Thermal Material Properties: density as function of temperature

[gforney]

Application Version: FDS 6

In the development of FDS6 is it possible to implement the temperature-dependence (see
chapt. 8.3.2 User's Guide) also for DENSITY to describe solid materials? 

CONDUCTIVITY and SPECIFIC_HEAT could define as a function of temperature since a long
time. Why does DENSITY cannot define until now? Is there a reason?

Gregor

Original issue reported on code.google.com by gregor.jaeger on 2012-04-14 12:31:37

[gforney]

The reason for a constant density is that if the density were to change with temperature,
we have to regrid the solid phase mesh for the finite difference calculation of heat
conduction. This is more complicated than changing k or c, but it is possible. I'll
ask Simo et al. at VTT to address this.

Original issue reported on code.google.com by mcgratta on 2012-04-14 13:41:32

[gforney]

Kevin mentioned one reason. Another reason may be that I don't really understand what
temperature-dependent density means. I know this concept is sometimes used, but what
does it mean? 

Mass should be conserved. So, a change in density means 
1) material is removed from a control volume to gas phase or brought into a control
volume from gas phase by chemical reactions,
2) the volume of the sample is changed by some physical process.

If both options can be calculated as reactions, as we do now in FDS5, why should we
need the density change by temperature? Do you have some practical example that you
cannot calculate with the current model?

(Honestly, I don't reject your idea, I just want to understand it first)

Original issue reported on code.google.com by shostikk on 2012-04-16 21:06:05

[gforney]

The European Standard EN 1992-1-2 "Design of concrete structures - Part 1-2 General
Rules - Structual fire design" applies to the design of buildings and civil engineering
works in concrete. For concrete with siliceous and calcareous aggregates the specific
heat c_p(theta) of dry concrete may be determined from the siliceous and calcareous
aggregates, also the thermal conductivity. The variation of density with temperature
is influenced by water loss and is defined in chapter 3.3.2 (3). 

Some people mean that you must take into account the variation of density if you calculate
the temperature profiles for slabs, beams and columns. It would be interesting how
far are the results from each other if the variation of density is not considered.

Original issue reported on code.google.com by gregor.jaeger on 2012-04-23 20:46:44

[gforney]

You can add water to your solid if you want. Define your concrete to be a mixture of
cement, water, rocks, etc. Create a reaction for water so that it boils off at 100
C.

Original issue reported on code.google.com by mcgratta on 2012-04-23 20:58:13

[gforney]

In the European Standard the variation of density with temperature is defined as follows

rho(theta) = rho(20°C) for 20°C ≤ theta ≤ 115°C
rho(theta) = rho(20°C)⋅(1 - 0,02(theta - 115)/85) for 115°C < theta ≤ 200°C
rho(theta) = rho(20°C)⋅(0,98 - 0,03(theta - 200)/200) for 200°C < theta ≤ 400°C
rho(theta) = rho(20°C)⋅(0,95 - 0,07(theta - 400)/800) for 400°C < theta ≤ 1200°C

I'll check the variation with the reaction for water.

Original issue reported on code.google.com by gregor.jaeger on 2012-04-23 21:05:21

[gforney]

The decrease in density is due to water and other substances "boiling" off. We prefer
that these reactions be included rather than programming a change in density. This
is really not a change in the density of the little rocks, but rather the boiling off
of additives like water.

Original issue reported on code.google.com by mcgratta on 2012-04-23 21:09:37

[gforney]

I close this. Decision was: density cannot be explicit function of temperature. Density
variations should be implemented as reactions.

Original issue reported on code.google.com by shostikk on 2012-11-05 07:31:13