FDS Version 4

[Kevin McGrattan <kevin.mcgrattan@nist.gov>]

All,

A few more notes about the release of FDS 4 that I announced
last week:

1. Until the manuals are through review and the Windows
install program
   is assembled, download version 4 manuals, executables,
samples, etc.
   from the FTP site
(ftp://ftp.nist.gov/pub/bfrl/mcgratta/FDS4). Some
   of you have reported trouble getting these files. If you
are using
   Windows Explorer, you may be experiencing a security
feature that 
   prevents downloading from untrusted sites. Try using some
other FTP
   client to get to the site. If prompted for a user name
and password, 
   the user name is anonymous, and the password is just your
email address.

2. There are a number of issues that still need to get
sorted out in FDS 4.
   Probably in a few months we will release a 4.1 that will
include
   a smoke detector feature, a fix to the multiblock
function, and a fix
   to the combustion routine. There is a known bug that when
using
   multiblocks in a relatively sealed domain, like a tunnel,
there can
   arise spurious pressure fluctuations. We know why this is
occurring, but
   have not found a satisfactory fix. Another known bug
occurs when 
   an external agent is added to the domain during a
calculation that uses
   the mixture fraction combustion model. The dilution and
combustion 
   suppression are not being handled properly. We think we
have a solution
   to the problem, but it will require some time. 

3. Some of you have tried using the FDS/SV executables for
other types of
   operating systems, like Linux or Mac OSX. Keep in mind
that these
   executables are not as portable as the Windows .exe
files. This is
   especially true of the parallel version. Our Linux
executables were
   compiled using Lahey/Fujitsu Fortran running on RedHat
Linux. The parallel
   version uses LAM/MPI. On our Mac, we compiled using IBM's
Fortran 
   compiler and LAM/MPI. Sometimes these executables will
work, sometimes
   not, depending on your version of the OS, libraries, etc.
Let us know
   if you are successful or not running these.

4. The newly released version of FDS is called 4.00. Minor
bug fixes will
   be applied to 4.01, 4.02, etc. More serious changes will
result in
   4.10, 4.20, etc. Note when running FDS the version number
and compile
   date. Try to keep up with newer releases because often
reported bugs
   have been fixed and posted on the FTP site. I do not have
the time to
   fully document the bug fixes, so be flexible and always
consider trying
   a newer version to see if a bug has been fixed.

5. Versions of FDS and Smokeview are stamped with numbers
and compile dates.
   References to the programs are best made with the version
number, unless
   it is a test version, in which case the date is
preferable. However,
   the database file (database4.data, etc) is NOT
referencable. I have
   seen too often in papers and conference proceedings
references to such
   and such material from the NIST database. The database is
meant to 
   give examples of different types of materials to
illustrate how you 
   provide input to FDS. The numbers have been pulled from
handbooks, papers,
   bench-scale tests, etc. Some of the numbers are just
order of magnitude
   guesses. We do not track changes that we make to the
database file. If we
   find a number that makes no sense, we just change it. If
you are 
   reporting an FDS simulation, you must confirm all the
numbers that you
   have pulled from the database. You must understand what
they mean and
   how they are used in the model.

6. The sample cases being released with FDS 4 have changed,
so do not compare
   output from previous versions. We have also added more
comments in the
   sample files to help new users.

That's all for now. Let me know if there are any problems
with the new
version.

Kevin


-- 
Kevin McGrattan
National Institute of Standards and Technology
100 Bureau Drive, Stop 8663
Gaithersburg MD 20899-8663

Telephone: (301) 975 2712
FAX:       (301) 975 4052