FEPC - FINITE ELEMENT PERSONAL COMPUTER PROGRAM
FEPCIP - INPUT PROCESSOR
FEPCOP - OUTPUT PROCESSOR
SHAREWARE VERSION 3.2
CHARLES E. KNIGHT
DEPARTMENT OF
MECHANICAL ENGINEERING
VIRGINIA POLYTECHNIC INSTITUTE AND STATE UNIVERSITY
BLACKSBURG, VIRGINIA 24061
COPYRIGHT 1989,
1990, 1992 BY C. E. KNIGHT
February 19, 1992
PREFACE
This user's
guide is written
to describe the operation of the
FEPCIP, FEPC, and
FEPCOP programs. It assumes the user has
been in-
structed in use of
the finite element method to solve stress
analy-
sis problems. The
programs will be
described with their
capabilities and
general sequence of use for solving a problem.
The
first time user
should refer to the SETUP section at the end of this
guide for initial
setup, or the README file on the program disk.
These FEPC
programs are provided as shareware for
educational pur-
poses. They are copyrighted programs and you are
not authorized to
sell or
distribute them for COMMERCIAL purposes.
You are free to
use, copy and distribute them for NONCOMMERCIAL
uses only if no fee
is charged for
use, copying or distribution.
Specifically the pro-
grams were
designed for use by students in university courses.
REGISTRATION
If you
are an instructor and use the programs in a univer-
sity course, I would appreciate a
simple registration (no
fee) of
your university, course name or
description, and
approximate number of students taking the course.
If you
use the programs for COMMERCIAL purposes, i.e., em-
ployed
engineering, consulting, sponsored research, etc, a
partial registration fee of $40
would be appreciated. If
you
send a registration of $75, you will receive a copy of
a 600
node version which runs on an IBM PC
or compatible
with
640K RAM. Please state the current
version number of
the
software you are presently using. Send
check or money
order
to:
Dr. C.
E. Knight
Professor of Mechanical Engineering
914
Ballard Ct.
Blacksburg, VA 24060
The programs
are continually under development and your comments
concerning
present features or future enhancements would be appreci-
ated.
Preface ii
TABLE OF CONTENTS
1.0 USING FEPC, FEPCIP, AND FEPCOP . . . . . . . . . . . . . . . 1
2.0 ENTERING THE MODEL IN FEPCIP . . . . . . . . . . . . . . . . 2
2.1 FILES
. . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2 MODEL DATA
. . . . . . . . . . . . . . . . . . . . . . . . . 4
2.2.1 ELEMent TYPE
. . . . . . . . . . . . . . . . . . . . . . 5
2.2.2 MATeriaL PROPerties . . . . . . . . . . . . . . . . . . 6
2.2.2.1 INPUT material properties . . . . . . . . . . . . . 7
2.2.2.2 QUERY material properties . . . . . . . . . . . . . 7
2.2.3 NODE DEFinition . . . . . . . . . . . . . . . . . . . . 7
2.2.3.1 DEFINE node
. . . . . . . . . . . . . . . . . . . . 8
2.2.3.2 GENerate ROW of nodes . . . . . . . . . . . . . . . 9
2.2.3.3 MOVE node
. . . . . . . . . . . . . . . . . . . . . 9
2.2.3.4 DELETE node
. . . . . . . . . . . . . . . . . . .
10
2.2.3.5 QUERY node
. . . . . . . . . . . . . . . . . . . .
10
2.2.4 ELEMent DEFinition . . . . . . . . . . . . . . . . . . 10
2.2.4.1 SELECt MATeriaL set . . . . . . . . . . . . . . . 10
2.2.4.2 DEFINE element . . . . . . . . . . . . . . . . . . 11
2.2.4.3 GENerate ROW of elements . . . . . . . . . . . . . 11
2.2.4.4 MODIFY element . . . . . . . . . . . . . . . . . . 12
2.2.4.5 DELETE element . . . . . . . . . . . . . . . . . . 13
2.2.4.6 QUERY element . . . . . . . . . . . . . . . . . . 13
2.2.5 RESTRAINTS
. . . . . . . . . . . . . . . . . . . . . . 13
2.2.5.1 SET VALUES
. . . . . . . . . . . . . . . . . . . .
14
2.2.5.2 DELETE restraint . . . . . . . . . . . . . . . . . 16
2.2.6 LOADS
. . . . . . . . . . . . . . . . . . . . . . . . 16
2.2.6.1
NODE FORCE . . . . . . . . . . .
. . . . . . . . . 16
2.2.6.2 PRESSURE
. . . . . . . . . . . . . . . . . . . . . 17
2.2.6.3 DELETE FORC
. . . . . . . . . . . . . . . . . . .
17
2.2.6.4 DELETE PRES
. . . . . . . . . . . . . . . . . . .
17
2.2.6.5 QUERY FORC
. . . . . . . . . . . . . . . . . . . .
18
2.2.6.6 QUERY PRES
. . . . . . . . . . . . . . . . . . . .
18
2.3 2D AUTOMESH GENERATION . . . . . . . . . . . . . . . . . . 18
2.3.1 POINT
. . . . . . . . . . . . . . . . . . . . . . . . 20
2.3.1.1 CREATE point
. . . . . . . . . . . . . . . . . . .
21
2.3.1.2 MODIFY point
. . . . . . . . . . . . . . . . . . .
21
2.3.1.3 DELETE point
. . . . . . . . . . . . . . . . . . .
21
2.3.2 LINE .
. . . . . . . . . . . . . . . . . . . . . . . .
21
2.3.2.1 CREATE line
. . . . . . . . . . . . . . . . . . .
22
2.3.2.2 MODIFY line
. . . . . . . . . . . . . . . . . . .
22
2.3.2.3 DELETE line
. . . . . . . . . . . . . . . . . . .
22
2.3.3 ARC .
. . . . . . . . . . . . . . . . . . . . . . . .
23
2.3.3.1 CREATE arc
. . . . . . . . . . . . . . . . . . . .
23
2.3.3.2 MODIFY arc
. . . . . . . . . . . . . . . . . . . .
24
2.3.3.3 DELETE arc
. . . . . . . . . . . . . . . . . . . .
24
2.3.4 GENerate MESH . . . . . . . . . . . . . . . . . . . . 24
2.4 VIEW OPTIONS
. . . . . . . . . . . . . . . . . . . . . . . 25
2.4.1 AUTOSCALE
. . . . . . . . . . . . . . . . . . . . . . 26
2.4.2 ZOOM .
. . . . . . . . . . . . . . . . . . . . . . . .
26
2.4.3 MAGNIFY
. . . . . . . . . . . . . . . . . . . . . . . 26
Table of
Contents ii
2.4.4 CENTER
. . . . . . . . . . . . . . . . . . . . . . . . 27
2.5 DISPLAY OPTIONS . . . . . . . . . . . . . . . . . . . .
. 27
2.5.1 ENTITY SWitch . . . . . . . . . . . . . . . . . . . . 27
2.5.2 LABEL SWitch
. . . . . . . . . . . . . . . . . . . . . 28
2.5.3 MONO/COLOR
. . . . . . . . . . . . . . . . . . . . . . 29
3.0 THE ANALYSIS BY FEPC . . . . . . . . . . . . . . . . . . . 30
3.1 RUNNING FEPC
. . . . . . . . . . . . . . . . . . . . . . . 30
3.2 THE FEPC OUTPUT, FILENAME.LST, FILE . . . . . . . . . . . 31
3.3 FEPC ERROR MESSAGES . . . . . . . . . . . . . . . . . . . 32
4.0 GRAPHIC RESULTS USING FEPCOP . . . . . . . . . . . . . . . 34
4.1 DEFORMED . . . . . . . . . . . . . . . . . . . . . . .
. . 35
4.2 X-STRESS
. . . . . . . . . . . . . . . . . . . . . . . . . 36
4.3 Y-STRESS
. . . . . . . . . . . . . . . . . . . . . . . . . 36
4.4 XY-STRESS
. . . . . . . . . . . . . . . . . . . . . . . . 36
4.5 T-STRESS
. . . . . . . . . . . . . . . . . . . . . . . . . 36
4.6 VON MISES
. . . . . . . . . . . . . . . . . . . . . . . . 36
4.7 TRUSS STRS
. . . . . . . . . . . . . . . . . . . . . . . . 37
4.8 BEAM STRS
. . . . . . . . . . . . . . . . . . . . . . . . 37
4.9 OPTIONS
. . . . . . . . . . . . . . . . . . . . . . . . . 37
4.10 EXIT
. . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.0 SETUP OF THE PROGRAMS . . . . . . . . . . . . . . . . . . 39
Table of
Contents iv
1.0 USING FEPC, FEPCIP, AND FEPCOP
FEPC is
a program that performs finite
element stress analysis of
two-dimensional
truss, beam, plane solid,
or axisymmetric solid
structures. There are two companion programs. FEPCIP is the FEPC
INPUT PROCESSOR
which is used to input and check a model and prepare
data files for
FEPC. FEPCOP is the FEPC
OUTPUT PROCESSOR which
reads FEPC output
data files and produces graphic displays.
The shareware
version of the FEPC programs are currently dimensioned
to run in
a PC with 256K memory. The dimension
limits in the pro-
grams are 250
nodes, 250 elements, 10 materials,
100 points, 30
lines, and 20
arcs. Also, the overall model size is
limited in FEPC
based on number of nodes and average nodal
bandwidth. For example,
250 nodes with an
average nodal bandwidth of 7, 200 with 10, or
150
with 14 are all
maximum model sizes that can be run in FEPC.
Dimen-
sion limits of
the automatic mesh generation grid in FEPCIP are I=15
by J=25.
Therefore a model can be built in FEPCIP even within the
250 node and
element limit which is too large to
run in FEPC,
so
plan carefully.
The procedure
for solving a problem is to run FEPCIP to create the
model, run FEPC
to solve the equations, and run
FEPCOP to display
the results.
The FEPCIP program presents the
user with menus for
interactive
input, checking and storing a
model. This creates
an
analysis file
used as input for FEPC. Running FEPC
produces a list-
ing file of printout results and files of results
used as input for
FEPCOP. Graphic displays of deformed shape and
stress plots may
then be produced
by FEPCOP.
Using FEPC,
FEPCIP, AND FEPCOP 1
2.0 ENTERING THE MODEL IN FEPCIP
Before starting to enter the model, develop a node
and element num-
bering plan,
boundary conditions, and the load placement
for the
model.
With the
FEPCIP.EXE file in the current drive and directory, begin
by typing
FEPCIP<CR>
where <CR>
means press the enter or return key.
After the FEPCIP
logo appears the
program will continue after a short pause.
The screen will clear and the program will
automatically detect the
proper graphics
mode for the supported graphics
cards. The sup-
ported cards
are IBM CGA, EGA, VGA, and MCGA
of the PS/2 model 30
along with the
Hercules graphics card.
After making the
selection the main menu and graphics
windows will
then appear along
with a prompt to SELECT A FUNCTION KEY.
INPUT
PROCESSOR FINITE ELEMENT PERSONAL
COMPUTER DATE TIME
TITLE:
+---------------++--------------------------++---------------------+
| || || |
|F1 FILES || || MODEL |
|F2 MODEL
DATA || || SUMMARY |
|F3 2D
AUTOMSH || || WINDOW |
|F4 TITLE || || |
| ||
|+---------------------+
| ||
|+---------------------+
| || MODEL || |
|F6 CLEAR
MEM || GRAPHICS || |
|F7 EXIT || WINDOW || |
|F8 VIEW
OPTS || || |
|F9 DSPLY
OPTS || || |
| || || |
| || || |
|SELECT A || || |
|FUNCTION
KEY || || |
| || || |
+---------------++--------------------------++---------------------+
Selection of
a menu item
by tapping a function key brings
up a
branch menu.
Key F1 branches
to a menu for recalling a previously stored model,
storing a new
model, or adding a title.
Key F2 branches
to a menu for entering or editing all
data re-
quired for the
model.
Entering the
model in FEPCIP 2
Key F3
branches to a menu for two-dimensional area mesh gener-
ation.
Key F4 prompts
the user to input a title for the current model.
Key F6 will clear all the current model data from
memory in order
to start
entering a new model.
Key F7 exits
the program.
Key F8 branches
to a menu to change the current view of the model.
Key F9 branches
to a menu to change the visibility
of entities
(nodes, elements,
loads, etc.) or labels (node
numbers, element
numbers) on the
next redraw of the model.
Every branch menu
has a function key selection to return to the pre-
vious menu. Many of the selections on the branch menus
will branch
to additional
menus. In each case, following
completion of tasks on
the current
menu, use the
previous menu selection to step
back
through the menus
until the modeling is complete.
The general
procedure for entering a model is to use the MODEL DATA
function key to access the menu for selecting the
element type, de-
fining the
material properties, defining nodes and elements, setting
node displacement
restraints, and applying loads.
For truss and
beam element
models all the model data are
entered from this menu
and its branch
menus.
Two-dimensional
solid models using plane stress,
plane strain, or
axisymmetric
elements may first use the 2D AUTOMSH selection to gen-
erate the model
mesh of nodes and elements. Once the
nodes and ele-
ments are
defined return to the
model data menu to complete the
model by material
definition, setting node restraints and loads.
The model must be
stored on a disk file before exiting the
program.
The model
data may be saved to disk at any time in the progress of
building the
model. Two files are stored for all
complete models
under a
user specified filename with file extensions of .MOD AND
.ANA. If the model is incomplete only the .MOD file
is stored and
messages denoting
the yet to be defined data for the .ANA file are
displayed. All the current model data is saved in the
.MOD file.
The program
operates by using the function keys to select the opera-
tion from the
menu. When data is required a prompt
appears on the
data entry line
just below the TITLE: header. The user
types in the
requested data
separated by commas or spaces
followed by the car-
riage return or
enter key.
WHEN THE PROMPT
TO DETECT AN ENTITY APPEARS ON THE DATA ENTRY LINE A
CURSOR WILL
APPEAR. IF A MOUSE EXISTS AND ITS DRIVER
IS LOADED THEN
USE THE MOUSE TO
POSITION THE CURSOR AND PRESS THE
LEFT BUTTON TO
DETECT OR
THE RIGHT BUTTON TO ABORT. OTHERWISE WITH NO MOUSE USE
THE ARROW KEYS TO
POSITION THE CURSOR AT THE ENTITY
LOCATION AND
PRESS THE
SPACE BAR TO DETECT OR PRESS THE
RETURN OR ENTER KEY TO
ABORT THE DETECT
AND TERMINATE THE CURRENT OPERATION.
Entering the
model in FEPCIP 3
2.1 FILES
Selecting FILES
from the main menu branches to the submenu below.
+---------------+
| |
| F1 RCL FN.MOD|
| F2 STO FN.MOD|
| & FN.ANA|
| |
| |
| |
| |
| |
| |
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
A previously
formed and stored model may be
recalled from disk by
selecting F1
RCL FN.MOD. The
user is prompted
to enter the
filename, FN, without its .MOD extension. The filename may include
the drive
designation and path, but it may be a maximum of 20 char-
acters long
including the drive designator characters.
DO NOT enter
any leading blank spaces in the input of the
filename. If the file
cannot be found
an error message is displayed.
The current model
may be stored on disk by selecting F2 STO FN.MOD &
FN.ANA. The model currently in memory will be stored
in FN.MOD as-
suming no
errors. Also, if the model is complete and ready for
analysis the
input file for the FEPC program
will be stored
in
FN.ANA If the model is incomplete the user is given
messages indi-
cating which data
are missing. The store operation may
be done
at
any time
during the progressive construction of the model in order
to have a place
to restart in case of destruction of
the current
model data in memory. If the files FN.MOD and FN.ANA already exist
on the disk they
may be overwritten with the user's consent
by the
current data in
memory each time the store function is executed.
After completing
use of this branch menu select F10 PREV MENU to re-
turn to the main
menu.
2.2 MODEL DATA
Begin entering a new model by selecting F2 MODEL
DATA on the FEPCIP
main menu. This produces the branch menu shown
below. If a truss
Entering the
model in FEPCIP 4
or beam element model is to be entered then all
the model data will
be entered from
this menu. If a two-dimensional solid is
to be en-
tered then
all the data
may be entered from this menu or the 2D
AUTOMSH selection
may be used to generate the nodes and elements for
the mesh.
Once these are generated they may
be edited from this
menu. If 2D AUTOMSH is to be used it should be done
first or after
the element type is selected and material
property sets are defined
since it will
overwrite any existing node and element definitions.
+---------------+
| |
| F1 ELEM TYPE |
| F2 MATL PROP |
| F3 NODE DEF
|
| F4 ELEM DEF
|
| F5 RESTRAINTS|
| F6 LOADS
|
| |
| F8 VIEW OPTS |
| F9 DSPLY OPTS|
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
2.2.1 ELEMENT TYPE
This selection
displays the list of available elements shown
below.
Use the indicated function key to select the
element for the model.
Only one element
type may be used in a model. After
selection the
program returns
to the previous menu.
Entering the
model in FEPCIP 5
+---------------+
| |
| F1 TRUSS
|
| F2 BEAM
|
| F3 PLN STRESS|
| F4 PLN STRAIN|
| F5 AXISYMMETR|
| |
| |
| |
| |
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
If the element type has been previously
selected, then a prompt de-
notes that and
requests the user's consent to change.
Two node
truss or beam elements may not be interchanged with
any of the two-
dimensional
quadrilateral 4 node elements if any elements have been
defined. Truss and beam elements may be interchanged
provided the
material property
sets are redefined and displacement boundary
con-
ditions are
checked for validity. Plane
stress, plane strain, or
axisymmetric
element types may be interchanged without
any other
data changes,
but with obvious
effects on the
overall model
behaviour.
2.2.2 MATERIAL PROPERTIES
This selection
branches to the menu below.
+---------------+
| |
| F1 INPUT
|
| F2 QUERY
|
| |
| |
| |
| |
| |
| F10 PREV MENU|
|
|
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Entering the
model in FEPCIP 6
2.2.2.1 INPUT MATERIAL PROPERTIES
The F1 INPUT key
produces prompts for material set definition.
The
first prompt
on the data entry line requests
the integer material
set number to be
defined.
ENTER MATERIAL
SET NUMBER #
Up to 10 material
property sets may be defined and should be defined
in numerical order. Material property sets may include some
phys-
ical properties
depending on the element type. Each element
in the
model has a
material set number associated with it which defines its
material properties.
If different material or physical
properties
exist in
different parts of the structure then
multiple material
sets should be defined before elements are defined
so that the cor-
rect assignments
may be made at the time of element definition.
The next prompt
requests the property values which are dependent on
the element type
that has been selected. The prompts are:
for truss
elements, ENTER E, A FOR MATERIAL # M
for beam
elements, ENTER E, A, I, C FOR MATERIAL
# M
for 2-D plane
and
axisymmetric
elements, ENTER E, NU FOR MATERIAL # M
where,E is the
modulus of elasticity,
A is the
member cross section area,
I is the
member area moment of inertia,
C is the
beam section distance from neutral axis to
surface
for bending stress computation,
NU is the
poisson's ratio, and
M is the
material set number.
Any material set
may be changed or corrected by reusing the F1 INPUT
key and entering
the material set number and new data
2.2.2.2 QUERY MATERIAL PROPERTIES
This selection prompts for the material set number
and follows with
a list of the
current data values for that material.
2.2.3 NODE DEFINITION
This selection
branches to a menu to perform node operations.
Nodes
for truss and
beam element models will all be defined in
this sec-
Entering the
model in FEPCIP 7
tion. If
the 2D automesh
option is used for the 2D plane
and
axisymmetric
models then that should be done
first and any
addi-
tional node
operations will be done in this section.
The menu is
+---------------+
| |
| F1 DEFINE
|
| F2 GEN ROW
|
| F3 MOVE
|
| |
| F5 DELETE
|
| F6 QUERY
|
| |
| F8 VIEW OPTS |
| F9 DSPLY OPTS|
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
2.2.3.1 DEFINE NODE
A node
is defined by its number and coordinate position. A prompt
will appear on
the data entry line to
ENTER NODE #,X,Y
=
To input a node,
simply enter the node number, and its X and Y coor-
dinates. The data must be separated by commas or
spaces. All lead-
ing blanks are
ignored, but do not enter any trailing
blanks. If
the node is
generated properly, it will be displayed on the graphics
screen if it lies
inside the current window. The starting
window is
10 units by 10
units, but it will change automatically if any of the
view options
are exercised. Autoscale will resize the window so
that all
currently defined nodes fit inside. The
prompt recycles so
that the next
node may be input. Terminate input by
tapping the re-
turn or enter
key.
Nodes do not have
to be defined in any specific order.
The number-
ing plan should
be made to maximize utilization of row generation if
2D automesh
is not used. The highest node
number defined will ap-
pear as the
number of nodes in the model summary
displayed in the
upper right
corner box. Any nodes which are
undefined and unused in
element
definitions when the model is finished will be listed in the
model input file to FEPC, but they will have fixed
displacements so
that they do not
cause the model solution to fail.
Entering the
model in FEPCIP 8
If a node number
is entered which is already defined, a program mes-
sage appears that
the node is already defined and then requests
the
user's consent
to change to a new definition.
This prevents acci-
dental
redefinition, but also allows for correction of node coordi-
nates.
2.2.3.2 GENERATE ROW OF NODES
If a
row or line of equally spaced
nodes need to be defined, they
may be program
generated by defining the end nodes
and then using
this F2 GEN ROW
selection. The first prompt is
DETECT START
NODE
The lowest
numbered node of the row is found by positioning the cur-
sor at the
node and detecting it. Pressing the
right mouse button
or return key in
response to either of the detect prompts will abort
the
generation. The second prompt is
DETECT END NODE
Following the end
node detection the prompt is to
ENTER NODE
NUMBER INCR [1] #
The node
increment is the integer value added to the start node num-
ber to define the
first generated node number, and then it is
added
again for the
next generated node, etc. The number
interval between
the start
node and end node must be evenly divisible by the node
number
increment. The default value is 1 if
the return key is
pressed without
entering a numerical value.
2.2.3.3 MOVE NODE
This selection
allows any node to be moved by repositioning on the
graphics
screen. This is useful if an element is
badly shaped and
moving a node
will improve its shape. It should be
used primarily
for nodes
internal to the structure boundary because it
is not as
accurate as input of numerical values for the boundary
points. The
first prompt is
DETECT NODE
After detecting
the node the prompt is
DETECT NEW
POSITION
Move the cursor
to the new location of the node and detect it.
Entering the
model in FEPCIP 9
2.2.3.4 DELETE NODE
This key produces
a prompt to detect the node to be deleted.
Fol-
lowing detection
the node is deleted, and all elements which use
this node in
their definition will also be deleted.
When an element
is deleted all
higher numbered elements have their number reduced by
1.
2.2.3.5 QUERY NODE
This key
produces a prompt to detect the
node and then its number
and coordinates
are displayed.
2.2.4 ELEMENT DEFINITION
This selection
branches to a menu to perform
element operations.
Elements for
truss and beam element models
will all be defined in
this
section. If the 2D automesh option is
used for the 2D
plane
and axisymmetric
models then that should be done first and any addi-
tional element
operations will be done in this section.
The menu is
+---------------+
| |
| F1 SELEC MATL|
| F2 DEFINE
|
| F3 GEN ROW
|
| F4 MODIFY
|
| F5 DELETE
|
| F6 QUERY
|
| |
| F8 VIEW OPTS |
| F9 DSPLY OPTS|
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
2.2.4.1 SELECT MATERIAL SET
This function
selects the material set assignment for elements to be
defined after this selection and is identified as the
current mate-
Entering the
model in FEPCIP 10
rial in the model
summary window. The initial material
set assign-
ment for all elements is 1. This value is assigned to all elements
as they are
created. If a different assignment is
needed for some
elements then the
current material must be changed prior to defining
the
elements. The prompt is
ENTER MATERIAL
SET NUMBER #
Enter the
integer set number to be assigned to the next elements
created. If more
than one material is to be assigned the
material
set properties
must already be
defined for the program to allow
change of the current
material.
2.2.4.2 DEFINE ELEMENT
This selection
defines single elements by user
selection of nodes
for each element. The user is prompted to detect each node
needed
for the element
definition. The order of node selection on
two node
elements is of no
consequence.
THE NODES FOR
FOUR NODE ELEMENTS MUST BE DEFINED IN A
COUNTERCLOCK-
WISE ORDER
SURROUNDING THE ELEMENT AREA.
Elements are
numbered in numerical order as they are defined. Their
material set assigned is the current material
set. Each element is
drawn on the
graphics screen as it is defined. The
node prompts re-
cycle to define
the next element and will continue until
the right
mouse button is
pressed or the return or enter key is pressed at the
node detect
prompt to terminate element definition.
2.2.4.3 GENERATE ROW OF ELEMENTS
If a row
or sequence of elements can be defined by adding a single
integer value to
the node numbers of one element to define the
next
element then this
function will do the generation. The
first prompt
is
DETECT LEAD
ELEMENT
Position the cursor at the center of the element which
has the node
pattern to be
incremented to generate the next elements
and detect
it. The next prompt is
NO. OF ELEMS TO
GENERATE #
Enter the number of elements to be generated from
the lead element.
The last prompt
is
Entering the
model in FEPCIP 11
NODE NO. INCR TO GEN ELEMS [1]#
Enter the integer
value to be added to the lead element node numbers
to define the first generated element and
successively all the ele-
ments to be
generated. The return key will give a
default node in-
crement value of
1.
2.2.4.4 MODIFY ELEMENT
After an
element has been defined it may
be modified by change of
material or
change of node definition. The submenu
is
+---------------+
| |
| F1 CHG MATL
|
| F2 CHG NODES |
| |
| |
| |
| |
| |
| F10 PREV MENU|
| |
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Function key F1
produces a prompt to
ENTER MATL #
Enter the new
material set number for the elements to
be selected.
The properties
for the set must have already been input. The next
prompt is to
detect the element. Position the cursor
at the element
center and detect
it. The prompt cycles for additional
elements to
change to the
new material until the right mouse button is pressed
or a return key
entry terminates input.
Function key F2
produces a prompt to detect the element and
follow-
ing element
detection, prompts appear to detect the new nodes defin-
ing the element.
Entering the
model in FEPCIP 12
2.2.4.5 DELETE ELEMENT
This function prompts to detect the element and
following detection
deletes it. When an element is deleted all higher
numbered elements
have their
label(number) reduced by 1. The
prompt recycles until
the right
mouse button is pressed or a return key entry terminates
deletions.
2.2.4.6 QUERY ELEMENT
This function prompts to detect the element and
following detection
lists the element
number, its node numbers and material set
number.
Pressing the
right mouse button or a return key entry terminates
queries.
2.2.5 RESTRAINTS
This section is
for applying node displacement boundary
conditions.
By default
all nodes displacement components are free to take on
nonzero values
appropriate to the structure response
under load.
The components
which must be zero for the model to behave properly
are specified to
be fixed. The following menu is
displayed.
+---------------+
| |
| F1 SET VALUES|
| |
| |
| |
| F5
DELETE|
| |
| |
| F8 VIEW OPTS |
| F9 DSPLY OPTS|
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Entering the
model in FEPCIP 13
2.2.5.1 SET VALUES
This function
sets the restraint condition to be
applied to the
nodes which are
to be detected. A prompt appears to
SET
X-TRANSLATION BOUNDARY CONDITION
followed by the
submenu
+---------------+
| |
| F1 FREE
|
| F2 FIXED
|
| F3 ANGLE
|
| |
|
|
| |
| |
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Selection of F1 or F2 sets the x displacement
component. Selection
of F3 will prompt
to
ENTER INCLINED
BOUNDARY ANGLE (-89.99 TO +89.99 DEG. FROM X) =
Enter an angle in
decimal degrees from the x-axis in
the range of
-89.99 to +89.99 degrees. This will slave the x and y displacement
components
together such that the node can only move along the line
making this angle
with the x-axis. Neither the x nor y
displacement
will be fixed to
zero, but it constrains node motion to a line which
is not
parallel to the x or y axis. After the angle is specified
the program
prompts to detect nodes which are
to have this
re-
straint.
If the x
displacement component is free or fixed the next prompt is
SET Y-TRANSLATION
BOUNDARY CONDITION
followed by the
submenu
Entering the
model in FEPCIP 14
+---------------+
| |
| F1 FREE
|
| F2 FIXED
|
| |
| |
| |
| |
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Select the y
displacement condition.
If the element
type is a beam then the next prompt is
SET Z-ROTATION
BOUNDARY CONDITION
followed by the
submenu
+---------------+
| |
| F1 FREE
|
| F2 FIXED
|
| |
| |
| |
| |
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
The z-rotation boundary condition for beam
elements is the node ro-
tation component
whose vector points in the z direction for beam el-
ements lying in
the x-y plane.
After setting the
restraint values, the prompt appears to
detect a
node. Following detection the restraint is
plotted graphically by
triangles
pointing in the direction of translation restraint at the
node. The
graphic representation of the rotation restraint for
beams is a
leftward descending line from the node with an arc at its
end. Continue to detect all nodes to which this
restraint condition
is to be
applied. Terminate by pressing the right
mouse button or a
return key entry,
and reset the
restraint conditions for
other
nodes.
Entering the
model in FEPCIP 15
2.2.5.2 DELETE RESTRAINT
The restraint condition on any node may be
removed by this function
following the
node detection.
2.2.6 LOADS
This section is for applying loads to the
model. The menu for load
application is
+---------------+
| |
| F1 NODE FORCE|
| F2 PRESSURE
|
| |
| F4 DELET FORC|
| F5 DELET PRES|
| F6 QUERY FORC|
| F7 QUERY PRES|
| F8 VIEW OPTS |
| F9 DSPLY OPTS|
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
2.2.6.1 NODE FORCE
This function
accepts node force component input and applies
it to
detected
nodes. The first prompt is
ENTER X FORCE =
Input the value
of the x-component of force. The next
prompt is
ENTER Y FORCE =
Input the value
of the y-component of force. If the
element type is
the beam the next
prompt is
ENTER Z MOMENT =
At the
detect node prompt detect all the nodes which carry these
force
components. When each node is detected
graphic symbols of the
Entering the
model in FEPCIP 16
loads are plotted
at the node. The forces are represented
with a
hollow arrow
scaled relative to the maximum value of load and point-
ing in the direction
corresponding with the load's
sign. If a
higher load value
is entered later these vectors will rescale to the
higher value.
Moment loads on beam elements are represented by a
rightward
ascending line with a ccw curved hollow arrow at the end.
It is not scaled nor sign dependent. Terminate detection by press-
ing the right
mouse button or a return key entry.
Select F1 again
to change the load
set and pick other nodes.
2.2.6.2 PRESSURE
This selection
calls for input of a uniform pressure load and then
applies it to all
detected element edges. Only models
using the 2-D
solid plane
stress, plane strain or axisymmetric
elements are al-
lowed to be
pressure loaded. For a new model the first
action is to
determine all
the surface element edges. A message will appear
while the list of
edges is being computed. Only
surface edges may
have an
applied pressure load. The
pressure load is converted to
node forces by
the program so only the resultant node forces will be
printed in the
*.LST file. A negative value may be entered
for a
'tensile'
pressure. The prompt is
ENTER ELEMENT
EDGE PRESSURE =
Input the
value of the pressure. The next
prompt is to detect the
element
edge. At each detection two lines
parallel to the edge will
be drawn
symbolizing the pressure load.
Continue detecting until
all edges with the given pressure load are
detected. Select the F2
PRESSURE key
again to enter a new value of pressure to be applied to
other element
edges in the model.
2.2.6.3 DELETE FORC
This selection
calls the detect node prompt to delete the forces ap-
plied to detected
nodes.
2.2.6.4 DELETE PRES
This selection
calls the detect edge prompt to delete
the pressure
applied to
detected edges.
Entering the
model in FEPCIP 17
2.2.6.5 QUERY FORC
This selection
calls the detect node prompt to
list the values of
applied loads at
the detected node.
2.2.6.6 QUERY PRES
This selection
calls the detect edge prompt to
list the values of
pressure at the
detected edge.
2.3 2D AUTOMESH GENERATION
This section of
the program is used for area mesh generation of two-
dimensional plane
stress, plane strain, or axisymmetric models.
The
principle of the
approach is a mapping of an integer area grid
into
the geometrical area of the model. The geometrical area is defined
using point
locations, lines, and arcs. The
perimeter of the ge-
ometrical area
is defined by the
complete set of lines and arcs
which enclose the
area.
The integer area
grid will have lines which correspond to the
lines
and arcs of the geometrical area. Plan the correspondence by imag-
ining or
physically sketching on square grid paper the perimeter in
the integer
area. Use integer coordinates I and J
with a range of 1
to IMAX and 1 to
JMAX respectively. IMAX and JMAX values
are listed
in the first section of this guide. A 1 by 1 square in the integer
area grid will
map to an element in the geometrical
area. Grid
points in the
integer area grid will map to node points in the ge-
ometrical area
model.
Lines in the
integer area can only be lines of constant I
or Lines
of constant
J. The perimeter must be defined by a
head-to-tail con-
nection of
lines in a counterclockwise(ccw) direction around the
area. The length of line in the integer area is
equal to the number
of elements
desired along the corresponding line or arc in
the ge-
ometrical area.
The process
involves defining the geometrical points needed to de-
scribe the model
area, then defining lines
or arcs using
those
points which
complete the model
perimeter. Next plan the corre-
sponding integer
area grid to be mapped into the geometry
of the
model.
After all
the geometric points, lines, and arcs have been entered,
area mesh
generation can begin. The genmesh function
presents a
prompt to
pick the starting point of the
area. This point on the
geometry will
correspond to the 1,1 I,J coordinate location
on the
Entering the
model in FEPCIP 18
integer
area. A series of prompts then proceed
for the detection of
a line
or arc, the number of elements on that line or arc, and the
direction of the
corresponding line in the integer I,J area.
The first line or
arc detected must have the selected starting point
as one of its end
points. The next line or arc picked must
have the
other end
point of the first line or arc as one of its end points.
Each successive
line or arc picked must then connect
to the other
end point
of the previous line or arc. This continues until the
perimeter of the
geometry is closed and the end point
of the last
line in the
integer area must be back at the starting point, i.e.,
the perimeter in
the geometry area and the perimeter in the
integer
area must
close simultaneously. Both
of these perimeters must
progress ccw
around the area.
When a line or
arc is picked the entry of number of elements
deter-
mines the
length of the line in the integer area. The direction
entry chooses one
of four allowable line directions in
the integer
area. The directions are labeled 1, 2, 3, and 4,
which correspond
to right(+I),
up(+J), left(-I), and down(-J) respectively in the I,J
coordinates.
The integer area
of the model must lie in the positive
quadrant of
I,J
coordinates. Since the starting point in
the integer area is at
1,1, and the
perimeter must be ccw, the direction for the first line
or arc
must be 1. The direction for the
second line or arc picked
may be 1 or
2. Successive lines may have any direction
values as
long as some lines with directions 1 and 2 are
used before any with
directions 3 or 4
so that the I and J coordinate values
always re-
main positive.
The total number of elements on all lines in the 1
direction must
match the total number in the 3
direction, and the
total number
in the 2 direction must match the number in the 4 di-
rection.
The bandwidth of
the structure stiffness matrix is minimized by mak-
ing the number of
elements in the I direction smaller than in the
J
direction. The
limits are IMAX-1 elements in the I direction and
JMAX-1 elements
in the J direction. However, no model
may have more
than the maximum
number of nodes or elements listed in
the first
section of this
guide.
Mapping is
an iterative process of
distorting the integer area to
fit in the
geometry area. After a few
iterations a mesh
will be
drawn on the
screen. If it appears to be suitable
then it can be
accepted or more
iterations may be requested to make it
smoother.
If it is
unacceptable then a different integer area may be tried.
The menu of functions
for mesh generation is
Entering the
model in FEPCIP 19
+---------------+
| |
| F1 POINT
|
| F2 LINE
|
| F3 ARC
|
| F4 GENMESH
|
| |
| |
| |
| F8 VIEW OPTS |
| F9 DSPLY OPTS|
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
2.3.1 POINT
Points are
used to define lines and arcs which make up the model's
geometric
perimeter. Two points are needed to
define a
line, and
three points along the arc are needed to define an
arc. Points are
input by their
coordinate location. Selection of key F1
produces a
submenu.
+---------------+
| |
| F1 CREATE
|
| F2 MODIFY
|
| F3 DELETE
|
| |
| |
| F6 QUERY |
| |
| F8 VIEW OPTS |
| F9 DSPLY OPTS|
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Entering the
model in FEPCIP 20
2.3.1.1 CREATE POINT
A point is
created in response to the prompt
ENTER X,Y FOR
POINT #N
The nth
point is plotted on the graphics screen using the + symbol
following the x,y
coordinate entry. The prompt
recycles for addi-
tional point
entries until a return key entry terminates point in-
put. Additional points may be entered later by
accessing the point
function again.
2.3.1.2 MODIFY POINT
This function
allows a defined
point to be moved to another lo-
cation. A detect point prompt picks the point
to move, and
the
prompt for
coordinate input appears. Modifying
points automatically
modifies any
existing lines or arcs using the modified point.
2.3.1.3 DELETE POINT
A detect
point prompt picks the point to delete, and following de-
tection it is
deleted. The space in the program array
is not recov-
ered, however, so
it is preferable to use the modify
function to
move any unused points to a location where they
are usable. Delet-
ing points
automatically deletes any existing lines
or arcs using
the deleted
point.
2.3.2 LINE
A straight line
may be used to represent all or part of any straight
edge on the
model. More than one line on an edge
might be used to
produce different
element spacings along the edge. If
two or
more
lines are
used on any single edge, they should be connected in se-
ries with no
overlap. Selection of key F2 produces a
submenu.
Entering the
model in FEPCIP 21
+---------------+
| |
| F1 CREATE
|
| F2 MODIFY
|
| F3 DELETE
|
| |
| |
| F6 QUERY
|
| |
| F8 VIEW OPTS |
| F9 DSPLY OPTS|
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
2.3.2.1 CREATE LINE
A line
is defined by picking two points at the ends of the
line.
The first prompt
is
DETECT POINT 1
Position the
cursor at the point and detect it. The
next prompt is
DETECT POINT 2
Following detection
of the second point the line is
drawn. The
prompts recycle until terminated by pressing the
right mouse button
or without a
mouse a return key entry at the detect point 1 prompt.
2.3.2.2 MODIFY LINE
If a line was
defined incorrectly or no longer useful it may be mod-
ified by this
function. The first prompt is to
detect the line.
Position the
cursor at the center of the line to detect it.
Follow-
ing detection are
the prompts to detect the two new points.
2.3.2.3 DELETE LINE
A line may be
deleted by detecting the line at the prompt.
However,
the space in the program array is not recovered
so it is preferable
to modify it if
another line is needed.
Entering the
model in FEPCIP 22
2.3.3 ARC
An arc may be
used to represent all or part of any circular
arc on
the model of 180
degr. or less included angle. If more
than one arc
is used on a
circular arc of the model then they should be connected
in series.
Three points along the arc are needed for the defi-
nition. They are the two end points and an
intermediate point. An-
other point
is created during definition of the arc at the arc's
center of curvature. This may cause the autoscale function to re-
duce the
model scale substantially if the arc radius is very large
in order to fit
all the points on the graphics screen.
Selection of
key F3 produces a
submenu.
+---------------+
| |
| F1 CREATE
|
| F2 MODIFY
|
| F3 DELETE
|
| |
| |
| F6 QUERY
|
| |
| F8 VIEW OPTS |
| F9 DSPLY OPTS|
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
2.3.3.1 CREATE ARC
An arc is defined
by picking three points sequentially
along the
arc. The first prompt is
DETECT START
POINT 1
Position the
cursor at the point and detect it. The
next prompt is
DETECT MID POINT
2
Position the
cursor at the point and detect it. The
next prompt is
DETECT END POINT
3
Following detection
of the third
point the arc is drawn. The
prompts recycle
until terminated by the right mouse button or a
re-
turn key entry at
the detect start point 1 prompt.
Entering the
model in FEPCIP 23
2.3.3.2 MODIFY ARC
If an arc was
defined incorrectly or no longer useful it may be mod-
ified by this
function. The first prompt is to detect
the arc. Po-
sition the
cursor at the mid point of the arc and detect it. Then
the prompts to
detect the new three points will follow.
2.3.3.3 DELETE ARC
An arc may be deleted by detecting the arc at
the prompt. However,
the space in the
program array is not recovered so it is
preferable
to modify it if
another arc is needed.
2.3.4 GENERATE MESH
Selection of the
GENMESH function begins a series of prompts and in-
puts to
define the meshing area. If the
element type has not been
selected the
element menu will be presented for a choice.
If more
than one material
set has been defined then the prompt to
ENTER MATERIAL
SET NUMBER #
will appear. Enter the set number which will be assigned
to all the
elements defined
using the mesh generator.
Following these
conditional entries the prompt is to
DETECT START POINT
This is a geometric point on the model which
corresponds to the 1,1
point in the I,J
integer area. Next the prompt to
DETECT LINE OR
ARC
begins the
sequence of perimeter definition. Detect
the line or arc
which connects to
the starting point and starts on
the ccw path
around the
perimeter by positioning the
cursor on the line or arc
center and
pressing the spacebar. Following
detection
ENTER NUMBER OF
ELEMENTS #
along the line or
arc. Then
ENTER DIRECTION
#
of the line in
the integer area(1, 2, 3, or 4). The set
of prompts
to detect line or
arc, enter number of elements, and enter direction
Entering the
model in FEPCIP 24
all cycle
until the user terminates input by entry of the return
key. The user should be sure the geometry
perimeter is closed be-
fore terminating. The
program checks that the integer area is
closed, and if so begins iterating on the
mapping. This may take a
few minutes.
If the integer
area is not closed a program message
reports this
condition and
the geometry is redrawn. Another
trial to input the
perimeter may
begin with selection of the automesh function.
If the integer
area did close, after a few iterations
the mesh is
drawn on the
screen with the prompt
MORE MESH
ITERATION (Y OR N)?
If it needs additional smoothing answer Y. More iterations will be
done and the prompt
will reappear. If it looks acceptable or
it is
to be redone
differently then answer N.
The next question
is
OK TO KEEP (Y OR
N)?
Enter Y to
keep the mesh, or enter N to discard this mesh and redo
the GENMESH
function with another plan.
Once an
acceptable mesh of nodes and elements is kept, return to the
model data menu
to apply the displacement boundary conditions
and
loads, and
perhaps define a material property set.
When the
model data is complete go to the files menu and store the
model and
analysis files to disk. This should also
be done period-
ically during
building of the model in case of unexpected termi-
nation of the
session or a different path is chosen to complete the
model.
2.4 VIEW OPTIONS
This selection
appears on many of the branch menus to allow exercis-
ing the view
options without retracing the menus to reach them. The
following menu
appears on selection of view opts.
Entering the
model in FEPCIP 25
+---------------+
| |
| F1 AUTOSCALE |
| F2 ZOOM
|
| F3 MAGNIFY
|
| F4 CENTER
|
| |
| |
| |
| |
| |
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
2.4.1 AUTOSCALE
Selection of
autoscale automatically scales the graphics window to
include all currently
defined points and nodes.
2.4.2 ZOOM
This function
allows the user to select a portion
of the graphics
window which
is then scaled
to fit the full window. The
first
prompt is
SELECT NEW
WINDOW CORNER
Position the
cursor at one corner of the zoom area
and detect it.
The next prompt
is
SELECT OPPOSITE
CORNER
Position the cursor at the opposite corner of the zoom
area and de-
tect it.
2.4.3 MAGNIFY
This option
changes the size of the model displayed.
The prompt is
ENTER THE
MAGNIFICATION FACTOR =
Entering the
model in FEPCIP 26
A positive value
must be entered; values larger than
one will in-
crease the size of the drawing and values smaller
than one will de-
crease the size.
Subsequent use of
the magnify command will enlarge (or decrease) the
model display
with respect to its current size. For
example, magni-
fying your
model by two
and then by three produces an image six
times larger than
the original.
2.4.4 CENTER
The model may be
moved by selecting a new center of
the graphics
window. The prompt is
LOCATE NEW
CENTER, PRESS SPACE BAR
Position the
cursor where the graphics window
center is to be and
detect it.
2.5 DISPLAY OPTIONS
Display options
control which entities and labels are visible when a
graphics plot is
done. A branch menu appears.
+---------------+
| |
| F1 ENTITY SW |
| F2 LABEL SW
|
| |
| F4 MONO/COLOR|
| |
| |
| |
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
2.5.1 ENTITY SWITCH
An entity switch
setting is off or on to control the individual
en-
tity's
visibility. Selection of this function
brings up a submenu.
Entering the
model in FEPCIP 27
+---------------+
| |
| F1 POINTS
|
| F2 LINES
|
| F3 ARCS
|
| F4 NODES
|
| F5 ELEMENTS
|
| F6 RESTRAINTS|
| F7 LOADS
|
| |
| |
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Selection of a
function key will produce a prompt to change its cur-
rent setting by
default. If the visibility is 'on' the
prompt is
OK TO SWITCH OFF
[Y]#
A Y or return key
entry will switch the visibility 'off'.
2.5.2 LABEL SWITCH
This function
controls the display
of labels(numerals) on the
graphic
model. The submenu is
+---------------+
| |
| F1 NODE NOS
|
| F2 ELEMNT NOS|
| |
| |
| |
| |
| F10 PREV MENU|
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Selection of a
function key will produce a prompt to change its cur-
rent setting by
default. If the display is 'off' the
prompt is
Entering the
model in FEPCIP 28
OK TO SWITCH ON
[Y]#
A Y or return key
entry will switch the display 'on'.
2.5.3 MONO/COLOR
This function switches the display between black
and white or color
when the computer
has a VGA graphics board.
On other graphics
boards the display is always black and white. Switching the VGA to
black and white
allows the screen graphics to be dumped to
a black
and white
printer without loss of character intensity as sometimes
happens when
screen dumping color graphics to a
black and white
printer. Since only the drawing color palette is
changed with this
switch the change
occurs when the next drawing is done
after the
switch. Execute
the function key again to return to a color dis-
play.
Entering the
model in FEPCIP 29
3.0 THE ANALYSIS BY FEPC
When a model has been developed and saved, it is
complete and ready
to be processed
by the finite element processor, FEPC.
After exit
from FEPCIP, and
before starting FEPC, be sure that the filename.ANA
file can be
accessed by FEPC by copying it to the FEPC diskette in
the same
directory where FEPC.EXE resides, or by using the drive and
path designation
in the filename.
3.1 RUNNING FEPC
With the FEPC.EXE
file in the current drive and directory, begin
by
typing
FEPC<CR>
After the FEPC
logo appears, a prompt will appear to enter the model
filename (with
drive and/or path designation but
without the .ANA
extension),(20
characters max).
As the computations
proceed, messages will appear on the screen
re-
porting the
computation step in
progress. If errors occur, error
messages will
also appear on the screen. FEPC
creates some other
files as it
runs. There is a listing file of all the
printed output
labeled
filename.LST. This file should be
studied by the user after
an analysis to check the input data interpreted
by FEPC and all the
numerical
output. A file labeled filename.MSH
stores the node and
element data
for FEPCOP. A file labeled filename.NVL stores the
node displacement
and element stress data for FEPCOP.
Some other files
are also created during the FEPC run which are
de-
leted upon
normal termination of
the program so the disk which
stores the .ANA
file should have some excess space for
these files
during runtime.
If the run terminates abnormally some of these files
may still
be on the disk with extensions of
.ELM and .LOD. These
and other output
files will be overwritten when running a model with
the same
filename.
If the FEPC run
was successful then FEPCOP may be
used to display
the results in
graphic form. If the run was not
successful then ex-
amine the
filename.LST file for data errors or error messages that
may help to
correct the model.
THE ANALYSIS BY
FEPC 30
3.2 THE FEPC OUTPUT, FILENAME.LST, FILE
The output from a
FEPC run is stored
in a listing
file called
filename.LST, where the filename
is the same as the model file name
entered when
beginning the FEPC analysis. This file
includes a
listing of all
the input data as well as all the numerical results.
It begins
with the job title followed by a
summary of some of the
control
data. The number of node points in the
model is given along
with the number
of element groups and load cases. All
models pre-
pared by
FEPCIP will have only one element group and one load case
although the FEPC
program allows more. The location and
displace-
ment boundary conditions for all nodes are listed
next. The z dis-
placement
boundary condition should be fixed for all truss and
2-D
solid models.
For beam element models it corresponds to the node
rotation
component and may be fixed or free.
Then a group headed
EQUATION NUMBERS follows. Each degree-of-freedom in the model has
an associated
equation in the set of simultaneous equations to solve
for the
displacement components of every node.
Every displacement
component which is fixed by a boundary condition is
then eliminated
from the set of
equations. This equation number list
is useful if
the program fails to execute and reports error
message 10 described
in the next
section.
The load case
data list all the node point loads by node number, di-
rection, and
magnitude. These should agree with the
input done in
FEPCIP. However,
if a pressure load was input in FEPCIP then the
node forces are
those calculated to give the correct pressure appli-
cation to the
model. Next the element data begins with
the defi-
nition of material property sets. The property values for each set
are listed. Then the element definition of all
elements follows.
The elements are defined by the list of node
numbers which form the
geometry of the
element and the material property set which
matches
the material
forming the element. This concludes the input data
listing of the
model.
The output data
begins with a summary of the system equations.
The
number of
equations, the number of terms inside the profile of the
structure
stiffness matrix, the maximum bandwidth, and
the average
bandwidth are
reported. This gives a feeling
for the size of the
problem and the
effectiveness of the node numbering pattern in mini-
mizing the
bandwidth.
The displacement
values for all node components are
listed. Dis-
placements for
all fixed node components should be zero.
The z dis-
placement components
will generally be zero, except for the beam
element models in
which they are the rotation components.
Stress output is
related to the element type used in the analysis.
In truss element
models the axial force and stress are reported.
In
beam element models the axial, flexure and average
shear stress are
reported,
followed by a similar report of the axial force, the bend-
ing moments and
lateral force. The sign conventions are
that the
THE ANALYSIS BY
FEPC 31
force and moment vectors are positive when pointing
in positive co-
ordinate
directions. A local coordinate system
for each element may
be set up using
the first node I as the origin and pointing the
ax-
ial coordinate toward the second node J. If that coordinate points
to the viewer's
right then the lateral coordinate is upward and
the
third coordinate
points toward the viewer.
In this local
coordinate system a positive axial force at node I re-
sults in a
compressive axial stress which is constant along the ele-
ment length.
A positive moment
at node I results in a tensile
stress on the top
surface which is the value printed
in the file.
The moment varies
linearly from node I to node J. A
positive moment
at node J results in a compressive stress on the
top surface of the
beam
element. This is consistent with simple
beam theory in which a
positive moment
vector on the right end and an equal negative moment
vector on the
left end produce a compressive stress on the top sur-
face. A positive shear force at node I balances
an equal negative
shear force at
node J, and this is usually defined as
a positive
shear stress in
simple beam theory conventions.
Stress components
for 2-D plane stress or plane strain analyses that
are listed
are the two normal stresses and the shear stress in the
x,y plane. Also, the Von-Mises equivalent or effective
stress is
calculated. For axisymmetric solid models the same
inplane compo-
nents are given
along with the hoop stress component.
The x compo-
nent is in the radial direction, the y component
is parallel to the
axis of symmetry
and the t component is the hoop value.
The Von-
Mises equivalent
stress is also calculated in this case.
3.3 FEPC ERROR MESSAGES
1 - OUT OF SPACE,
MODEL IS TOO LARGE (I)
The model
is too large to run in FEPC.
Reduce the model size
in FEPCIP
and try again. Consult the program
limits given in
section 1.
2 - NODE 'n' HAS BEEN PLACED ON AN INCLINED
BOUNDARY, BUT IT IS
ALREADY
CONSTRAINED AGAINST X OR Y DISP. OR BOTH
An inclined
boundary angle is specified for node n, but an x or
y restraint
was also specified which is incompatible.
Edit the
model in
FEPCIP.
3 - FEPC.EXE file
not found
The FEPC.EXE
file must reside in the current drive
and direc-
tory to
execute.
6 - 'm' ELEM IS
HIGHER THAN NO. OF ELEMENTS IN THE GROUP
The
filename.ANA file has been corrupted because element number
m is higher than the total number of elements. The *.ANA file
is an ASCII
file so it may be printed or edited.
Examine its
THE ANALYSIS BY
FEPC
32
contents in comparison
with the data
printout in the
filename.LST
file.
7 - ELEMENT NO 1
IS NOT DEFINED FIRST
The
filename.ANA file has been corrupted because element number
1 is not
defined first in the list of element definitions. The
*.ANA file
is an ASCII file so it may be printed or edited.
Examine its
contents in comparison with the
data printout in
the
filename.LST file.
8 - YOU HAVE A
ZERO LENGTH ELEMENT #'m'
A beam or
truss element is defined using the same node for both
ends
or the two nodes defining the element have coincident co-
ordinate
locations.
9 - BAD ELEMENT
#'m'
A
quadrilateral element is improperly defined or
is too dis-
torted. Check for cw node order definition around the
element(
it should be
ccw), inside angles between sides greater than 180
degrees, butterfly shaped
element, or a triangle formed by us-
ing one node
for two corners (this is legal if the
last two
nodes in the
element definition are the same).
10 - STIFFNESS
MATRIX NOT POSITIVE DEFINITE,
NEGATIVE STIFFNESS
DIAGONAL TERM FOR
EQUATION 'n' VALUE = '#'
During solution
of the system equations a negative diagonal
term is
found which means that the equations cannot be
solved.
The equation
number corresponds to the free node degree-of-
freedom in
the system ordered consecutively with node
numbers.
These are listed in the filename.LST file produced
in the FEPC
run.
Find the node number from this list then examine the ele-
ments which
are defined using this node number for errors.
If
the equation
number is 1 or the last equation number then the
error is
probably due to lack of sufficient displacement
re-
straints to
prevent rigid body motion.
11 - INCLINED
BOUNDARY ANGLE MUST BE BETWEEN -89.99 AND +89.99 DEGR
The inclined
boundary angle input
is outside the allowable
range.
THE ANALYSIS BY
FEPC 33
4.0 GRAPHIC RESULTS USING FEPCOP
After a
successful run by
FEPC, results files, filename.MSH and
filename.NVL,
will have been created on the disk.
These are the in-
put files for
output processing by FEPCOP.
With the
FEPCOP.EXE file in the current drive and
directory, begin
by typing
FEPCOP<CR>
where <CR>
means press the enter or return
key. After the FEPCOP
logo appears the
program continues after a short pause.
The screen will
clear and a prompt will appear to
ENTER MODEL FILE
NAME (NO EXT) -
Enter the
filename (with drive and/or path designation
but without
the .MSH or .NVL
extension),(20 characters max). Some
messages will
appear noting the
progress of calculations, and then the screen will
clear and
the FEPCOP MAIN MENU is displayed along with a prompt to
SELECT A FUNCTION
KEY.
OUTPUT
PROCESSOR FINITE ELEMENT PERSONAL
COMPUTER DATE TIME
TITLE:
+---------------++--------------------------++---------------------+
| || || |
| F1 DEFORMED
||
|| |
| F2 X-STRESS
||
|| |
| F3 Y-STRESS
||
|| |
| F4 XY-STRESS || || |
| F5 T-STRESS
|| || |
| F6 VON MISES || || |
| F7 TRUSS STRS|| || |
| F8 BEAM STRS || || |
| F9 OPTIONS
||
|| |
| F10 EXIT
||
|| |
| || || |
| || || |
| SELECT A
||
|| |
| FUNCTION KEY || || |
| || || |
+---------------++--------------------------++---------------------+
GRAPHIC RESULTS
USING FEPCOP 34
4.1 DEFORMED
Selection of F1
DEFORMED brings up a branch menu.
+---------------+
| |
| F1 PLOT
|
| F2 ANIMATE
|
| |
| |
| |
| |
| |
| F10 PREV MENU|
| |
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Selection of F1
PLOT produces a deformed shape plot of the element
mesh superimposed
over the undeformed model. This
plot shows the
finite element
mesh when the node displacements are scaled and added
to the
node coordinates so that the deformed shape is exaggerated.
In truss and beam
models the deformed mesh is superimposed over
the
undeformed mesh
plot. In 2-D solid models the deformed mesh is
superimposed over
the outer boundary of the undeformed shape.
The
displacement scale factor may be changed in the OPTIONS
menu to in-
crease or
decrease the plotted deformation. An
additional submenu
appears for
modifying the view of the plot.
+---------------+
| |
| F1 AUTOSCALE |
| F2 ZOOM
|
| F3 MAGNIFY
|
| F4 CENTER
|
| |
| |
| |
| F10 PREV MENU|
| |
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Selection of F2
ANIMATE produces a sequential mesh plot of truss and
beam models
or a boundary outline plot of 2-D models showing
the
GRAPHIC RESULTS
USING FEPCOP 35
progressive
deformation as the load is cyclicly applied.
Press any
key to terminate
the animation.
The next five function key selections show the
stress contour plots
developed in
2-D solid models
for the indicated
components of
stress.
4.2 X-STRESS
This function
draws a contour plot of the X direction normal stress.
It is
accompanied by a legend of the contour values and the view
modification menu.
4.3 Y-STRESS
This function
draws a contour plot of the Y direction normal stress.
4.4 XY-STRESS
This function
draws a contour plot of the XY shear stress.
4.5 T-STRESS
This function draws
a contour plot of the T direction normal stress.
This stress
component is nonzero only for the
axisymmetric element
models and
represents the hoop stress in the axisymmetric structure.
4.6 VON MISES
This function
draws a contour
plot of the Von Mises equivalent
stress. This stress is calculated based on the
distortion energy
failure theorem
using all the stress components
calculated in the
loaded model.
GRAPHIC RESULTS
USING FEPCOP 36
4.7 TRUSS STRS
This function is
used to display the results in truss
element mod-
els. The user may select a plot of the axial force
or stress in all
truss elements.
The plot is in a bar chart format with the heights
scaled to the
maximum value in any element. Plus or
minus signs are
drawn on the
bar near the top to indicate whether the member is in
tension or
compression.
4.8 BEAM STRS
This function is
used to display the results in beam element models.
The user may
select to plot the axial, flexure,
average transverse
shear, or the
maximum combined axial plus flexure stress.
These are
also in bar
chart format with signs indicated near the top of each
bar. The axial and transverse shear stresses are
constant along an
element length
so one bar per element is
sufficient. The flexure
stress component
varies linearly along the element length so
a bar
is plotted for
the value at each end. Two bars are also
plotted for
the combined
axial plus flexure stress. The sign of the combined
stress is the
same as the sign of the axial stress which is the com-
bination producing
the largest magnitude
4.9 OPTIONS
Function key F9
OPTIONS produces a submenu.
+---------------+
| |
| F1 NODE SW
|
| F2 ELEM SW
|
| F3 DISP SCALE|
| |
| |
| F10 PREV MENU|
| |
| |
| SELECT A
|
| FUNCTION KEY |
| |
+---------------+
Selecting F1 adds
node symbols to the 2-D stress plots, and F2
adds
element outlines inside the 2-D boundary for stress
plots. Both of
these selections
produce a prompt to switch the current
setting to
off or on.
Answering Y or by default a return key entry will make
GRAPHIC RESULTS
USING FEPCOP 37
the change. Selection F3 allows the scale factor for the
deformed
shape plots
to be changed by prompting for a new scale factor with
the current scale
factor shown as the default value. Enter
a larger
value to
increase the exaggeration or a
smaller value to decrease
it.
4.10 EXIT
Selecting F10
exits the FEPCOP program.
GRAPHIC RESULTS
USING FEPCOP 38
5.0 SETUP OF THE PROGRAMS
The programs
executable files, a README file, this user's guide file
and some sample
model and analysis files are supplied on one 3.5 in.
diskette. First,
make backup copies of all the files.
Then copy
the *.EXE files
to any drive and directory you wish to operate
them
from. The data files may be in another
directory. Files that are
created during
execution of FEPC will reside in the
same directory
with the
*.ANA file. The programs will
also work when copied over
to a hard disk
drive. There are also some sample model,
*.MOD, and
analysis, *.ANA, data files on the disks which can be
recalled into
FEPCIP or run in
FEPC.
Tidak ada komentar:
Posting Komentar