Thanks, Eyore. You are right that it is not possible to reliably make
PREDICTIONS of burning rates with FDS, for a variety of reasons that
have been discussed in this forum. However, one of the frustrations
for me is that very few people use a simple technique that people like
Simo Hostikka, Chris Lautenburger, Guillermo Rein, and others, use to
quickly assess the validity of a given set of material properties.
Given that most, if not all, pyrolysis models assume that the heat
conduction is 1-D, normal to the surface, you can easily calculate, in
seconds or a few minutes, the response of your "virtual material" (as
Nick Dembsey calls it) to a given external heat flux. If you have cone
data, you can then compare your model prediction against an
experiment. Sure, it's not a perfect comparison, like the fire in the
cone changes the flux a bit, but at least this is a starting point. At
least you have some confidence that your assumptions about the
properties have some validity, and you also have a better sense of
what kind of error you can expect when you put your material
properties into a full-scale fire calc.
Even though we threw away the old FDS material database, I still get
the impression, based on the initial posting of this thread, that
practicing engineers still think that material properties for real, or
even imaginary, things exist, and it's just a matter of searching. But
the story of Pooh and Piglet searching in vain for the heffalump
should be seen as a metaphor for fire protection engineers searching
in vain for this fictional "database". It's not the data itself, but
rather the measurement techniques to get the data, that we need to
work on. There are as many materials in this world as there are plants
and animals found in that natural history museum. We'll never have a
comprehensive list of material properties, nor should we even try to
compile one. We should just make sure that the measurements are
clearly defined, relatively inexpensive to perform, and easily
transformed into model inputs.