Moving Objects
475 views
Skip to first unread message
Marco C
Aug 3, 2010, 12:28:09 PM8/3/10
to FDS and Smokeview Discussions
Hi All,
I need to study the air velocity along a tunnel while trains are
running along it. It is well known that FDS cannot prescribe moving
object, so I would like to know if somebody already tried to
"simulate" this issue in some alternative way.
Long ago I already tried to have some answer about it, so, after long
time maybe some new feauture of FDS or a "techique" from some user can
help me.
By my side, I tried to prescribe some vertical vents with FAN SURF_ID,
simulating the flux originating from the running train; of course such
a way to simulate it determines a different flux, so I could prescribe
such a model just in order to evaluate some features related to a
special configuration of a tunnel, where the smoke originating from a
fire in a tunnel could spread to an adjacent tunnel, joined by a
trasverse passageway, where the train is running (Venturi effect).
In order to evaluate appropriate ventilation strategies aimed to
mantain eg a certain longitudinal velocity I guess the FDS it is not
the appropriate tool to use, but any suggestion is really welcome.
Thanks
Marco
I need to study the air velocity along a tunnel while trains are
running along it. It is well known that FDS cannot prescribe moving
object, so I would like to know if somebody already tried to
"simulate" this issue in some alternative way.
Long ago I already tried to have some answer about it, so, after long
time maybe some new feauture of FDS or a "techique" from some user can
help me.
By my side, I tried to prescribe some vertical vents with FAN SURF_ID,
simulating the flux originating from the running train; of course such
a way to simulate it determines a different flux, so I could prescribe
such a model just in order to evaluate some features related to a
special configuration of a tunnel, where the smoke originating from a
fire in a tunnel could spread to an adjacent tunnel, joined by a
trasverse passageway, where the train is running (Venturi effect).
In order to evaluate appropriate ventilation strategies aimed to
mantain eg a certain longitudinal velocity I guess the FDS it is not
the appropriate tool to use, but any suggestion is really welcome.
Thanks
Marco
Franck Didieux @ LNE
Aug 4, 2010, 7:15:38 AM8/4/10
to FDS and Smokeview Discussions
Hi Marco,
I have been facing the same kind of problem. Right now, the best
approximation I have achieved is to simulate wind in outdoors by
prescribing vents at all open boundaries, and insuring stable wind
conditions are achieved within the whole domain before starting any
combustion related model.
However, tunnel is an even more complicated topic, as there are
interactions between air and train on one hand, and simultaneously
between air and tunnel on the other hand which cannot be handled with
that solution. Transverse passageways add a new level of complexity,
and I have no tenable advice with FDS alone.
Moreover, one specific problem arises also because of these
interactions between walls and air flow : as far as I know, FDS cannot
handle that properly on the friction side. A RANS model would be more
effective to take into account friction. I have still not tried, but
planed to try a two step modeling : solving propagation of fire and
generation of HRR and smoke through FDS, then using something like
Comsol Multiphysic with FDS output for smoke and HRR as an input for
Comsol.
Has anybody tried this ?
I have been facing the same kind of problem. Right now, the best
approximation I have achieved is to simulate wind in outdoors by
prescribing vents at all open boundaries, and insuring stable wind
conditions are achieved within the whole domain before starting any
combustion related model.
However, tunnel is an even more complicated topic, as there are
interactions between air and train on one hand, and simultaneously
between air and tunnel on the other hand which cannot be handled with
that solution. Transverse passageways add a new level of complexity,
and I have no tenable advice with FDS alone.
Moreover, one specific problem arises also because of these
interactions between walls and air flow : as far as I know, FDS cannot
handle that properly on the friction side. A RANS model would be more
effective to take into account friction. I have still not tried, but
planed to try a two step modeling : solving propagation of fire and
generation of HRR and smoke through FDS, then using something like
Comsol Multiphysic with FDS output for smoke and HRR as an input for
Comsol.
Has anybody tried this ?
rmcdermo
Aug 4, 2010, 8:06:15 AM8/4/10
to FDS and Smokeview Discussions
FDS handles the friction law quite well. See the latest Verification
Guide. I think I somewhat disagree with Franck in that I feel we have
a better handle of tunnel flows than outdoor flows at the moment.
As for moving objects, the new immersed boundary method will handle
this. You may want to play with the following, which should work with
5.5.1. Unfortunately, we do not have a good visualization for the
'box' right now. So, you can just see its effect on the velocity and
pressure fields.
&HEAD CHID='train', TITLE='train moving through a tunnel' /
&MESH IJK=100,16,16, XB=0,100,-8,8,0,16/
&TIME T_END=10.0 /
&MISC FDS6=.TRUE.,IMMERSED_BOUNDARY_METHOD=2/
&VENT MB='XMIN', SURF_ID='OPEN'/
&VENT MB='XMAX', SURF_ID='OPEN'/
&SLCF PBY=0,QUANTITY='VELOCITY',VECTOR=.TRUE./
&SLCF PBY=0,QUANTITY='PRESSURE'/
&SLCF PBZ=8,QUANTITY='VELOCITY',VECTOR=.TRUE./
&SLCF PBZ=8,QUANTITY='PRESSURE'/
&GEOM ID='train'
XB=80,100,-4,4,4,12
VELOCITY=-20,0,0
SHAPE='BOX'/
&TAIL /
On Aug 4, 7:15 am, "Franck Didieux @ LNE" <franck.didi...@lne.fr>
wrote:
Guide. I think I somewhat disagree with Franck in that I feel we have
a better handle of tunnel flows than outdoor flows at the moment.
As for moving objects, the new immersed boundary method will handle
this. You may want to play with the following, which should work with
5.5.1. Unfortunately, we do not have a good visualization for the
'box' right now. So, you can just see its effect on the velocity and
pressure fields.
&HEAD CHID='train', TITLE='train moving through a tunnel' /
&MESH IJK=100,16,16, XB=0,100,-8,8,0,16/
&TIME T_END=10.0 /
&MISC FDS6=.TRUE.,IMMERSED_BOUNDARY_METHOD=2/
&VENT MB='XMIN', SURF_ID='OPEN'/
&VENT MB='XMAX', SURF_ID='OPEN'/
&SLCF PBY=0,QUANTITY='VELOCITY',VECTOR=.TRUE./
&SLCF PBY=0,QUANTITY='PRESSURE'/
&SLCF PBZ=8,QUANTITY='VELOCITY',VECTOR=.TRUE./
&SLCF PBZ=8,QUANTITY='PRESSURE'/
&GEOM ID='train'
XB=80,100,-4,4,4,12
VELOCITY=-20,0,0
SHAPE='BOX'/
&TAIL /
On Aug 4, 7:15 am, "Franck Didieux @ LNE" <franck.didi...@lne.fr>
wrote:
Kevin
Aug 4, 2010, 8:09:31 AM8/4/10
to FDS and Smokeview Discussions
You can specify the DYNAMIC_PRESSURE at OPEN vents to create an air
flow in a tunnel due to external winds.
Why do say that FDS cannot properly handle the wall friction, and that
a RANS model would be more effective? The FDS Verification Guide has a
section 3.4, "FDS Wall Flows, Part 1, Straight Channels," that shows
how FDS does handle wall friction properly.
flow in a tunnel due to external winds.
Why do say that FDS cannot properly handle the wall friction, and that
a RANS model would be more effective? The FDS Verification Guide has a
section 3.4, "FDS Wall Flows, Part 1, Straight Channels," that shows
how FDS does handle wall friction properly.
Bryan Klein
Aug 4, 2010, 2:21:41 PM8/4/10
to FDS and Smokeview Discussions
Randy, that is excellent!
I posted a note about it and uploaded a short video to YouTube.
http://bryanklein.com/2010/08/fds-demo-of-a-moving-box-in-a-tunnel/
Keep up the great work,
-Bryan
I posted a note about it and uploaded a short video to YouTube.
http://bryanklein.com/2010/08/fds-demo-of-a-moving-box-in-a-tunnel/
Keep up the great work,
-Bryan
Marco C
Aug 4, 2010, 4:25:14 PM8/4/10
to FDS and Smokeview Discussions
thank you all
Well, I didnt know about this new feature, that's excellent, but as I
read in the video the feature isnt yet fully implemented, what is the
limitation? It means that the predictions are not yet reliable at the
moment?
> I posted a note about it and uploaded a short video to YouTube.http://bryanklein.com/2010/08/fds-demo-of-a-moving-box-in-a-tunnel/
Well, I didnt know about this new feature, that's excellent, but as I
read in the video the feature isnt yet fully implemented, what is the
limitation? It means that the predictions are not yet reliable at the
moment?
> I posted a note about it and uploaded a short video to YouTube.http://bryanklein.com/2010/08/fds-demo-of-a-moving-box-in-a-tunnel/
rmcdermo
Aug 4, 2010, 5:26:20 PM8/4/10
to FDS and Smokeview Discussions
Well, let's just say the new IBM is in 'alpha' testing. But I think
it would yield better results than an ad hoc fan model. You should be
extremely careful here no matter what you try. Neither approach has
been validated.
If you would like to pursue the moving boundary method, send me an
email off line and I can help.
it would yield better results than an ad hoc fan model. You should be
extremely careful here no matter what you try. Neither approach has
been validated.
If you would like to pursue the moving boundary method, send me an
email off line and I can help.
Franck Didieux @ LNE
Aug 5, 2010, 2:55:39 AM8/5/10
to FDS and Smokeview Discussions
Well, considering your remarks, I think my notions about walls, flow
and friction with FDS are a little bit too old...
Have to update ^^
and friction with FDS are a little bit too old...
Have to update ^^
Kevin
Aug 5, 2010, 7:57:04 AM8/5/10
to FDS and Smokeview Discussions
Read the recent discussion thread
http://groups.google.com/group/fds-smv/browse_thread/thread/94c0e385c4fd1cb9#
The reason that we have such strong opinions about the traditional
system of scholarly publication is that these misconceptions about FDS
continue long after we have addressed the issue. I still receive
papers to review about FDS 4, and the papers are filled with
descriptions of algorithms and assumptions that are no longer used. If
such a paper is published in 2010, then it is not surprising that
people still believe that these assumptions are still used.
We understand that it is difficult to read the FDS V&V Guides each
time we release the software. But it is very easy for us to respond to
questions posted to the Discussion Group. The V&V Guides are on-line,
so we can point you to a particular section to learn more about the
current state of the model.
On Aug 5, 2:55 am, "Franck Didieux @ LNE" <franck.didi...@lne.fr>
wrote:
http://groups.google.com/group/fds-smv/browse_thread/thread/94c0e385c4fd1cb9#
The reason that we have such strong opinions about the traditional
system of scholarly publication is that these misconceptions about FDS
continue long after we have addressed the issue. I still receive
papers to review about FDS 4, and the papers are filled with
descriptions of algorithms and assumptions that are no longer used. If
such a paper is published in 2010, then it is not surprising that
people still believe that these assumptions are still used.
We understand that it is difficult to read the FDS V&V Guides each
time we release the software. But it is very easy for us to respond to
questions posted to the Discussion Group. The V&V Guides are on-line,
so we can point you to a particular section to learn more about the
current state of the model.
On Aug 5, 2:55 am, "Franck Didieux @ LNE" <franck.didi...@lne.fr>
wrote:
S.Desanghere
Aug 6, 2010, 4:02:00 AM8/6/10
to FDS and Smokeview Discussions
I agree with Kevin: as users, we should try to forget our point of
view made with older versions of FDS because it can lead to
misconceptions concerning current capabilities of FDS. Furthermore,
that might be frustrating for FDS developers !
But, from the user side, it is not so easy. I did my PhD using FDS 3
and 4 and thus sometimes I can have bad interpretation of FDS 5
behavior. Things are evolving fast because (we are lucky) FDS
developers are very efficient. So, we (the users) have to be very
careful when talking about latest FDS capabilities ! Moreover, I think
we should forget general assumptions like "RANS is better than LES for
wall friction". It may be written somewhere in books but that is not
really of interest for doing fire simulations with FDS.
I have a question about modeling strategy : do you plan in a near
future to implement "volume of fluid" method or something like that,
in order to simulate spill movements from tank leakage for instance ?
Cheers
> http://groups.google.com/group/fds-smv/browse_thread/thread/94c0e385c...
view made with older versions of FDS because it can lead to
misconceptions concerning current capabilities of FDS. Furthermore,
that might be frustrating for FDS developers !
But, from the user side, it is not so easy. I did my PhD using FDS 3
and 4 and thus sometimes I can have bad interpretation of FDS 5
behavior. Things are evolving fast because (we are lucky) FDS
developers are very efficient. So, we (the users) have to be very
careful when talking about latest FDS capabilities ! Moreover, I think
we should forget general assumptions like "RANS is better than LES for
wall friction". It may be written somewhere in books but that is not
really of interest for doing fire simulations with FDS.
I have a question about modeling strategy : do you plan in a near
future to implement "volume of fluid" method or something like that,
in order to simulate spill movements from tank leakage for instance ?
Cheers
> http://groups.google.com/group/fds-smv/browse_thread/thread/94c0e385c...
Kevin
Aug 6, 2010, 8:27:41 AM8/6/10
to FDS and Smokeview Discussions
We have used lagrangian particles to represent liquid droplets
spreading out over a horizontal plane. It is assumed that the droplets
move at a constant velocity. If there were some simple empirical
correlation with which we could predict the flow speed based on the
local depth (which can be calculated from the concentration of
droplets at a given point), we could then just set the particle
velocity accordingly. This is about as far as we would want to go.
As far as different versions of FDS -- the reason we decided to
develop the FDS V&V Guides was because of the problem you mention.
Before these Guides were published, it was very difficult for a model
user to know the capabilities of the model. If you just read the FDS
Technical Guide called "Mathematical Model," it was difficult to know
whether or not the various sub-grid scale models were really working.
Whenever I read a CFD paper, the theoretical development always seems
OK, but I cannot tell if it really works until I see good V&V cases.
> > > Have to update ^^- Hide quoted text -
>
> - Show quoted text -
spreading out over a horizontal plane. It is assumed that the droplets
move at a constant velocity. If there were some simple empirical
correlation with which we could predict the flow speed based on the
local depth (which can be calculated from the concentration of
droplets at a given point), we could then just set the particle
velocity accordingly. This is about as far as we would want to go.
As far as different versions of FDS -- the reason we decided to
develop the FDS V&V Guides was because of the problem you mention.
Before these Guides were published, it was very difficult for a model
user to know the capabilities of the model. If you just read the FDS
Technical Guide called "Mathematical Model," it was difficult to know
whether or not the various sub-grid scale models were really working.
Whenever I read a CFD paper, the theoretical development always seems
OK, but I cannot tell if it really works until I see good V&V cases.
>
> - Show quoted text -
rmcdermo
Aug 6, 2010, 8:49:02 AM8/6/10
to FDS and Smokeview Discussions
Regarding VOF, I've toyed with the idea. But it is low on the
priority list at the moment, unless our users start asking for it.
One of the reasons it looks quite possible is that Ruddy implemented a
level-set method for front tracking of fire lines. The same machinery
could be used for tracking the fluid surface in VOF.
priority list at the moment, unless our users start asking for it.
One of the reasons it looks quite possible is that Ruddy implemented a
level-set method for front tracking of fire lines. The same machinery
could be used for tracking the fluid surface in VOF.
Franck Didieux @ LNE
Aug 9, 2010, 12:07:22 PM8/9/10
to FDS and Smokeview Discussions
@Sylvain :
to date on that point may be a full time job too. A good compromise
might be found.
wall friction) ?
Regards
>Furthermore, that might be frustrating for FDS developers !
Totally agree ^^
> So, we (the users) have to be very careful when talking about latest FDS capabilities !
This is a general statement, not only for FDS... However, always be up
to date on that point may be a full time job too. A good compromise
might be found.
> Moreover, I think we should forget general assumptions like "RANS is better than LES for wall friction". It may be written somewhere in books but that is not really of interest for doing fire simulations with FDS.
Most probably. Could you please develop on that point (RANS vs LES for
wall friction) ?
Regards
Marco C
Aug 27, 2010, 7:05:53 AM8/27/10
to FDS and Smokeview Discussions
thanks Randy,
for sure I'll send you an e mail in order to try this feature;
nevertheless I didnt found on the User Guide some explanation about
it, do you have some draft that you can send me?
thanks
Marco
for sure I'll send you an e mail in order to try this feature;
nevertheless I didnt found on the User Guide some explanation about
it, do you have some draft that you can send me?
thanks
Marco
Bryan Klein
Aug 27, 2010, 5:59:15 PM8/27/10
to FDS and Smokeview Discussions
Hello Marco,
The most current bleeding edge draft version of the FDS User Guide
(unpublished draft) is available in the FDS-SMV Project SVN
repository. It is a LaTeX document, and will need to be built along
with the necessary includes (images, etc.)
This means you must have a LaTeX engine installed ( Windows - http://miktex.org/
| OSX - http://www.tug.org/mactex/ ).
If you have a Subversion client installed (including command line
utility) you can always get the document and subdirectories with the
following command.
svn export http://fds-smv.googlecode.com/svn/trunk/FDS/trunk/Manuals/FDS_5_User_Guide/
Be sure to run this svn command in the directory where you want the
FDS_5_User_Guide folder to appear.
There are also a couple of good tips in this Wiki page.
http://code.google.com/p/fds-smv/wiki/Compiling_FDS_Manuals
(You will want to get the SCRIPT_FIGURES directory from the zip file
and drop it into the FDS_5_User_Guide directory.)
Best of luck,
-Bryan Klein
The most current bleeding edge draft version of the FDS User Guide
(unpublished draft) is available in the FDS-SMV Project SVN
repository. It is a LaTeX document, and will need to be built along
with the necessary includes (images, etc.)
This means you must have a LaTeX engine installed ( Windows - http://miktex.org/
| OSX - http://www.tug.org/mactex/ ).
If you have a Subversion client installed (including command line
utility) you can always get the document and subdirectories with the
following command.
svn export http://fds-smv.googlecode.com/svn/trunk/FDS/trunk/Manuals/FDS_5_User_Guide/
Be sure to run this svn command in the directory where you want the
FDS_5_User_Guide folder to appear.
There are also a couple of good tips in this Wiki page.
http://code.google.com/p/fds-smv/wiki/Compiling_FDS_Manuals
(You will want to get the SCRIPT_FIGURES directory from the zip file
and drop it into the FDS_5_User_Guide directory.)
Best of luck,
-Bryan Klein
Message has been deleted
Bryan Klein
Aug 27, 2010, 6:03:18 PM8/27/10
to FDS and Smokeview Discussions
Not a big fan of the line truncation for the svn link...
Click it to see the full URL in the browser and use it.
-Bryan
On Aug 27, 4:59 pm, Bryan Klein <kl...@thunderheadeng.com> wrote:
> Hello Marco,
>
> The most current bleeding edge draft version of the FDS User Guide
> (unpublished draft) is available in the FDS-SMV Project SVN
> repository. It is a LaTeX document, and will need to be built along
> with the necessary includes (images, etc.)
> This means you must have a LaTeX engine installed ( Windows -http://miktex.org/
> | OSX -http://www.tug.org/mactex/).
>
> If you have a Subversion client installed (including command line
> utility) you can always get the document and subdirectories with the
> following command.
>
> svn exporthttp://fds-smv.googlecode.com/svn/trunk/FDS/trunk/Manuals/FDS_5_User_...
Click it to see the full URL in the browser and use it.
-Bryan
On Aug 27, 4:59 pm, Bryan Klein <kl...@thunderheadeng.com> wrote:
> Hello Marco,
>
> The most current bleeding edge draft version of the FDS User Guide
> (unpublished draft) is available in the FDS-SMV Project SVN
> repository. It is a LaTeX document, and will need to be built along
> with the necessary includes (images, etc.)
> This means you must have a LaTeX engine installed ( Windows -http://miktex.org/
> | OSX -http://www.tug.org/mactex/).
>
> If you have a Subversion client installed (including command line
> utility) you can always get the document and subdirectories with the
> following command.
>
>
> Be sure to run this svn command in the directory where you want the
> FDS_5_User_Guide folder to appear.
>
> There are also a couple of good tips in this Wiki page.http://code.google.com/p/fds-smv/wiki/Compiling_FDS_Manuals
> Be sure to run this svn command in the directory where you want the
> FDS_5_User_Guide folder to appear.
>
Marco C
Aug 28, 2010, 8:39:16 AM8/28/10
to FDS and Smokeview Discussions
Thanks Randy,
u will hear me soon...
Marco
u will hear me soon...
Marco
CFD modeling Engineering design/calculations
Apr 13, 2017, 11:07:43 AM4/13/17
to FDS and Smokeview Discussions
среда, 4 августа 2010 г., 15:06:15 UTC+3 пользователь Randy McDermott написал:
Randy McDermott
Apr 13, 2017, 11:18:14 AM4/13/17
to FDS and Smokeview Discussions
Sorry, most of us don't speak Russian. I see 6.5.3 and exclamation points, but I am not sure how to answer this. Can one our Russian collaborators please translate? Thanks.
--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+unsubscribe@googlegroups.com.
To post to this group, send email to fds...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/3f3d24bf-397b-4099-af51-56152d73e21f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
dr_jfloyd
Apr 13, 2017, 11:24:39 AM4/13/17
to FDS and Smokeview Discussions
Did you try to run the example that Randy had posted in 2010? If so that example was prior to the release of FDS6 and the FDS6 keyword on MISC was to enable new routines being tested for inclusion in FDS6. There is no longer an FDS6 keyword on MISC as we have now released FDS6 and permanently enabled those features we decided to include.
Pavel_Ch
Apr 14, 2017, 2:29:47 AM4/14/17
to FDS and Smokeview Discussions
Hello! Sorry I was not able to quickly respond. translation of the original question below.
Maybe my question will seem not a lot of inappropriate/stupid, but I can't run the code in FDS 6.5.3, an error (in the picture) line MISK, MISK removed error Geom line. If I understand correctly, at the moment the program does not support it, only the demo...
четверг, 13 апреля 2017 г., 18:18:14 UTC+3 пользователь Randy McDermott написал:
четверг, 13 апреля 2017 г., 18:18:14 UTC+3 пользователь Randy McDermott написал:
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
Pavel_Ch
Apr 14, 2017, 2:43:44 AM4/14/17
to FDS and Smokeview Discussions
Thanks in advance.
четверг, 13 апреля 2017 г., 18:24:39 UTC+3 пользователь dr_jfloyd написал:
четверг, 13 апреля 2017 г., 18:24:39 UTC+3 пользователь dr_jfloyd написал:
Salah Benkorichi
Apr 14, 2017, 4:44:11 AM4/14/17
to fds...@googlegroups.com
You need to delete the FDS from the MISC in order to run it.
Test this
&HEAD CHID='train', TITLE='train moving through a tunnel' /
&MESH IJK=100,16,16, XB=0,100,-8,8,0,16/
&TIME T_END=10.0 /
/&MISC IMMERSED_BOUNDARY_METHOD=2/
&VENT MB='XMIN', SURF_ID='OPEN'/
&VENT MB='XMAX', SURF_ID='OPEN'/
&SLCF PBY=0,QUANTITY='VELOCITY',VECTOR=.TRUE./
&SLCF PBY=0,QUANTITY='PRESSURE'/
&SLCF PBZ=8,QUANTITY='VELOCITY',VECTOR=.TRUE./
&SLCF PBZ=8,QUANTITY='PRESSURE'/
&GEOM ID='train'
/XB=80,100,-4,4,4,12
/VELOCITY=-20,0,0
/SHAPE='BOX'/
&TAIL /
--
You received this message because you are subscribed to the Google Groups "FDS and Smokeview Discussions" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fds-smv/fee5d216-47ec-4d12-bc8c-ba37c3d6e8cb%40googlegroups.com.
Pavel_Ch
Apr 14, 2017, 6:25:28 AM4/14/17
to FDS and Smokeview Discussions
Everything works. Thank you very much.
пятница, 14 апреля 2017 г., 11:44:11 UTC+3 пользователь Salah Benkorichi написал:
пятница, 14 апреля 2017 г., 11:44:11 UTC+3 пользователь Salah Benkorichi написал:
To unsubscribe from this group and stop receiving emails from it, send an email to fds-smv+u...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages