1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366
|
This is the README file for the alpha release of Felt - a package for
introductory finite element analysis.
FElt (Finite ELemenT) is a mathematical package intended for introductory
finite element work. It is not intended as a replacement for cutting
edge or real-world research code. At this stage FElt consists of
of three components. The first is the basic mathematical engine (the actual
felt command-line application) which includes an intelligent parser and
straightforward inputfile syntax. Because FElt is intended to be fairly simple
(at this stage of development at least) and because the authors' backgrounds
are largely in structural engineering, the element library is fairly small at
this point. Truss (bar), two- and three-dimensional beam, CST (plane stress
and plane strain) and iso-parametric (four to nine node) two-dimensional
elements are all currently supported. Adding additional elements to the
library is intended to be a straightforward exercise, if for example, such
a project were to be given as an exercise in an introductory finite element
course. As of this release, only linear static analysis is supported
(sorry ... linear dynamic analysis is in the works but we wanted to get a
version out the door and stabile before the fall term started).
In addition to the command line application version, there is xfelt,
an X11 based encapsulator for felt which provides a graphical text-editor
and two- and three-dimensional viewing of the structure. You can edit
the actual text file, solve the problem (xfelt actually execs the felt
application), graphically view the structure and save and print results
from within the xfelt gui.
Velvet is a full-featured X11 interface to the two-dimensional functionality
of felt and corduroy. Unlike xfelt, velvet is fully self-contained - it is
intended as the primary graphical user interface and along with the
mathematical engine is where most future development will occur. Within
velvet a user can interactively set-up, model, and solve a FElt problem
through a menu driven, CAD-like interface with no reliance on traditional
text based input files (as in felt, corduroy and even xfelt). Velvet does
not currently provide any real access to the three-dimensional aspects of
FElt.
Features:
o Easy to use. Intelligent parsers mean that there are no files
of comma delimited lists of numbers like you see in a lot of
other FE codes used in introductory courses. Velvet provides
a full-featured, interactive problem editor.
o Portable. Felt and corduroy should compile on any reasonable
Un*x system. There is also a DOS version of the felt program
only (including a simple utility for structural visualization,
drop one of us a note if you're interested). Velvet is built
around the public domain Athena widget set using
X11R5 ... R4 should work as well. Either regular or 3d Athena
widgets work (we prefer the 3d). The package was
developed on 386's running Linux; it has also been tested on 386's
running SysV R3.2, HP 700's, and Suns. We can provide binaries
for any of the above systems if there is an interest.
Caveats:
o Remember, this is meant largely as a teaching tool. In most
cases in the code we have sacrificed efficiency in favor of
readability. It is not our intention to be solving anyone's
cutting edge research problems ... the project was born out of
a fit of "we can do better than that" based on some antiquated
Fortran code we had been using.
o On that same note (and this is only a caveat to devoted
believers in the Fortran way of life) the entire package
is coded in C (with the exception of the Barry Joe's Geompk
routines ... they are in Fortran and we have not modified
them except where the interface is concerned). This really
shouldn't be a problem; we think the code is very readable and C's
data structures lend themselves quite readily to keeping
track of nodes and elements and forces and constraints and such.
Also, we can do dynamic memory allocation without a lot of
hocus pocus and have long variable names, so the code is
usually pretty easy to follow
The file felt.ps included in this distribution is a postscript document
providing a user's manual for the entire package, including mathematical
and algorithmic details and programming notes. Man pages are available
which briefly describe each individual application.
For installation instructions see the file INSTALL. The best way to get
a feel for the package is to use it. Several examples are provided
in the etc/Tests/ directory.
The FElt package is free software. This means that you can redistribute it
and/or modify it under the terms of the GNU General Public License, a copy of
which is provided in the file COPYING. (Excluding the Geompk routines which
are covered under their own terms, see the README file in lib/Generate/Geompk
for details.) Given the way in which it was developed, we wouldn't
want it any other way ... Linux is a Unix-like OS, based on a free kernel,
free development tools (gcc) and hundreds of free utilities (GNU, etc.).
(And not only is it free, but it runs circles around most commercial Un*ces
available for the 386). The entire graphical environment (X11 and the
Athena widget set) is also free. What a concept ...
Lastly, we would like to stress that feedback would be greatly appreciated ...
this is an alpha release and we welcome all comments as to usability and
functionality. Mail bug reports, comments, hints, suggestion, complaints,
etc. to one of us at the email addresses below. Our one wish in life is
that people find the package useful (well, maybe not, but you get
the point ...)
Jason Gobat Darren Atkinson
jgobat@ucsd.edu atkinson@ucsd.edu
-------------------------------------------------------------------------------
This is the README file for version 1.3 of FElt - a package for
introductory level finite element analysis.
See the file README.alpha for a basic description of the package. See
the file CHANGELOG for a detailed description of what the changes are
from the public release of 1.0a and this release of 1.3. This file will
simply bring you up to speed on a few things.
The most visible changes are to the velvet GUI. Things should be a lot
easier to use now. Most of the dialogs have keyboard interfaces (tab,
space, return keys etc.) which is nice and you can keep them up while
you work, which is very nice. There is also a real file selection
mechanism. For a nice gee whiz, from FElt/bin/Velvet execute "velvet
+numbers ../../etc/Tests/wrench.flt", solve it and then select plot
stresses from the problem menu ... wait a while and see the new color
stress output (assuming you have an 8-bit screen of course). The other
post-processing addition is basic displaced structure plotting but this
still needs some work. Other important, but not so visible, changes
mostly relate to distributed loads on beams and beam3ds and several bug
fixes to beam3d elements (they might even produce the right answers
now!) and constraints (displacement boundary conditions and hinges).
Some of the changes to distributed loads are not backward compatible.
If you have an input file from 1.0a that used direction=x y or z, you'll
have to change it to work with the new version. Sorry. The only new
element is a beam element that uses Timoshenko beam theory, but it is
pretty much just for instructional purposes at this stage, see Chapter 9
of the User's Guide and Reference manual on how to add elements to see
what we mean.
The file felt.ps ("FElt User's Guide and Reference Manual") in this
distribution bears little resemblance to the document in the alpha
release. The manual has tripled in size and includes much better
discussions of elements, algorithms, adding elements and using velvet ...
lots of cool new figures even. Most importantly, we did this one in
LaTeX and produced postscript with dvips so everyone should be able to
print it :-). Updated man pages are also available which briefly describe
each individual application.
For installation instructions see the file INSTALL. The best way to get
a feel for the package is to use it. Several examples are provided
in the etc/Tests/ directory.
Finally, as you are probably aware, FElt has found a permanent home via
anonymous ftp at cs.ucsd.edu in /pub/felt. Given this, we are going to
go to a system of smaller releases (e.g., 1.31, 1.32). In all
probability we will not announce each such revision very broadly. If
you want to be made aware of such updates, please let us know and we
will put your name on a mailing list whereby we will notify you every
time a new release becomes available. The latest version (including bug
fix "patchlevel" changes) will always be available on cs.ucsd.edu. We
will only be uploading major new versions (this one, the one that
finally has dynamic analysis in it, etc.) to sunsite.unc.edu.
As usual, we would like to stress that feedback is always appreciated ...
we welcome all comments as to usability and functionality. If you're using
the package please drop us a note and let us know what you think, even if
you've got no complaints whatsoever. Mail bug reports, comments, hints,
suggestion, complaints, etc. to one of us at the email addresses below.
Jason Gobat Darren Atkinson
jgobat@ucsd.edu atkinson@ucsd.edu
-------------------------------------------------------------------------------
This is the README file for version 2.0 of FElt - a package for
introductory level finite element analysis.
See the file README.alpha for a basic description of the package. See the
file CHANGELOG for a detailed description of what the changes are from the
public release of 1.3 to this release. This file will simply bring you up to
speed on a few things.
Major changes include the incorporation of transient analysis. In general
this means you can solve a structural dynamics problem using the Hilbert-
Hughes-Taylor alpha method of numerical integration (a variant of
Newmark integration). You can specify dynamic nodal forces with a continuous
function in time and they will be dynamically evaluated throughout the
course of the simulation. Velvet includes an animation capability to model
the dynamic behavior of the complete structure given this new analysis
capability. Old fashioned tabular and ascii based time-displacement plots
are also available in both felt and velvet. Gibbs-Poole-Stockmeyer/Gibbs-King
automatic node renumbering is also now available for badnwidth/profile
reduction. The matrix assembling routines now assemble directly into
a compact column representation. This is in itself can save major amounts
of memory. Coupled with renumbering, savings in both speed and memory can
be dramatic.
To keep people better informed a mailing list has been setup. We'll try to
make bug reports and fixes, announcements, etc. available through this
channel. Hopefully, it will also turn into a forum for FElt discussions in
general - including usage questions, how do I do xyz questions and discussions
of what capabilities people would like to see in the package. If you
are using FElt we urge you to subscribe if for no other reason than to keep
track of bug fixes.
To subscribe send a message to listserv@mecheng.fullfeed.com with the line
subscribe felt-l
as the only line in the _body_ of the message (the server ignores subject
headings). To send a question or comment to the list just set it up as a
regular email and send it to felt-l@mecheng.fullfeed.com. Thanks to
Greg Jackson at the ASME BBS for setting this up for us.
felt.ps ("FElt User's Guide and Reference Manual") has been updated to
reflect the new changes - as always look there for the real details. Updated
man pages are also available which briefly describe each individual
application.
For installation instructions see the file INSTALL. The best way to get a
feel for the package is to use it. Several examples are provided in the
etc/Tests/ directory. A simple dynamic example can be found in
etc/Tests/dynframe.flt.
As usual, we would like to stress that feedback is always appreciated ... we
welcome all comments as to usability and functionality. If you're using the
package please drop us a note and let us know what you think, even if you've
got no complaints whatsoever. Mail bug reports, comments, hints, suggestion,
complaints, etc. to one of us at the email addresses below or even better.
Jason Gobat Darren Atkinson
jgobat@mit.edu atkinson@ucsd.edu
-------------------------------------------------------------------------------
This is the README file for version 2.30 of FElt - a package for
introductory level finite element analysis.
See the file README.old for a basic descriptions of previous versions of FElt
and what it's all about. See the file CHANGELOG for a detailed list of
what's new/fixed in this release. The list of frequently asked questions
(and their answers) is in FAQ.
Major changes include the incorporation of modal analysis and the addition
of a brick element (8 node solid). In general modal analysis means that
FElt can figure out the natural frequencies and mode shapes for an arbitrary
structure. Velvet can actually give you a graphical rendering of mode shapes.
Support for the brick is still pretty weak, but at least we've officially
crossed the line into solid elements now.
For those of you who don't know there is a mailing list for FElt. We try to
make bug reports and fixes, announcements, etc. available through this
channel. We'd also like for it to turn into a forum for FElt discussions in
general - including usage questions, how do I do xyz questions and discussions
of what capabilities people would like to see in the package. If you
are using FElt we urge you to subscribe if for no other reason than to keep
track of bug fixes.
To subscribe send a message to listserv@mecheng.asme.org with the line
subscribe felt-l
as the only line in the _body_ of the message (the server ignores subject
headings). To send a question or comment to the list just set it up as a
regular email and send it to felt-l@mecheng.asme.org. Thanks to
Greg Jackson at the ASME BBS for setting this up for us.
felt-2.30.ps.gz ("FElt User's Guide and Reference Manual") is the updated
documentation for FElt. It's up to 151 pages or so. As always look
there for the real details. Updated man pages are also available which briefly
describe each individual application.
For installation instructions see the file INSTALL. The best way to get a
feel for the package is to use it. Several examples are provided in the
etc/Tests/ directory. A simple modal example can be found in
etc/Tests/modal.flt and simple example using a brick element is in
etc/Tests/brick.flt.
As usual, we would like to stress that feedback is always appreciated ... we
welcome all comments as to usability and functionality. If you're using the
package please drop us a note and let us know what you think, even if you've
got no complaints whatsoever. Mail bug reports, comments, hints, suggestion,
complaints, etc. to one of us at the email addresses below or even better.
Jason Gobat Darren Atkinson
jgobat@mit.edu atkinson@ucsd.edu
-----------------------------------------------------------------------------
This is the README file for version 3.00 of FElt - a package for
introductory level finite element analysis.
See the file README.old for a basic descriptions of previous versions of FElt
and what it's all about. See the file CHANGELOG for a detailed list of
what's new/fixed in this release - there are a fairly large number of
substantial changes with this release. The list of frequently asked
questions (and their answers) is in FAQ.
The big jump from 2.30 to 3.00 is based on the incorporation of burlap in
this release. burlap is an entirely new appliation designed to combine
the ease of use of FElt with the power and flexibility of command-line
driven, scripting environments like Matlab. burlap allows the easy
incorporation and testing of new element types, new analysis types, and
new ways to use the existing analysis types. There are two brand new chapters
in the manual that describe it and we'd really like people to give it a try
and let us know what they think.
Based largely on the two new burlap chapters, the manual has grown to 220
something pages; part of the growth is also due to a new chapter near the
front that tries to give some of the mathematical basis for each of the
analysis types. We've also done some extensive proofreading for the first
time in several releases. The xfelt chapter is gone because xfelt is gone
from the source and binary distributions. You can still get it from old
ones if you really want it (and don't for some reason already have it).
Three new analysis types have been added: static and transient thermal
analysis and spectral analysis (transfer functions, spectral response).
Modal analysis has some new capabilities (like the ability to do actual
modal analysis as opposed to just calculating eigenvalues and vectors).
The only new element is a one-dimensional spring.
The configure process and Makefile scheme is all new. We're still sort
of fine tuning it so if you have any problems or complaints we'd really
like to hear about them.
For installation instructions see the file INSTALL. The best way to get a
feel for the package is to use it. Several examples are provided in the
etc/Tests/ directory. A simple modal example can be found in
etc/Tests/modal.flt and simple example using a brick element is in
etc/Tests/brick.flt.
As usual, we would like to stress that feedback is always appreciated ... we
welcome all comments as to usability and functionality. If you're using the
package please drop us a note and let us know what you think, even if you've
got no complaints whatsoever. Mail bug reports, comments, hints, suggestion,
complaints, etc. to one of us at the email addresses below.
Jason Gobat Darren Atkinson
jgobat@mit.edu atkinson@ucsd.edu
-----------------------------------------------------------------------------
This is the README file for version 3.02 of FElt - a package for
introductory level finite element analysis.
See the file README.old for a basic descriptions of previous versions of FElt
and what it's all about. See the file CHANGELOG for a detailed list of
what's new/fixed in this release. This is basically an interim bug
fix release - there are lots of new features but for the most part
they are only partially implemented ... The list of frequently asked
questions (and their answers) is in FAQ (which has not been updated
for this release). The documentation (felt-3.02.ps.gz) has also
only been marginally updated. It now includes an (incomplete)
chapter describing the operation of WinFElt.
For installation instructions see the file INSTALL. The best way to get a
feel for the package is to use it. Several examples are provided in the
examples/ directory. Nonlinear, loadcase and load range examples
include truss_nl.flt beam_lr.flt, beam_lc.flt, etc.
As usual, we would like to stress that feedback is always appreciated ... we
welcome all comments as to usability and functionality. If you're using the
package please drop us a note and let us know what you think, even if you've
got no complaints whatsoever. Mail bug reports, comments, hints, suggestion,
complaints, etc. to one of us at the email addresses below.
Jason Gobat Darren Atkinson
jgobat@mit.edu atkinson@ucsd.edu
|