Soot yield selection

[Cian Davis]

Hi All,
When modelling smoke control systems, visibility is regularly used as an acceptance criteria. When modelling with FDS, the visibility is affected by the selected values for reaction soot yield, mass extinction coefficient and visibility factor.

Both the mass extinction coefficient and visibility factor have reasonably solid basis in literature (8700 m^2/kg and 3, respectively).

While we could get into a debate about whether the above figures are always appropriate, the bigger question for me is selection of the soot yield.

While the default of 0.01 might be appropriate for a well ventilated, clean burning, methane fire for example, it obviously isn't for polyurethane. However, using published yields for polyurethane would like be over-conservative since often, polyurethane doesn't make up a significant portion of combustibles.

I've had a look back through the archives and while this issue has been raised, there has been no suggestions as to how to select a soot yield for a non-experimental fire.

I generally use 0.07 for a "general" fire - what I would class as either an office or general retail. I would say this takes into account good ventilation (a critical factor for soot formation in real-life fires) and a mixed fire load that is unlikely to contain large amounts of heavy soot-producing materials. From reviewing reports of other Fire Engineers who model similar smoke control systems in building, the range is similar (0.05-0.15).

How does anyone else select a soot yield for a "general" fire and is it a finger-in-the-air judgement or have more solid backing?

Regards,
Cian

[S.Desanghere]

Dear Cian,

I suggest you to try different values to make your own sensivity
analysis study.

I don't agree about the fact that visibility distance has "solid
basis". It is a extremy simplified approach to consider it like it is
done in FDS. But a realistic calculation would be very complicated.
Visibility distance is introduced because it can be easily understood
by people who are not experts in fire simulation.

In fact, using the default mixture fraction model, with a given value
of the mixture fraction, the soot mass fraction is directly
proportional to the soot yield you give in the input file. For this
reason, it may be more interesting to study qualitatively the temporal
evolution of the soot mass fraction (or visibility distance or mixture
fraction) that represent "smoke filling" process, rather than focusing
on a particular threshold value.

Cheers
Sylvain

[PeterS]

In fact the soot yield has obviously great impact on the results
relating visibility, optical density etc.
For the simulation of an auditorium in a University I used also
0,07(well-ventilated) as a soot yield, under the circumstances of
wooden seat rows and a bit of mixed combustibles (Jackets etc.). My
fire was only well-ventilated in the beginning, but as I took a look
"only" on the person safety I think this assumption is OK. I "created"
my soot yield by combining the soot yield for cellulose and for
plastics, both for a well-ventilated, medium sized fire.

Values(Table 4.9, Page 81) from a free guide (German Language) of my
Institue/University:
http://www.vfdb.de/download/Leitfaden_Ingenieurmethoden_Juni_2009.pdf

[Chris]

Soot yield is also dependant on fire size.
Chris

> > > Cian- Zitierten Text ausblenden -
>
> - Zitierten Text anzeigen -