Issue 591 in fds-smv: FDS+Evac and flow fields with thin obstructions

5 views
Skip to first unread message

codesite...@google.com

unread,
Jan 9, 2009, 5:58:13 AM1/9/09
to fds-smv...@googlegroups.com
Updates:
Status: Fixed

Comment #2 on issue 591 by tkorhon1: FDS+Evac and flow fields with thin
obstructions
http://code.google.com/p/fds-smv/issues/detail?id=591

I just posted to SVN

Compilation Date : Fri, 09 Jan 2009
Version : 5.3.0 Serial
SVN Revision No. : 2984

The changes:
http://virtual.vtt.fi/virtual/proj6/fdsevac/documents/Readme.txt
(Well, I forgot to change evac version no 2.1.0 to 2.1.1, but this will
change
next time I'll commit an evac.f90 update.

Just a question for other droplets: AFILL2 and thin walls? Is it working
there correctly or not?

==========================
Latest changes, Evac 2.1.1
==========================

9.1.2009: The prefered walking direction field v0 is interpolated
differently than in v2.1.0. Old version used function AFILL,
which used the closest four velocity (x,y directions, z is not
needed for 2-dim case) values. There was a problem if there was
thin walls. Then some of these four values used to interpolate
the velocity might be on the other side of the thin wall, i.e.,
interpolation used the values at the cell, where the agent is, and
the values form an adjacent cell, but there was no check for
walls. The present version uses just the values in the current
cell to interpolate the direction field. (Note: the u-velocity is
given at the cell faces -1,+1, the v-velocity is given at the cell
faces -2,+2, see the FDS Tech. Guide)

This modification will change the FDS+Evac results slightly,
but one can use the old evacuation flow fields (CHID_evac.eff).
These fields are not modified, but they are used differently,
i.e., interpolation is done differently.

9.1.2009: v2.1.1 uses less memory than v2.1.0, because radiation
related arrays are not allocated for the solid state baundaries
for the evacuation meshes. For the example case
"evac_example2a.fds" the savings are about 18 % less RAM.

Ciao,
TimoK


--
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

Reply all
Reply to author
Forward
0 new messages