This research deals with moisture safety concerns of cross-laminated timber (CLT). In case of water contact under a CLT wall panel edge, the panels will absorb water along both the transverse and longitudinal grain direction of the timber boards which CLT is comprised of. Due to the anisotropic nature of wood, the moisture transport characteristics differ depending on the cell direction. Water uptake is more intense along the longitudinal grain direction which greatly increases the moisture risks of vertical water uptake. What’s more, often there are moisture trapping conditions and moisture dry-out will only be possible through the side faces of CLT panels (if not covered). My experience from numerous construction sites reaffirm the criticality of end-grain wetting of CLT.
Researchers suggest erecting CLT buildings dry, but this is hard to achieve. Even brief rain can raise moisture content (MC) in the longitudinally oriented timber boards (with regard to water uptake) above recommended limits. However, localized high MC may not pose immediate risks. This discrepancy: excessive MC without negative consequences, creates contention at construction sites, hindering efficient construction process.
I wanted to solve this problem by providing practical MC limits specifically for the end-grain wetting situation. For that I needed a reliable hygrothermal simulation model, which turned out to be more difficult to produce than initially thought. The vast majority of previous CLT moisture simulations are one-dimensional and some which are two-dimensional ignore the anisotropic properties of wood. However, for end-grain wetting, it’s important to consider moisture transport in different directions within the same wood layer (e.g., water uptake along the grain and simultaneously water vapour transport perpendicular to the grain in the same layer).
Therefore, I composed the hygrothermal simulation model (in IBK Delphin), the necessary anisotropic material files (which were not provided in the material database as of 2024 due to the anisotropic model still being in development) and validated the results against three different experiments achieving a very good fit. Variations in material properties required the use of multiple material definitions and moisture storage and liquid conductivity significantly impacted the results.
I then employed the validated model to analyse which MC distribution in the CLT wall panel bottom area is still permissible in terms of mould growth and based on this, formulated a novel two-step MC performance criterion: MC ≤ 16% at 30 mm and MC ≤ 25% at 10 mm from the end grain surface to prevent mould growth. The outcome is important for me (the first author) personally, because I have a background in prefabricated timber building industry (and was part of the team which built the first CLT building in Estonia twelve years ago, thus having a fondness for CLT) and I always look for practical solutions which are backed by research. My supervisor, professor Kalamees has greatly helped me in connecting my practical experience with scientific research methods.
Keywords: hygrothermal performance, anisotropic material properties, moisture management
Authors
Kristo Kalbe
Nearly Zero Energy Buildings Research Group, Tallinn University of Technology, Estonia
Targo Kalamees
Nearly Zero Energy Buildings Research Group, Tallinn University of Technology, Estonia
Login to download the PDF