Adding a liner to the wall

[Vera-MVD]

Hi everyone.

I have another question. I want to model a liner on the inner surface of my compartment. Say, I have a 10mx10mx10m cubical compartment with 1m thick concrete walls/floor/ceiling. But there is also a 2 cm thick liner on the inside surface of the compartment (on every surface, i.e., walls, floor and ceiling) made of a different material. For the moment, let's say there is no gap between the liner and concrete. The initial T inside the cube (and throughout the simulation) will be higher than outside.

I am trying to model this as a compartment inside compartment. When I input the first compartment (outer cube) I just use width(X) of 10, Depth (Y) of 10, and Height (Z) of 10; and Position X, Y, and Z all 0s. But then when I input the inner compartment (liner) do I just use Position X, Y, and Z as 1m because that is how thick the walls of the first compartment (outer cube) are? I've tried it this way and the Smokeview animation shows extra space between two compartments that should not be there. Plus there is a significant P difference between outer compartment and the inner, that also should not be there, T difference yes (T inside the liner is the highest, due to heat transfer some of that is going to the outside surface of the liner and that is what the concrete inner surface of the outer compartment is getting in). The only way I see to not have this extra space is to put the inner compartment (liner) right in the same position (position X, Y, Z all zeros), but that does not make much physical sense.

Question: is this even possible to add a liner to the inner surface of a compartment?

As an alternative, I was thinking about using some modified material data for the compartment surfaces to account for the liner, but wanted to ask here first in case any of you have done a similar modeling before or have an idea of how to do this in cfast.

Thanks in advance for any ideas,
Vera

[mcgratta]

This is not possible. I suggest that you simply use the properties of the 2 cm liner, or develop some sort of effective properties.

[drjfloyd]

You could look at the thermal penetration time of the liner and compare with the time range of interest for your cases. Since there is an air gap, if the penetration time is greater than or similar to the time range of interest, then only the liner matters.

[Vera-MVD]

Yes, that's what I am planning to do. Since one of the output I am interested in is the inside T and P profiles (over a days long run) I will replace my compartment walls thermal properties with modified ones representing the (liner-air gap-compartment walls) system to account for a proper heat loss through the compartment walls. Thank you!

[Vera-MVD]

Just wanted to verify something. While working on the homogenizing my compartment walls for my model (steel liner, air gap and the concrete wall) I saw that CEdit 7.7.5 UI for the Compartments has three rows to specify compartment materials for ceiling, wall and floor. If I enter multiple materials in there, like shown in the fist screenshot, does cfast treat it as a multilayer surfaces or somehow homogenize them? For the case shown, I also have an electric heater modeled (second screenshot). The last screenshot shows a plot of an average compartment temperature profile for different compartment surfaces material/s tested. The temperature averaged using layer volumes.

Vera

[mcgratta]

CFAST treats the surfaces as multi-layer -- not homogenous

[Vera-MVD]

CFAST treats the surfaces as multi-layer -- not homogenous

Thank you!