Version 1.

User’s Manual
Written and Edited by

Steve Maas and Jeff Weiss

Musculoskeletal Research Laboratories

Department of Bioengineering, and
Scientific Computing and Imaging Institute
University of Utah

72 S. Central Campus Drive, Room 2646

Salt Lake City, Utah

[email protected]

[email protected]

Last Updated: August 29, 2007

 1

Table of Contents
CHAPTER 1. Introduction .......................................................................................................... 2
1.1. Overview of FEBio .............................................................................................................. 2
1.2. About this document ............................................................................................................ 2
CHAPTER 2. Running FEBio ..................................................................................................... 4
2.1. The Command Line ............................................................................................................. 4
2.2. Advanced Options................................................................................................................ 5
CHAPTER 3. Free Format Input................................................................................................ 6
3.1. Free format overview........................................................................................................... 6
3.2. Control Section .................................................................................................................... 8
3.3. Material Section ................................................................................................................. 10
3.4. Geometry Section............................................................................................................... 22
3.5. Boundary Section............................................................................................................... 24
3.6. LoadData Section............................................................................................................... 27
3.7. Example Files..................................................................................................................... 28
CHAPTER 4. Fixed Format Input ............................................................................................ 29
4.1. Control Section .................................................................................................................. 30
4.2. Material section.................................................................................................................. 38
4.3. Nodal coordinates section .................................................................................................. 42
4.4. Element connectivity section ............................................................................................. 43
4.5. Rigid node and facet section .............................................................................................. 44
4.6. Contact section................................................................................................................... 45
4.7. Load curve section ............................................................................................................. 47
4.8. Concentrated nodal force section....................................................................................... 48
4.9. Pressure boundary section.................................................................................................. 49
4.10. Prescribed displacement section ...................................................................................... 50
4.11. Body force section ........................................................................................................... 51
CHAPTER 5. References ........................................................................................................... 52
 2

CHAPTER 1. Introduction
1.1. Overview of FEBio
FEBio is a nonlinear finite element solver that is specifically designed for biomechanical
applications. It offers modeling scenarios, constitutive models and boundary conditions that are
relevant to numerous research areas in biomechanics. This section briefly describes the available
features of FEBio. All features can be used together seamlessly, giving the user a powerful tool
for solving 3D problems in computational biomechanics.

FEBio supports two analysis types, namely quasi-static and quasi-static-poroelastic. In a quasi-
static analysis the (quasi-) static response of the system is sought; inertial terms are ignored. In a
quasi-static-poroelastic analysis a coupled solid-fluid problem is solved. This analysis type is
useful for modeling tissues that have a mobile fluid phase and the accurate modeling of the fluid
behavior is important.

Several nonlinear constitutive models are available, allowing the user to model the often
complicated biological tissue behavior. Several isotropic material models are supported such as
Neo-Hookean, Mooney-Rivlin and Veronda-Westmann. All these models have a non-linear
stress-strain response. A linear elastic model is also available for small strain scenarios and
validation problems. In addition to the isotropic models there are several transversely isotropic
constitutive models available. These models exhibit anisotropic behavior in a single preferred
direction and are useful for representing biological tissues such as tendons, muscles and other
tissues that contain fibers. FEBio also contains a rigid body constitutive model. This model can
be used to represent materials or structures whose deformation is negligible compared to that of
other materials in the overall model.

Biological tissues can interact in very complicated ways. Therefore FEBio supports a wide range
of boundary conditions to model these interactions. These include prescribed displacements,
nodal forces, and pressure forces. Deformable models can be connected to rigid bodies. With this
feature, the user can model prescribed rotations and torques for rigid segments, thereby allowing
the coupling of rigid body mechanics with deformable continuum mechanics. FEBio provides
the ability to represent contact between rigid and/or deformable materials using sliding
interfaces. A sliding surface is defined between two surfaces that are allowed to separate and
slide across each other but are not allowed to penetrate. Finally, the user may specify a body
force to model the effects of, for instance, gravity or base acceleration.

FEBio does not have any mesh generation capabilities. Therefore the input files, which are
described in detail in this document, need to be generated by preprocessing software. The
preferred preprocessor for FEBio is called PreView. PreView can convert some other formats to
the FEBio input specification. For instance, NIKE3D files can be imported in PreView and can
be exported as a FEBio input file.

1.2. About this document

This document is a part of a set of three manuals that accompany FEBio: the User’s manual,
describing how to use FEBio (this manual), a Developer’s Manual for user who wish to modify
 3

or add features to the code, and a Theory Manual, which describes the theory behind the FEBio
algorithms.

This document discusses how to use FEBio and describes the input file formats in detail. Chapter
2 describes how to run FEBio and explains the various command line options. It also discusses
the different files that are required and created by FEBio. FEBio supports two input formats.
Chapter 3 describes the free input format. This input format is an XML-based format that
organizes the data in a convenient hierarchical structure. In chapter 4 the details of the fixed
format input file are described. This format lists the problem data in a predefined sequential list.
The latter format is almost identical to the input specification of NIKE3D. Therefore FEBio can
run NIKE3D input files, however with some limitations as discussed in chapter 4.

Although this document describes some of the theoretical aspects of FEBio, a complete
theoretical development can be found in the FEBio Theory Manual. Developers who are
interested in modifying or extending the FEBio code will find the FEBio Developer’s Manual
very useful.
 4

CHAPTER 2. Running FEBio

2.1. The Command Line
FEBio runs on several different computing platforms including Windows XP, Mac OSX and
Linux. FEBio is started from a shell window (also known as the command prompt in Windows).
The command line is the same for all platforms:

febio [-o1 [name1] | -o2 [name2] | …]

Where -o1, -o2 are options and name1, name2, … are filenames. The different options (of
which most are optional) are given by the following list.

• -i : name of input file in fixed format

• -r : restart file name
• -g[n] : debug flag (does not require a file name)
• -p: plot file name
• -d: dump file name
• -o: log file name

The -i option is used to specify the name of the input file. Note that the format of the input file
is determined by the file extension. A “.n” extension or no extension indicates to FEBio that the
input file is formatted in the fixed format, while the extension “.feb” or “.xml” implies that the
input file is formatted in the free format.

Example: > febio -i input.n

The -r option allows you to restart a previously started analysis. The filename that must follow
this option is the name of a binary “dump” file that was created by a previous call to FEBio. The
-i and -r options are mutually exclusive; only one of them may appear on the command line.

Example: > febio -r f3dmp

The –g[n] option runs FEBio in debug mode. The n must be 0, 1 or 2. See the next section for
more information on running FEBio in debug mode.

Example: > febio -i input.n –g2

After running FEBio, two or three files are created: the log file, the plot file and optionally the
dump file. The log file is a text file that contains the same output that was written to the screen.
The plot file contains the results of the analysis. Since this file is binary, the results must be
analyzed using postprocessing software such as PostView. In some cases the user may wish to
request the creation of a dump file. This file contains temporary results of the run. In case the run
terminates unexpectedly, this file can be used to restart the analysis from the last converged time
 5

step. The names of these files can be specified with the command options –p (plot file), -d
(dump file), -o (log file). If one or more of the file names following these flags are omitted, then
the omitted file name(s) will be given a default name. If the input file was in the fixed format, the
following default names are used: f3plot for the plot file, f3log.txt for the log file and f3dmp for
the dump file. If the input file was in the free format the default file names are derived from the
input file name. For example, if the input file name is input.feb the logfile will have the name
input.log, the plot file is called input.plt and the dump file is called input.dmp.

2.2. Advanced Options

Interrupting a run
The user can pause the run by pressing ctrl+z. This will bring up a menu, and the user can pick
from several options. The user can enter his choice by typing one of the following commands on
the command line.

• cont: continue run without change

• conv: force the current time step to converge
• help: list the available commands with a short explanation
• plot: dump current state to plot database and continue
• quit: exit the application

Note that it may take a while before the command prompt is displayed after FEBio detects the
ctrl+z interruption.

Debugging a run
As stated in section 2.1 you can run FEBio in debug-mode by specifying the –g[n] option on
the command line. Three debug levels are available.

• -g0: debug mode is off. No information is saved to the log file and only the converged
states are written to the plot file.
• -g1: write quasi-Newton convergence information to screen and file and save converged
states to file. (default mode)
• -g2: same as –g1 but also save plot state for every quasi-Newton iteration.

The last option is useful for debugging an input file that passes the syntax check but does not run
to completion (e.g., “error termination”). When no option is entered on the command line, the
default option is –g1.
 6

CHAPTER 3. Free Format Input

In this chapter we describe the free input format used by FEBio. This format is based on the
popular XML format that is used as a standard in web based data representation (ADD REF).
The free input format follows the standard XML conventions and therefore can be viewed with
any file viewer that supports XML files. Since the free format input file is a text file, it can be
edited with any text editor.

Before we start with the format specification it is useful to review the syntax of XML and to
define some terminology. An XML file is composed of a hierarchical list of elements. The first
element is called the root element. Elements can have multiple child elements. All elements are
enclosed by two tags: a tag defining the element and an end tag. A simple example of an XML
file might look like this.

<root>
<child>
<subchild> ... </subchild>
</child>
</root>

The value of an element is enclosed between the name and the end tag.

<element> here is the value </element>

Note that the XML format is case-sensitive.

XML elements can also have attributes in name/value pairs. The attribute value must always be
quoted.

<element attr=”value”>...</element>

Comments can be added as follows.

<!-- This is a comment -->

The first line in the document – the XML declaration – defines the XML version and the
character encoding used in the document. An example can be,

<?xml version="1.0" encoding="ISO-8859-1"?>

3.1. Free format overview

The free format organizes the FEBio input data into hierarchical XML elements. The root
element is called febio_spec. The different sections introduced in this chapter are child elements
of this root element. The following sections are currently defined:

• Control: specifies control and solver parameters.

• Material: Specifies the materials used in the problem and the material parameters.
 7

• Geometry: Defines the geometry of the problem, such as nodal coordinates and element
connectivity.
• Boundary: Defines the boundary conditions that are applied on the geometry.
• LoadData: Defines the load curve data.

Example input files are provided in section 3.7.

 8

3.2. Control Section

The control section is defined by the Control element. This section defines all parameters that are
used to control the evolution of the solution as well as solver parameters. These parameters are
defined as child elements of the Control element. The parameters are grouped into two groups:
mandatory parameters that must be specified and optional parameters.

Mandatory Parameters
The following parameters must be specified in every input file.

Parameter Description
<time_steps> Total number of time steps that will be taken.
<step_size> The initial time step size. (= dt)

Optional Parameters
The following parameters are optional. If not specified they are assigned the default values,
which are found in the last column.

Parameter Description Default

<title> Title of problem (none)
<dtol> Convergence tolerance on displacements 0.001
<etol> Convergence tolerance on energy 0.01
<rtol> Convergence tolerance on residual 1e+10
<lstol> Convergence tolerance on line search 0.9
<time_stepper> Enable the auto time stepper (off)
<max_refs> Max number of stiffness reformations 15
<max_ups> Max number of BFGS stiffness updates 10

If the time_stepper parameter is defined it will enable the auto time-stepper, which will adjust the
time step size based on convergence information. The following sub-elements may also be
defined, although all are optional. Note that these are sub-elements of the time_stepper element
and not of the Control element.

Parameter Description Default

<dtmin> Minimum time step size dt/3
<dtmax> Maximum time step size dt*3
<max_retries> Maximum number of retries allowed per time step 5
<opt_iter> Optimal number of iterations 11

The dtmin and dtmax values are used to constrain the range of possible time step values. The
opt_iter defines the optimal number of quasi-Newton iterations. If the actual number of iterations
is less than or equal to this value the time step size is increased, otherwise it is decreased. The
 9

max_retries parameter determines the maximum number of times a timestep may be retried
before FEBio error terminates.
 10

3.3. Material Section

The material section is defined by the Material element. This section defines all the materials
and material parameters that are used in the model. A material is defined by the material child
element. This element has two attributes: id, which specifies a number that is used to reference
the material, and type, which specifies the type of the material. The material element can also
have a third optional attribute called name, which can be used to identify the material by a text
description. A material definition might look like this:

<material id=”1” type=”elastic”>

Or, if the name attribute is present,

<material id=”2” type=”rigid body” name=”femur”>

The material parameters that have to be entered depend on the material type. The following
material types are available.

• linear elastic: isotropic linear elastic

• St-Venant-Kirchhoff: St Venant-Kirchhoff elastic
• neo-Hookean: Neo-Hookean hyperelastic
• Mooney-Rivlin: Mooney-Rivlin hyperelastic
• Veronda-Westmann: Veronda-Westmann hyperelastic
• trans iso Mooney-Rivlin: transversly isotropic Mooney-Rivlin
• trans iso Veronda-Westmann: transversely isotropic Veronda-Westmann
• rigid body: rigid body material
• poroelastic: poroelastic material

The following sections describe the material parameters for each of the available constitutive
models, along with a short description of each material. A more detailed theoretical description
of the constitutive models can be found in the FEBio Theory Manual.
 11

Linear Elastic
The material type for isotropic linear elasticity is “linear elastic”. The following material
parameters must be defined:

<E> Young’s modulus

<v> Poisson’s ratio

This is the classical material model used in linear elasticity theory. The Cauchy stress is given
by,
σ ij = λ ( tr ε ) δ ij + 2με ij .
2

Here, λ and μ are the Lamé parameters and are related to the more familiar Young’s modulus E
and Poisson’s ratio ν as follows,
νE
λ= ,
(1 + ν )(1 − 2ν )
E
μ= .
2 (1 + ν )
This material model is only valid for small strains and small rotations. It is often convenient to
express the material properties with the bulk modulus K and shear modulus G. To convert to
Young’s modulus and Poisson’s ratio you can use the following formulas.

9 KG
E=
3K + G
3K − 2G
ν=
6 K + 2G

Example:

<material id=”1” type=”linear elastic”>

<E>1000.0</E>
<v>0.45</v>
</material>
 12

St. Venant-Kirchhoff
The material type for a St. Venant-Kirchhoff material is St. Venant-Kirchhoff [1]. The following
material parameters must be defined.

<E> Young’s modulus

<v> Poisson’s ratio

This is an alternative to linear elasticity. It can only be used for small strains but allows large
rotations. Like most of the models used in FEBio it can also be derived from a hyperelastic
strain-energy function, namely
1
W = λ ( tr E ) + μ E : E .
2

2
Here, λ and μ are the Lamé parameters and are related to the more familiar Young’s modulus E
and Poisson’s ratio ν as follows,
νE
λ= ,
(1 + ν )(1 − 2ν )
E
μ= .
2 (1 + ν )

Example:
<material id=”1” type=”St. Venant-Kirchhoff”>
<E>1000.0</E>
<v>0.45</v>
</material>
 13

Neo-Hookean
The material type for a Neo-Hookean material is neo-Hookean. The following parameters must
be defined.

<E> Young’s modulus

<v> Poisson’s ratio

This model describes a compressible Neo-Hookean material [1]. It has a non-linear stress-strain
behavior, but reduces to the classical linear elasticity model for small strains and small rotations.
It is derived from the following hyperelastic strain-energy function:
μ λ
W= ( I1 − 3) − μ ln J + ( ln J )
2

2 2
Here, I1 and I 2 are the first and second invariants of the right Cauchy-Green deformation tensor
C and J is the determinant of the deformation gradient tensor.

This material model uses a standard displacement-based element formulation, so care must be
taken when modeling materials with nearly-incompressible material behavior to avoid element
locking. For the latter case it is recommended to use the Mooney-Rivlin material.

Example:

<material id=”1” type=”neo-Hookean”>

<E>1000.0</E>
<v>0.45</v>
</material>
 14

Mooney-Rivlin
The material type for incompressible Mooney-Rivlin materials is Mooney-Rivlin. The following
material parameters must be defined:

<c1> Coefficient of first invariant term

<c2> Coefficient of second invariant term
<k> Bulk modulus

This material model is a hyperelastic Mooney-Rivlin type with uncoupled deviatoric and
volumetric behavior. The strain-energy function is given by,
( ) ( ) 1
W = C1 I1 − 3 + C2 I2 − 3 + K ( ln J ) .
2
2

C1 and C2 are the Mooney-Rivlin material coefficients. The variables I1 and I2 are the first and
second invariants of the deviatoric right Cauchy-Green deformation tensor C  . The coefficient K
is a bulk modulus-like penalty parameter and J is the determinant of the deformation gradient
tensor. When C2 = 0 , this model reduces to an uncoupled version of the neo-Hookean
constitutive model.

This material model uses a three-field element formulation, interpolating displacements as linear
field variables and pressure and volume ratio as piecewise constant on each element [2].

This material model is useful for modeling certain types of isotropic biological tissues. For
example, the isotropic collagen matrix can be described quit well with this material model.

Example:
<material id=”2” type=”Mooney-Rivlin”>
<c1>1000.0</c1>
<c2>2000.0</c2>
<k>1000</k>
</material>
 15

Veronda-Westmann
The material type for incompressible Veronda-Westmann materials is Veronda-Westmann. The
following material parameters must be defined:

<c1> First VW coefficient

<c2> Second VW coefficient
<k> Bulk modulus

This model is similar to the Mooney-Rivlin model in that it also uses an uncoupled deviatoric
dilatational strain energy:
(C ( I − 3))
⎢⎣ (
⎥⎦
CC
2
)
W = C1 ⎡e 2 1 − 1⎤ − 1 2 I2 − 3 + U ( J )

The dilatational term is identical to the on used in the Mooney-Rivlin model. This model can be
used to describe certain types of biological materials that display exponential stiffening with
increasing strain. It has been used to describe the response of skin tissue [3].

Example:
<material id=”2” type=”Veronda-Westmann”>
<c1>1000.0</c1>
<c2>2000.0</c2>
<k>1000</k>
</material>
 16

Transversely Isotropic Mooney-Rivlin

The material type for transversely isotropic Mooney-Rivlin materials is “trans iso Mooney-
Rivlin”. The following material parameters must be defined:

<c1> Mooney-Rivlin coefficient 1

<c2> Mooney-Rivlin coefficient 2
<c3> Exponential stress coefficient
<c4> Fiber uncrimping coefficient
<c5> Modulus of straightened fibers
<k> Bulk modulus
<lam_max> Fiber stretch for straightened fibers
<fiber> Fiber distribution option

This constitutive model can be used to represent a material that has a single preferred fiber
direction and was developed for application to biological soft tissues [4-6]. It can be used to
model tissues such as tendons, ligaments and muscle. The elastic response of the tissue is
assumed to arise from the resistance of the fiber family and an isotropic matrix. It is assumed that
the strain energy function can be written as follows:
( ) ( ) K
W = F1 I1 , I2 + F2 λ + ⎡⎣ln ( J ) ⎤⎦ .
2

2
 
Here I1 and I 2 are the first and second invariants of the deviatoric version of the right Cauchy
Green deformation tensor C and λ is the deviatoric part of the stretch along the fiber direction
( λ 2 = a0 ⋅ C ⋅ a0 , where a0 is the initial fiber direction), and J = det ( F ) is the Jacobian of the
deformation (volume ratio). The function F1 represents the material response of the isotropic
ground substance matrix and is the same as the Mooney-Rivlin form specified above, while
F2 represents the contribution from the fiber family. The strain energy of the fiber family is as
follows:
∂F
λ 2 = 0, λ ≤ 1
∂λ
∂F
( )
λ 2 = C3 e 4 ( ) − 1 , 1 < λ < λm
∂λ
C λ −1

∂F
λ 2 = C5 + C6λ, λ ≥ λm
∂λ
Here, C1 and C2 are the Mooney-Rivlin material coefficients, lam_max ( λm ) is the stretch at
which the fibers are straightened, C3 scales the exponential stresses, C4 is the rate of uncrimping
of the fibers, and C5 is the modulus of the straightened fibers. C6 is determined from the
requirement that the stress is continuous at λm .

This material model uses a three-field element formulation, interpolating displacements as linear
field variables and pressure and volume ratio as piecewise constant on each element [2].
 17

The element fiber direction is specified using the fiber parameter. This parameter has an attribute
named type that can take the following values.

• local: material axis determined by local element nodes. The value is interpreted as the
local node numbers of the nodes that define the direction of the axis. The following
example defines a local fiber axis by local element nodes 1 and 2.

<fiber type=”local”>1,2</fiber>

4 3

1 2
Figure 3-1. local fiber direction option .

• spherical: material axis is determined by a point in space and the global location of each
element integration point. The value is the location of the point. The following example
defines a spherical fiber distribution centered at [0,0,1].

<fiber type=”spherical”>0,0,1</fiber>

4 3
a
x
a

1 2
Figure 3-2. spherical fiber direction option.

• vector: material axis is specified by vector. The value is the direction of the material axis.
The following defines all element fiber directions in the direction of the vector [1,0,0].

<fiber type=”vector”>1,0,0</fiber>
 18

4 3

a
r

1 2
Figure 3-3. vector fiber direction option.

If the type attribute is omitted the fiber distribution follows the local element nodes 1 and 2. So,
this would be the same as adding a fiber parameter such as the example above under local.

Active stress along the fiber direction can be simulated using an active contraction model. To use
this feature you need to define the active_contraction parameter. This parameter takes an
optional attribute, lc, which defines the loadcurve. There are also several options.

<ca0> Intracellular calcium concentration

<beta> tension-sarcomere length relation constant
<l0> Unloaded sarcomere length
<refl> No tension sarcomere length

Example:
This example defines a transversely isotropic material with a Mooney-Rivlin basis. It defines a
homogeneous fiber direction and uses the active fiber contraction feature.

<material id=”3” type=”trans iso Mooney-Rivlin”>

<c1>13.85</c1>
<c2>0.0</c2>
<c3>2.07</c3>
<c4>61.44</c4>
<c5>640.7</c5>
<lam_max>1.03</lam_max>
<fiber type=”vector”>1,0,0</fiber>
<active_contraction lc=”1”>
<ca0>4.35</ca0>
<beta>4.75</beta>
<l0>2.04</l0>
<refl>1.58</refl>
</active_contration>
</material>
 19

Transversely Isotropic Veronda-Westmann

The material type for transversely isotropic Veronda-Westmann materials is “trans iso Veronda-
Westmann”. The following material parameters must be defined:

<c1> Veronda-Westmann coefficient 1

<c2> Veronda-Westmann coefficient 2
<c3> Exponential stress coefficient
<c4> Fiber uncrimping coefficient
<c5> Modulus of straightened fibers
<k> Bulk modulus
<lam_max> Fiber stretch for straightened fibers
<fiber> Fiber distribution option.

This material differs from the Transversely Isotropic Mooney-Rivlin model in that it uses the
Veronda-Westmann model for the isotropic matrix. The interpretation of the material parameters,
except C1 and C2 is identical to this material model.

The fiber distribution option is the same as the Mooney-Rivlin transversely isotropic model. An
active contraction model can also be defined for this material. See the transversely isotropic
mooney-rivlin model for more details.

Example:
This example defines a transversely isotropic material model with a Veronda-Westmann basis.
The fiber direction is implicitly implied as local.

<material id=”3” type=”trans iso veronda-Westmann”>

<c1>13.85</c1>
<c2>0.0</c2>
<c3>2.07</c3>
<c4>61.44</c4>
<c5>640.7</c5>
<lam_max>1.03</lam_max>
</material>
 20

Poroelasticity
The material type for poroelastic materials is poroelastic. The following material parameters
must be defined:

<perm> Permeability
<solid_id> Material ID of the solid phase

This material model uses the biphasic theory for describing the time-dependent material behavior
of materials that consist of both a solid and fluid phase [7, 8]. We refer to the FEBio Theory
Manual for a more indebt description of the biphasic theory.

This material model uses a “displacement-pressure”, or “u-p” finite element formulation [9].
Solid nodal displacements and fluid nodal pressures are interpolated using trilinear shape
functions. Details of this finite element formulation can be found in [10]. At this point the
boundary conditions that apply to the pressure degree of freedom are limited to prescribed
pressures. See section 3.5 for more details.

A major strength of this implementation of poroelasticity is the seamless integration of solid

elements (with degrees of freedom only for the solid displacement) and biphasic elements (which
also include degrees of freedom for fluid pressure), thanks to a judicious choice of primary
variables and natural boundary conditions in the formulation of the governing equations. Thus,
at the interface of a solid element and a biphasic element, continuity of the solid displacement is
automatically satisfied, while the fluid flux across this interface automatically reduces to zero,
since the solid element is impermeable by definition.

The parameter perm defines the permeability of the solid. The solid_id parameter is the material
number that is used to describe the solid phase. In other words, to use this material model you
need to define at least two materials: a material for the solid phase and a poroelastic material.
The solid phase may be described by any of the available material models except rigid body and
poroelastic.

Example:
This example defines a poroelastic material that uses a neo-Hookean model for the solid phase.

<material id=”1” type=”neo-Hookean”>

<E>1.0</E>
<v>0.4</v>
</material>
<material id=”2” type=”poroelastic”>
<perm>0.001</perm>
<solid_id>1</solid_id>
</material>
 21

Rigid Body
A rigid body can be defined using the rigid material model. The material type for a rigid body is
“rigid body”. The following parameters are defined:

<density> Density of rigid body

<center_of_mass> Position of the center of mass
<trans_x> x-translation code. Attribute type specifies the type of the translation.
<trans_y> y-translation code. Attribute type specifies the type of the translation.
<trans_z> z-translation code. Attribute type specifies the type of the translation.
<rot_x> x-rotation code. Attribute type specifies the type of the rotation.
<rot_y> y-rotation code. Attribute type specifies the type of the rotation.
<rot_z> z-rotation code. Attribute type specifies the type of the rotation.

If the center_of_mass parameter is omitted then FEBio will calculate the center of mass
automatically. In this case you must specify a density. If the center_of_mass is defined the
density parameter may be omitted. Omitting both will generate an error message.

The translation and rotation boundary codes are used to specify the motion of the rigid body. The
full syntax for e.g. the x-translation is as follows.

<trans_x type=”<type>” [lc=”<lc>”]> 1.0 </trans_x>

Here, the type can be either free, fixed or prescribed. If the type is prescribed then the lc attribute
can be used to specify a load curve. The value is then interpreted as a scale factor. For all other
types the value is ignored. Omitting this translation code is the same as specifying a fixed code.
The syntax and interpretation is the same for the other translation and rotation codes.

Example:
<material id=”5” type=”rigid body”>
<center_of_mass>1.0,2.0,3.0</center_of_mass>
<trans_x type=”prescribed” lc=”2”>2.0</trans_x>
</material>

In this example the center of mass is located at (1,2,3) and the x-displacement of the rigid body is
prescribed by a load curve. The value of the load curve is scaled by 2. The other translational and
rotational degrees of freedom are fixed.
 22

3.4. Geometry Section

The geometry section contains all the geometry data including nodal coordinates and element
connectivity. It has the following sub-sections.

• Nodes: contains nodal coordinates.

• Elements: contains element connectivity

Nodes Section
The nodes section contains all the nodal coordinates. Repeat the following XML-element for
each node.

<node id=”n”>x,y,z</node>

The id attribute is a unique number that identifies the node. This id is used as a reference in the
element connectivity section. The ids do not have to be in order but the lowest id must be 1 and
numbers can not be skipped. (So if there is a node 4 and a node 6, there must also be a node 5
somewhere).

Elements Section
The element section contains all the element connectivity data. Different element types have
different tags. The following solid elements are available in FEBio.

hex8: 8-node hexahedral element

penta6: 6-node pentahedral element
tet4: 4-node tetrahedral element

The value for the XML element is the nodal connectivity.

<hex8 id=”n” mat=”m”>n1,n2,n3,n4,n5,n6,n7,n8</hex8>

<penta6 id=”n” mat=”m”>n1,n2,n3,n4,n5,n6</penta6>
<tet4 id=”n” mat=”m”>n1,n2,n3,n4</tet4>

Elements have two attributes. First, there is a unique id that can be used to reference the element.
The same limitations to this id apply as to the nodal ids. And second a material number. This
number is the material id that was defined in the material section.

Surface Elements
In many cases the surface (or part of it) of the geometry is required. For example, pressure forces
are applied to the surface. For this reason surface elements have to be defined. Two surface
elements are available.

• quad4: 4-node quadrilateral element

• tri3: 3-node triangular element

The value for the surface element is the nodal connectivity.

 23

<quad4 id=”n”>n1,n2,n3,n4</quad4>
<tri3 id=”n”>n1,n2,n3</tri3>

Surface elements can not overlap element boundaries. That is, the surface element must belong
to a specific element. Surface elements do not contribute to the total number of elements in the
mesh. They are also not to be confused with shell elements; currently only rigid shell elements
are supported in FEBio.
 24

3.5. Boundary Section

The Boundary section defines all boundary conditions that are applied to the geometry. Several
boundary conditions are available: nodal displacements, nodal forces, pressure forces, contact
interfaces and rigid joints.

Prescribed Nodal Degrees of Freedom

Nodal degrees of freedom (dof) can be prescribed using the prescribe sub-element.

<prescribe>
<node id=”n” bc=”x” [lc=”l”]>1.0</node>
...
</prescribe>

The id attribute indicates to which node this prescribed dof is applied. The bc attribute gives the
degree of freedom. The following values are allowed.

x: apply displacement in x-direction

y: apply displacement in y-direction
z: apply displacement in z-direction
p: apply prescribed fluid pressure (only used in poroelastic analysis)

An optional loadcurve can be given as well with the lc attribute. If no loadcurve is present the
value will be automatically ramped from 0 to the value specified in the xml file.

Degrees of freedom that are fixed (in other words are always zero) can be defined by the fix
element.
<fix>
<node id=”n” bc=”x”></node>
...
</fix>

In this case the value of the element is ignored. Of course, the prescribe element with a value of
zero for the node tags can also be used to constrain a certain nodal degree of freedom. The
advantage of the fix element is that the equation corresponding to the constrained degree of
freedom is removed from the linear system of equations. This reduces the run time of the FE
analysis.

Nodal forces
Nodal forces are applied by the force element. These forces always point in the same direction
and do not deform with the geometry.

<force>
<node id=”n” bc=”x” [lc=”l”]>1.0</node>
...
</force>
 25

The id attribute indicates to which node this prescribed dof is applied. The bc attribute gives the
degree of freedom. The following values are allowed.

x: apply force in x-direction

y: apply force in y-direction
z: apply force in z-direction

An optional load curve can be given as well with the lc attribute. If no loadcurve is present the
value will be automatically ramped from 0 to the value specified in the xml file.

Pressure forces
Pressure forces are applied to the surface of the geometry and are defined by the pressure
element.

<pressure>
<quad4 id=”n” [lc=”l” scale=”1.0”]>n1,n2,n3,n4</quad4>
...
</pressure>

The sub-elements define the geometry of the pressure boundary surface. Each element must be
either quad4 for a four-node quadrilateral or tri3 for a 3-noded triangular element. The id n
references the surface element. These pressure forces are also known as follower forces; they
change direction as the body is deformed and, in this case, are always oriented along the local
surface normal. The sign convention is so that a positive pressure will act opposite to the normal,
so it will compress the material. The optional parameter lc defines a loadcurve for the pressure
evolution and scale defines a scale factor. The default scale factor is 1.0 and if lc is not defined
the scale factor is automatically ramped from 0 to the value specified here.

Contact Interfaces
Contact boundary conditions are defined with the contact element. The type attribute specifies
what type of contact interface you wish to define.

<contact type=”sliding_with_gaps”> ... </contact>

The type can be one of the following options:

• sliding_with_gaps: defines a sliding interface that may separate

• rigid: defines a rigid interface
• rigid_joint: defines a joint between two rigid bodies

Sliding Interfaces
This type of contact interface demands the declaration of a slave and master surface, in addition
to some control parameters. The control parameters are given by.

tolerance augmentation tolerance 0.01

 26

penalty penalty factor 1.0

The slave and master surfaces are entered by specifying the surface element. The type attribute is
used to specify whether it is a slave or master surface. The surface tag is followed by the
definition of the surface elements.

<surface type=”master”>
<quad4 id=”n”>n1,n2,n3,n4</quad4>
...
</surface>

Instead of quadrilateral surface elements (quad4) you may also use triangular elements (tri3).

Rigid Interfaces
A rigid interface defines a list of nodes that are attached to a rigid body. These nodes will move
with the rigid body.

<contact type=”rigid”>
<node id=”n1” rb=”1”></node>
...
<node id=”n2” rb=”1”></node>
</contact>

The id attribute identifies the node and rb is the material id of the rigid body. The value of the
node is ignored.

Rigid Joints
A rigid joint connects two rigid bodies at a point in space.

<contact type=”rigid joint”>

<tolerance>0.1</tolerance>
<penalty>2</penalty>
<body1>1</body1>
<body2>2</body2>
<joint>0,0,0</joint>
</contact>

The tolerance element defines the augmentation tolerance. That is, when the relative change in
the constraint forces (the Lagrange multipliers) are less than this value. The body1 and body2
elements are the material numbers of the two rigid bodies. The joint element defines the position
of the joint in world coordinates at the start of the analysis. Note that this point does not have to
be inside or on the surface of either of the two bodies.
 27

3.6. LoadData Section

The LoadData section contains the loadcurve data. A loadcurve is defined by the loadcurve
element. Each loadcurve is defined by repeating the loadpoint element for all data points.

<loadcurve id=”1”>
<loadpoint> 0, 0 </loadpoint>
...
<loadpoint> 1, 1 </loadpoint>
</loadcurve>

The id attribute is the loadcurve number and is used in other sections of the input file as a means
to reference this curve.

FEBio also defines a “zero” loadcurve. This loadcurve is used for all time-dependent parameters
that have no loadcurve specified (that is, the lc attribute is omitted). This loadcurve will have the
effect that parameters are ramped up linearly from zero to their specified value. Imagine for
instance, a rigid body defined as follows.

<material id=”1” type=”rigid body”>

<density>1.0</density>
<trans_x type=”prescribed”>2.0</trans_x>
</material>

The x degree of freedom is prescribed but no loadcurve is specified. FEBio will automatically
use the “zero” loadcurve and ramp up the value from 0 at the start time to 2 at the end time.
 28

3.7. Example Files

<?xml version="1.0" encoding="ISO-8859-1"?>

<!-- This is an example of the free format input for FEBio -->
<febio_spec version="0.1">
<Control>
<time_steps>10</time_steps>
<step_size>0.1</step_size>
</Control>
<Material>
<material id="1" type="neo-Hookean">
<E>100.0</E>
<v>0.45</v>
</material>
</Material>
<Geometry>
<Nodes>
<node id="1">0,0,0</node>
<node id="2">1,0,0</node>
<node id="3">1,1,0</node>
<node id="4">0,1,0</node>
<node id="5">0,0,1</node>
<node id="6">1,0,1</node>
<node id="7">1,1,1</node>
<node id="8">0,1,1</node>
</Nodes>
<Elements>
<hex8 id="1" mat="1">1,2,3,4,5,6,7,8</hex8>
</Elements>
</Geometry>
<Boundary>
<fix>
<node id=”1” bc=”xyz”></node>
<node id=”4” bc=”xyz”></node>
<node id=”5” bc=”xyz”></node>
<node id=”8” bc=”xyz”></node>
</fix>
<prescribe>
<node id="2" bc="x">1.0</node>
<node id="3" bc="x">1.0</node>
<node id="6" bc="x">1.0</node>
<node id="7" bc="x">1.0</node>
</prescribe>
</Boundary>
</febio_spec>
 29

CHAPTER 4. Fixed Format Input

This chapter describes the fixed format for the FEBio input file. By “fixed”, we mean that each
section of the input file must follow the order specified in this document. This also applies for
each line of each section and for each entry on each line. In addition, every entry on each line has
a fixed number of columns where its value may appear on the line. The user must pay careful
attention to these restrictions, since FEBio may not be able to find all input mistakes related to
incorrectly placed parameters. This format is identical to the format used by NIKE3D [11],
developed and maintained by the Methods Development Group at Lawrence Livermore National
Laboratory, in the sense that FEBio can read and run most NIKE3D input files, though FEBio
does not support all NIKE3D features. Although our experience has shown that all FEBio input
files using this format can be run with NIKE3D, we have not performed exhaustive testing. The
input format for NIKE3D is detailed in the NIKE3D User’s Manual. In this manual we only
describe the features supported by FEBio.

The fixed format for the input file is composed of a series of sections:

1. Control section
2. Material section
3. Nodal coordinates section
4. Element connectivity section
5. Rigid node and facet section
6. Contact section
7. Load curve section
8. Concentrated nodal force section
9. Pressure boundary section
10. Prescribed displacement section
11. Body force section

Of all these sections, only the first four sections will appear in every input file. The other
sections are optional and only need to be specified if the corresponding entry in the control
section is non-zero. Note that all sections must appear in the order they are described in this
manual. The rest of this chapter describes the sections in detail.

Note that any line that contains an asterisk (“*”) in the first column is interpreted as a comment
and is ignored by FEBio.
 30

4.1. Control Section

The control section contains entries that describe the size of the problem as well as the solver
parameters. It is composed of ten lines, and although the last three lines are ignored in FEBio,
their presence is required for compatibility with NIKE3D.

Control Line 1
Colums _____________________________________________ Format
1 – 72 heading to appear on output. (1) A256

Comments.
1. This line contains the problem title. In NIKE3D this line is limited to 72 characters, but in
FEBio this line can be up to 256 characters. Note however that in FEBio (as well as in
NIKE3D) only the first forty characters are saved to the plot file, a limitation that is
imposed by the TAURUS file format used by NIKE3D and other LLNL codes including
DYNA3D.
 31

Control Line 2
Colums _____________________________________________ Format
1-2 Input format (1) A2

3-5 Number of materials I3

6-15 Number of node points I10

16-25 Number of solid elements (2) I10

26-35 (not used)

36-45 Number of shell elements (3) I10

46-50 (not used)

51-55 Number of sliding interfaces I5

56-65 (not used)

66-70 Number of rigid nodes and facet lines I5

Comments
1. The input format must be “FL”, which corresponds to NIKE3D version 3.0 and later.
2. This number includes all hexahedral (“brick”), pentahedral (“wedge”) and tetrahedral
elements.
3. Currently FEBio only supports rigid shells. These are shells that are assigned to a rigid
material. FEBio also ignores shell thicknesses.
 32

Control Line 3
Colums _____________________________________________ Format
1-10 Number of time or load steps (ntime) I10

11-20 Time or load step size (dt) E10.0

21-25 Auto time/load step control flag (1) 1x,A4

EQ. “ “: use fixed time step size
EQ. “AUTO”: enable standard automatic time step control

26-30 Maximum number of retries allowable per step (2) I5

EQ.0: default = 5

31-35 Optimal number of iterations per step (3) I5

EQ.0: default = 11

36-45 Minimum allowable step/load size E10.0

EQ.0: default = dt / 3

46-55 Maximum allowable step/load size E10.0

EQ.0: default = dt*3

Comments
1. The automatic time stepping algorithm will adjust the time step size based on
convergence information returned by the quasi-Newton solver. The time step size is
bracketed by the minimum and maximum values that are input on this line. The algorithm
also ensures that FEBio attempts to obtain a converged state at the desired termination
time, determined by (ntime × dt).
2. This is the maximum number of attempts that the auto time stepper retries a failed quasi-
Newton iteration. After each attempt the auto time stepper decreases the time step and
restarts the iteration. When the minimum time step size is reached before the maximum
number of iterations, the program ends in error termination.
3. The optimal number of iterations is the expected average number of iterations for each
time step. By comparing this number with the actual number of iterations, the auto time
stepper decides whether to increase or decrease the time step size.
 33

Control Line 4
Colums _____________________________________________ Format
1-5 Number of load curves (1) I5

6-10 Maximum number of points defining any load curve (2) I5

11-15 Number of concentrated nodal loads I5

16-20 Number of element surfaces with applied pressure loading I5

21-25 Number of displacement boundary conditions I5

26-35 (not used)

36-40 Body force loads due to base acceleration in x-direction I5

EQ.0: no x-acceleration
NE.0: x-acceleration

41-45 Body force loads due to base acceleration in y-direction I5

EQ.0: no y-acceleration
NE.0: y-acceleration

46-50 Body force loads due to base acceleration in z-direction I5

EQ.0: no z-acceleration
NE.0: z-acceleration

Comments
1. A loadcurve is simply a list of data pairs that represents a curve of time versus load. The
load can be interpreted in different ways. For example, it might represent a rigid body
displacement, a concentrated nodal force, or any other variable that may depend on time.
2. This number is ignored in the current version of FEBio.
 34

Control Line 5
Colums _____________________________________________ Format
1-35 (not used)

36-40 Dump file creation flag (1) I5

EQ.0: no dump file will be created
EQ.1: dump file will be created

Comments
1. The dump file allows the user to restart an unfinished problem at a later time. When the
flag is set to 1, FEBio will create a dump file (called f3dump) that can be used to restart
the analysis from the last converged time step.
 35

Control Line 6
Colums _____________________________________________ Format
1-30 (not used)

31-35 Maximum number of Quasi-Newton (1) equilibrium iterations I5

permitted between stiffness matrix reformations.
EQ.0: default set to 10

36-40 Maximum number of stiffness matrix reformations per time step (2) I5
EQ.0: default set to 15

41-50 Convergence tolerance on displacements E10.0

EQ.0: default set to 0.001

51-60 Convergence tolerance on energy E10.0

EQ.0: default set to 0.01

61-70 Convergence tolerance on residual E10.0

EQ.0: default set to 1E+10 (i.e. deactivated)

71-80 Convergence tolerance on line search E10.0

EQ.0: default set to 0.9

Comments
1. FEBio uses the BFGS method [12] to solve the nonlinear finite element equations. In this
method the stiffness matrix does not get recalculated during every iteration (such as in the
Full-Newton method). In stead it calculates an approximation of the (inverse of the)
stiffness matrix, which is called a stiffness update. The second parameter on this control
line sets the maximum number of such updates (one during every iteration) before the
exact stiffness matrix is reformed.
2. This parameter sets the maximum number of stiffness reformations per time step. If this
number is reached the time step fails and, if the auto-time stepper is activated, a smaller
time step will be tried.
 36

Control Line 7
Colums _____________________________________________ Format
1-5 analysis type I5
EQ. 0: quasi-static analysis
 37

Control Line 8
This line is not used in FEBio.

Control Line 9
This line is not used in FEBio.

Control Line 10
This line is not used in FEBio.
 38

4.2. Material section

The Material Section contains all the material parameters of all the materials used in the
problem. Repeat the following set of eight lines for each material. If the material is used for shell
elements then an additional two lines are expected. Although FEBio reads those lines in, it
currently ignores the values. The total number of materials is entered on control line 2. For
detailed descriptions of the materials, please see Section 3.3 above.

Line 1
Colums _____________________________________________ Format
1-5 Material identification number I5

6-10 Material Type I5

EQ.1: neo-Hookean material
EQ.15: Mooney-Rivlin
EQ.18: Transversely isotropic Mooney-Rivlin
EQ.20: Rigid body

11-20 Density E10.0

21-25 Element class for which this material is used I5

EQ.20: brick element (also pentahedral and tetrahedral element)
EQ.2: shell element

Line 2
Colums _____________________________________________ Format
1-72 Material title
 39

Material Type 1 – Neo-Hookean

Colums _____________________________________________ Format

1-10 Line 3 Young’s modulus E10.0

1-10 Line 4 Poisson’s ratio E10.0

Line 5-8 blank

Material Type 15 – Mooney Rivlin Hyperelastic

Colums _____________________________________________ Format
1-10 Line 3 Coefficient of first invariant A E10.0

1-10 Line 4 Coefficient of second invariant B E10.0

1-10 Line 5 Poisson’s ratio ν E10.0

Line 6-8 Blank

Material Type 18 – Transversely Isotropic Hyperelastic

Colums _____________________________________________ Format
1-10 Line 3 Isotropic material coefficient C1 E10.0
11-20 Isotropic material coefficient C2 E10.0
21-30 Exponential stress coefficient C3 E10.0
31-40 Fiber uncrimping coefficient C4 E10.0
41-50 Modulus of straightened fibers C5 E10.0

1-10 Line 4 Bulk modulus K E10.0

11-20 Fiber stretch for straightened fibers λm E10.0

1-10 Line 5 (not used)

1-10 Line 6 Material axes option, AOPT E10.0

Fiber axis is always aligned with load axis a

EQ.0: local material axes given by local element

Nodes specified on line 7 below. Line 8 is blank

EQ.1: local material axes determined by a point in

space and global location of each element integration
 40

point. Line 8 is blank.

EQ.2: local material axes determined by normalized

vector a.

1-30 Line 7 AOPT.EQ.0: local element nodes (default=1,2,4) 3E10.0

AOPT.EQ.1: x p , y p , z p 3E10.0
AOPT.EQ.2: a1 , a2 , a3 3E10.0

1-30 Line 8 (not used) 3E10.0

31-40 AC – (active contraction) load curve number E10.0

41-50 AC – intracellular calcium concentration E10.0

51-60 AC – tension-sarcomere length relation constant E10.0

61-70 AC – Unloaded sarcomere length E10.0

71-80 AC – no tension sarcomere length E10.0

The local fiber direction is specified using the AOPT parameter. For AOPT=0 the material axis
is defined by the local element node numbering specified on line 7. For AOPT=1 the material
axis is defined a fixed point in space and the global position of the element Gauss points. The
normalized vector connecting these two points defines the axis. For AOPT=2 the material axis is
defined by a global vector specified on line 7.
 41

Material type 20 – Rigid Body

Colums _____________________________________________ Format
Line 3 Blank

Line 4 Blank

1-10 Line 5 X-translational boundary code E10.0

11-20 Y-translational boundary code E10.0

21-30 Z-translational boundary code E10.0

31-40 X-rotational boundary code E10.0

41-50 Y-rotational boundary code E10.0

51-60 Z-rotational boundary code E10.0

1-10 Line 6 Center of mass input flag E10.0

NE.1: FEBio computes center of mass
EQ.1: read user supplied coordinates below

11-20 X – coordinate of center of mass E10.0

21-30 Y – coordinate of center of mass E10.0

31-40 Z – coordinate of center of mass E10.0

Line 7-8 Blank

Nodal constraints and displacement boundary conditions on rigid body nodes are ignored.
Instead, constraints are applied at the rigid body center of mass. The coordinates of the center of
mass are either computed by FEBio or may be input by the user. The boundary condition codes
are interpreted as:

code < 0: fixed

code = 0: free
code > 0: number of load curve for prescribed displacement or
rotation (in radians).
 42

4.3. Nodal coordinates section

The following line is repeated for every node of the mesh. The total number of nodes is entered
on line 2 of the Control section.

Colums _____________________________________________ Format

1-8 Node number I8

9-13 Displacement boundary condition code I5

EQ.0: no constraints
EQ.1: constrained x displacement
EQ.2: constrained y displacement
EQ.3: constrained z displacement
EQ.4: constrained x and y displacement
EQ.5: constrained y and z displacement
EQ.6: constrained z and x displacement
EQ.7: constrained x, y and z displacement

14-33 x – coordinate E20.0

34-53 y – coordinate E20.0

54-73 z – coordinate E20.0

74-78 (ignored) I5
 43

4.4. Element connectivity section

The following line is repeated for every element in the mesh. The total number of elements is
entered on line 2 of the Control section.

Colums _____________________________________________ Format

1-8 Element number I8

9-13 Material number I5

14-21 Node point 1 I8

22-29 Node point 2 I8

… …

70-77 Node point 8 I8

Nodes 1 through 8 define the corner nodes of the 8-node hexahedral or “brick” element. In the
case of a pentahedral or “wedge” element the sixth node is repeated for entry 7 and 8. In the case
of a tetrahedral element, the fourth node is repeated four times.

• hex : n1 , n2 , n3 , n4 , n5 , n6 , n7 , n8
• penta : n1 , n2 , n3 , n4 , n5 , n6 , n6 , n6
• tet : n1 , n2 , n3 , n4 , n4 , n4 , n4 , n4
 44

4.5. Rigid node and facet section

The number of rigid node and facet lines is defined in the control section on line 2.

Colums _____________________________________________ Format

1-5 Rigid body number I5

6-13 Node point 1 I8

14-21 Node point 2 I8

22-29 Node point 3 I8

30-37 Node point 4 I8

Rigid nodes and facets move as part of the indicated rigid body. The input is a very general list
of node points, four (or less) per line. The nodes need not be unique and need not form a facet of
an element, although this method of input is often convenient. This feature gives a convenient
way to attach a portion of a deformable mesh to a rigid body without merging node points or
using a tied interface. Note that the nodes in this section must be part of a deformable mesh; this
section cannot be used to define new geometry.
 45

4.6. Contact section

Contact between objects is defined via sliding interface surfaces. The number of sliding
interfaces is defined on control line 1 of the control section. Each sliding interface consists of a
master surface and a slave surface. When using a two-pass algorithm for enforcement of the
contact constraint (REF), the definition of master and slave surfaces is arbitrary. When using the
single-pass algorithm, the results can be influenced by the choice of slave and master surfaces. It
is best to use the most tessellated surface as the master. The master and slave contact surfaces are
described by “facets” of existing elements in the model. A facet is defined by the node numbers
that construct either a quadrilateral or triangular polygon. For each sliding interface the
following lines are repeated.

Control line
Colums _____________________________________________ Format
1-8 Number of slave facets I8

9-16 Number of master facets I8

17-20 Interface type I4

EQ.3: two-pass algorithm
EQ.-3: single-pass algorithm

21-30 Penalty factor (1) E10.0

31-75 (not used)

76-80 Auxiliary interface control line flag (IAUG) I5

EQ.1: read an auxiliary interface control line flag
immediately following this line.

Comments
1. The penalty factor must be entered as a negative number to be consistent with Nike3D,
since in Nike3D a positive number is interpreted as a scale factor to an automatically
calculated penalty factor. If a negative number is entered, the absolute value is used as
the actual penalty factor. In FEBio the absolute value of this number is always used as the
actual penalty factor.

Auxiliary control line

Add this control line immediately after the control line if IAUG equal 1.

Colums _____________________________________________ Format

1-5 (not used)
6-15 Normal direction convergence tolerance for augmentations (ALTOL) E10.0
GT.0: converged when force norm < ALTOL
EQ.0: DEFAULT = 0.1
 46

Slave facet cards

For each slave facet repeat the following line. There must be at least three non-repeated node
numbers but no more than four node numbers per line. Triangular facets are entered by repeating
the last node.

Colums _____________________________________________ Format

1-8 slave facet number I8

9-16 node point n1 I8

17-24 node point n2 I8

25-32 node point n3 I8

33-40 node point n4 I8

Master facet cards

The format for entering master facets is the same as for the slave facets. For each master facet
repeat the following line.

Colums _____________________________________________ Format

1-8 slave facet number I8

9-16 node point n1 I8

17-24 node point n2 I8

25-32 node point n3 I8

33-40 node point n4 I8

If there are multiple sliding interfaces, the order of the cards is as follows.

Control line 1 – sliding interface 1

Control line 2 – sliding interface 1 (optional)
...
Control line 1 – sliding interface n
Control line 2 – sliding interface n (optional)
Slave surface – sliding interface 1
Master surface – Sliding interface 1
...
Slave surface – sliding interface n
Master surface – Sliding interface n
 47

4.7. Load curve section

A load curve is a list of data pairs that represent a discretized function of time. They are used to
specify the magnitude and/or time variation of many types of data, including boundary
conditions, material properties and motion of rigid materials. Any boundary condition or
property may reference any load curve, and a load curve may be referenced more than once.

Define the number of load curves on Control line 4. Repeat the following lines for every load
curve:

Line 1
Colums _____________________________________________ Format
1-5 Load curve number I5
6-10 Number of points in load curve (pts) I5

Line 2,…,pts
Colums _____________________________________________ Format
1-10 Time E10.0

11-20 Load curve value E10.0

The value of a time point in between two load points is calculated using linear interpolation. The
value of a time point outside the domain of the load curve is clamped to the nearest data point.
 48

4.8. Concentrated nodal force section

Define the number of concentrated nodal loads as specified on control line 4. Repeat the
following line for every nodal load.

Colums _____________________________________________ Format

1-8 Node point number on which this load acts I8

9-13 Direction in which load acts I5

EQ.1: x – direction force
EQ.2: y – direction force
EQ.3: z – direction force

14-18 Load curve number I5

19-28 Scale factor E10.0

EQ.0: default set to “1.0”

Concentrated nodal loads are defined in global coordinates. In contrast to “follower forces”,
concentrated nodal loads act in a fixed direction as over the course of the analysis. Hence, loads
that were defined normal to a surface in the undeformed configuration may not remain normal if
the surface undergoes large deformation. Pressure loads may be used to maintain normality
during large deformation.
 49

4.9. Pressure boundary section

Define the number of lines as specified on control card 4. Repeat the following line for every
pressure load. For triangular facets repeat the third point.

Colums _____________________________________________ Format

1-5 Load curve number I5

6-13 Node point 1 I8

14-21 Node point 2 I8

22-29 Node point 3 I8

30-37 Node point 4 I8

38-47 Scale factor at node 1 E10.0

48-57 Scale factor at node 2 E10.0

58-67 Scale factor at node 3 E10.0

68-77 Scale factor at node 4 E10.0

Pressure forces are “follower forces”. They always remain normal to the surface even under
large deformations. The magnitude of the pressure is determined by the load curve value and the
scale factor. When a scale factor is zero, it is not replaced by the default value of 1.0. Only when
all four scale factors are zero, then they are replaced with the default values.
 50

4.10. Prescribed displacement section

Define the number of lines as specified on control line 4.

Colums _____________________________________________ Format

1-8 Node point number to which this node is applied I8

9-13 Global direction in which node is displaced I5

EQ.1: translation in x – direction
EQ.2: translation in y – direction
EQ.3: translation in z – direction

14-18 Load curve number I5

19-28 Scale factor on load curve E10.0

EQ.0: default set to “1.0”
 51

4.11. Body force section

The body force due to base acceleration is specified on Control card 4. Skip lines 1, 2, and/or 3 if
the corresponding flag is zero. The sign convention for body force loads is understood by
considering that the coordinate system, or base, is accelerated in the direction specified. Thus, a
structure with its base fixed in the X-Y plane and extending upward in the +Z direction will be
crushed by a +Z acceleration, and stretched by a –Z acceleration. Material mass density must be
specified for all analyses using body forces.

Line 1
Colums _____________________________________________ Format
1-5 Load curve number I5

6-15 Scale factor on X acceleration E10.0

EQ.0: default set to “1.0”

Line 2
Colums _____________________________________________ Format
1-5 Load curve number I5

6-15 Scale factor on Y acceleration E10.0

EQ.0: default set to “1.0”

Line 3
Colums _____________________________________________ Format
1-5 Load curve number I5

6-15 Scale factor on Z acceleration E10.0

EQ.0: default set to “1.0”
 52

CHAPTER 5. References
[1] Bonet, J., and Wood, R. D., 1997, Nonlinear continuum mechanics for finite element
analysis, Cambridge University Press.
[2] Simo, J. C., and Taylor, R. L., 1991, "Quasi-incompressible finite elasticity in principal
stretches: Continuum basis and numerical algorithms," Computer Methods in Applied Mechanics
and Engineering, 85, pp. 273-310.
[3] Veronda, D. R., and Westmann, R. A., 1970, "Mechanical Characterization of Skin -
Finite Deformations," J. Biomechanics, Vol. 3, pp. 111-124.
[4] Puso, M. A., and Weiss, J. A., 1998, "Finite element implementation of anisotropic quasi-
linear viscoelasticity using a discrete spectrum approximation," J Biomech Eng, 120(1), pp. 62-
70.
[5] Quapp, K. M., and Weiss, J. A., 1998, "Material characterization of human medial
collateral ligament," J Biomech Eng, 120(6), pp. 757-763.
[6] Weiss, J. A., Maker, B. N., and Govindjee, S., 1996, "Finite element implementation of
incompressible, transversely isotropic hyperelasticity," Computer Methods in Applications of
Mechanics and Engineering, 135, pp. 107-128.
[7] Mow, V. C., Kuei, S. C., Lai, W. M., and Armstrong, C. G., 1980, "Biphasic creep and
stress relaxation of articular cartilage in compression: Theory and experiments," J Biomech Eng,
102(1), pp. 73-84.
[8] Mow, V. C., Kwan, M. K., Lai, W. M., and Holmes, M. H., 1985, "A finite deformation
theory for nonlinearly permeable soft hydrated biological tissues," Frontiers in Biomechanics, G.
a. W. Schmid-Schonbein, SL-Y and Zweifach, BW, ed., pp. 153-179.
[9] Wayne, J. S., Woo, S. L., and Kwan, M. K., 1991, "Application of the u-p finite element
method to the study of articular cartilage," J Biomech Eng, 113(4), pp. 397-403.
[10] Ateshian, G. A., Ellis, B. J., and Weiss, J. A., 2007, "Equivalence between short-time
biphasic and incompressible elastic material response," J Biomech Eng, In press.
[11] Maker, B. N., 1995, "NIKE3D: A nonlinear, implicit, three-dimensional finite element
code for solid and structural mechanics," Lawrence Livermore Lab Tech Rept, UCRL-MA-
105268.
[12] Matthies, H., and Strang, G., 1979, "The solution of nonlinear finite element equations,"
Intl J Num Meth Eng, 14, pp. 1613-1626.