forked from AkMecha/felt
-
Notifications
You must be signed in to change notification settings - Fork 0
/
README.old
366 lines (303 loc) · 19.3 KB
/
README.old
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