SURF_ID6 and layered materials

[gforney]

Please complete the following lines...

FDS Version:5.5.3
SVN Revision Number:7031
Compile Date:Fry, 29 October 2010
Smokeview Version/Revision:5.6
Operating System:WIN64 (Intel C/C++)

FDS Version:6.0 Release Candidate 3
SVN Revision Number:14299
Compile Date:Thu, 20 December 2012
Smokeview Version/Revision:6.0.11
Operating System:WIN64 (Intel C/C++)

Describe details of the issue below:

Very often I have to deal with compounds and therefore I use SURF_ID6 to assign reciprocal
stacking sequences to the back and front of the wall; which surf should I assign to
the remaining faces?
At the beginning I used 'INERT' (due to a misunderstanding), then I passed to defining
an ‘ADIABATIC’  ad hoc surf but it is still an error (I was thinking to FEM symmetry
condition in thermal analysis): how could I get a good approximation?
In addition, I would like to use BURN_AWAY in order to make that the fuel disappears
but, till now, I never used it because of what it is written in the user’s guide about
different SURFs assigned to the same obstruction. Could a sort of ‘neutral’ SURF, to
be assigned to the side faces of the wall, be a reliable workaround? In this case,
should I calculate the BULK_DENSITY as a weighted average of densities of each layer?
In attachment 2 files for FDS5.5 and FDS6.0 respectively. It looks like the latter
doesn’t work due to the presence of BURN_AWAY.

Thank you for your attention and best regards.

Marco

Original issue reported on code.google.com by spagnolo.marco on 2013-03-14 16:37:46

---

• _Attachment: SURF_ID6_FDS_5.fds_
• _Attachment: SURF_ID6_FDS_6.fds_

[gforney]

I'll take a look at it.

Original issue reported on code.google.com by mcgratta on 2013-03-14 17:25:45

[gforney]

Use BULK_DENSITY whenever there are multiple SURF_IDs applied to a solid. In the FDS
6 case, FDS saw that the side boundary was massless and set the block to be massless,
which then burned away immediately.

Original issue reported on code.google.com by mcgratta on 2013-03-14 18:55:04

[gforney]

Thank you! I'll try immediately.

Best regards,

Marco

Original issue reported on code.google.com by spagnolo.marco on 2013-03-15 08:08:09

[gforney]

Hi,

I used BULK_DENSITY and it works, thank you.

I'd like to know what "WontFix" stands for as far as issues status is
concerned.
Thank you.

Original issue reported on code.google.com by spagnolo.marco on 2013-03-18 13:51:31

[gforney]

It means that we are not taking any action on this. The User Guide recommends using
BULK_DENSITY whenever there is a possible inconsistency in the specification of boundary
conditions.

Original issue reported on code.google.com by mcgratta on 2013-03-18 14:16:23

[gforney]

Thank you for your fast reply.
However, the first question is still open: when I use SURF_ID6 to
assign reciprocal stacking sequences to the back and front of a wall,
which surf should I assign to the remaining faces?
What do you think to use a uniform surf with the more combustible
material between the layers and a thickness which gives a negligible
contribution to the whole fire load (respect to the back and front of
the wall)?
Thank you for your attention.

Original issue reported on code.google.com by spagnolo.marco on 2013-03-18 16:27:20

[gforney]

FDS makes the assumption that the heat transfer within a solid is one dimensional. This
is an assumption, and assumptions have limits. You are pushing those limits, and therefore
I don't have a good answer for you. You need to try your idea on your case, and verify
that the error associated with your assumption is tolerable. If there is something
we can do to the FDS code to improve its accuracy, we will consider it. But this sounds
more like a modeling decision that is yours to make.

Original issue reported on code.google.com by mcgratta on 2013-03-18 16:37:01

[gforney]

You are right.
As soon as I'm able to get some experimental data about something more
than cone calorimeter ones (e.g. a flame penetration test) I'll
perform a numerical-experimental comparison and let you know the
results.
In the meanwhile, thank you for your availability.

Original issue reported on code.google.com by spagnolo.marco on 2013-03-18 16:59:25