WUFI

Hello,
I have a question about the Thermal Conductivity, moisture-dependent data.

In your materials database your data for maximum value of Thermal Conductivity for glass wool is 0.6 W/(mK), which is logical considering that in glass insulation there is about 95% of air which is replaced by water and this has a value of 0.6 W/(mK).

However, for wood fibre you state a value of 0.12 which I find strange considering that even there is about 85% air which water will eventually fully replace, can you please explain this to me. In my opinion, if the maximum for Normalized Water Content is 1 should also be Thermal Conductivity for Wood fiber also 0.6 W/(mK).

With glass wool, cellulose etc it is maximum Thermal Conductivity 0.6 W/(mK), but I don't know why it is different with wood fibre, it doesn't make sense to me.

Thanks

Hi K-A-R-E-L,

the maximum value of the thermal conductivity for the glass wool is indeed the thermal conductivity of water, for exactly the reason you state.

There are two ways in WUFI to enter a moisture-dependent thermal conductivity: entering a detailed tabulation of the conductivities, or entering the "moisture-dependent thermal conductivity supplement", a number which states by how many percent the thermal conductivity increases for each mass-percent increase of moisture content.

The curve for the glass wool is tabulated. This curve was originally created for the "Mineral Wool (heat cond.: 0,04 W/mK)", one of the very first materials in WUFI's database. For this material we had a few measurements of the moisture-dependent thermal conductivity, but only for low water contents. We derived a polynomial curve whose initial slope agreed with those measurements and which ended at 0.6 W/mK (the value for water). This means that the curve is realistic at the beginning and at the end, but the middle part it is not much more than a guess.

The curves for many other mineral wool or glass wool materials have been derived by adjusting this "grandfather curve". The glass wool you are looking at is such a case.

.
The other option, entering the "supplement" number, results in a strictly linear increase of the thermal conductivity. This should be realistic at the beginning of the curve, but most likely not in the later parts of the curve.

In other words: If the "supplement" is used, you should trust only the initial part of the curve. You can use this material for hygrothermal simulations where this material is exposed to "normal" or "slightly elevated" moisture contents. Do not use this material if you intend a detailed simulation of the heat flows in a material which is filled up with condensate, for example. For such a demanding simulation, you would need to determine an appropriate and expensive set of material data tailored to the specific case.

Kind regards,
Thomas

Hello Thomas,
Thank you for the explanation, but you should correct the end of the curve (maximum) for the wood fibre anyway, similar to the cellulose, so it doesn't look weird.

I have also one thesis on this subject, I can possibly provide it. Admittedly wood fibers have not been measured, but it may be useful for other insulation.

Kind regards,
Karel