An introduction to fire simulation with FDS and Smokeview: new and improved release!

[Emanuele Gissi]

I am really proud to announce the new a vastly improved release of the
manual *An introduction to Fire Simulation with FDS and Smokeview*. As
usual, you can freely download it from: http://www.corbezzolo.org

The main goal of this manual is to introduce the student to the
complex world of fire simulation with NIST Fire Dynamic Simulator and
complements the official documentation.

Thanks to user feedback, there are many improvements: new sections,
new figures, new examples, hundreds of bugs fixed, converted to
LaTeX...

After vacations I will publish the LaTeX sources.

Thanks to everybody,
Emanuele Gissi.

[Chris]

Very nice. Thanks.
Just one thing of the parking-input-file: Your grid-cells are of the
order of 0.265 x 0.27 x 0.24 m, which is not an really nice hexahedron
whith exactly the same edge length! Take care, your geometry of your
OBSTs is adaptad to that grid!
Cheers

Chris

[Emanuele Gissi]

That's a feature, not a bug ;-) Cells do not need to be perfect cubes.
Grid adaption does not seem a problem to me in that case.
What do you suggest?

Emanuele.

[Chris]

Hi Emanuele,

It is not a critic, but for my calculations I always use perfect
cubes. And as I wrote you should adapt your OBST at your grid: If you
have a 0.25 grid your car should not be at 0.2 at the height. As it is
a guide it should be precise, that's my opinion. But I really
appreciate your work, it has some nice features. Thanks for your time
for that document!

Cheers
Chris

> > > Emanuele Gissi.- Zitierten Text ausblenden -
>
> - Zitierten Text anzeigen -

[Emanuele Gissi]

Hi Chris,

In general I prefer to use real dimensions for the geometry, as at
start I am not always sure of the grid size to use. My grid size comes
from a grid sensitivity analysis.
I am not sure that perfect cubes would improve results dramatically.
But I am ready to change my mind if it does!

Thank you for your comments, I sincerely appreciate feedback.

Best regards,
Emanuele.

Do you think that perfect cubes will improve the results?

> > - Zitierten Text anzeigen -- Hide quoted text -
>
> - Show quoted text -

[Guillermo Rein]

Congratulations Dr Gissi. I celebrate the released of this document
for many reasons, but specially for two. First that it is an open and
transparent document on how to run FDS from the point of view of an
independent and non-commercially biased advanced used. And second, and
more importantly, because it is written by an active fire-fighter
whose team's health and life are put at risk by fire as a professional
duty.

I have many questions, but I do not have the time now (it's Sunday!)
and I shall not flood the discussion here. But I was expecting more on
the limitations of FDS stemming from your personal experience on the
differences between what FDS simulations could says and what fire-
fighters face in real fire emergencies. The current version on
limitations seems mainly theoretical, more based on known
approximations of the mathematical engine and shortcoming of the
physical mechanisms.

Best Regards, and congratulations again.
G.

[Chris]

Hi Emanuele

Your grid should stick with your geometry. If you use real cubes this
has the advanteage you have less problems to stick your geometry to
your grid. And as I know there is a slight advantage for the
calculations as well (but for this question the developpers can answer
better). I would avoid grid-sizes of 0.26777 (no exact number). Let
say if you want a 0.25 grid than you should just have a MESH of
X*0.25, Y*0.25, Z*0.25 in each direction. That is what I do.

What I like is Figure 6.1 Combustion is not Pyrolysis. It is a good
describtion of a common question!

Kind regards
Chris

> > - Show quoted text -- Zitierten Text ausblenden -

[Emanuele Gissi]

Thank you for the feedback Guillermo,

the part on the limitations of the model is copy-pasted and adapted
from the official documentation. I wish I had the time to develop
that.

I focused my work on the part of the official doc that are more
lacking (from my point of view).

For what regards the other part of your email, smoke is the most
important problem we face in real fires, especially in SAR operations.
And FDS overprediction of smoke is on the safe side for us.

That said, we are setting up a community repository where every
advanced user can give its contribution (and be credited for that) on
the manual. The result will be much more valuable that the sum of all
the contributions: 2 + 2 = 5...

Best regards,
Emanuele Gissi

> > Emanuele Gissi.- Hide quoted text -