Extra species boundary condition
52 views
Skip to first unread message
Stephen Olenick
Apr 24, 2009, 9:54:52 AM4/24/09
to FDS and Smokeview Discussions
I am trying to simulate a vent blowing smoke at a prescribed
velocity. The velocity is known. The temperature is known. The
amount of smoke is known. But I am running into problems on how to
actually code it in. My initial thought was something like this:
&SPEC ID='Smoke',MW=28.8,MASS_EXTINCTION_COEFFICIENT=8700. /
&VENT
XB=25.171,25.171,9.81701,11.1886,15.6975,16.8405,RGB=255,0,0,IOR=-1,SURF_ID='hotsmoke' /
area 1.5677 m2
&SURF ID='hotsmoke',RGB=255,255,0,TMP_FRONT=600.0,VEL=-6.7056,MASS_FLUX
(1)=0.0001 /
My thinking is that this should set a 600C vent blowing a mixture of
smoke (at mass flux of 0.0001) at 6.7056 m/s and fill in the rest with
background species (I think a mix of air (mass frac 0.992) and water
vapor (mass frac 0.008)).
The manual, though, indicates that velocity and mass flux cannot be
set together because the program calculates the velocity from the mass
flux. So I guess I am looking for suggestions on how to do this. I
could use velocity along with mass fractions, but how to calculate the
mass fraction of 'Smoke' when I don't know the mass fraction of
whatever else is going to be included in the calculation by FDS? Or I
could use mass fluxes, but how to know what the velocity? I am
probably making this harder than it has to be.
Any suggestions?
velocity. The velocity is known. The temperature is known. The
amount of smoke is known. But I am running into problems on how to
actually code it in. My initial thought was something like this:
&SPEC ID='Smoke',MW=28.8,MASS_EXTINCTION_COEFFICIENT=8700. /
&VENT
XB=25.171,25.171,9.81701,11.1886,15.6975,16.8405,RGB=255,0,0,IOR=-1,SURF_ID='hotsmoke' /
area 1.5677 m2
&SURF ID='hotsmoke',RGB=255,255,0,TMP_FRONT=600.0,VEL=-6.7056,MASS_FLUX
(1)=0.0001 /
My thinking is that this should set a 600C vent blowing a mixture of
smoke (at mass flux of 0.0001) at 6.7056 m/s and fill in the rest with
background species (I think a mix of air (mass frac 0.992) and water
vapor (mass frac 0.008)).
The manual, though, indicates that velocity and mass flux cannot be
set together because the program calculates the velocity from the mass
flux. So I guess I am looking for suggestions on how to do this. I
could use velocity along with mass fractions, but how to calculate the
mass fraction of 'Smoke' when I don't know the mass fraction of
whatever else is going to be included in the calculation by FDS? Or I
could use mass fluxes, but how to know what the velocity? I am
probably making this harder than it has to be.
Any suggestions?
drjfloyd
Apr 24, 2009, 10:05:50 AM4/24/09
to FDS and Smokeview Discussions
For a boundary condition MASS_FRACTION applies to what is being blown
in at the VENT not what is already in the gas cells next to the VENT.
in at the VENT not what is already in the gas cells next to the VENT.
Stephen Olenick
Apr 24, 2009, 10:21:22 AM4/24/09
to FDS and Smokeview Discussions
Agreed. But how would you suggest injecting smoke at a specified
velocity and temperature while ensuring a particular mass flux of
'Smoke'? I think it will require dilution with another species (i.e.
'AIR'). For example, the users manual section 12.2.7 suggests:
&SPEC ID='MY SMOKE', MW=29.,MASS_EXTINCTION_COEFFICIENT=8700. /
&SURF ID='NO FIRE',TMP_FRONT=1000.,MASS_FLUX(1)=0.0001,COLOR='RED' /
The program will then calculate the velocity based on the mass_flux.
No problem there. But what if you wanted the same situation, but
wanted to specify the velocity and just dilute it with background
species until the velocity is attained?
velocity and temperature while ensuring a particular mass flux of
'Smoke'? I think it will require dilution with another species (i.e.
'AIR'). For example, the users manual section 12.2.7 suggests:
&SPEC ID='MY SMOKE', MW=29.,MASS_EXTINCTION_COEFFICIENT=8700. /
&SURF ID='NO FIRE',TMP_FRONT=1000.,MASS_FLUX(1)=0.0001,COLOR='RED' /
The program will then calculate the velocity based on the mass_flux.
No problem there. But what if you wanted the same situation, but
wanted to specify the velocity and just dilute it with background
species until the velocity is attained?
Kevin
Apr 24, 2009, 10:37:28 AM4/24/09
to FDS and Smokeview Discussions
That over-specifies things. If you know your mass flux, and you want
that to be injected exactly, then FDS has to figure out the normal
velocity at the vent so that the sum of the convective and diffusive
transport of the species is what you want. Diffusion is small in
reality, but remember in FDS that we use "artificial" diffusion to
handle subgrid-scale turbulence. So specify the mass flux as input and
check the velocity that FDS picks. It should be fairly close to the
ideal, but not exactly.
On Apr 24, 10:21 am, Stephen Olenick <Stephen.Olen...@gmail.com>
>
> - Show quoted text -
that to be injected exactly, then FDS has to figure out the normal
velocity at the vent so that the sum of the convective and diffusive
transport of the species is what you want. Diffusion is small in
reality, but remember in FDS that we use "artificial" diffusion to
handle subgrid-scale turbulence. So specify the mass flux as input and
check the velocity that FDS picks. It should be fairly close to the
ideal, but not exactly.
On Apr 24, 10:21 am, Stephen Olenick <Stephen.Olen...@gmail.com>
wrote:
> Agreed. But how would you suggest injecting smoke at a specified
> velocity and temperature while ensuring a particular mass flux of
> 'Smoke'? I think it will require dilution with another species (i.e.
> 'AIR'). For example, the users manual section 12.2.7 suggests:
>
> &SPEC ID='MY SMOKE', MW=29.,MASS_EXTINCTION_COEFFICIENT=8700. /
> &SURF ID='NO FIRE',TMP_FRONT=1000.,MASS_FLUX(1)=0.0001,COLOR='RED' /
>
> The program will then calculate the velocity based on the mass_flux.
> No problem there. But what if you wanted the same situation, but
> wanted to specify the velocity and just dilute it with background
> species until the velocity is attained?
>
> On Apr 24, 10:05 am, drjfloyd <drjfl...@gmail.com> wrote:
>
>
>
> > For a boundary condition MASS_FRACTION applies to what is being blown
> > in at the VENT not what is already in the gas cells next to the VENT.
>
> > On Apr 24, 9:54 am, Stephen Olenick <Stephen.Olen...@gmail.com> wrote:
>
> > > I am trying to simulate a vent blowing smoke at a prescribed
> > > velocity. The velocity is known. The temperature is known. The
> > > amount of smoke is known. But I am running into problems on how to
> > > actually code it in. My initial thought was something like this:
>
> > > &SPEC ID='Smoke',MW=28.8,MASS_EXTINCTION_COEFFICIENT=8700. /
>
> > > &VENT
> > > XB=25.171,25.171,9.81701,11.1886,15.6975,16.8405,RGB=255,0,0,IOR=-1,SURF_ID='hotsmoke' /
> Agreed. But how would you suggest injecting smoke at a specified
> velocity and temperature while ensuring a particular mass flux of
> 'Smoke'? I think it will require dilution with another species (i.e.
> 'AIR'). For example, the users manual section 12.2.7 suggests:
>
> &SPEC ID='MY SMOKE', MW=29.,MASS_EXTINCTION_COEFFICIENT=8700. /
> &SURF ID='NO FIRE',TMP_FRONT=1000.,MASS_FLUX(1)=0.0001,COLOR='RED' /
>
> The program will then calculate the velocity based on the mass_flux.
> No problem there. But what if you wanted the same situation, but
> wanted to specify the velocity and just dilute it with background
> species until the velocity is attained?
>
> On Apr 24, 10:05 am, drjfloyd <drjfl...@gmail.com> wrote:
>
>
>
> > For a boundary condition MASS_FRACTION applies to what is being blown
> > in at the VENT not what is already in the gas cells next to the VENT.
>
> > On Apr 24, 9:54 am, Stephen Olenick <Stephen.Olen...@gmail.com> wrote:
>
> > > I am trying to simulate a vent blowing smoke at a prescribed
> > > velocity. The velocity is known. The temperature is known. The
> > > amount of smoke is known. But I am running into problems on how to
> > > actually code it in. My initial thought was something like this:
>
> > > &SPEC ID='Smoke',MW=28.8,MASS_EXTINCTION_COEFFICIENT=8700. /
>
> > > &VENT
> > > area 1.5677 m2
>
> > > &SURF ID='hotsmoke',RGB=255,255,0,TMP_FRONT=600.0,VEL=-6.7056,MASS_FLUX
> > > (1)=0.0001 /
>
> > > My thinking is that this should set a 600C vent blowing a mixture of
> > > smoke (at mass flux of 0.0001) at 6.7056 m/s and fill in the rest with
> > > background species (I think a mix of air (mass frac 0.992) and water
> > > vapor (mass frac 0.008)).
>
> > > The manual, though, indicates that velocity and mass flux cannot be
> > > set together because the program calculates the velocity from the mass
> > > flux. So I guess I am looking for suggestions on how to do this. I
> > > could use velocity along with mass fractions, but how to calculate the
> > > mass fraction of 'Smoke' when I don't know the mass fraction of
> > > whatever else is going to be included in the calculation by FDS? Or I
> > > could use mass fluxes, but how to know what the velocity? I am
> > > probably making this harder than it has to be.
>
> > > Any suggestions?- Hide quoted text -
>
> > > &SURF ID='hotsmoke',RGB=255,255,0,TMP_FRONT=600.0,VEL=-6.7056,MASS_FLUX
> > > (1)=0.0001 /
>
> > > My thinking is that this should set a 600C vent blowing a mixture of
> > > smoke (at mass flux of 0.0001) at 6.7056 m/s and fill in the rest with
> > > background species (I think a mix of air (mass frac 0.992) and water
> > > vapor (mass frac 0.008)).
>
> > > The manual, though, indicates that velocity and mass flux cannot be
> > > set together because the program calculates the velocity from the mass
> > > flux. So I guess I am looking for suggestions on how to do this. I
> > > could use velocity along with mass fractions, but how to calculate the
> > > mass fraction of 'Smoke' when I don't know the mass fraction of
> > > whatever else is going to be included in the calculation by FDS? Or I
> > > could use mass fluxes, but how to know what the velocity? I am
> > > probably making this harder than it has to be.
>
>
> - Show quoted text -
Stephen Olenick
Apr 24, 2009, 1:59:26 PM4/24/09
to FDS and Smokeview Discussions
Ok. Makes sense. So here is what I did (using fds 5.3.0 svn 3193,
although also did it with one of the 5.2.1 modules with similar
results):
&SPEC ID='Smoke',MW=28.8,MASS_EXTINCTION_COEFFICIENT=8700. /
&VENT
XB=25.171,25.171,9.81701,11.1886,15.6975,16.8405,RGB=255,0,0,IOR=-1,SURF_ID='hotsmoke' /
area 1.5677 m2
&SURF ID='hotsmoke',RGB=255,255,0,TMP_FRONT=500.0,MASS_FLUX(1)
=0.000876,MASS_FLUX(0)=4.650924 /
By doing this, I have created a vent blowing smoke at a mass flux of
0.000876 kg/s and background species at 4.650924 kg/s. All fine, and
what I plan on doing is adjusting the mass flux of the background
species as necessary to match whatever overall velocity I want.
Again, all fine. To complicate things, though, I put the following
lines in my input file to check what the vent is doing:
&DEVC
ID='input2_smoke',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='MASS
FLUX X',STATISTICS='AREA INTEGRAL',SPEC_ID='Smoke'/
&DEVC
ID='input2_mass',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='MASS
FLOW -' /
&DEVC
ID='input2_volume',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='VOLUME
FLOW -' /
&DEVC
ID='input2_heat',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='HEAT
FLOW -' /
&DEVC ID='input2_vel',XYZ=25.1,10.5028,16.269,QUANTITY='VELOCITY' /
&DEVC ID='input2_uvel',XYZ=25.1,10.5028,16.269,QUANTITY='U-
VELOCITY' /
&DEVC ID='input2_vvel',XYZ=25.1,10.5028,16.269,QUANTITY='V-
VELOCITY' /
&DEVC ID='input2_wvel',XYZ=25.1,10.5028,16.269,QUANTITY='W-
VELOCITY' /
After the flow has started up and stabilized, I get the following
output:
kg/s/m2 kg/s m3/s kW m/s m/s m/s m/s
input2_smoke input2_mass input2_volume input2_heat input2_vel
input2_uvel input2_vvel input2_wvel
-0.0006756 7.1811069 15.6978490 3548.5069000 9.4148583 -9.3886389
0.1028782 0.2237604
There are a few things I don't understand about this output. First,
the mass flux of smoke is 0.0006756 kg/s/m2 * 1.5677 m2 = 0.001059
which is higher than what I specified. Second, the total mass flow is
more than the smoke and background species input to flow out of the
vent. Potentially from whatever was already in the cells. Of further
concern is the volume flow is 15.69785 m3/s which is a velocity of
10.01 m/s but the velocity from the prob there indicates the velocity
is 9.41 m/s.
So I guess I have 2 questions. First, is my strategy for input make
sense? Secondly, can you explain the output in terms of the input?
Perhaps I am misunderstanding, otherwise it may be a bug in which case
I'll submit to the issue tracker. Thanks for all the help, the
discussions are quite useful.
although also did it with one of the 5.2.1 modules with similar
results):
&SPEC ID='Smoke',MW=28.8,MASS_EXTINCTION_COEFFICIENT=8700. /
&VENT
area 1.5677 m2
&SURF ID='hotsmoke',RGB=255,255,0,TMP_FRONT=500.0,MASS_FLUX(1)
=0.000876,MASS_FLUX(0)=4.650924 /
By doing this, I have created a vent blowing smoke at a mass flux of
0.000876 kg/s and background species at 4.650924 kg/s. All fine, and
what I plan on doing is adjusting the mass flux of the background
species as necessary to match whatever overall velocity I want.
Again, all fine. To complicate things, though, I put the following
lines in my input file to check what the vent is doing:
&DEVC
ID='input2_smoke',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='MASS
FLUX X',STATISTICS='AREA INTEGRAL',SPEC_ID='Smoke'/
&DEVC
ID='input2_mass',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='MASS
FLOW -' /
&DEVC
ID='input2_volume',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='VOLUME
FLOW -' /
&DEVC
ID='input2_heat',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='HEAT
FLOW -' /
&DEVC ID='input2_vel',XYZ=25.1,10.5028,16.269,QUANTITY='VELOCITY' /
&DEVC ID='input2_uvel',XYZ=25.1,10.5028,16.269,QUANTITY='U-
VELOCITY' /
&DEVC ID='input2_vvel',XYZ=25.1,10.5028,16.269,QUANTITY='V-
VELOCITY' /
&DEVC ID='input2_wvel',XYZ=25.1,10.5028,16.269,QUANTITY='W-
VELOCITY' /
After the flow has started up and stabilized, I get the following
output:
kg/s/m2 kg/s m3/s kW m/s m/s m/s m/s
input2_smoke input2_mass input2_volume input2_heat input2_vel
input2_uvel input2_vvel input2_wvel
-0.0006756 7.1811069 15.6978490 3548.5069000 9.4148583 -9.3886389
0.1028782 0.2237604
There are a few things I don't understand about this output. First,
the mass flux of smoke is 0.0006756 kg/s/m2 * 1.5677 m2 = 0.001059
which is higher than what I specified. Second, the total mass flow is
more than the smoke and background species input to flow out of the
vent. Potentially from whatever was already in the cells. Of further
concern is the volume flow is 15.69785 m3/s which is a velocity of
10.01 m/s but the velocity from the prob there indicates the velocity
is 9.41 m/s.
So I guess I have 2 questions. First, is my strategy for input make
sense? Secondly, can you explain the output in terms of the input?
Perhaps I am misunderstanding, otherwise it may be a bug in which case
I'll submit to the issue tracker. Thanks for all the help, the
discussions are quite useful.
Kevin
Apr 24, 2009, 2:34:29 PM4/24/09
to FDS and Smokeview Discussions
'MASS FLUX X' is usually used in the gas phase only, not against a
surface. I'd have to check what it's actually measuring.
In the meantime, seal your compartment, run the case with
MASS_FILE=.TRUE. on the DUMP line. This will produce a spreadsheet of
the total mass of the various species. Plot it and calculate the slope
of the line. That's the true mass flux for that species.
On Apr 24, 1:59 pm, Stephen Olenick <Stephen.Olen...@gmail.com> wrote:
> Ok. Makes sense. So here is what I did (using fds 5.3.0 svn 3193,
> although also did it with one of the 5.2.1 modules with similar
> results):
>
> &SPEC ID='Smoke',MW=28.8,MASS_EXTINCTION_COEFFICIENT=8700. /
>
> &VENT
> FLOW -' /
> &DEVC
> ID='input2_volume',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='VOLUME
> FLOW -' /
> &DEVC
> ID='input2_heat',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='HEAT
> > > - Show quoted text -- Hide quoted text -
surface. I'd have to check what it's actually measuring.
In the meantime, seal your compartment, run the case with
MASS_FILE=.TRUE. on the DUMP line. This will produce a spreadsheet of
the total mass of the various species. Plot it and calculate the slope
of the line. That's the true mass flux for that species.
On Apr 24, 1:59 pm, Stephen Olenick <Stephen.Olen...@gmail.com> wrote:
> Ok. Makes sense. So here is what I did (using fds 5.3.0 svn 3193,
> although also did it with one of the 5.2.1 modules with similar
> results):
>
> &SPEC ID='Smoke',MW=28.8,MASS_EXTINCTION_COEFFICIENT=8700. /
>
> &VENT
> XB=25.171,25.171,9.81701,11.1886,15.6975,16.8405,RGB=255,0,0,IOR=-1,SURF_ID='hotsmoke' /
> area 1.5677 m2
> area 1.5677 m2
> &SURF ID='hotsmoke',RGB=255,255,0,TMP_FRONT=500.0,MASS_FLUX(1)
> =0.000876,MASS_FLUX(0)=4.650924 /
>
> By doing this, I have created a vent blowing smoke at a mass flux of
> 0.000876 kg/s and background species at 4.650924 kg/s. All fine, and
> what I plan on doing is adjusting the mass flux of the background
> species as necessary to match whatever overall velocity I want.
> Again, all fine. To complicate things, though, I put the following
> lines in my input file to check what the vent is doing:
>
> &DEVC
> ID='input2_smoke',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='MASS
> =0.000876,MASS_FLUX(0)=4.650924 /
>
> By doing this, I have created a vent blowing smoke at a mass flux of
> 0.000876 kg/s and background species at 4.650924 kg/s. All fine, and
> what I plan on doing is adjusting the mass flux of the background
> species as necessary to match whatever overall velocity I want.
> Again, all fine. To complicate things, though, I put the following
> lines in my input file to check what the vent is doing:
>
> &DEVC
> FLUX X',STATISTICS='AREA INTEGRAL',SPEC_ID='Smoke'/
> &DEVC
> ID='input2_mass',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='MASS
> &DEVC
> FLOW -' /
> &DEVC
> ID='input2_volume',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='VOLUME
> FLOW -' /
> &DEVC
> ID='input2_heat',XB=25.1,25.1,9.81701,11.1886,15.6975,16.8405,QUANTITY='HEAT
Stephen Olenick
May 5, 2009, 9:13:12 AM5/5/09
to FDS and Smokeview Discussions
Still have an issue - output is not what was input. I have posted to
the issue tracker. Thanks.
the issue tracker. Thanks.
Reply all
Reply to author
Forward
0 new messages