Exhaust fans and openings in FDS5

[Tom]

I'd like to get my head around how to properly setup smoke exhaust
fans in FDS5 and also how to create openings (to e.g. simulate a
window breaking, or to model a closed fan before it ramps up).

As I've understood, with the advent of ZONEs, the use of VENTs as
exhaust fans is no longer desirable? Additionally, the use of VENTs
that open to create openings in walls is no longer permitted either?

From what I've gathered:

1. To model a ceiling mounted exhaust fan, you would create a HOLE in
the ceiling OBST, place a thin obstruction over that same area, and
apply a SURFace to that obstruction that is POROUS=.TRUE. and has a
specified VOLUME_FLUX, say 5 cubic metres per second for example.

Then, if you would like the fan to be "closed" (i.e. not permeable to
smoke) until it ramps up, you would need to place another OBST over
that space, and remove it at the same time the fan starts? That would
be two OBSTs at the same coordinates, is that desirable? Or would you
simply create the hole when the fan starts instead?

2. To model a breaking window or door that opens, I suppose my
questions is the same as how to model the closed fan above. Is the way
to go to create a HOLE in the wall/ceiling or to remove an obstruction
placed over a hole?

Thanks for developing a truly useful and impressive tool, and for
maintaining this very useful forum.

Tom

[Jason]

Tom,

According to my experiences with FDS5 for the exhaust fans, it seems
there are some problems in this application. For example, if I
increase the fan exhaust rate by ten times, the smoke removal can not
be significantly improved. Furthermore, if I use FDS5 for a Deem-to-
Satisfy case, the FDS result can not pass the assessment against the
standard, which is strange. Therefore I recommend some validation is
required for exhaust fans.

Yunlong

[andrew]

"......，if I increase the fan exhaust rate by ten times, the smoke removal
can not be significantly improved ......",this could not be the evidence
that FDS has a problem with exhausting fans.Under some conditions,the
smoke removal efficiency increases little as the exhaust rate is too
large.

> > Tom- Hide quoted text -
>
> - Show quoted text -

[Kevin]

This fan discussion is beginning to sound like urban legend.

Seriously, if the information you are looking for is not in the User's
Guide, then it should be. Plus, for these situations, we want to add
sample cases that are described in the User's Guide and provided along
with the downloads.

In FDS 5 there are various options for venting, fans, and other HVAC
features. If you just have a specified exhaust flow rate (m3/s or cfm
properly converted to m3/s), you vent just as in past versions -- you
put a VENT on a solid obstruction and set SURF parameters of either
VEL or VOLUME_FLUX. You can ramp these in time, and you can trigger
them with whatever device. See the sample case "activate_vents.fds"
for some options.

As for the other cases you describe, we just have not considered every
conceivable venting configuration. So the best advice is to create a
very simple little test case to do what you want. If it doesn't do
what you want, or if something is not working, it's best to post it on
the Issue Tracker, so we can resolve the problem. You can take a look
at the current sample cases and work from there.

[Kevin]

Well, is this a bug? What is the "standard"? Are you comparing FDS
against a set of experiments, or an engineering correlation?

[Jason]

Kevin,

The "Standard" I refered to is the Clause Specification E2.2b (Smoke
Exhaust System) of Building Code Australia (BCA) 2007.
Under this Clause, the Deemed-to-Satisfy fan exhaust rate is given for
a certain fire HRR under certain ceiling height. The clause is
uploaded as pdf file (file name: DtS_specE2.2b.pdf).

Thanks.

Yunlong

[N Kuzewicz]

Yunlong

Not sure if I have missed the point but the BCA is an 'accpetable
benchmark' level of rather than 'ultimate' (zero fatality) level of
safety. Many areas within the BCA have been shown to be not adequate
for there purpose (ie DTS spandrels in highrise, etc), but they still
provide an acceptable design.

Nick

[Kevin]

My concern is not necessarily the Building Code of Australia, but
rather the accuracy of FDS. I'd much prefer to compare FDS simulations
with experimental data rather than regulatory curves or standards.

> > against a set of experiments, or an engineering correlation?- Hide quoted text -

[Bak, Daniel]

Kevin,

I totally agree with you. The use of any fire modeling tool is not to
support minimum requirement of any Code but rather to simulate
particular conditions. However, it is clear, at least to me that there
is some confusion about proper uses of venting in a compartment. I
believe that the user guide should further "guide" the user as to
tail-tale tools to use to verify desired outcomes.

For example, I have been trying to do the classical smoke management in
a large compartment. I have progressively complicated my scenario with
smoke detection, then both smoke detection and sprinklers. But, it
seems that every time I have added a level of complexity, another
surprise popped up in my output.

I understand that you and your team wish for simple scenarios to debug
the issues or to guide users to appropriate suggestions.

I wish that you address this issue in more depth including the reason
for higher exhaust rates which do not seem to exhaust as much smoke as
one would expect. Another issue that I wish to obtain better clarity is
the progressive control to the point of total extinguishment of a fire.

I do apologize in advance for some elementary suggestions and request,
but you must keep in mind that new FDS users are getting in the fire
modeling world everyday.

Nonetheless, I have found the Users group extremely valuable! Thank you
for your efforts and patience.

Daniel

[Kevin]

Daniel

Exactly how many people do you think are currently working on the FDS
source code, verification, validation, compiling, testing on the
various platforms, MPI and parallel processing, debugging, answering
these questions and posts, and on and on.

At the moment -- two.

You seem to be missing the whole point of these on-line resources.
NIST and VTT do not have armies of programmers dedicated to this
effort. We keep saying over and over that we need the help of the FPE
community to maintain FDS (Smokeview is maintained by one person only,
Glenn Forney). Those "surprises" that you talk about should be
diagnosed as either bugs or model limitations or legitimate fire
phenomena. And if you just get frustrated and shrug your shoulders and
curse this wretched software, then you are not helping. If we are to
continue to provide this software for free, we absolutely need users
to be more active in its support.

If something is not explained in the Users Guide, either let us know
or suggest some text that can be inserted. Better yet, develop a
simple test or verification case that demonstrates how to do
something. There are only so many hours in the day for us to do these
sorts of things.

> > - Show quoted text -- Hide quoted text -

[Bak, Daniel]

Kevin,

I have never cursed at any program that NIST has produced and made
available to the entire world for free!!! I can assure you that I have
made countless praises about the dedication of most NIST people and
their invaluable work. One of the reasons for my strong belief is the
fact that I was extremely fortunate to conduct my doctorate research at
NIST under the guidance of "Bud" Nelson back in 1986. My experience at
NIST was absolutely rewarding in every aspects and it gave me insights
about fire science I could not have learned and appreciated anywhere
else. I am a human being and I do get frustrated sometimes with some
"disconnect" between the scientific community and practitioners.

I am all ears to help make models and modeling as accurate and
respectable by many skeptics. I do not claim to be a guru in modeling.
I have learned heaps from religiously following discussion threads.

I think that I understand what you are asking in terms of help and I can
promise that I am going to provide you with feedback as to more clarity
with content of the User's Guide and some of the practical issues
practicing FPEs deal with. But, please bear in mind that some of my
questions and requests may sometimes be already answered in some past or
elementary. Kevin, I really like to help!

Thank you again for taking the time to answer my comments.

[Kevin]

OK, thanks for the clarification. Even though we've had these on-line
services up for about 8 months, I still get contacted directly by
people, who complain about something not working right, but then say
that they did not want to bother me, or did not have the time to
submit the report. I just want to emphasize again that it is vitally
important that any "surprise" be checked out, as that is how we are
going to make the software better.

[Tom]

Jason,

I have to date not done any tests with smoke exhaust in FDS5. I will
report my findings if there seems to be something amiss.

Kevin,

Thanks for your answer. I was obviously a bit confused as I for some
reason was under the impression VENTs should not be used for actual
smoke extraction (or whatever you want to extract or vice versa) at
all... In retrospect, and after reading some posts as well as the
guide a bit closer, I'm more in tune with the logic to replace free
standing VENTs in FDS4 with porous OBSTs in FDS5.

[Lenny]

In regards the to smake exhaust rate there is alot more to effective
smoke exhaust extraction than setting the smoke exhaust rate then
x10.

In my experince , the most significant impact is in the make up air.
If you are not supplying the sytem with the right volume of air the
exhaust system will be constrained to the amount of air that is
introduced to the model. I know that I often over look this. i try to
set the make up air velovity over the grill / opening to be around 1 m/
s to a max of 2.5 m/s. These numbers have always served me well.

Also a localised high extraction rate fans will increase the local
velocity that will generate turbulence - resulting in lost engery to
the trubulence and this will impact the exhaust systems capacity to
extract smoke.

I have found that FDS 5 models the smoke exhaust rate faily accurately
based on my modeling work in both previous versions of fds and other
CFD code.

[Kevin]

The term VENT has always been a problem. It is really nothing more
than a way to specify a BC on a "patch" of the boundary. It is like
drawing a rectangle on the wall and telling FDS to do something
different there, like suck or blow air/gas, heat it up, burn it,
whatever. VENT was just a conventient keyword from pre-FDS days.

[Jason]

Is there anybody ever conducted validation of mechanical ventilation
for FDS5?
Some research on this topic would certainly be valuable.

Jason

[Kevin]

FDS was part of a V&V study by the US Nuclear Regulatory Commission

http://www.nrc.gov/reading-rm/doc-collections/nuregs/staff/sr1824/

Although this study was done with FDS 4, all the cases were rerun with
FDS 5 with similar results. This new Validation Guide will be released
soon.