Condensation and evaporation process in the roof cavity

[WEI LI]

Hi experts,

As for the metal cladding roof system, water may condense on the cold metal surfaces. After that, the ventilation flow through the roof cavity can somehow evaporate these condensation. I want to investigate how fast the ventilation flow to dry out the condensation on the metal surfaces. I am wondering if WUFI can take this phenomance into account?

Thanks for your reply.
Cheers

[Christian Bludau]

Dear Wei Li,

it is possible calculate that in WUFI with some limitations.
I would set up a splitted air layer for that using two 1mm layers for the absorption of the dew water:
eg. like this:

1 mm "air layer 50 mm"
48 mm "air layer 50 mm without additional moisture capacity"
1 mm "air layer 50 mm"

The ventilation source is placed in the middle layer. The two air layers differ in their moisture storage function.

Please note:

• Dew water is not running of in WUFI. It is forming up to the available pore volume, larger quantities are taken out from the calculation. So you should stay in ranges where no dew water can run off (eg. <200 g/m²).
• The air layers in WUFI contain effective values for convection and diffusion (see also material info text). So a constant heat transfer coefficient is included that applies for a stagnant layer of air and does not change with the air flow (expressed by the air exchange source).
• The effective values for the air layers in WUFI are set up for non metallic surfaces. If you have blank metals at the surface you have calculate new values (see WUFI program help topic "air layers" on how to do that). In the database there is one example: "Air layer 10mm; metallic".
• The moisture storage function in WUFI is not temperature dependent. They are measured / created for 20°C. For air layers that will have an influence, especially for short time assessments.

So the results will be probably more a rough estimation than completely close to reality.

Best regards
Christian

[WEI LI]

Dear Chris,

Thanks a lot for the reply.
In my mind, there is another limitation of such a method.
The condensation on the metal surface will be in form of droplets, but we assume it is a layer, or film.
Do you have any idea about the differnece it will bring in terms of evaporation rate?

Thank you!

[Thomas]

Hi Wei Li,

please keep in mind that WUFI has been developed mainly to simulate heat and moisture transport in porous materials. Air layers and metals are not porous materials, however. They can be taken into account, but only in a simplified manner. In particular, air layers are only considered as layers in the building component which act as a certain resistance to the heat and moisture flows going across them. What occurs within the air layers is not modelled in detail. For example, the free or forced air flow within the air layer is not modelled directly, only its effect on the heat and moisture resistance of the air layer as a whole is taken into account. Also, the heat and moisture transfer resistances at the two surfaces of the air layer are not considered separately, they are taken into account as part of the total heat and moisture resistance.

Therefore, WUFI can not simulate the deposition of condensate on the metal surfaces in a detailed way, taking account of all the details, such as the air speed in the cavity, the air-speed-dependent surface transfer resistances and so on. A simplified treatment is possible, as described by Christian. By using an air-change source in the air layer, the amount of moisture is regulated by the properties of that source, and some of the moisture transported into the air layer is deposited as "condensate" on the sorptive layers covering the metal. But apart from that, no detailed simulation of the condensation process itself takes place, and Christian has already described other caveats.

So if you want to take cavity condensation into account as a contributing background process which acts as a kind of temporary moisture reservoir, WUFI can do so in principle and in a strongly simplified manner, but not in detail. If the condensation is your primary interest and you wish to simulate it so accurately that the distinction between full and partial wetting of the surface makes a difference, this is beyond WUFI's scope.

(To answer the question: Yes, if the surface is only partially wetted, the evaporation rate per square meter will also be reduced in proportion to the covered/uncovered area ratio. But if you are aiming at such accuracy that these details need to be taken into account, you would also need to account for the facts that the curved evaporation surface of a droplet is slightly larger than its "footprint" area (slightly modifying the area ratio mentioned above), that the saturation vapor pressure of a curved surface is slighty higher than that of a plane surface (and you certainly did correct for the influence of the atmospheric pressure on the saturation vapour pressure), that in addition to the latent heat of evaporation or condensation the heat of adsorption of the water on the metal surface has to be figured in, that the saturation vapor pressure of the droplets may also be affected by dissolved substances the water may have picked up from the surfaces, that the metal surface may be hydrophobic, retarding the onset of condensation, that the low long-wave emissivity of the metal surfaces is strongly modified by the presence of a (full or partial) water film, possibly changing the temperature conditions, and so on. All of this is beyond WUFI's hygrothermal model.)

Kind reagrds,
Thomas

[WEI LI]

Thanks for the reply. Really appreciate that.

[WEI LI]

Hi Thomas and Christian,

To investigate the amount of condensation of the air near the metal deck, Christian suggests I spit a 1 mm air layer and investigate its water content. However, I checked the tutorial "assessment of condensation in hydrophobic mineral wool", and it suggests that for material without measured moisture storage function, a 1 cm layer is more suitable. In the WUFI material database, obviously, the air (the one with extra moisture capacity) does not have measured moisture storage.
So which one should I use? A 1mm air layer or a 1 cm air layer?

Thank you!