INTRODUCTION 1-1

1 INTRODUCTION

1.1 Overview

FLAC is a two-dimensional explicit finite difference program for engineering mechanics compu-
tation. This program simulates the behavior of structures built of soil, rock or other materials
that may undergo plastic flow when their yield limits are reached. Materials are represented by
elements, or zones, which form a grid that is adjusted by the user to fit the shape of the object to
be modeled. Each element behaves according to a prescribed linear or nonlinear stress/strain law
in response to the applied forces or boundary restraints. The material can yield and flow, and the
grid can deform (in large-strain mode) and move with the material that is represented. The explicit,
Lagrangian calculation scheme and the mixed-discretization zoning technique used in FLAC en-
sure that plastic collapse and flow are modeled very accurately. Because no matrices are formed,
large two-dimensional calculations can be made without excessive memory requirements. The
drawbacks of the explicit formulation (i.e., small timestep limitation and the question of required
damping) are overcome to some extent by automatic inertia scaling and automatic damping that do
not influence the mode of failure.
Though FLAC was originally developed for geotechnical and mining engineers, the program offers
a wide range of capabilities to solve complex problems in mechanics. Several built-in consti-
tutive models are available that permit the simulation of highly nonlinear, irreversible response
representative of geologic, or similar, materials. In addition, FLAC contains many special features
including:
• interface elements to simulate distinct planes along which slip and/or separa-
tion can occur;
• plane-strain, plane-stress and axisymmetric geometry modes;
• groundwater and consolidation (fully coupled) models with automatic phreatic
surface calculation;
• structural element models to simulate structural support (e.g., tunnel liners,
rock bolts, or foundation piles);
• extensive facility for generating plots of virtually any problem variable;
• optional dynamic analysis capability;
• optional viscoelastic and viscoplastic (creep) models;
• optional thermal (and thermal coupling to mechanical stress and pore pressure)
modeling capability;
• optional two-phase flow model to simulate the flow of two immiscible fluids
(e.g., water and gas) through a porous medium; and

FLAC Version 4.0

1-2 User’s Guide

• optional facility to add new, user-defined, constitutive models written in C++

and compiled as dynamic linked libraries (DLLs) that can be loaded when
needed.
FLAC also contains the powerful built-in programming language FISH (short for FLACish). With
FISH, you can write your own functions to extend FLAC ’s usefulness and even implement your own
constitutive models if so desired. FISH offers a unique capability to FLAC users who wish to tailor
analyses to suit specific needs. You will soon see that with all of these capabilities, FLAC can be
an indispensable analysis-and-design tool in a variety of fields in civil and mechanical engineering.
FLAC can be operated as either a menu-driven or a command-driven computer program. The menu-
driven mode provides easy-to-use mouse access to FLAC operation by generating and applying
all the input required for a FLAC simulation, in response to point-and-click operations. This
mode allows first-time or occasional users a simple means to begin solving problems with FLAC
immediately.
The command-driven mode requires knowledge of the word-command “language” used by FLAC,
which can be more difficult for new users to master than the menu-driven mode. However, it offers
several advantages when applied to engineering problems.
1. The input language is based upon recognizable word commands that allow you
to identify the application of each command easily and in a logical fashion (e.g.,
the APPLY command applies boundary conditions to the model).
2. Engineering simulations usually consist of a lengthy sequence of operations
— e.g., establish in-situ stress, apply loads, excavate tunnel, install support
and so on. A series of input commands (from a file or from the keyboard)
corresponds closely with the physical sequence that it represents.
3. A FLAC data file can easily be modified with a text editor. Several data files
can be linked to run a number of FLAC analyses in sequence. This is ideal for
performing parameter sensitivity studies.
4. The word-oriented input files provide an excellent means to keep a documented
record of the analyses performed for an engineering study. Often, it is con-
venient to include these files as an appendix to the engineering report for the
purpose of quality assurance.
5. The command-driven structure allows you to develop pre- and post-processing
programs to manipulate FLAC input/output as desired. For example, you may
wish to write a mesh-generation function to create a special grid shape for a
series of FLAC simulations. This can readily be accomplished with the FISH
programming language and incorporated directly in the input data file.
When operated from the menu-driven mode, FLAC commands are created and applied automatically.
Also, a record of the commands are kept and can be saved to provide a documented listing of the
commands used in the analysis. This command record can be used to drive FLAC in command-
driven mode.

FLAC Version 4.0

 INTRODUCTION 1-3

Dr. Peter Cundall developed FLAC in 1986 specifically to perform engineering analyses on an
IBM-compatible microcomputer. The software is designed for high-speed computation of models
containing several thousand elements. With the advancements in floating-point operation speed and
the ability to install additional RAM at low cost, increasingly larger problems can be solved with
FLAC. For example, FLAC can solve a model containing up to 30,000 elements of Mohr-Coulomb
material on a microcomputer with 8 MB RAM. The solution speed for a model of this size is roughly
7 calculation steps per second on a 400 MHz Pentium microcomputer.* The speed is essentially
a linear function of the number of elements; a model of 15,000 elements would require half the
runtime to process the same number of calculation steps.
For typical models, consisting of 15,000 elements or fewer, the explicit solution scheme in FLAC
requires approximately 4000 to 6000 steps to reach a solved state.† Thus, a 15,000 element model
run on the Pentium described above would require roughly 6 minutes to perform 5000 calculation
steps. Consequently, typical engineering problems involving several thousand elements to model,
which once required access to a mainframe computer to solve, can be solved with FLAC on a
microcomputer in a matter of minutes.
A comparison of FLAC to other numerical methods, a description of general features and new
updates in FLAC Version 4.0, and a discussion of fields of application are provided in the following
sections. If you wish to try FLAC right away, the program installation instructions and simple
menu-driven and command-driven tutorials are provided in Section 2.

* See Section 5 for a comparison of FLAC runtimes on various computer systems.

† This can vary but, typically, a problem solution can be reached between 4000 and 6000 steps for
models containing up to 15,000 elements, regardless of material type. The explicit scheme is
explained in Section 1 in Theory and Background.

FLAC Version 4.0

1-4 User’s Guide

1.2 Comparison with Other Methods

How does FLAC compare to the more-common method of finite elements for numerical modeling?
Both methods translate a set of differential equations into matrix equations for each element, relating
forces at nodes to displacements at nodes. Although FLAC ’s equations are derived by the finite
difference method, the resulting element matrices, for an elastic material, are identical to those
derived by using the finite element method (for constant strain triangles). However, FLAC differs
in the following respects:
1. The “mixed discretization” scheme (Marti and Cundall 1982) is used for ac-
curate modeling of plastic collapse loads and plastic flow. This scheme is be-
lieved to be physically more-justifiable than the “reduced integration” scheme
commonly used with finite elements.
2. The full dynamic equations of motion are used, even when modeling systems
that are essentially static. This enables FLAC to follow physically unstable
processes without numerical distress.
3. An “explicit” solution scheme is used (in contrast to the more-usual implicit
methods). Explicit schemes can follow arbitrary nonlinearity in stress/strain
laws in almost the same computer time as linear laws, whereas implicit solu-
tions can take significantly longer to solve nonlinear problems. Furthermore,
it is not necessary to store any matrices, which means that: (a) a large number
of elements may be modeled with a modest memory requirement; and (b) a
large-strain simulation is hardly more time-consuming than a small-strain run,
because there is no stiffness matrix to be updated.
4. FLAC is robust in the sense that it can handle any constitutive model with no
adjustment to the solution algorithm; many finite element codes need different
solution techniques for different constitutive models.
5. FLAC numbers its elements in a row-and-column fashion rather than in a
sequential fashion. For many problems, this method makes it easier to identify
elements when specifying properties and interpreting output.
These differences are mainly in FLAC ’s favor, but there are two disadvantages:
1. Linear simulations run slower with FLAC than with equivalent finite element
programs; FLAC is most effective when applied to nonlinear or large-strain
problems, or to situations in which physical instability may occur.
2. The solution time with FLAC is proportional to the ratio of the longest natural
period to the shortest natural period in the system being modeled. This point
is discussed in more detail in Section 1 in Theory and Background, but
certain problems are very inefficient to model (e.g., beams, represented by
solid elements rather than structural elements, or problems that contain large
disparities in elastic moduli or element sizes).

FLAC Version 4.0

INTRODUCTION 1-5

1.3 General Features

1.3.1 Basic Features

FLAC is primarily intended for geotechnical engineering applications; the code embodies special
numerical representations for the mechanical response of geologic materials. FLAC has ten built-in
material models: the “null” model, which represents holes (excavations) in the grid; the isotropic
elastic model; the transversely isotropic elastic model; and seven plasticity models (Drucker-Prager,
Mohr-Coulomb, ubiquitous-joint, strain-hardening/softening, bilinear strain-hardening/softening
ubiquitous-joint, double-yield, and modified Cam-clay). You can also create your own constitutive
models by using the FISH programming language. Each zone in a FLAC grid may have a different
material model or property, and a continuous gradient or statistical distribution of any property may
be specified.
Additionally, an interface, or slip-plane, model is available to represent distinct interfaces between
two or more portions of the grid. The interfaces are planes upon which slip and/or separation are
allowed, thereby simulating the presence of faults, joints or frictional boundaries.
The basic formulation for FLAC is for a two-dimensional plane-strain model. This condition is
associated with long structures or excavations with constant cross section and acted on by loads in
the plane of the cross section. In addition, FLAC offers a plane-stress option for elastic and Mohr-
Coulomb plasticity analysis. In this condition, the stresses normal to the cross section are zero. This
is encountered, for example, in thin plates loaded only in their plane. Finally, an option to model
axisymmetric geometry also exists. In this case, cylindrical coordinates are used; x = 0 is the axis
of symmetry, the positive x-direction corresponds to the radial coordinate, the y-direction to the
axial coordinate and the out-of-plane direction (the z-direction) to the circumferential coordinate.
This geometry applies, for example, to problems involving cylindrical test specimens or cylindrical
and spherical holes in a continuum. The out-of-plane stress, σzz , is calculated in FLAC for all three
solution conditions, and σzz is taken into account in plastic yield calculations for plane-strain and
axisymmetry conditions.
Either velocity (and displacement) boundary conditions or stress (and force) boundary conditions
may be specified at any boundary orientation. Initial stress conditions, including gravitational
loading, may be given, and a water table may be defined for effective stress calculations. All
conditions may be specified with gradients.
FLAC incorporates the facility to model groundwater flow and pore pressure dissipation, and the
full coupling between a deformable porous solid and a viscous fluid flowing within the pore space.
(The coupled interaction is described further in Section 1.3.3.) The fluid may be assumed to obey
either the isotropic or anisotropic form of Darcy’s law and is considered to be deformable. Non-
steady-flow is modeled, with steady-flow treated as an asymptotic case. Fixed pore pressure and
constant-flow boundary conditions may be used, and sources and sinks (wells) may be modeled.
The flow model can also be run independently from the mechanical calculation, and both confined
and unconfined flow can be simulated, with automatic calculation of the phreatic surface. Various
features are available in FLAC to monitor flow, including streamline plots and fluid-particle tracking.

FLAC Version 4.0

1-6 User’s Guide

Structures, such as tunnel liners, piles, sheet piles, cables, rock bolts or yielding props, that interact
with the surrounding rock or soil may be modeled with the structural element logic in FLAC. It
is possible to examine the stabilizing effects of supported excavations or to examine the effects of
soil or rock instability on surface structures.
FLAC contains a powerful built-in programming language, FISH, which enables the user to define
new variables and functions. FISH is a compiler; programs entered via a FLAC data file are
translated into a list of instructions stored in FLAC ’s memory space; these are executed whenever
a FISH function is invoked. FISH was developed in response to requests from users who wanted
to do things with FLAC that were either difficult or impossible with the existing code. Users can
write their own functions to extend FLAC ’s usefulness. For example, FISH permits:
• user-prescribed property variations in the grid (e.g., nonlinear in-
crease in modulus with depth);
• plotting and printing of user-defined variables (custom-designed
plots);
• implementation of special grid generators;
• servo-control of numerical tests;
• specification of unusual boundary conditions; variations in time and
space;
• automation of parameter studies; and
• specification of user-defined constitutive models.
An extensive plotting facility is built directly into FLAC. This allows the user to generate plots (either
on the screen or to a hardcopy device) of virtually any problem variable in the FLAC model. Several
variables can be plotted as overlays on a plot of the model, or histories of their change as a function
of calculation step can be plotted. The history plots are especially helpful to ascertain when an
equilibrium or failure state has been reached, and to monitor the change in variables during transient
calculations, such as groundwater flow, or dynamic calculations, such as earthquake motion. As
mentioned above, plots can be custom-designed via FISH to meet the user’s need. Examples are
given in Section 3 in the FISH volume.
The basic version of FLAC can be operated in menu-driven or command-driven mode, and the user
can easily switch back and forth between the two modes if desired. The menu-driven mode is easy
to use with point-and-click mouse operation and is designed to emulate expected Windows features.
This mode provides access to all commands and facilities available in FLAC.
Two executable codes are provided as part of the basic version of FLAC: a single-precision version
and a double-precision version. The single-precision version is more efficient for most analyses.
However, for situations in which the accumulated value of a variable after many thousands of calcu-
lational steps is much larger than the incremental change in the variable, single-precision limitations
will prevent further changes in the variable. The double-precision version is recommended for this
case.

FLAC Version 4.0

INTRODUCTION 1-7

1.3.2 Optional Features

Five optional features (for dynamic analysis, thermal analysis, two-phase flow analysis, modeling
creep material behavior, and adding new constitutive models as dynamic linked libraries(DLLs))
are available as separate modules that can be included in FLAC at an additional cost per module.
Dynamic analysis can be performed with FLAC, using the optional dynamic calculation module.
User-specified acceleration, velocity or stress waves can be input directly to the model either as
an exterior boundary condition or an interior excitation to the model. FLAC contains absorbing
and free-field boundary conditions to simulate the effect of an infinite elastic medium surrounding
the model. The dynamic calculation can be coupled to the groundwater flow model; the level of
coupling is discussed in Section 1.3.3.
There is a thermal analysis option available as a special module in FLAC. This model simulates the
transient flux of heat in materials and the subsequent development of thermally induced stresses.
The thermal model can be run independently or coupled to the mechanical stress calculation or pore
pressure calculation, either in static or dynamic mode. (The coupling interactions are described in
Section 1.3.3.)
FLAC can perform fluid flow and fully coupled simulations in which two immiscible fluids (with
optional capillary pressure) are present within a porous medium. This optional feature extends the
facility of the basic groundwater flow model; the components of the formulation are described in
Section 1.3.3.
There are six optional material models available that simulate viscoelastic and viscoplastic (creep)
behavior: the classical viscoelastic (Maxwell) model; a two-component power law; a reference
creep formulation (the WIPP model) implemented for nuclear waste isolation studies; a Burger-
creep viscoplastic model; a WIPP-creep viscoplastic model; and a crushed-salt constitutive model.
All six models are available in the creep module. A FLAC grid can be configured for both a creep
calculation and a dynamic calculation. However, both modes cannot be active simultaneously
because of the widely different timesteps.
New constitutive models can be added to FLAC as dynamic linked libraries (DLLs) that are written
in C++ and compiled as separate files. The DLLs can be loaded in FLAC whenever needed, via the
MODEL load command. By implementing this optional feature, users can access new constitutive
models from Itasca’s web site. An advantage of these models is that they run at nearly the same
speed as built-in models, and noticeably faster than FISH constitutive models.

1.3.3 Modeling Physical Processes and Interactions

The default calculation mode in FLAC is for static mechanical analysis. Alternatively, a groundwater
flow analysis or a heat transfer analysis can be performed by itself, independently of the mechanical
calculation. In addition, both the groundwater flow and thermal models may be coupled to the
mechanical stress model and to each other. Because the full equations of motion are used in FLAC,
the coupling interactions can be applied for dynamic analysis as well as static analysis.

FLAC Version 4.0

1-8 User’s Guide

The coupling mechanisms are divided into three types of interaction: mechanical and groundwater
flow; mechanical and thermal; and thermal and groundwater flow. The level of interaction modeled
in FLAC for each type is described below.
Mechanical-Groundwater Flow Coupling — Several types of fluid/solid interaction can be specified
in FLAC. One type of interaction is consolidation, in which the slow dissipation of pore pressure
causes displacements to occur in the solid (e.g., soil). Two mechanical effects are at work in this
case: (1) the fluid in a zone reacts to mechanical volume changes by a change in the pore pressure;
and (2) the pore pressure change causes changes in the effective stress that affect the response of
the solid (e.g., a reduction in effective stress may induce plastic yield).
FLAC can calculate pore pressure effects, with or without pore pressure dissipation, simply by
setting the flow calculation on or off. Also, dynamic pore pressure generation (e.g., related to
liquefaction) can be modeled by accounting for irreversible volume strain in the constitutive model.
This can be done using a built-in constitutive model, the Finn model, provided with the dynamic
option.
By default, porosity is assumed constant; however, porosity can be made a function of volumetric
strain as prescribed by a look-up table. Likewise, permeability is not adjusted by default, but can
also be made a table function of volumetric strain, for inclusion as a coupling mechanism. As a
consequence, two-way coupling of mechanical stress and groundwater flow can be modeled with
FLAC.
Two-phase flow processes and capillary effects can be modeled with the optional two-phase flow
model. With this model, the capillary pressure and relative permeability laws are built-in functions
of the van Genuchten form. The mechanical coupling uses Bishop’s effective stresses, and solid
grains are assumed to be incompressible, as in the basic groundwater formulation.
Other types of interaction, such as electrical or chemical forces between particles of a partially
saturated material, are not modeled directly by FLAC, but can also be included by writing a suitable
FISH function. Similarly, a FISH function may be used to vary the local fluid modulus as a function
of other quantities such as pressure or time.

Thermal-Mechanical Coupling — The thermal-mechanical coupling in FLAC is one-way: tem-

perature change may induce a mechanical stress change as a function of the thermal-expansion
coefficient. Mechanical changes in the body, however, do not result in temperature change, or
changes to thermal properties.
Additionally, mechanical properties can be made a function of temperature change since FISH
permits access to both temperatures and properties.

Thermal-Groundwater Flow Coupling — The thermal calculation may be coupled to the ground-
water flow calculation by making pore pressures a function of temperature change. Volumetric
strain can arise from thermal expansion of both the fluid and the grains within a saturated matrix.
Pore pressure change results from this volumetric strain, as well as mechanical volumetric strain.
This interaction is one-way: groundwater flow does not influence heat transfer in any way. Also,
there is no representation of convection or heat transfer between fluid and solid.

FLAC Version 4.0

INTRODUCTION 1-9

As with mechanical properties, groundwater properties can be made a function of temperature

change by accessing temperature and property values via FISH.

FLAC Version 4.0

1 - 10 User’s Guide

1.4 Summary of Updates from Version 3.4

FLAC 4.0 contains several improvements; the new features are summarized in the following sections.
Existing data files created for Version 3.4 should still operate as before; new commands have been
added for the updates in Version 4.0. You should be aware, however, that FLAC 4.0 will not be able
to restore files saved by an earlier version of FLAC.

1.4.1 Full Graphics Interface

A new graphical interface, the GIIC (Graphical Interface for Itasca Codes), makes all features in
FLAC available by using graphical tools. (The FLAC command-line input is still available.) New
features in the GIIC include: a grid library that helps users create common types of geo-engineering
grids (e.g., retaining walls, slopes, tunnels, dams); use of digitized plots or imported graphics files
to guide the grid generation graphically; and access to a database of material properties.
All commands that are available to users in previous versions of FLAC are now accessible via
menu- and mouse-driven operations in the GIIC. The command-line output of the code is available
for viewing via a pane in the GIIC. Through the use of multi-paned windows that may be re-sized
or minimized, users can customize the modeling environment to match their individual needs and
preferences. See Section 2.2.2 for an introduction to the GIIC.
The GIIC is a JAVA-based application that runs independently of the modeling code; data are
exchanged directly between FLAC and the GIIC, allowing the user to obtain and manipulate model
results without interfering with the solution process. Eventually, we plan to make all Itasca codes
accessible via the GIIC.

1.4.2 Double-Precision Arithmetic Version

FLAC 4.0 includes a double-precision executable code as well as a single-precision code. The
double-precision version provides more-accurate solutions for cases in which the accumulated
value of a variable after many thousands of timesteps is much larger than the incremental change
in the variable (e.g., an accumulated value for displacement that is roughly six orders of magnitude
larger than the incremental value in a creep calculation). The double-precision version is also
recommended for models with grids containing many zones with coordinates that are large compared
to typical zone dimensions. See Section 2.9 for a discussion on precision limits in FLAC. Note
that the FISH optimizer does not work in the double-precision version.

1.4.3 Factor of Safety Calculation

The strength reduction method for determining factor of safety is now implemented through the
SOLVE fos command. See note 12 in Section 3.8 for a description. A FISH scalar variable, fos f,
is available to monitor the current value of the factor of safety.

FLAC Version 4.0

INTRODUCTION 1 - 11

1.4.4 Two-Phase Flow and Capillary Effects

FLAC can now perform fluid flow simulations involving displacement of a fluid by another within
a porous medium. Two-phase flow processes and capillary effects have important applications in
fields as diverse as the oil industry (gas drive), agriculture (flow in the vadoze zone), pollution
control, radioactive waste disposal and slope stability.
In the FLAC formulation, the two fluids are considered immiscible and slightly compressible (i.e.,
spacial variations of densities are neglected). The capillary pressure and relative permeability
laws are built-in functions of the van Genuchten form. The nonlinear partial differential equations
for two-phase flow are solved using an explicit finite difference scheme with upstream weighting
technique. The mechanical coupling uses Bishop’s effective stresses, and the assumption of solid
grain incompressibility is adopted. See Section 2 in Fluid-Mechanical Interaction for a detailed
description of the formulation and example applications.

1.4.5 Ability to Link FLAC to Other Itasca Codes via TCP/IP Connection

With FLAC 4.0, data can be passed rapidly between two or more Itasca codes running on the
same machine. The upcoming versions of PFC 2D and FLAC 3D also allow such coupling, and the
same facility will exist in new versions of other codes, as they become available. One example
application involves embedding a PFC 2D particle assembly within a FLAC grid, thus providing an
efficient continuum boundary condition for the PFC 2D assembly. The data transmission between
codes is invoked by FISH functions that allow large arrays of data to be exchanged with single
function calls. This facility is described and an example application given in Section 2.7 in the
FISH volume.

1.4.6 User-Defined Constitutive Models in C++

External constitutive models can now be written in C++ and compiled as DLL (dynamic link library)
files that can be loaded whenever needed in a FLAC simulation. C++ DLL models run at nearly the
same speed as built-in models in FLAC. This facility is provided as an optional feature in FLAC
(accessed via the CONFIG cppudm and MODEL load commands). DLL models can be accessed from
Itasca’s web site and used with FLAC 4.0 when needed.
Eventually, we plan to make it possible for users to write and compile their own models as DLLs.
Also, it is planned that the same DLLs will be used by upcoming versions of Itasca codes (FLAC 3D
and UDEC). Thus, it will be possible for a single user-written model to be used by several codes.

1.4.7 New Printing Facilities

FLAC 4.0 can generate graphics files in several industry standard formats: BMP, EMF, JPEG, DXF
and PCX, in addition to PostScript. These formats are invoked through the SET plot command.

FLAC Version 4.0

1 - 12 User’s Guide

1.4.8 New Movie Facility

A new movie viewer program (“MOVIE.EXE”) is provided as a Windows application. The MOVIE
command now works with the Windows version of FLAC, and controls the capture of screen plots as
a set of PCX images that are viewed with “MOVIE.EXE.” The “MOVIE.EXE” program is located
in the “\ITASCA\Utility” directory.

1.4.9 Enhancements for Dynamic Analysis

The Finn model to simulate dynamic pore pressure generation has been modified to include the
formula by Byrne (1991), which is a simpler relation than that given by Martin et al. (1975), and
has input that can be estimated from standard penetration test (STP) blow counts. The Finn model
is now provided as a dynamic link library (DLL) that runs at essentially the same speed as a built-
in model. The model is accessed via the MODEL finn command. See Section 3.4.4 in Optional
Features for a discussion on the model and example application.
Von Neumann and Landshoff artificial viscosity terms have been implemented in FLAC to control
damping in dynamic analyses involving sharp fronts. This form of damping is accessed with the
SET dy damp avisc command. See Section 3.4.2.6 in Optional Features for a description.

1.4.10 Structural Element Enhancements

The keyword szz on/off is now available with the STRUCT prop command so that the out-of-plane
stress, σzz , can be included or excluded from the calculation of mean effective confining stress, σc ,
for cable elements.
The effect of thermal expansion is implemented for beam and cable elements. A linear thermal
expansion coefficient can be assigned with the STRUCT prop thexp command.
The identification numbers for structural elements are now assigned in ascending order to simplify
plotting variables for individual structural elements.
The plastic moment calculation for beam and pile elements has been modified to include shear force
yielding when the limiting moment is reached.

1.4.11 Modifications for Model Ranges

A new command, GROUP, is now available to identify sets of zones in a FLAC model by a group
name. The name is assigned to zones in a selected range; the range can be in several forms (see
Section 1.1.3). A group keyword can be used in the same manner as the usual range keywords to
limit the action of specific commands, as described in Section 1.1.3. A FISH variable, z group,
is available to access group names.
Two additional range keywords have been added: notnull and nmregion. notnull restricts the oper-
ations of commands such as MODEL and PROPERTY to only non-null zones. nmregion allows the
action of the INITIAL command to apply to both marked and unmarked gridpoints.

FLAC Version 4.0

INTRODUCTION 1 - 13

1.4.12 New Utility Features

The SOLVE command has been modified for more-efficient calculations of models including ground-
water or structural elements. A model will now come to equilibrium in one step for the case of
initialized pore pressures already in balance with a gravity gradient. Convergence is improved for
the case of a beam with zero moment of inertia connected by interfaces on each side to the grid
(e.g., to simulate a geo-grid).
A SOLVE elastic command is provided to perform mechanical calculations in two steps: first
assuming elastic behavior and then using the actual strength values of the material.
For mechanical-groundwater coupled analyses, the command CONFIG ats can be given to automat-
ically update total stresses when there is an instantaneous change in pore pressure by the INITIAL
pp, APPLY pp or WATER table command. This is particularly useful when simulating dewatering
— e.g., see Section 1.10.9 in Fluid-Mechanical Interaction.
A PRINT model command is now available to provide a list of currently active constitutive models,
including FISH models and DLL models, in the FLAC model.
The SYSTEM command can now access the DOS command, CD, in order to change directories.
An optional message can now be assigned to a SAVE file. The message can be a single word, a
string of words enclosed by single quotes, or the contents of a FISH string variable.

1.4.13 Mass Density Modification

The mass density is now updated in large-strain mode, such that the mass remains constant as
zones change volume. As a result, gravity body forces will now remain unchanged for very large
deformations. The density is not changed for small-strain calculations.

1.4.14 Interface Modifications

A new command, SET tolint, is provided to change the tolerance for recognizing interface contact
in small-strain mode. The tolerance may need to be changed for problems in which the coordinate
values are very large and zone sizes are very small. If interfaces are not recognized in small-
strain mode, or if initializing stresses do not install interface stresses, then the tolerance should be
increased. The default is 10−4 .
The interface constitutive model has been modified so that if the tensile bond is not exceeded, then
sliding is prevented (i.e., both the shear and normal springs will act elastically). Optional keywords,
bslip and sbratio, are available to control the slip condition for a bonded interface.

FLAC Version 4.0

1 - 14 User’s Guide

1.4.15 New Features in FISH

The “APP.FIN” file has been modified to describe the pointers appgw pnt and appth pnt, used for
accessing the APPLY data for groundwater and thermal modes.
A new FISH intrinsic, fstring, is available to print additional significant figures for floating-point
numbers.
K , used in the Burger-creep viscoplastic model, can be inspected and
The internal Kelvin strains, eij
modified as FISH property variables k exx, k eyy, k ezz and k exy.

FLAC Version 4.0

INTRODUCTION 1 - 15

1.5 Fields of Application

FLAC was developed primarily for geotechnical engineering applications. Section 6 contains a
bibliography of reports published since 1986 on the application of FLAC to geotechnical problems
in the fields of mining, underground engineering, rock mechanics, and research. Some example
applications are summarized below.
The program has been used primarily for analysis and design in mining engineering and underground
construction. The explicit, time-marching solution of the full equations of motion (including inertial
terms) permits the analysis of progressive failure and collapse, which are important phenomena in
studies related to mine design.
The interface logic has been used to evaluate the influence of fault structures in mine design.
Structural elements have been employed to simulate various rock reinforcement systems such as
grouted rockbolting and shotcrete. The creep model option in FLAC has been applied for studies
in salt and potash mine design. Through the use of the thermal model option, FLAC has been used
to study the performance of deep underground repositories for high-level radioactive waste.
FLAC has also been used extensively in research areas related to studies of the process of localization
and evolution of shear bands in frictional materials. The results of these studies have helped elucidate
the mechanisms involved in localization and provide insight into phenomena such as the genesis of
fault systems and families of faults.
Additionally, FLAC has potential for application in a variety of fields in civil and mechanical engi-
neering. Several applications are described in the following paragraphs and listed in the bibliography
in Section 6.
With the incorporation of groundwater flow logic and coupled mechanical-fluid flow (consolidation)
analysis, FLAC offers a robust capability for applications in soil mechanics. Potential applications
include analyses of earth-retaining structures and earthen slopes under drained and undrained load-
ing, and calculations of bearing capacity and settlement of foundations. With the axisymmetry
geometry in FLAC, design calculations can be made for cofferdams or caissons, including the ef-
fects of layered materials on lateral loading. The structure element logic in FLAC also permits the
simulation of soil reinforcement such as tiebacks or soil nailing.
With the dynamics analysis option in FLAC, the code can be applied to various engineering dynamics
problems. Analyses can be performed in earthquake engineering — e.g., studies of dam stability,
soil-structure interaction and liquefaction. Calculations can also be made to evaluate effects of
explosive loading, such as underground blasting.
One drawback of early versions of FLAC was the inability to present results from FLAC in the
conventional format used in soil mechanics and foundation design. FISH overcomes this problem
because, now, the user can define the format desired. For example, a p-q diagram can be plotted
or factors of safety calculated for a specific design analysis; the factor may be defined by the user
in terms of the resistance of the structure to the applied load for the specified problem conditions.
These calculations now can be done directly in FLAC (see Section 3 in the FISH volume for
examples).

FLAC Version 4.0

1 - 16 User’s Guide

It is possible for users to create their own constitutive models via FISH. Researchers can test their
assumptions of material behavior directly in the code, bypassing the effort of writing a separate
numerical program to test their theory. FISH provides the user with great flexibility in designing
models to represent complex material behavior.
In addition, analyses of highly deformable engineering materials can be made with the large-strain
logic in FLAC. For example, the problem of extrusion of a copper billet through a die has been
studied using FLAC (see the reference for Cundall and Board (1988) in Section 6).

FLAC Version 4.0

INTRODUCTION 1 - 17

1.6 Guide to the FLAC Manual

The FLAC Version 4.0 manual consists of thirteen documents. This document, the User’s Guide,
is the main guide to using FLAC and contains descriptions of the features and capabilities of
the program, along with recommendations on the best use of FLAC for problem solving. The
remaining documents cover various aspects of FLAC, including theoretical background information,
verification testing and example applications. The complete manual is available in electronic format
on the FLAC CD-ROM (viewed with Acrobat Reader), as well as in paper format.
The organization of the eleven documents and brief summaries of the contents of each section
follows. Please note that if you are viewing the manual in the Acrobat Reader, by double-clicking
on a section number given below, you will immediately open that section for viewing.
User’s Guide
Section 1 Introduction
This section introduces you to FLAC and its capabilities and features. An overview
of the new features in the latest version of FLAC is also provided.
Section 2 Getting Started
If you are just beginning to use FLAC, or are only an occasional user, we recommend
that you read Section 2. This section provides instructions on installation and opera-
tion of the program, as well as simple tutorials on menu-driven and command-driven
operation to guide the new user through a FLAC analysis.
Section 3 Problem Solving
Section 3 is a guide to practical problem solving. Turn to this section once you are
familiar with the program operation. Each step in a FLAC analysis is discussed in
detail, and advice is given on the most effective procedures to follow when creating,
solving and interpreting a FLAC model simulation.
Section 4 FISH Beginner’s Guide
Section 4 provides the new user with an introduction to the FISH programming
language in FLAC. This includes a tutorial on the use of the FISH language. FISH
is described in detail in Section 2 in the FISH volume.
Section 5 Miscellaneous
Various information is contained in Section 5, including the FLAC runtime bench-
mark and procedures for reporting errors and requesting technical support. Descrip-
tions of utility files to assist with FLAC operation are also given.
Section 6 Bibliography
Section 6 contains a bibliography of published papers describing some uses of FLAC.

FLAC Version 4.0

1 - 18 User’s Guide

Command Reference
Section 1 Command Reference
All the commands that can be entered in the command-driven mode in FLAC are
described in Section 1 in the Command Reference.

FISH in FLAC
Section 1 FISH Beginner’s Guide
Section 1 in the FISH volume provides the new user with an introduction to the
FISH programming language in FLAC. This includes a tutorial on the use of the
FISH language.
Section 2 FISH Reference
Section 2 in the FISH volume contains a detailed reference to the FISH language.
All FISH statements, variables and functions are explained and examples given.
Section 3 Library of FISH Functions
A library of common and general purpose FISH functions is given in Section 3 in
the FISH volume. These functions can assist with various aspects of FLAC model
generation and solution.
Section 4 Program Guide
Section 4 in the FISH volume contains a program guide to FLAC ’s linked-list data
structure. This is provided for advanced users to have more-direct access to FLAC
variables.

FLAC Version 4.0

INTRODUCTION 1 - 19

Theory and Background

Section 1 Background — The Explicit Finite Difference Method
The theoretical formulation for FLAC is described in detail in Section 1 in Theory
and Background. A FISH tutorial example of the components of the explicit finite
difference method is also given.
Section 2 Constitutive Models: Theory and Implementation
The theoretical formulation and implementation of the various constitutive models
are described in Section 2 in Theory and Background.
Section 3 Grid Generation
Section 3 in Theory and Background presents a general guide to the principles of
grid generation in FLAC. This includes procedures to adjust and grade the mesh and
to attach sub-grids to produce an accurate solution. Several examples are also given.
Section 4 Interfaces
The interface logic is described and example applications are demonstrated in Sec-
tion 4 in Theory and Background. A discussion on interface properties is also
provided.
Fluid-Mechanical Interaction
Section 1 Fluid-Mechanical Interaction
The formulation for the groundwater flow model is described, and the various ways
to model groundwater flow, both with and without solid interaction, are illustrated
in Section 1 in Fluid-Mechanical Interaction.
Section 2 Two-Phase Flow
Section 2 in Fluid-Mechanical Interaction contains a description of the formulation
and example applications of the optional two-phase flow model.
Structural Elements
Section 1 Structural Elements
Section 1 in Structural Elements describes the various structural element models
available in FLAC.

FLAC Version 4.0

1 - 20 User’s Guide

Optional Features
Section 1 Thermal Option
Section 1 in Optional Features describes the thermal model option, and presents
several verification problems that illustrate its application both with and without
interaction with mechanical stress and pore pressure.
Section 2 Creep Material Models
The different creep material models available as an option in FLAC are described, and
verification and example problems are provided in Section 2 in Optional Features.
Section 3 Dynamic Analysis
The dynamic analysis option is described, and considerations for running a dynamic
model are provided in Section 3 in Optional Features. Several verification examples
are also included in this section.

Verification Problems
This volume contains a collection of FLAC verification problems. These are tests
in which a FLAC solution is compared directly to an analytical (i.e., closed-form)
solution. See Table 1 in the Verifications volume for a list of the verification
problems.

Example Applications
This volume contains example applications of FLAC that demonstrate the various
classes of problems to which FLAC may be applied. See Table 1 in the Examples
volume for a list of the example applications.

Command and FISH Reference Summary

A quick summary of all FLAC commands and FISH statements is contained in the
Command and FISH Reference Summary.

FLAC-GIIC Reference
All the components of the GIIC (the graphical interface for FLAC) are described in
the FLAC-GIIC Reference.

FLAC Version 4.0

INTRODUCTION 1 - 21

FLAC/Slope User’s Guide

This volume provides a user’s guide to FLAC/Slope, a mini-version of FLAC that
is designed specifically to perform factor-of-safety calculations for slope-stability
analysis.

Index
A subject index is provided in the Index volume for all of the volumes of the FLAC
manual.

FLAC Version 4.0

1 - 22 User’s Guide

1.7 Itasca Consulting Group, Inc.

Itasca Consulting Group, Inc. is more than a developer and distributor of engineering software.
Itasca is a consulting and research firm comprised of a specialized team of civil, geotechnical and
mining engineers with an established record in solving problems in the areas of:
Civil Engineering
Mining Engineering and Energy Resource Recovery
Nuclear Waste Isolation and Underground Space
Defense Research
Software Engineering
Groundwater Analysis and Dewatering

Itasca was established in 1981 to provide advanced rock mechanics services to the mining industry.
Today, Itasca is a multidisciplinary geotechnical firm with 53 professionals and offices worldwide.
The corporate headquarters for Itasca is located in Minneapolis, Minnesota. Worldwide offices
of Itasca are operated as subsidiaries of HCItasca, Inc.: Hydrologic Consultants, Inc. (Denver,
Colorado); Itasca Geomekanik AB (Stockholm, Sweden); Itasca Consultants S.A. (Ecully, France);
Itasca Consultants GmbH (Gelsenkirchen, Germany); Itasca Consultores S.L. (Llanera, Spain);
Itasca S.A. (Santiago, Chile); Itasca Africa (Johannesburg, South Africa); and Itasca Consultants
Canada Inc. (Sudbury, Canada).
Itasca’s staff members are internationally recognized for their accomplishments in geological, min-
ing and civil engineering projects. Itasca staff consists of geological, mining, hydrological and
civil engineers who provide a range of comprehensive services such as (1) computational anal-
ysis in support of geo-engineering designs, (2) design and performance of field experiments and
demonstrations, (3) laboratory characterization of rock properties, (4) data acquisition, analysis,
and system identification, (5) groundwater modeling, and (6) short courses and instruction in the
geomechanics application of computational methods. If you should need assistance in any of these
areas, we would be glad to offer our services.
Itasca Consulting Group is a subsidiary of HCItasca, Inc. HCItasca was formed in 1999 with
the merger of Hydrologic Consultants, Inc. (HCI) of Denver, Colorado with Itasca Consulting
Group, Inc. of Minneapolis, Minnesota. HCI adds advanced groundwater modeling and dewatering
expertise to Itasca.

FLAC Version 4.0

INTRODUCTION 1 - 23

1.8 User Support

We believe that the support that Itasca provides to code users is a major reason for the popularity
of our software. We encourage you to contact us when you have a modeling question. We provide
a timely response via telephone, electronic mail or fax. General assistance in the installation of
FLAC on your computer, plus answers to questions concerning capabilities of the various features
of the code, are provided free of charge. Technical assistance for specific user-defined problems
can be purchased on an as-needed basis.
If you have a question, or desire technical support, please contact us at:

Itasca Consulting Group, Inc.

Mill Place
111 Third Avenue South, Suite 450
Minneapolis, Minnesota 55401 USA

Phone: (+1) 612-371-4711

Fax: (+1) 612·371·4717
Email: software@itascacg.com
Web: www.itascacg.com

We also have a worldwide network of code agents who provide local technical support. Details
may be obtained from Itasca.

FLAC Version 4.0

1 - 24 User’s Guide

1.9 References

Byrne, P. “A Cyclic Shear-Volume Coupling and Pore-Pressure Model for Sand,” in Proceedings:
Second International Conference on Recent Advances in Geotechnical Earthquake Engineering
and Soil Dynamics (St. Louis, Missouri, March, 1991), Paper No. 1.24, 47-55.
Marti, J., and P. A. Cundall. “Mixed Discretization Procedure for Accurate Solution of Plasticity
Problems,” Int. J. Num. Methods and Anal. Methods in Geomech., 6, 129-139 (1982).
Martin, G. R., W. D. L. Finn and H. B. Seed. “Fundamentals of Liquefaction Under Cyclic Loading,”
J. Geotech., Div. ASCE, 101(GT5), 423-438 (1975).

FLAC Version 4.0