Representative data sets of wood-based materials created for moisture control analysis by hygrothermal simulation.

For wooden constructions, questions concerning moisture control design based on heat and moisture transfer are crucial, since wood and wood-based materials are particularly sensitive to moisture. Hygrothermal simulation models are state of the art, especially for the design of moisture-sensitive building components. However, the relevant material properties of specific wood-based products like moisture storage or liquid water transport parameters are often not available. Therefore, within a research project, generic data sets for eight product groups have been created. With the help of hygrothermal simulations, the impact of the generated generic data sets is compared to that of simulations with real measured data sets. The results showed that under normal and uncritical moisture conditions, the generic data sets well represent the behavior of the components and are therefore well suited for design purposes. Peer-review under the responsibility of the organizing committee of the ICMB23.


Background and aim
In wooden constructions, questions concerning moisture control design based on heat and moisture transfer are of particular importance, since wood and wood-based materials are subject to biological decomposition processes and loss of mechanical strength if local humidity levels are too high.Also important changes in humidity resulting in shrinkage or swelling can affect the construction, e.g. by causing delamination or cracks.
Hygrothermal simulation models are state of the art for the design of moisture-sensitive building components.Since only these models consider all relevant influencing parameters, they present the best tools for a reliable and sustainable building envelope design.In addition to the mostly known water vapour diffusion resistance, also moisture storage, free water saturation and liquid water transport parameters are relevant to define the material properties of wood-based products.However, the material properties of specific products are often unavailable and the property range of the products from different manufacturers or production lines may be quite large.This leads to uncertainty concerning hygrothermal evaluations.Therefore, it would be useful to revert to generic data that cover the usual property range of common wood-based products on the market.The aim of an industry funded research project [1] was therefore to create hygrothermal data sets, representing different product groups of wood-based building materials.

Performed measurements
In the laboratory of the Fraunhofer-Institute for Building Physics, the hygrothermal properties of commercially available wood-based panels were determined.It became clear that individual material properties could vary significantly more than expected, even within the same product groups.Apart from usually well-documented parameters, such as density, porosity and thermal conductivity, the data required for hygrothermal simulation generally include specific heat capacity, vapour diffusion resistance, sorption isotherm and moisture dependent liquid diffusivity functions.Detailed material property tables can be found in [1] and in the WUFI ® material database.

Generation of generic data sets and their validation
Within the research project, generic data sets for eight product groups, e.g.Oriented Strand Board (OSB), wood fiberboards and batts, are developed.Hygrothermal calculations are performed to check the extent to which these generic data sets can represent the respective material group.Parametric studies using WUFI ® [2] to calculate various constructions that are rather  Corresponding author.+49 8024 643-229, daniel.zirkelbach@ibp.fraunhofer.de The Author(s).This is an open access article distributed under the terms of the Creative Commons Attribution Licence (CC-BY) 4.0 https://creativecommons.org/licenses/by/4.0,which permits unrestricted use, distribution and reproduction in any medium, provided the original author and source are credited.DOI: 10.14293/ICMB230039 critical in terms of moisture conditions, evaluate the influence of the generated generic data sets versus that of measured data sets.Thereby, in each case, using the generic data set should provide calculation results being somewhat on the safe side.However, during these analyses it became clear that critical limit cases can be both overestimated or underestimated depending on the position of the material within the assembly.
For example, when used in a pitched, ventilated roof, the moisture in the materials results mainly from the high humidity level of the outdoor climate in winter, which means that the water content becomes higher when the sorption isotherm level rises.In flat roofs, on the other hand, the moisture migrates through the vapour retarder from the interior and is also influenced by the material's vapour permeability.On the other hand, high level of vapour capacity (steep sorption isotherm), which is more critical in the ventilated roof, leads to a smaller increase in relative humidity in the flat roof and is therefore more favorable.Figure 1(a) and 1(b) show the generic data set "AiF Flexible Wood-Fiber Insulation WF" compared to the measured data set "Insulation A" applied to typical pitched roof and flat roof constructions.This shows that the generic data set once produces more favorable and once less favorable results than the measured data set.Obviously, realistic material data sets that produce critical results in all cases do not exist.
However, relevant discrepancies only occur when the hygrothermal performance of a specific construction comes close to the failure criteria anyway.Under normal and uncritical moisture conditions, the generic data sets well represent the behavior of the components.To cover also more extreme cases two data sets were created for product groups with a particularly large variation in material properties.The two data sets represent the range of a material group with properties being either on the upper or the lower end of the spectrum.

Application limits
If the simulation results, obtained with the generic data, approach the failure criteria, the behavior of the envelope assembly should be examined in more detail.The first step would be to vary the material properties within a reasonable range and to rerun the simulation.This will show whether the properties of the considered material layer have a large impact on the calculation results.If this is the case, products specific data are necessary for the design analysis.Another reason for using product specific instead of generic data is the analysis of existing constructions showing signs of damage or degradation, as the characteristics of wood-based materials also depend on the manufacturing process and the employed adhesives.Finally, also new material types or recipes could lead to the need to revise or renew the existing generic data sets.

Conclusions
The advantage of using generic materials data is obvious.There is no need to search for specific product data or determine those prior to doing a moisture control analysis.Only if the simulation results show that the considered construction reaches a critical moisture range, the use of specific material data may become necessary for a more detailed verification.However, in summary, the design of moisture tolerant constructions is clearly facilitated by the new generic data.

Figure 1 .
Figure 1.Calculated course of the relative humidity on the outside of the insulation of a pitched roof (a) and a flat roof (b).