many fire cases will need a specific fire science expertise: fire modeling.
For attorneys, obtaining adequate expertise in fire modeling is not a
trivial issue. The reason is that this is a highly specialized technical
area, yet one without clear means of qualifying the individuals. Especially,
it is important to note that professional qualifications such as: fire
investigator, fire protection engineer, fire marshal, etc. etc., do not
imply expertise in fire modeling.
Part of the
problem arises from the fact that it is quite easy to acquire the trappings
of a fire modeler, without actually having the background necessary for
a full understanding of the subject. For a very nominal fee one can obtain
NIST's HAZARD, CFAST, and FPETOOL. The programs are sufficiently user-friendly
that answers may be cranked out by an individual who is not at all aware
of the physics that he is actually representing. When this author was
with NIST, we agonized quite a bit over the release of the first version
of HAZARD: how to avoid putting tools into unqualified hands. The issue
was, unfortunately, never resolved. The attorney in a fire case, however,
has to be able to determine whether the individual is qualified or not.
To this end,
we offer a list of questions which can be helpful in determining the individual's
courses have you taken in heat transfer and combustion?
Computer fire models are exceedingly difficult to adequately comprehend
for a person who has not studied either of these topics on a graduate
level. At least several courses in heat transfer (convection, conduction,
radiation) and at least an introductory course on combustion should
be looked for.
fire models do you have available on your computer?
There are several dozen useful models to choose from. It would be hard
to claim expertise in fire modeling if one had only the models from
NIST in one's toolkit. The person should have at least some models from
abroad (Sweden, Finland, UK, Australia, etc.).
you ever commissioned heat release rate fire tests for use as input
data to fire models?
Today's fire models require that heat release rate data (from small-scale
or large-scale tests) be supplied as part of the input to the problem.
A certain amount of public-domain HRR data exists. This, however, is
not likely to solve very many needs in actual fire cases, since the
data do not refer to the specific products which burned in this particular
fire. Thus, someone who uses fire models in a meaningful way is going
to need to obtain lab test results for HRR on exemplar materials.
HRR tests have you commissioned?
Appropriate tests would include the Cone Calorimeter (small-scale tests),
a large-scale open-burning calorimeter, and a large-scale room calorimeter.
They would not include tests such as UL 94, Steiner Tunnel, or
various Bunsen burner type ignitability tests.
do you represent flame spread in a fire model?
Appropriate answers would include: (a) by conducting a full-scale sectional
mockup on the geometry in question; or (b) by using a fire model which
can represent flame spread.
you use a fire model with flame spread features, what computation is
the model making?
There are only a very limited choice of conditions where today's fire
models can represent flame spread; the details should be understood
by the expert.
you ever use fire models to predict the smoke in fires? Can you explain
the units that are used to represent smoke input data for smoke in fire
Most models use either m2/kg or m2 as the basic
smoke characteristic of the material. The former is known as the specific
extinction area while the latter is known as the smoke production.
is the fire plume an important feature of room fire models?
Because this is the mechanism for moving gas from the lower (cold) to
the upper (hot) layer.
determines the amount of carbon monoxide being produced in a fire?
Of most importance: ventilation (i.e., air supply). Of secondary importance:
chemical nature of the combustible being burned.
you double the height of window opening from a room on fire, will you
double the outflow from it?
No, you will more than double it. The flow rate through an opening is
not linearly proportional to the height.