Soot Yield

[DDH]

Hi

The issue of soot yield in FDS has come up time and again. It's an
input which can vary depending on the radiant heat, oxygen
concentration and ventilation conditions. From what I've seen, the
value applied for typical fires in offices, residential, shopping
centres and even residential is generally in the region of 5-10%.

In my view, soot yield must be considered in the context of the
building being designed and judged on whether it provides results
expected from a real fire. Furthermore, where sprinklers have not
been modelled in FDS, they must be taken into account when selecting
input parameters such as soot yield.

So for instance, 7% soot yield seems reasonable and adequately
conservative for a residential occupancies (See BRANZ Study for
residential occupancies). This takes into account the potential fire
load e.g. a living room which could contain several pieces of
upholstered furniture, plastics including TV, stereo systems. In
addition to this, applying a conservative soot yield is reasonable in
residential based on the risk associated with this type of occupancy
i.e. most deaths caused by fire occur in residential.

So if designing for a sprinklered office, what is an appropriate soot
yield? Again, 5% is generally used although I have recently been
asked to model 10% because the Fire Brigade 'didn't like' the extract
rates. This has massive implications where smoke extract is required
either on each floor or at the top of an atrium. One issue I am
seeing is that conditions on the fire floor in terms of visibility
become untenable very quickly. Even with 5% soot yield, it is
virtually impossible to get a postive ASET/RSET result. That's before
the recommended safety factor of 1.5 is applied. In my view, the soot
yield results in conditions which are not a realistic representation
of those expected in a sprinklered building.

Does anybody have any other references or experimental data on soot
yield, particularly for offices? I have seen one reference from the
Board of Housing, Building and Planning in Sweden which recommends 3%
for offices. NIST have carried out real fire test and equivalent FDS
modelling on offices (See Cork County Admin Fire Test). What did they
use?

Look forward to hearing from you.

Thanks

[Franck Didieux @ LNE]

Hi !

I agree with the difficulty to state a soot yield that is conservative
without being to much conservative, and the direct or indirect links
that exist between this input data and the output data that are
compared to the acceptation threshold, for a given performance
criterion.

There is a reference document in Switzerland, which states that the
Smoke Production is found to be between 0.01 g/g and 0.2 g/g.

It also states that it is admitted, for simulations, to use a 4% (0.04
g/g) yield, and link to another document called the SICC BT 101-04
directive (I don't have this document, but it is related to smoke
extractors engineering).

The field of application is not clearly stated (residential,
offices ?), but seems to cover offices since it is stated that FSE may
be used wherever there is no mandatory requirement in the regulation,
rather recommendations. It also lists some of the concerned areas such
as smoke control.

Hope this help ^^