integrated Radiant Intensity

[kam]

Hi,
I am trying to get the net radiant flux on an ambient target at 1.5 m.
For that I am getting the integrated RADIANT_INTENSITY from FDS and
subtract from it the reflected part of radiation ( (1-ε) x
RADIANT_INTENSITY ) and the radiation emitted by the target ( ε x σ x
Tambient^4 ).
As I understood from the description in the user guide that
RADIANT_INTENSITY is the integrated intensity on a point in the 4π
directions per unit area (kW/m2). But something is confusing, as in
the user guide also there is a similar energy equation for the
thermocouple temperature and the integrated radiant intensity is used
there but it is divided by (4).
I would like to know if I am using the integrated radiant intenity
properly.

Many Thanks,

Kam

[dr_jfloyd]

The integrated intensity will sum the flux from all directions,
whereas, an actual target will only see those directions with a normal
component towards the target's surface. One way to what you wish would
be to place a small obstacle at the desired location and then on the
surface put a DEVC to measure the flux.

There has been discussion amongst the developers about adding a gas-
phase DEVC that would measure the radiant flux for a desired
orientation without having to place an obstacle.

[kam]

Thank you for your reply,

It is important for me to have this radiant flux at every point or
second point of the grid at 1.5 m level (I am using SLCF and
fds2ascii). This is because I am intending to ingegrate the results
with an evacuation model to calculate the resultant fatalities in a
fire incident. In using DEVC it is quite possible to place as many as
I want and to have it in an appropriate file.
Is there any other way to do it?
I am assuming that the flux is radiated on a sphere with body
temperature (human head) at 1.5 m, so it is ok to assume it from all
directions. Is the way I described in the previous message to obtain
the NET radiant flux right?
And why divivding by "4" in the energy equation of Thermocouple
Temperature?

Many Thanks in advance.

Kam

[Kevin]

We've thought about this often in the past few years. However, it is
difficult because the radiation solver tracks thermal radiation in
about 100 different directions. Thus, in every gas phase cell, you'd
need to save 100 real (8 bytes) numbers. Too much memory. So the
radiation solver does not save gas phase directional fluxes for post-
processing. It only saves information at boundaries. That is why
Jason was recommending the DEVC.

We'd like to be able to extract radiation information at discrete
points and orientations without the need for little solid cells. This
is possible. Making the points "move" like a person is more
difficult, and maybe too much for now.

K

[kam]

Thanks again and sorry for these too many questions.

As I understood; If I have been able to extract the radiant flux in
gas phase without the need of little solids, I would use &SLCF to
obtain it at a horizontal plane (z=1.5m) in the whole domain.

But in the absence of this feature now, I would like to know if in my
approach described in my first message, I am using the integrated
RADIANT_INTENSITY available for gas phase properly. In eq 4.16 of the
users guide there is similar use of the integrated radiant intensity
but it is divided by 4. Can you please clarify this issue for me.

Regards
Kam

[Kevin]

The radiation transport equation is complicated, but let me at least
explain where that factor of 4 comes from. Ultimately, it ties in
with 4*pi, the area of the unit sphere, or the "solid angle." The
source term in the radiative transport equation is

I_b = sigma*T^4/pi

U is the integral I over 4*pi solid angles, I is the radiant
intensity, so U is comparable to 4*pi*sigma*T^4/pi or 4*sigma*T^4. So
when you divide U by 4, you get something like sigma*T^4, which is
what I think of (crudely) when I assess radiation.

All this being said, I think we need to either remove
RADIANT_INTENSITY as an ouput, or explain it better. I think many
have misinterpreted it as you have.

K