 5 m.
Next, the boundary conditions for the problem are set. In this problem, we want to place roller
boundaries on the bottom and sides, apply gravitational forces to the zones, and allow the in-situ
stresses to develop as they occur in nature. To fix these boundaries (i.e., no displacement or velocity
in the specified direction), use the following commands:
fix
fix
fix
y
x
x
j=1
i=1
i=6
* The printer type can be changed with the SET plot command; for example, type SET plot post before
entering the COPY command to direct plots to a PostScript-compatible printer. The output port can
be changed or a filename can be specified with the SET output command see Section 1 in the
Command Reference.
GETTING STARTED
2 - 13
grav=9.81
where 9.81 m/sec2 is the acceleration due to gravity. Here, gravity is taken as positive downward
and negative upward. (If gravity is set negative, objects will rise!)
We wish to see a history of the displacement of a gridpoint on the model to indicate equilibrium or
collapse. Type:
his nstep=5
his ydis i = 2 j = 6
Here, we choose to monitor the y-displacement every five timesteps for a point at the top of the
ground surface. Now, we are ready to bring the initial model to equilibrium. Because FLAC is an
explicit dynamic code, we step the model through time,* allowing the kinetic energy of the mesh to
damp out (thus providing the static solution we seek). To allow gravity to develop within the body,
we timestep the simulation to equilibrium. Here, the SOLVE command is used to detect equilibrium
automatically.
Type the commands
set force=100
solve
The calculation process will begin and the timestep number, maximum unbalanced force and equilibrium ratio will be displayed on the screen. When the unbalanced force falls below the limiting
value (a limiting force of 100 N is specified with the SET command), the run will stop. Other
options for solution limits, such as equilibrium ratio, are discussed later in Section 2.6.4.
Now, we can see what has occurred within the model. Examine the y-displacement history requested
earlier:
plot his 1
A screen plot will be shown which indicates that the model came to equilibrium within 108 timesteps.
The final y-displacement at equilibrium is -0.881 103 m, due to the gravitational loading. For
a screen listing of this history, type
hist dump 1
2 - 14
Users Guide
Lets examine the gravitational stresses developed in the body. The window was automatically
defined but, if we wish to enlarge or shrink the plot, we can reset it with the WINDOW command.
Now, give the plot a title* by typing
title
a simple trench excavation example
gre
This will create a plot (Figure 2.2) of the yy -stresses in yellow-brown and the boundary in green.
Similarly, the xx -stresses can be plotted by typing
plot
hold
sxx
yel
bou
gre
LEGEND
12-Apr- 0 16:33
step
108
-2.500E-01 <x< 5.250E+00
-2.500E-01 <y< 5.250E+00
YY-stress contours
Contour interval= 5.00E+03
A: -4.000E+04
H: -5.000E+03
Boundary plot
0
4.500
3.500
2.500
1E 0
C
1.500
A
0.500
Figure 2.2
1.500
2.500
3.500
4.500
* The title and legend appear on hardcopy plots as well as screen plots. There are slight differences
between the legend shown on the screen plot and that shown on hardcopy plots.
The color switch also controls the line style on hardcopy plots. See Table 1.6 in the Command
Reference to select line styles based on color keywords. Note that all plots shown in the figures in
Section 2 have a line style set by the default line style, which is a solid line.
GETTING STARTED
2 - 15
We note that the gravitational stresses increase linearly with depth. The values can be printed by
typing
print
sxx
syy
It is wise to save this initial state so that we can restart it at any time for performing parameter
studies. To save this, type
save
trench.sav
A save file will be created on the default drive. The FLAC prompt will then return.
Now we can excavate a trench in the soil. First, type
prop coh=0
With a zero cohesion and vertical, unsupported trench walls, collapse will certainly occur. Because
we want to examine this process realistically, the large-strain logic must be set in the code. This is
done by typing
set large
Finally, for plotting purposes, we wish to see only the change in displacements from the trench
excavation and not the previous gravitational setting so we can zero out the x- and y-displacement
components:*
init
xdis=0
ydis=0
null
i=3
j=3,5
Since we purposely set the cohesion low enough to result in failure, we do not want to use the SOLVE
command with a limit for out-of-balance forces (which checks for equilibrium). Our simulation
will never converge to the equilibrium state. Instead, we can step through the simulation process
one step at a time and plot and print the results of the collapse as it occurs. This is the real power
of the explicit method. The model is not required to converge to equilibrium at each calculation
cycle, because we never have to solve a set of linear algebraic equations simultaneously, as is the
case in the implicit codes with which many engineers are familiar.
In FLAC, we use the STEP command:
step 100
FLAC will now step through 100 timesteps. When it is finished, the prompt will reappear. Now,
examine the results thus far by plotting some variables e.g.,
plot
hold
plastic
boundary
* This will not affect the calculations since the model does not require displacements in the calculation
sequence. They are kept only as a convenience to the user.
2 - 16
Users Guide
The present state of each zone will be indicated by symbols which represent the type of failure condition. This plot indicates that the zones adjacent to the trench are actively yielding in shear.* The
Mohr-Coulomb failure model is discussed in detail in Section 2.4.2 in Theory and Background.
Now, try plotting some parameters:
plot
hold
grid
This will produce a plot of the x-velocity contours (in yellow, contour interval of 5 106 m, zero
contours removed) overlaid by the displacement vectors (in red, scaled to a maximum vector length
of 1 102 m) and the boundary (in green). This is shown in Figure 2.3. The velocity contours
are given here to help visualize those areas of active yield, because this material is flowing.
The collapse process can be examined as it occurs by timestepping 100 steps at a time. We encourage
you to step ahead in this fashion, creating plots at each stage and experimenting with the max, int
and color keywords at each stage. Try plotting the stresses, velocities and displacements to produce
meaningful results. In this example, we will jump ahead to a convenient spot in the collapse process:
step 400
Again, try
plot hold grid
There is a drastically different picture at this stage as the trench collapses (Figure 2.4).
By typing
plot
hold
plas
bou
we note that the zones are still at the yield failure point. Examine the yy -state and displacements
by requesting
plot hold syy zero int=2500 disp max=0.2 mage bou gree
* Note that we have made our boundaries on this problem small in order to speed operation; thus,
some boundary interference occurs.
GETTING STARTED
2 - 17
4.500
13-Apr- 0 11:50
step
208
-2.675E-01 <x< 5.267E+00
-2.688E-01 <y< 5.266E+00
R
Q
3.500
X-velocity contours
Contour interval= 5.00E-06
(zero contour omitted)
N: 5.000E-06
S: 3.000E-05
Displacement vectors
Scaled to Max = 1.000E-02
Max Vector = 6.177E-03
O
N
2.500
2E -2
1.500
Boundary plot
0
1E 0
0.500
Figure 2.3
1.500
2.500
3.500
4.500
LEGEND
12-Apr- 0 16:40
step
608
-2.500E-01 <x< 5.250E+00
-2.711E-01 <y< 5.229E+00
3.500
Grid plot
0
1E 0
2.500
1.500
0.500
Figure 2.4
1.500
2.500
3.500
4.500
2 - 18
Users Guide
We observe distortion of the stress contours due to the excavation and an increase in magnitude (by
approximately 100 times) of the displacement vectors (Figure 2.5). Also note that stress contours,
unlike displacement and velocity contours, are not plotted to the external and excavation boundaries,
because stresses are constant within a zone.*
JOB TITLE : a simple trench excavation example
LEGEND
N
12-Apr- 0 16:28
step
608
-2.628E-01 <x< 5.263E+00
-2.844E-01 <y< 5.241E+00
M
L
3.500
K
YY-stress contours
Contour interval= 2.50E+03
B: -3.500E+04
N: -5.000E+03
Displacement vectors
Scaled to Max = 2.000E-01
Max Vector = 1.338E-01
0
J
I
H
5E -1
Boundary plot
0
2.500
1.500
E
D
1E 0
C
B
0.500
Figure 2.5
1.500
2.500
3.500
4.500
From this point, you may wish to play with the various features of FLAC in an attempt to stabilize
the excavation. Try restarting the previous file you created by entering
rest
trench.sav
Excavate the trench as before, but try using the structural element logic described in Section 1 in
Structural Elements to model bracing or tieback anchors.
You will see that FLAC is virtually bullet-proof an error-trapping function recognizes most
commonly-occurring errors.
This ends the command-driven tutorial. We recommend that you now try running FLAC in menudriven mode, following the instructions in Section 2.2.2. Then, for comparison of command-driven
versus menu-driven operation, try the GIIC tutorial given in Section 2.2.2.4. After that, read the
rest of Section 2 for a beginners guide to the mechanics of using FLAC. As you become more
familiar with the code, turn to Section 3 for additional details on problem solving with FLAC.
* It is possible to use an interpolation function to extend the stress contour lines to the boundaries,
using the FISH built-in language (see EXTRAP.FIS in Section 3 in the FISH volume).
GETTING STARTED
2 - 19
from the flac: prompt if you are in the text mode after loading FLAC. The GIIC command is
automatically included in the FLAC.INI file (see Section 2.1.7) when you install FLAC from the
Itasca CD, so that the graphical interface will start up every time FLAC is loaded. The GIIC main
window is shown in Figure 2.6.
The code name and current version number are printed in the title bar at the top of the window,
and a main menu bar is positioned just below the title bar. Beneath the main menu bar are two
windows: a resources pane and a model-view pane. The resources pane contains two tabbed panes
with text-based information. A Console pane shows text output and allows command-line input
(at the bottom of the pane). A Record pane shows a record of commands needed to generate the
current model project state. This record can be exported to a data file as a set of FLAC commands
that represent the problem being analyzed.
The model-view pane shows a graphical view of the model. Additional tabbed views can be added
to this window which display user-defined plots. At the top of the model-view pane is a tab bar
containing modeling-stage tabs. When you click on a modeling-stage tab, a tool bar will open; this
contains buttons that access model tool panes. The tool bar for the model Build tool is shown in
Figure 2.6. When you click on a button, this opens a modeling-stage pane; these panes contain all
the tools you will need to create and run your model.
You can use the View menu to manipulate any view pane (e.g., translate or rotate the view, increase
or decrease the size of the view, turn on and off the model axes). The View menu is also available as
a tool bar that can be turned on from the Show menu. The View tool bar is shown on the model-view
pane in Figure 2.6.
An overview of the GIIC operation is provided in the Help menu. The menu also contains a list of
Frequently Asked Questions about the GIIC and an index to all GIIC Help files.
2 - 20
Users Guide
Figure 2.6
The text field with the flac: prompt located at the bottom of the Console pane allows you to enter
FLAC commands directly from the GIIC. The Console pane will echo the commands that you enter.
You should not need to use the command line at all; it is provided as a shortcut if you prefer to type
a command rather than use the graphical interface. A status bar is located at the bottom of the main
window and displays information related to the currently active view or tool.
There is also a Fish Editor pane available in the GIIC that allows you to create new FISH functions,
edit existing functions and specify FISH parameters. This window can be opened from the Show
menu.
A Model Options dialog box will appear every time you start the GIIC or begin a new model
project. The dialog is shown in Figure 2.6. This dialog identifies which optional modes of analysis
are available to you in your version of FLAC. (Note that dynamic analysis, thermal analysis, twophase flow analysis, creep models and C++ user-defined models are separate modules that can be
activated at an additional cost per module.) The FLAC Configuration Options must be selected at
the beginning of a new analysis, while the User Interface Options (structural elements, advanced
material models and factor-of-safety calculation) can be included at any time in the model run.
You can select a system of units for your analysis in the Model Options dialog. Many parameters
will then be labeled with the corresponding units, and predefined values, such as gravitational
magnitude and properties within the material database, will be converted to the selected system.
The selection for system of units must be done at the beginning of the analysis.
GETTING STARTED
2 - 21
If you are a new user, or only intend to perform a simple static analysis, we recommend that you
click the Ok button in the Model Options dialog to access the basic FLAC features. In this case, only
the null, isotropic elastic and Mohr-Coulomb models are active, and a static, plane-strain analysis
is performed in the GIIC. If you wish to come back later in the analysis and, for example, add
structural elements, click File / Model Options in the main menu. This will reopen the Model
Options dialog. Check Include Structural Elements? and click Ok . A Structure tab will be added to
the modeling-stage tab bar, and structural elements can now be included in your model.
The final model option that can be selected is the format for the project record that is used in the
Record pane. Two types of format are provided: a List Record format and a Project Tree Record
format. The List format is a simple record with independent save files. Each save file includes a
record of all the commands needed to generate the state. The Project Tree format shows changes
between save files. Save states are displayed in a tree structure.
The Model Options dialog is shown below in Figure 2.7 with the following model options selected:
groundwater configuration option with automatic adjustment of total stresses for external porepressure change (CONFIG gw ats), structural elements user-interface option, Project Tree Record
format and SI system of units.
Figure 2.7
2 - 22
Users Guide
Build
tab, and
Alter
tab.
Next, material models and properties are assigned to the zones in the model,
using the tools accessed from the Material tab.
Boundary and initial conditions are applied via the
In Situ
tab.
The Utility tab provides tools to monitor model variables and access existing
FISH functions.
The Settings tab allows model conditions to be set or changed during the analysis.
All plotting facilities in FLAC are accessible via the
Calculations are performed using tools from the
Run
Plot
tab.
tab.
Note that model conditions can be changed at any point in the solution process by re-entering a
modeling-stage tab. For example, model properties can be changed at any time via the Material
tab, and pressure or stress alterations can be made via the In Situ tab. Also, if you select structural
elements in the Model Options dialog, a Structure tab will be included in the modeling-stage tab bar
to access structural support for the model.
When you click on each of the modeling-stage tabs, a tool bar will appear that provides access to
model tool panes in which you can perform operations related to that tool. The Build tab tool bar is
shown in Figure 2.6. Next, a simple tutorial is given to provide an introduction to the model tools
and to help you become acquainted with the GIIC operation.
GETTING STARTED
2 - 23
Figure 2.8
2 - 24
Users Guide
We now begin the model creation. To set up the initial finite difference grid, we click on the Grid
button from the Build modeling-tool tab. This tool invokes the GRID command. We press Ok in the
How many zones? dialog to select the default grid of 10 zones in the i-direction by 10 zones in the
j -direction. A plot of the grid will immediately be shown in the Model-view pane. We will use SI
units for this example (see Section 2.8 for selection of system of units). The model domain is then
10 m by 10 m. Click on the View / Showaxis values menu item to show the x- and y-axes for the
model.
We next create the circular tunnel by shaping the grid to fit the tunnel boundary. To avoid errors in
calculation of gridpoint masses, all grid shaping should be done before the computational process
begins; these errors may occur if the grid is shaped after computational stepping (see Section 2.6.1
for further discussion). Grid shaping is done by clicking on the Shape button from the Alter modelingtool tab. A plot of the grid appears with a set of tools that we can use to add shapes to the grid. We
select the Circle radio button, move the mouse to a position on the grid corresponding to the tunnel
center, and press and hold the left mouse button while moving the mouse. A circle tool will appear
with two red boxes, one at the centroid and one along the circle periphery (see Figure 2.9). We can
move the circle and adjust its radius by pressing and holding the left mouse button while the mouse
is positioned within each box. Alternatively, we can select values for the centroid coordinates and
the circle radius with dialogs that open when we right-click the mouse while it is positioned within
each box. The circle in Figure 2.9 is centered at x = 5.0, y = 5.0 and has a radius of 2.0 m.
Figure 2.9
Circle
button active
GETTING STARTED
2 - 25
When we press Generate , the grid is deformed to fit the boundary of the circle, and the corresponding
GENERATE command is displayed in the Changes sub-pane to the right of the grid plot. Note
that this is a virtual grid: any alterations we make within this grid can be undone or changed.
We simply press one of the arrow keys above the Changes pane to remove (or add) a command
corresponding to the shape created in the virtual grid.
Once we are satisfied with the alteration, we press the Execute button. This sends the command(s)
to FLAC and returns to the Model-view pane. The FLAC commands are processed, and the altered
FLAC grid with marked gridpoints is displayed, as shown in Figure 2.10. The FLAC commands
created thus far are shown in the List Record pane in this figure.
Figure 2.10 FLAC grid with zones shaped for circular tunnel
2 - 26
Users Guide
We next move to the Material modeling-tool tab and press the Assign button to create and assign materials and their properties to the zones within the grid. Materials are created from a Material dialog
that is opened by pressing the Create button in the Assign pane. Within this dialog, we can assign
a classification and material name, prescribe a constitutive model type (elastic or Mohr-Coulomb)
and assign material properties. Soils and rocks can be divided into different classifications, such
as Tunnel rock, with separate material names within a classification, such as strong rock and
weak rock. The classification and material name are used to associate a GROUP name with each
material. We will create two different materials for this analysis: a strong rock and a weak rock.
The Material dialog with the selected properties for strong rock is shown in Figure 2.11. The dialog
for weak rock is similar except that the cohesion is zero.*
Assign
tool
We press Ok in the Material dialog to create the material. The material is added to a material List
shown on the right side of the Assign pane. Once all the materials required for an analysis have
been created and added to the list, they can be assigned to the grid. It is possible to assign different
materials to different zones in the grid, or to different marked regions of the grid, using the Range
tools provided in the Assign pane. In our example, we will evaluate the response of the tunnel
in strong rock versus weak rock, so we begin by assigning strong rock material to all zones. We
highlight the Tunnel:strong rock item and press the SetAll button to assign this material to all nonnull zones in the grid. Figure 2.12 shows the Assign pane with the strong rock material assigned.
GROUP, MODEL and PROPERTY commands are listed in the Changes pane when the materials are
assigned. We now press Execute to send these commands to FLAC.
* A database of common soil and rock materials and properties is also available by pressing the Database
button in the lower-right corner of the Assign pane. The database is divided into classification
groups and material names. You can also create your own database of common materials within
this database tool, which can be saved and loaded for other projects. Database materials are stored
in a file with extension .GMT see Section 2.10.
GETTING STARTED
Figure 2.12 Strong rock material assigned to all zones with the
2 - 27
SetAll
button
The next step is to assign the boundary conditions for our model. We select the Fix button from
the In Situ modeling-tool tab. We wish to have roller boundary conditions applied to the bottom
and sides of the model. To prescribe a roller boundary on the bottom, we press the Y radio button
to specify a fixed-gridpoint velocity in the y-direction. By default, the y-velocity is zero and, by
specifying that this velocity is fixed at the selected gridpoints, we are preventing any movement in
the y-direction. We hold down the left mouse button while dragging the mouse along the bottom
boundary. Gridpoints are marked and, when we release the button, a letter denoting the y-fixity
condition is printed at the selected gridpoints. We repeat the process using the X radio button
to specify a fixed-gridpoint velocity in the x-direction along the left and right boundaries. The
resulting boundary conditions are shown in Figure 2.13. Press Execute to send these commands to
FLAC.
2 - 28
Users Guide
Fix
tool
We can access different variables in our model with the Utility modeling-tool tab. We wish to
monitor the displacement at the ground surface as the tunnel is excavated. To do this we click on
the History button to open the History pane, and then click on the GP mode radio button. We select
the y-displacement history from the History Information sub-menu, and then we point the mouse
at a gridpoint on the top of the model. When we click on the gridpoint, a HISTORY command is
created for the y-displacement history at that gridpoint. Figure 2.14 shows the results of our action
in the History pane. Press Execute to send the command to FLAC.
GETTING STARTED
2 - 29
Hist
tool
Gravitational loading is specified as a global setting in our model via the Settings modeling-tool
tab. We click on the Gravity button to access the Gravity Settings dialog. Then, by clicking on the
globe icon in the dialog, the value of 9.81 m/sec2 will be listed as the magnitude of gravitational
acceleration. (You can also type in a different value for the magnitude.) The dialog is shown in
Figure 2.15. Note that the gravitational vector is shown by an icon in the model view.
2 - 30
Users Guide
We anticipate that large deformations will occur in this analysis, so we click on the Mech button from
the Settings tab to access the Mechanical Settings dialog. We press the Large-Strain radio button to
set the large-strain logic. Figure 2.16 shows the Mechanical Settings dialog.
Figure 2.16 Set global mechanical settings in the Mechanical Settings dialog
We are now ready to bring the model to an initial equilibrium state. We timestep the model to a force
equilibrium condition under gravity loading. The solution approach of timestepping to equilibrium
is described is Section 2.6.4. We press the Run modeling-tool tab and then the Solve button. This
opens a Solve tool; we press OK and invoke the SOLVE command to detect equilibrium automatically.
The Solve dialog appears and the timestep number, maximum unbalanced force and equilibrium
ratio are displayed. The equilibrium ratio is used to determine equilibrium (see Section 2.6.4 for
details). When the ratio falls below the default limiting value of 103 , the calculation stops. Other
limiting conditions can also be prescribed, as described in Section 2.6.4.
There are several ways to check that equilibrium has been reached. A quick check can be made
by plotting the change in maximum unbalanced force during stepping. Press the Plot modelingtool tab, then the Quick button, and finally the Unbalanced force item, and a plot of unbalanced force
versus accumulated timestep will appear. The plot given in Figure 2.17 shows that the maximum
unbalanced force is approaching zero, which indicates that an equilibrium state is reached.
GETTING STARTED
2 - 31
Quick
button
It is a good idea to save the project state at the different stages of our analysis. In this way we can
easily return to a given state and make modifications without the need to run the entire simulation
again. We can save our project model state at the initial-equilibrium stage by pressing the SaveState
button in the Run tool bar. This opens a dialog box that allows the user to give a descriptive title
to the saved state, select a directory in which to save the model-state file, and name the file. By
default, the file has the extension .SAV. The save file is described in Section 2.10.
We choose to save the model state as TUN1.SAV. We must save this file in the same directory as
the project file TUNNEL.PRJ so that the project can be opened later and list all associated save
files. The save file is added to a list at the top of the List Record pane. Each time we save the model
state, a new save file will be added to the list. We can click on any file in the list to open that saved
state.
We can create plots for a wide variety of variables in a FLAC model. Click on the Model button
from the Plot tab to open the FLAC Plot Items dialog box. The dialog is shown in Figure 2.18.
This dialog accesses most of the general plotting facilities in FLAC. (Note that separate tools are
provided for table, history, profile and failure plots in the Plot tool bar.) For example, if we wish
to examine the gravitational stresses that develop in the model, we can create a contour plot of
yy -stresses. Click on the Contour-Zone / Total Stress / syy plot item from the Plot Items tree,
and add this to the Add Plot Items list. Then click on the Geometry / boundary plot item and add
this to the list. We can either create a fill-contour plot or a line-contour plot. By default, a filled
contour plot is created with the contour range denoted by the fill colors. The resulting fill-contour
plot is shown in Figure 2.19.
2 - 32
Users Guide
GETTING STARTED
2 - 33
We can make a hardcopy plot of any FLAC model plot we choose. To do this, click on the File/Print
Plot menu item in the main menu. If the current Windows default printer is connected to the LPT1
port, we can send the currently active plot view directly to the printer by clicking this menu item. The
File/Print Plot Setup menu item can be used to change the printer device settings. Figure 2.20
shows the Plot Hardcopy Settings dialog with the selections for device settings.
2 - 34
Users Guide
GETTING STARTED
2 - 35
2 - 36
Users Guide
Next, we evaluate the tunnel response using the weak-rock material. We first need to return to a
previous model stage. We return to the state at which the tunnel is introduced (TUN2.SAV) by
clicking on this file name in the Record pane. (Note that we should actually return to the initial
equilibrium stage, TUN1.SAV, and re-calculate an initial stress state that corresponds to the weakrock strength properties and gravitational loading. In this simple exercise, this initial stress state is
the same as that for the strong rock.)
We return to the Assign pane and click on Tunnel: weak rock in the material list to highlight this
material. We click on the SetAll button to change all of the non-null zones from strong rock to weak
rock. The result is shown in Figure 2.24.
Figure 2.24 Weak rock material assigned to all zones with the
SetAll
button
For analyses in which we anticipate that material failure can occur and the simulation may never
reach an equilibrium state, we do not use the Solve tool. Instead, we use the Cycle tool in the Run tab
in order to step through the simulation and monitor the response as it occurs. After pressing Cycle ,
we enter 1000 cycles for the calculation duration and press Ok . FLAC will now step through 1000
timesteps. When stepping is finished, the model plot is refreshed automatically, and because we are
running in large-strain mode, we observe that the top of the grid has begun to deform downward.
See Figure 2.25. If we continue stepping, eventually FLAC will report an error message (Bad
Geometry) and the calculation will stop. This indicates that zones in the model have reached a
limiting distortion; the limiting conditions for zone distortion are described in Section 2.6.1.
GETTING STARTED
2 - 37
Figure 2.25 Weak rock: deformed grid at 1000 timesteps after tunnel excavation
There are different ways to monitor the collapse process. For example, if we plot the history of ydisplacement at gridpoint i = 6, j = 11, which we recorded at the beginning of the simulation, we can
identify collapse by the increasing displacement that is displayed. Press the History button in the Plot
tab, click on Item ID number 1 (which is the history number corresponding to the y-displacement
history we selected), and press Ok . A plot of the y-displacement history versus accumulated timestep
will appear, as shown in Figure 2.26. The displacement is increasing at a constant rate, indicating
collapse. (We made this plot view full-screen by clicking off Show / Resources.)
2 - 38
Users Guide
GETTING STARTED
2 - 39
2.3 Nomenclature
FLAC uses nomenclature that is consistent, in general, with that used in conventional finite difference
or finite-element programs for stress analysis. The basic definitions of terms are reviewed here for
clarification. Figure 2.27 is provided to illustrate the FLAC terminology.
water table
hydraulic
pressure
attached gridpoints
model
boundary
ter
fa
ce
horizontal
boundary
stress
in
structural cable
internal
boundary
(excavation)
zone
structural beam
gridpoint
fixed
bottom
boundary
2 - 40
Users Guide
GRIDPOINT Gridpoints are associated with the corners of the finite difference zones. There
are always four (4) gridpoints associated with each zone. In the FLAC model, a pair of x- and ycoordinates are defined for each gridpoint, thus specifying the exact location of the finite difference
zones. Other terms for gridpoint are nodal point and node.
FINITE DIFFERENCE GRID The finite difference grid is an assemblage of one or more finite
difference zones across the physical region which is being analyzed. Another term for grid is mesh.
MODEL BOUNDARY The model boundary is the periphery of the finite difference grid. Internal
boundaries (i.e., holes within the grid) are also model boundaries.
BOUNDARY CONDITION A boundary condition is the prescription of a constraint or controlled
condition along a model boundary (e.g., a fixed displacement or force for mechanical problems,
an impermeable boundary for groundwater flow problems, adiabatic boundary for heat transfer
problems, etc.).
INITIAL CONDITIONS This is the state of all variables in the model (e.g., stresses or pore
pressures) prior to any loading change or disturbance (e.g., excavation).
CONSTITUTIVE MODEL The constitutive (or material) model represents the deformation
and strength behavior prescribed to the zones in a FLAC model. Several constitutive models are
available in FLAC to assimilate different types of behavior commonly associated with geologic
materials. Constitutive models and material properties can be assigned individually to every zone
in a FLAC model.
SUB-GRID The finite difference grid can be divided into sub-grids. Sub-grids can be used to
create regions of different shapes in the model (e.g., the dam sub-grid on the foundation sub-grid
in Figure 2.27). Sub-grids cannot share the same gridpoints with other sub-grids; they must be
separated by null zones.
NULL ZONE Null zones are zones that represent voids (i.e., no material present) within the
finite difference grid. All newly created zones are null by default.
ATTACHED GRIDPOINTS Attached gridpoints are pairs of gridpoints that belong to separate
sub-grids that are joined together. The dam is joined to the foundation along attached gridpoints
in Figure 2.27. Attached gridpoints do not have to match between sub-grids, but sub-grids cannot
separate from one another once attached.
INTERFACE An interface is a connection between sub-grids that can separate (e.g., slide or
open). An interface can represent a physical discontinuity such as a fault or contact plane. It can
also be used to join sub-grid regions that have different zone sizes.
MARKED GRIDPOINTS Marked gridpoints are specially designated gridpoints that delimit a
region for the purpose of applying an initial condition, assigning material models and properties,
and printing selected variables. The marking of gridpoints has no effect on the solution process.
REGION A region in a FLAC model refers to all zones enclosed within a contiguous string of
marked gridpoints. Regions are used to limit the range of certain FLAC commands, such as the
MODEL command that assigns material models to designated regions.
GETTING STARTED
2 - 41
GROUP A group in a FLAC model refers to a collection of zones identified by a unique name.
Groups are used to limit the range of certain FLAC commands, such as the MODEL command that
assigns material models to designated groups. Any command reference to a group name indicates
that the command is to be executed on that group of zones.
STRUCTURAL ELEMENT Structural elements are linear elements used to represent the interaction of structures (such as tunnel liners, rock bolts, cable bolts or support props) with a soil or
rock mass. Some restricted material nonlinearity is possible with structural elements. Geometric
nonlinearity occurs in large-strain mode.
STEP Because FLAC is an explicit code, the solution to a problem requires a number of computational steps. During computational stepping, the information associated with the phenomenon
under investigation is propagated across the zones in the finite difference grid. A certain number
of steps is required to arrive at an equilibrium (or steady-flow) state for a static solution. Typical
problems are solved within 2000 to 4000 steps, although large complex problems can require tens
of thousands of steps to reach a steady state. When using the dynamic analysis option, STEP refers
to the actual timestep for the dynamic problem. Other terms for step are timestep and cycle.
STATIC SOLUTION A static or quasi-static solution is reached in FLAC when the rate of change
of kinetic energy in a model approaches a negligible value. This is accomplished by damping the
equations of motion. At the static solution stage, the model will be either at a state of force
equilibrium or at a state of steady-flow of material if a portion (or all) of the model is unstable
(i.e., fails) under the applied loading conditions. This is the default calculation in FLAC.* Static
mechanical solutions can be coupled to transient groundwater flow or heat transfer solutions. (As
an option, fully dynamic analysis can also be performed by inhibiting the static solution damping.)
UNBALANCED FORCE The unbalanced force indicates when a mechanical equilibrium state
(or the onset of plastic flow) is reached for a static analysis. A model is in exact equilibrium if the
net nodal force vector at each gridpoint is zero. The maximum nodal force vector is monitored in
FLAC and printed to the screen when the STEP or SOLVE command is invoked. The maximum nodal
force vector is also called the unbalanced or out-of-balance force. The maximum unbalanced force
will never exactly reach zero for a numerical analysis. The model is considered to be in equilibrium
when the maximum unbalanced force is small compared to the total applied forces in the problem.
If the unbalanced force approaches a constant nonzero value, this probably indicates that failure
and plastic flow are occurring within the model.
* The mistaken notion exists in some finite element (FE) literature that a dynamic solution method
cannot produce a true equilibrium state, compared to an FE solution, which is believed to satisfy
perfectly the set of governing equations at equilibrium. In fact, both methods only satisfy the
equations approximately, but the level of residual errors can be made as small as desired. In FLAC,
the level of error is objectively quantified as the ratio of unbalanced force at a gridpoint to the mean
of the set of absolute forces acting at the gridpoint. This measure of error is very similar to the
convergence criteria used in FE solutions. In both cases the solution process is terminated when
the error is below a desired value.
2 - 42
Users Guide
DYNAMIC SOLUTION For a dynamic solution, the full dynamic equations of motion (including
inertial terms) are solved; the generation and dissipation of kinetic energy directly affect the solution.
Dynamic solutions are required for problems involving high frequency and short duration loads
e.g., seismic or explosive loading. The dynamic calculation is an optional module to FLAC (see
Section 3 in Optional Features).
LARGE-STRAIN/SMALL-STRAIN By default, FLAC operates in small-strain mode: that is,
gridpoint coordinates are not changed, even if computed displacements are large (compared to
typical zone sizes). In large-strain mode, gridpoint coordinates are updated at each step, according
to computed displacements. In large-strain mode, geometric nonlinearity is possible.
GETTING STARTED
2 - 43
(*10^1)
LEGEND
13-Apr- 0 13:50
step
0
-3.333E+00 <x< 2.333E+01
-3.333E+00 <y< 2.333E+01
1.500
Grid plot
0
5E 0
1.000
0.500
0.000
0.500
1.000
(*10^1)
1.500
2.000
2 - 44
Users Guide
JOB TITLE :
1,7
2,7
3,7
4,7
5,7
6,7
7,7
6.500
1,6
2,6
3,6
4,6
5,6
6,6
7,6
5.500
1,5
2,5
3,5
4,5
5,5
6,5
7,5
4.500
1,4
2,4
3,4
4,4
5,4
6,4
7,4
3.500
1,3
2,3
3,3
4,3
5,3
6,3
7,3
2.500
1,2
2,2
3,2
4,2
5,2
6,2
7,2
1.500
1,1
2,1
3,1
4,1
5,1
6,1
7,1
0.500
LEGEND
13-Apr- 0 13:51
step
0
-1.000E+00 <x< 7.000E+00
-1.000E+00 <y< 7.000E+00
Grid plot
0
2E 0
Zone Numbers
-0.500
0.500
1.500
2.500
3.500
4.500
5.500
6.500
6.500
1,7
2,7
3,7
4,7
5,7
6,7
7,7
1,6
2,6
3,6
4,6
5,6
6,6
7,6
1,5
2,5
3,5
4,5
5,5
6,5
7,5
1,4
2,4
3,4
4,4
5,4
6,4
7,4
LEGEND
5.500
13-Apr- 0 13:51
step
0
-1.000E+00 <x< 7.000E+00
-1.000E+00 <y< 7.000E+00
4.500
Grid plot
0
2E 0
3.500
Gridpoint Numbers
2.500
1,3
2,3
3,3
4,3
5,3
6,3
7,3
1,2
2,2
3,2
4,2
5,2
6,2
7,2
1,1
2,1
3,1
4,1
5,1
6,1
7,1
1.500
0.500
-0.500
0.500
1.500
2.500
3.500
4.500
5.500
6.500
GETTING STARTED
2 - 45
In normal operation, the finite difference mesh origin is the lower left-hand corner of the grid. By
default, the x-coordinates of the gridpoints are 0, 1, . . . , p, and the y-coordinates are 0, 1, . . . , q.
The coordinates are indicated by the scales shown on the plots in Figures 2.28 and 2.29.
Grid generation with FLAC involves the shaping of the row-and-column grid to fit the shape of the
physical domain. Grid generation is described in Section 3.2.
The finite difference grid also identifies the storage location of all state variables in the model. The
procedure followed in FLAC is that all vector quantities (e.g., forces, velocities, displacements, flow
rates) are stored at gridpoint locations, while all scalar and tensor quantities (e.g., stresses, pressure,
material properties) are stored at zone centroid locations. There are three exceptions: saturation
and temperature are considered gridpoint variables; and pore pressure is stored at both gridpoint
and zone centroid locations.
2 - 46
Users Guide
* The commands and their meanings are presented in Section 1.3 in the Command Reference; a
summary is given in Section 1 in the Command and FISH Reference Summary.
GETTING STARTED
2 - 47
2 - 48
Users Guide
Start
MODEL SETUP
1. Generate grid, deform to desired shape
2. Define constitutive behavior and material properties
3. Specify boundary and initial conditions
Examine
the model response
Results unsatisfactory
Step to solution
More tests
needed
Examine
the model response
Acceptable result
Yes
Parameter
study needed
No
End
GETTING STARTED
2 - 49
icol jrow
where icol is the number of columns of zones, and jrow is the number of rows of zones in the mesh.
Be careful when selecting the number of zones for a model, because a balance must be struck
between the accuracy required and the solution speed. The calculation speed to reach a solution
varies directly as a function of the number of elements. As a rule of thumb, models containing up to
roughly 5000 elements will typically reach a solution state for a given alteration in approximately
2000 to 4000 steps. On a 300 MHz Pentium II microcomputer, the runtime for a 5000-element
model to perform 4000 steps is roughly 3 minutes. Check the speed of calculation on your computer
for the specific model to estimate the runtime required. A runtime benchmark test is provided in
Section 5.1.
It is best to start with a grid that has few zones (say, 100 to 500) to perform simple test runs and
make refinements to the model. Then, increase the number of zones to improve the accuracy.
Two commands are used in FLAC to shape the grid:
generate
initial
The GENERATE command creates regions of different shapes within the grid. The INITIAL command
changes the x- and y-coordinates of selected gridpoints. The complete descriptions for these
commands are given in Section 1.3 in the Command Reference. The following examples illustrate
their use.
Example 1 In its simplest form, the GENERATE command can supply new coordinates to a grid.
By entering the commands* in Example 2.2, a square grid of 10 zones by 10 zones (11 gridpoints
by 11 gridpoints) will be created, and each zone will be assigned the elastic material model.
Example 2.2 Generating a simple grid
grid 10 10
model elastic
If the coordinates of the grid are printed at this stage, by issuing the command
print x y
* If you want to try entering the command examples interactively from the text mode, type NEW
each time you start a new example. In the GIIC, press the File / New Project item in the main
menu. This will initialize FLAC without having to exit and reload the program for a new model. In
text mode, type PLOT grid after entering each example to view the result. The model view will be
displayed automatically in the GIIC.
2 - 50
Users Guide
you will see that both x and y run from 0.0 to 10.0 (i.e., FLAC assigns a square grid with 1 unit
spacing between gridpoints). Note that the MODEL command must be issued before the PRINT
command, otherwise the grid coordinates will not be displayed. This is also true for the PLOT
command. There must be a material present for information to be printed or plotted.
If the actual coordinates of the grid are to run from 0.0 to 500.0 in the x-direction and from 0.0 to
1000.0 in the y-direction, the GENERATE command is issued as follows:
gen
i=1,11 j=1,11
Note that the four corner coordinates for the portion of the mesh defined by i = 1,11, j = 1,11 start at
the lower left-hand corner of the grid and work around its outer corners in a clockwise fashion. All
gridpoints interior to these corner points will have their coordinates reassigned based on the corner
point coordinates. Now, print out the coordinates again to see that the coordinates have indeed
been changed. Note that just a portion of the grid can be given new coordinates. The portion of the
grid is defined by the i, j range (see Example 2.3). The corner coordinates must be specified in a
clockwise fashion.
Example 2 The GENERATE command can be used to create distortions in the grid. For example,
try the commands in Example 2.3.
Example 2.3 Distorting the grid
new
grid 20,20
model elas
gen 0,5 0,20 20,20 5,5 i=1,11
gen same same 20,0 5,0 i=11,21
plot hold grid
In this example, only a portion of the grid is distorted with each GENERATE command. The
first GENERATE command creates a distorted quadrilateral from half of the grid, while the second
command wraps the remainder of the grid around to form a rectangular opening. Successive
GENERATE commands are additive i.e., once changed, the coordinates of the grid remain at the
new coordinates until changed again by using the GENERATE or INITIAL commands. In the second
GENERATE command, the word same is used twice, which indicates that coordinates for the first
two corner points are not changed. When you type PLOT grid, the distorted grid shape should be
displayed.
GETTING STARTED
2 - 51
Example 3 The GENERATE command can be used to grade a mesh to represent far boundaries.
For example, in many cases, an excavation is to be created at a great depth in a rock mass. Detailed
information on the stresses and displacements is to be determined around the excavation, where the
disturbance is large, but little detail is necessary at greater distances. In the following example, the
lower left-hand portion of the grid is left finely discretized, and the boundaries are graded outward
in the x- and y-directions. Try issuing the commands in Example 2.4.
Example 2.4 Grading the mesh
new
grid 20,20
m e
gen 0,0 0,100 100,100 100,0 rat 1.25 1.25
plot hold grid
The GENERATE command forces the grid lines to expand to 100.0 units at a rate 1.25 times the
previous grid spacing in the x- and y-directions. (Example 2.4 also illustrates that command words
can be truncated: MODEL elas becomes M e.) Note that if the ratio entered on the GEN command is
between 0 and 1, the grid dimensions will decrease with increasing coordinate value. For example,
issue the commands in Example 2.5.
Example 2.5 Applying different gradients to a mesh
new
gr 10,10
m e
gen -100,0 -100,100 0,100 0,0 rat .80,1.25
plot hold grid
You will see a grid graded in the negative x- and positive y-directions.
Example 4 Excavations often need to be created in the grid. It is very tedious to create complex
excavation shapes, especially circular arcs, by simply moving individual gridpoints. Special shape
functions are built into the GENERATE command (e.g., circles, arcs and lines). An example is given
here for the creation of excavation shapes using the GENERATE command.
First, a circular excavation is created. Try the commands in Example 2.6.
Example 2.6 Creating a circular hole in a grid
new
grid 20,20
m e
gen circle 10,10 5
2 - 52
Users Guide
This command automatically creates a circular opening within the grid, centered at (x = 10, y = 10)
with a radius of 5.0. Note that the remainder of the mesh remains square (i.e., element corners are
at 90 degrees). Note, too, that the MODEL command must be specified first in order for the shape
functions (circle, arc, line) to work.
To cause the mesh to conform better to the new opening, type
gen adjust
plot grid
Successive GENERATE adjust commands will smooth the grid to increasingly greater levels.
When creating internal shapes within the grid using the GENERATE circle, GENERATE arc or GENERATE line commands, FLAC distinguishes between the various regions of the grid created by marking
closed paths. In the previous example, the GENERATE circle command creates two regions within
the grid created by the boundary of the circle: the region inside the boundary and that outside. If
you wish to see where the boundaries of the grid are, type
print mark
Those gridpoints which have been adjusted by FLAC to conform to boundaries are signified by an
M in the printout.
CAUTION: Two regions can only be formed if they are separated by closed contours. In other
words, a line segment which begins and ends within the grid and does not form a closed boundary,
subsequently will result in only one region.
Example 5 The INITIAL command can be used to move a point or a number of points from the
present location to a new one. The following commands in Example 2.7 create a grid and distort it
using the INITIAL command.
Example 2.7 Moving gridpoints with the INITIAL command
new
grid 5 5
model elastic
gen 0,0 0,10 10,10 10,0
ini x=-2 i=1 j=6
ini x=12 i=6
plot hold grid
The GENERATE command assigns coordinates to gridpoints from 0 to 10 in the x- and y-directions.
The first INITIAL command moves the upper left-hand corner horizontally by -2 units. The second
GETTING STARTED
2 - 53
INITIAL command moves the right-hand boundary gridpoints to the right by 2 units. Note that since
the j -range is not given, the entire range is assumed.
The INITIAL command can be used to move any gridpoint to any position. Of course, elements
cannot overlap. If this happens, a warning message referring to BAD GEOMETRY will be
given, and FLAC will not continue execution until the errors in grid construction are rectified.*
A practical limit on the aspect ratio of zones should be kept to about 1:10 or less for reasonable
solution accuracy.
During model solution, a quadrilateral may be deformed in any fashion, subject to the following
criteria:
(1) the area of the quadrilateral must be positive; and
(2) each member of at least one pair of triangular sub-zones which comprise the
quadrilateral must have an area greater than 20% of the total quadrilateral area
(see Section 1.3.2 in Theory and Background).
These criteria should be applied when creating zones to avoid bad geometry during model solution.
If either of these criteria is not met, FLAC will give a BAD GEOMETRY error message during
timestepping. Figure 2.31 illustrates possible zone deformations.
2
1
acceptable
deformed
geometry
initial
geometry
1
unacceptable
deformed
geometry
2 - 54
Users Guide
WARNING: All grid shaping to create holes or new boundaries (e.g., slope faces) that will be
removed, or excavated, at a later state in the solution must be performed before the computational
stepping begins. The GENERATE and INITIAL commands should not be used to adjust the grid after
the STEP or SOLVE command is issued. (These commands are described below.) This adjustment
can introduce an erroneous calculation for gridpoint masses in the model. If it is necessary to move
gridpoints after stepping has begun, a velocity can be applied to the gridpoint for a specified number
of steps to move the required displacement.
2.6.2 Assigning Material Models
Once the grid generation is complete, one or more material models and associated properties
must be assigned to all zones in the model. This is done by using two commands: MODEL and
PROPERTY. FLAC has ten (10) built-in material models; these are described in Section 2 in Theory
and Background. Three models are sufficient for most analyses the new user will make. These
are MODEL null, MODEL elastic and MODEL mohr.
MODEL null represents material which is removed or excavated from the model. MODEL elastic
assigns isotropic elastic material behavior, and MODEL mohr assigns Mohr-Coulomb plasticity
behavior.
MODEL elastic and MODEL mohr require that material properties be assigned via the PROPERTY
command. For the elastic model, the required properties are:
(1) density;
(2) bulk modulus; and
(3) shear modulus.
NOTE: Bulk modulus, K, and shear modulus, G, are related to Youngs modulus, E, and Poissons
ratio, , by
K =
E
3(1 2)
(2.1)
G =
E
2(1 + )
(2.2)
or
E =
9KG
3K + G
(2.3)
3K 2G
2(3K + G)
(2.4)
GETTING STARTED
2 - 55
10,10
elas j=6,10
den=2000 bulk=1e8 shear=.3e8 j=6,10
mohr j=1,5
den=2500 bulk=1.5e8 shear=.6e8 j=1,5
fric=30 coh=5e6 ten=8.66e6 j=1,5
hold model
Instead of using i, j indices to specify a range, the word (i.e., keyword) region can be used. For
example, to excavate the circular tunnel in Example 2.6, the following command can be applied:
model
By specifying one zone inside the marked tunnel region (e.g., zone i = 10, j = 10), then all zones
within the tunnel are set to null material (i.e., excavated).
The tunnel can be filled at a later stage by typing, for example,
model
Note that the excavation can be replaced by any model, and with properties consistent with the
model.
2 - 56
Users Guide
Command
Effect
APPLY
pressure
sxx
sxy
syy
xforce
yforce
xvel
yvel
FIX
pp
x
y
NOTE:
1. The FREE command is used to release the constraint set by the FIX command.
2. In order to assign a fixed-displacement boundary condition, only the FIX x
and/or FIX y commands are needed, provided that the velocity at the selected
gridpoint is zero.
3. See Section 1.3 in the Command Reference for a complete listing of APPLY
and FIX keywords.
GETTING STARTED
Table 2.4
2 - 57
Command
INITIAL
Effect
pp
sat
sxx
sxy
syy
szz
xvel
yvel
xdis
ydis
10 10
el
x i=1
x i=11
y j=1
press = 10 j=11
sxx=-10 syy=-10
hold bou fix apply stress
The grid has the left- and right-hand sides fixed from movement in the x-direction, and the bottom
fixed in the y-direction. A pressure is applied to the top boundary, and all zones in the model have
an initial stress: xx = yy = 10. In FLAC, compressive stresses have a negative sign, while
compressive pressure is positive. All these conditions are displayed with the PLOT command. The
applied pressure is displayed as force vectors. The stresses are shown as principal stress tensors.
The boundary of the grid is also shown.
2 - 58
Users Guide
unbal
Additionally, the history of selected variables (e.g., velocity or displacement at a gridpoint) may be
recorded. The following commands are examples:
hist
hist
The first history records x-velocity at gridpoint (5,5), while the second records y-displacement at
gridpoint (5,11).
After running several hundred (or thousand) calculation steps, a history of these records may be
plotted to indicate the equilibrium condition. The data file in Example 2.10 illustrates this process.
GETTING STARTED
2 - 59
10 10
el
d=1800 bulk=1e8 shear =.3e8
x i=1
x i=11
y j=1
pres=1e6 j=11
unbal
ydisp i=5 j=11
900
The initial maximum unbalanced force is 1 MN. After 900 steps, this force has dropped to approximately 100 N. By plotting the two histories, it can be seen that the maximum unbalanced force has
approached zero, while the displacement has approached a constant magnitude of approximately
0.07 m.
Type
plot hold hist 1
plot hold hist 2
to view these plots. The number following PLOT hist corresponds to the order in which the histories
are entered in the data file. Figures 2.32 and 2.33 show the unbalanced force and displacement
history plots.
2 - 60
Users Guide
JOB TITLE :
+05
LEGEND
14-Apr- 0 11:44
step
900
HISTORY PLOT
Y-axis :
Max. unbal. force
X-axis :
Number of steps
5.000
4.000
3.000
2.000
1.000
9
(10
+02
JOB TITLE :
-02
LEGEND
14-Apr- 0 11:45
step
900
HISTORY PLOT
Y-axis :
Y displacement( 5, 11)
X-axis :
Number of steps
-2.000
-3.000
-4.000
-5.000
-6.000
-7.000
9
(10
+02
GETTING STARTED
2 - 61
Normally, displacements are initialized to zero at the initial equilibrium stage. This can be done
now by typing
ini
xdis=0 ydis=0
Type
print
xdis ydis
to confirm this.
The SOLVE command can be used in place of STEP if the user wishes FLAC to stop automatically
when the maximum unbalanced force or equilibrium ratio falls below a specified limit. Replace
STEP 900 with SOLVE and repeat the above problem. This time, FLAC should stop the calculation
at step 664. If the plots are made again, essentially the same results as given in Figures 2.32 and
2.33 will be seen.
The SOLVE command is controlled by a limiting equilibrium ratio (103 ), a limiting unbalanced
force (100 force units), a limiting number of timesteps (100,000 steps), and a limiting computer
runtime (1440 minutes), where the default values are given in parentheses. The calculation will
stop when any one of these limits is reached. In the above example, the equilibrium ratio of 103
is reached first. In order for the unbalanced force to control stepping, the command SET sratio=0
should be given before the SOLVE command. Now, the calculation will stop at a force limit of 100.
Each of the solving limits can be changed with the SET command. For example, SET force=50
will change the unbalanced force limit to 50. (Alternatively, the SOLVE force=50 command can be
given.) The limit will remain in effect until changed again, or until a NEW command is issued, which
will reset the limits to their default values. When using the SOLVE command, it is important to
check that the calculation does not stop prematurely (e.g., if the calculation is expected to take more
than 100,000 steps to reach equilibrium, then the SET step command should be used to increase the
step limit).
For the above example, an initial equilibrium stage can be achieved without stepping by simply
inserting an INITIAL command:
ini
Now, the unbalanced force is exactly zero. Type SOLVE to confirm this. Note that, in this case,
the initial displacements in the model are automatically zero. Note also that, in this example, any
initial value for sxx or szz will give an initial equilibrium.
If the initial stage is subjected to gravitational loading, this may be added via
set
gravity=9.81
where a gravitational acceleration of 9.81 m/sec2 is applied in the negative y-direction. If the above
problem is continued with gravity loading, a maximum unbalanced force of approximately 18,000
N develops, and 720 steps are required (using SOLVE) to bring the model back to equilibrium. There
is a stress gradient now in the model which can be viewed by typing
print
syy
2 - 62
Users Guide
The values for yy range from 1 MPa in the top zones of the model, to 1.166 MPa in the bottom
zones. There is also a gradient in the x-direction z-direction; type
print
sxx szz
In an analysis, it is very important that the model be at equilibrium before alterations are made.
Several histories should be recorded throughout a model to ensure that a large force imbalance does
not exist. It does not affect the analysis adversely if more steps than needed are taken to reach
equilibrium, but it will affect the analysis if an insufficient number of steps are taken.
A FLAC calculation can be interrupted at any time during stepping by pressing the <Esc> key. It
often is convenient to use the STEP command with a high step number and periodically interrupt
the stepping, check the histories, and resume stepping (with STEP continue) until the equilibrium
condition is reached.
2.6.5 Performing Alterations
FLAC allows model conditions to be changed at any point in the solution process. These changes
may be of the following form:
excavation of material;
addition or deletion of gridpoint loads or pressures;
change of material model or properties for any zone; and
fix or free velocities for any gridpoint.
Excavation is performed with the MODEL null command. Gridpoint loads can be applied at any
gridpoint with the APPLY xforce and APPLY yforce commands. Pressure or stress alterations can be
made at model boundaries with the APPLY command, as discussed previously. Material models and
properties are changed with the MODEL and PROPERTY commands. Gridpoint velocities are fixed
or freed via the FIX/FREE commands. It should be evident that several commands can be repeated
to perform various model alterations.
Try the data file in Example 2.11.
Example 2.11 Excavating a tunnel and monitoring the response
grid
model
gen
plot
gen
plot
prop
set
fix
10,10
elastic
circle 5,5 2
hold grid
adjust
hold grid
s=.3e8 b=1e8 d=1600
grav=9.81
x i=1
GETTING STARTED
2 - 63
fix
x i=11
fix
y j=1
solve
pr
mark
ini sxx 0.0 syy 0.0 szz 0.0 region 5,5
prop s
.3e5 b 1e5 d 1.6
region 5,5
;mod
null region 5,5
plot hold grid
pause
solve
plot hold str bou
This problem illustrates the alteration on stress distribution due to excavation of a circular tunnel in
an elastic material. Because the grid cannot be altered after stepping begins, it must be deformed
to fit the boundaries of the tunnel before the initial stresses are equilibrated. After excavation (i.e.,
MODEL null), an unbalanced force results, and the model is stepped to equilibrium again. The plot
of principal stress tensors shows the stress distribution resulting from the excavation.
If model zones contain a plasticity material model (e.g., MODEL mohr), it is possible that an alteration
may be such that force equilibrium cannot be achieved. In other words, the unbalanced forces in
part or all of the model cannot approach zero in which case, the maximum unbalanced force will
approach a constant nonzero value, indicating that steady-state flow of material is occurring (i.e., a
portion, or all, of the model is failing).
Example 2.12 illustrates model failure.
Example 2.12 Excavate and fill in stages
grid 10,10
m e
prop s=5.7e9 b=11.1e9 d=2000
fix x i=1
fix y j=1
fix x i=11
apply syy -20e6 j=11
ini sxx -30e6 syy -20e6 szz -20e6
his unbal
his xdis i=4 j=5
solve
mod null i 4,7 j 3,6
plot hold grid
solve
plo hold his 1
plo hold his 2
plo hold grid str
mod mohr i 4,7 j 3,6
2 - 64
prop
mod
mod
ini
his
his
his
step
plot
plot
Users Guide
This is a simple analysis of cut-and-fill mining, where excavations are created and backfilled sequentially. The boundaries are too close for an accurate solution, but the simulation illustrates
FLAC s ability to change model conditions and calculate the results in this case, the backfill
fails upon excavation of the adjacent cuts. The region of failure is indicated by the x-displacement
contour plot. The history plot shows that the gridpoint (4,5) in the backfill zone is at a constantly
increasing steady-state displacement.
2.6.6 Saving/Restoring Problem State
Two other commands, SAVE and RESTORE, are helpful when performing analyses in stages. At the
end of one stage (e.g., initial equilibrium), the model state can be saved by typing
save
file.sav
where le is a user-specified filename. The extension .SAV identifies this file as a saved file (see
Section 2.10). This file can be restored at a later time by typing
rest
file.sav
and the model state at the point at which the model was saved will be restored. It is not necessary to
build the model from scratch every time a change is made; merely save the model before the change
and restore it whenever a new change is to be analyzed. For example, in the previous example,
the state should be saved after the initial equilibrium stage. Then, the effect of different backfill
properties can be evaluated by restoring this file, changing the properties, and calculating the result.
For example, insert the following
save
fill1.sav
after the MODEL mohr command. Then create a data file of the form shown in Example 2.13 to
study the influence of the backfill.
GETTING STARTED
2 - 65
This file should be created with a text editor and called into FLAC. After the run is completed, the
saved files can be restored and evaluated separately to study the effect of the backfill properties.
2.6.7 Summary of Commands for Simple Analyses
The major command words described in Section 2.6 are summarized in Table 2.5. These are all
that are needed to begin performing simple analyses with FLAC. Start by running simple tests
with these commands (e.g., uniaxial and confined compression tests or simple excavation stability
analyses). It may be helpful to review the detailed description of these commands in Section 1.3 in
the Command Reference. Then try adding more complexity to the model. Before running very
detailed simulations though, we recommend that you read Section 3, which provides guidance on
problem solving in general.
2 - 66
Users Guide
Table 2.5
Function
Command
Grid Generation
GRID
GEN
INITIAL
Boundary/Initial Conditions
APPLY
FIX
INITIAL
MODEL
PROPERTY
Initial Equilibrium
STEP
SOLVE
SET gravity
(with gravity)
Perform Alterations
MODEL
PROPERTY
APPLY
FIX
FREE
SAVE
RESTORE
GETTING STARTED
2 - 67
yx
xy
xy
yx
x
Figure 2.34 Sign convention for positive shear stress components
DIRECT STRAIN Positive strain indicates extension; negative strain indicates compression.
SHEAR STRAIN Shear strain follows the convention of shear stress (see above). The distortion
associated with positive and negative shear strain is illustrated in Figure 2.35.
PRESSURE A positive pressure will act normal to, and in a direction toward, the surface of a
body (i.e., push). A negative pressure will act normal to, and in a direction away from, the surface
of a body (i.e., pull). Figure 2.36 illustrates this convention.
2 - 68
Users Guide
Figure 2.35 Distortion associated with positive and negative shear strain
PORE PRESSURE Fluid pore pressure is positive in compression. Negative pore pressure
indicates fluid tension.
GRAVITY Positive gravity will pull the mass of a body downward (in the negative y-direction).
Negative gravity will pull the mass of a body upward.
GFLOW This is a FISH parameter (see Section 2 in the FISH volume) which denotes the net
fluid flow associated with a gridpoint. A positive gflow corresponds to flow into a gridpoint.
Conversely, a negative gflow corresponds to flow out of a gridpoint.
TFLOW This is also a FISH parameter which denotes net heat flux associated with a gridpoint.
The convention for heat flux at a gridpoint is the same as for fluid flow.
The x- and y-components of vector quantities such as forces, displacements, velocities, and flow
vectors are positive when pointing in the directions of the positive x- and y-coordinate space.
GETTING STARTED
2 - 69
INTERFACES Positive shear stresses are induced at interface nodes for the following direction
of relative movement:
Shear displacements in the sense depicted above are plotted as filled areas or curves to the right
of the interface, when looking along the Aside of the interface, in the direction in which it was
specified.
Normal stress is negative if the interface node is in compression.
Compressional displacements are plotted as filled areas or curves to the left of the interface, when
looking along the Aside of the interface, in the direction in which it was specified.
STRUCTURAL ELEMENTS Axial forces in structural elements are positive in compression.
Shear forces in structural elements follow the opposite sign convention as that given for zone shear
stress, illustrated in Figure 2.34. Moments at the end of beam and pile elements are positive in the
counterclockwise direction.
Translational displacements at nodes are positive in the direction of the positive coordinate axes,
and angular displacements are positive in the counterclockwise direction.
The shear force and shear displacement at a cable/grout interface-spring node, or a pile shear
coupling-spring node, are positive if the node displacement is in the direction of the specification
of the cable or pile (i.e., begin > end).
The normal force and normal displacement at a pile normal coupling-spring node are positive if the
coupling spring is in compression.
2 - 70
Users Guide
Length
Density
Force
Stress
Gravity
Stiffness
m
kg / m3
N
Pa
m / sec2
Pa / m
m
103 kg / m3
kN
kPa
m / sec2
kPa / m
m
106 kg / m3
MN
MPa
m / sec2
MPa / m
Imperial
cm
106 g / cm3
Mdynes
bar
cm / s2
bar / cm
ft
slugs / ft3
lbf
lbf / ft2
ft / sec2
lbf / ft3
in
snails / in3
lbf
psi
in / sec2
lb / in3
where
1 bar
1 atm
1 slug
1 snail
1 gravity
Table 2.7
=
=
=
=
=
Pa
kg / m3
m3 sec / kg
m2
m / sec
FLAC permeability
(in SI units)
bar
106 g / cm3
106 cm sec / g
cm2
cm / sec
Imperial
lbf/ft2
slugs / ft3
ft3 sec / slug
ft2
ft / sec
psi
snails / in3
in3 sec / snail
in2
in / sec
GETTING STARTED
2 - 71
Systems of units for parameters associated with structural elements and heat transfer are given in
Section 1 in Structural Elements and Section 1 in Optional Features, respectively.
Table 2.8
Property
Unit
SI
area
length2
m2
m2
m2
Imperial
cm2
ft2
in2
force/disp
N/m
kN/m
MN/m
Mdynes/cm
lbf /ft
lbf /in
bond stiffness
force/length/disp
N/m/m
kN/m/m
MN/m/m
Mdynes/cm/cm
lbf /ft/ft
lbf /in/in
bond strength
force/length
N/m
kN/m
MN/m
Mdynes/cm
lbf /ft
lbf /in
exposed perimeter
length
in
moment of inertia
m4
m4
cm
cm4
ft
length4
m
m4
ft4
in4
plastic moment
force-length
N-m
kN-m
MN-m
Mdynes-cm
ft-lbf
in-lbf
yield strength
force
kN
MN
Mdynes
lbf
lbf
Youngs modulus
stress
Pa
kPa
MPa
bar
lbf /ft2
psi
where
Systems of units for parameters associated with heat transfer are given in Section 1 in Optional
Features.
2.9 Precision Limits
When selecting a system of units, care should be taken to avoid calculations that approach the precision limits of the computer hardware. For 80386/387-based computers, the range is approximately
1035 to 1035 in single-precision. If numbers exceed these limits, it is likely that the program will
crash or, at least, produce artifacts in the model that may be difficult to identify or detect.
There are two versions of FLAC: a single-precision version and a double-precision version. In
the single-precision version (FLACW SP.EXE), calculations are primarily based upon singleprecision variables. Errors may be introduced for some variables (e.g., pore pressure) in cases
in which the accumulated value of the variable after many thousands of timesteps is much larger
than the incremental change in the variable (i.e., an accumulated value that is roughly six orders
of magnitude larger than the incremental value). In such a case, precision limitations will prevent
further change to the value of the variable. It is recommended that the double-precision version
(FLACW DP.EXE) be used for calculations involving variable changes of these magnitudes, or
for grids containing many zones with coordinates that are large compared to typical zone dimensions.
2 - 72
Users Guide
2.10 Files
There are nine types of files that are either used or created by FLAC. The files are distinguished by
their extensions and are described below.
INITIALIZATION FILE
FLAC.INI This is a formatted ASCII file, created by the user, that FLAC will automatically
access upon start-up or when a NEW command is issued. FLAC searches for the file FLAC.INI
in the directory in which the code is executed and, if not found, in the directory pointed to by the
ITASCA environment variable. The file may contain any valid FLAC command(s) (see Section 1 in
the Command Reference). Although this file does not need to exist (i.e., no errors will result if it
is absent), it is normally used to change default options in FLAC to those preferred by the individual
user each time a new analysis is run (see Section 2.1.7).
DATA FILES
The user has a choice of running FLAC interactively (i.e., entering FLAC commands while in the
FLAC environment) or via a data file (also called a batch file). The data file is a formatted ASCII
file created by the user which contains the set of FLAC commands that represents the problem being
analyzed. In general, creating data files is the most efficient way to use FLAC. To use data files
with FLAC, see the CALL command in Section 1 in the Command Reference. Data files can have
any filename and any extension. It is recommended that a common extension (e.g., .DAT for
FLAC input commands, and .FIS for FISH function statements) be used to distinguish these files
from other types of files.
SAVE FILES
FLAC.SAV This file is created by FLAC at the users request when issuing the command
SAVE. The default file name is FLAC.SAV, which will appear in the default directory when
quitting FLAC. The user may specify a different filename by issuing the command SAVE lename,
where lename is a user-specified filename. FLAC.SAV is a binary file containing the values of
all state variables and user-defined conditions. The primary reason for creating save files is to allow
one to investigate the effect of parameter variations without having to rerun a problem completely.
A save file can be restored and the analysis continued at a subsequent time (see the RESTORE
command in Section 1 in the Command Reference). If the save file is created in the GIIC, the
file will also include information that describes the state of the GIIC at the stage the file is saved.
Normally, it is good practice to create several save files during a FLAC run.
GIIC PROJECT SAVE FILES
This file is created at the users request when the File / Save Project menu item is pressed in the
GIIC. The file name has the extension .PRJ, which should not be changed. The file is an ASCII
file containing variables that describe the state of the GIIC at the stage that the project is saved, and
includes a link to the individual FLAC save files (.SAV) associated with the project.
GETTING STARTED
2 - 73
2 - 74
Users Guide
MOVIE FILES
FLAC.DCX This file is created by FLAC at the users request when issuing the command
MOVIE on. Its purpose is to capture graphics images for playback as a movie on the computer
monitor at a later time. The default file name is FLAC.DCX, which will appear in the default
directory when quitting FLAC. The user may specify a different filename by issuing the command
MOVIE le lename, where lename takes the place of FLAC.DCX. A DCX file format is used
for the movie file. DCX files are a collection of PCX files and include an index to the PCX files.
A DCX file can contain up to 1024 PCX images. See the MOVIE command in Section 1 in the
Command Reference.
GETTING STARTED
2 - 75
2.11 References
Journal of Petroleum Technology. The SI Metric System of Units and SPEs Tentative Metric
Standard, 1575-1616 (December, 1977).
2 - 76
Users Guide