H2O ambient levels

[GoogleCodeExporter]

Using CFASTv6.2 installed from this site.

Run any of the open door NIST_NRC cases.  For example test 3.  At the start of 
the simulation both the lower and the upper layer start at 1.4 % H2O.  Which 
seems reasonable as it is 30 C with 34 % humidity.  With the open door one 
would expect the lower layer to not change much from this value.  If anything 
one would expect maybe a slight increase in H2O due to some mixing of the upper 
layer (which should increase due to combustion products) into the lower from 
the door flow.  Instead what is seen is a continuous decrease of H2O from 1.4 % 
to 0 %.  This suggests that the ambient values of H2O are not being initialized 
correctly and the the compartment inflow has 0 % H2O rather than the expected 
34 % humidity.  

If I do a simple hand calc using the steady state layer height and inflow 
assuming I have an initial tracer mass fraction of 1 and the inflowing air is 
0, I get that I would reach near zero tracer in 500 s which matches closely the 
behavior seen in the results.

Original issue reported on code.google.com by drjfloyd on 4 Dec 2013 at 2:26

[GoogleCodeExporter]

Original comment by cfastdev@gmail.com on 14 Jan 2014 at 2:13

• Changed state: Accepted
• Added labels: ****
• Removed labels: ****

[GoogleCodeExporter]

You're correct, the water content was never set for the exterior ... this goes 
back at least to CFAST 3.1.7 and likely longer.  This should be fixed in r1610 
for the upcoming version of the model.

Original comment by cfastdev@gmail.com on 30 Dec 2014 at 3:15

• Changed state: Fixed
• Added labels: ****
• Removed labels: ****