sand, chat rock and gravel

[Rupert]

Hello,
what material properties should be used for filling materials such as sand, chat rock or gravel?.
So far I have been using concrete screed, but changed the porosity to 0,3 and the mu-value to 2 and disabled suction and redistribution. Will that be sufficiently accurate?
Thanks, Rupert

[Thomas]

Hello,
what material properties should be used for filling materials such as sand, chat rock or gravel?.
So far I have been using concrete screed, but changed the porosity to 0,3 and the mu-value to 2 and disabled suction and redistribution. Will that be sufficiently accurate?
Thanks, Rupert

Hi Rupert,

for general purposes, that should do.

Of course, you can't simulate situations like liquid water draining through the voids between the grains (a flow mechanism mainly driven by gravity) or a drainage pit slowly filling up with water. But if the filling material stays dry and just needs to be considered insofar as heat and vapor flows are passing through it, and as long as the behavior of the whole assembly does not depend critically on the properties of the filling material, your approach should be fine.

Here is what I would do to modify the screed or some similar material:

German standard DIN 4108 lists 1800 kg/m³ as the bulk density of sand or gravel. So I'd use that number or simply leave the original screed density untouched.

A porosity of 0.3 sounds reasonable for loose fill. Since the porosity only determines the maximum water content of the material and does not affect the other material properties, a rough guess is sufficient. After all, since situations where the fill is flooded to maximum water content cannot be simulated with WUFI (as mentioned above), no calculations should be done for such situations, and no high water contents then occur anyway.

The specific heat capacity can be left unchanged. Even if the new material had a markedly different density and thus a different heat capacity per cubic meter, the heat capacity per kilogram would be virtually the same, as long as it is still a mineral material.

Standard DIN 4108 gives 0.70 W/mK as the dry thermal conductivity of sand and gravel, so I'd adopt that number.

A mu-value of about 2 sounds reasonable for a coarse filling. I found measured values for fine fillings:
glass balls (grain diameter = 1.9 mm): porosity 0.365, mu 3.1
glass balls (d= 0.5 mm): porosity 0.37, mu 3.8
beach sand (d=0.2 mm): porosity 0.36, mu 4.7

The moisture storage function is somewhat difficult to transfer to a different material, but I'd try the following. Assume that the hygric properties of the new material are the same as those of the original material, but the new material (being not solid but loose fill) occupies only 70% of the original space and the hygric capacity of the air in the voids is negligible. Therefore reduce the moisture storage function to 70% of the original values, except the value for free saturation which should be 300 kg/m³ (for the filled voids) plus 70% of the original free saturation. Of course, caring about a value for free saturation is for completeness' sake only, since such high water contents cannot be simulated realistically (see above).

Set the liquid transport coefficients for suction and redistribution to zero, since liquid transport within the grains is quite irrelevant, the properties of liquid transport from grain to grain are unknown, and diffusion will be the dominant moisture transport mechanism anyway.

Regards,
Thomas