Summary

Fugtforhold i træbeklædninger. Laboratorieforsøg med accelereret ældning

Moisture conditions in timber claddings

Laboratory experiments with accelerated ageing

Denmark has not the same tradition for using timber for building and construction as other Nordic countries. This applies to timber used for loadbearing structures as well as for facade claddings. In the project described here which is part of a major international collaborative research effort, SBI has tested the moisture conditions and durability of some timber facade claddings in three of SBI's test facilities: "The Four Seasons", "The Driving Rain Chamber" and a Climate Simulator.

In "The Four Seasons", a climate simulator, small test specimens (19 × 85 × 361 mm) were tested with focus on the following test variables:

• Wood species/wood types
• Orientation/kerfing
• End face sealing
• Surface treatment
• Surface structure
• Fastening method
• Mounting

The test results from "The Four Seasons" show:

• Heartwood from larch and thuja is clearly the best type of wood of the wood species studied, which included six with heartwood and two with sapwood. They absorbed moisture more slowly, were more dimensionally stable and showed less cracking than other wood species. Douglas and pine/heartwood did not absorb a lot of moisture, but they were not as dimensionally stable. Spruce/heartwood and sapwood absorbed some moisture but was not especially dimensionally stable. Oak absorbed considerable moisture and was not dimensionally stable. Pine/sapwood absorbed unacceptably large amounts of moisture and was not dimensionally stable.
• Investigation of the effect of orientation/kerfing indicates that it is important to orientate the inside face towards the outdoor exposure, but that kerfing on the back can compensate for the test specimens being orientated with the outside face towards the outdoor exposure.
• The effect of end face sealing was examined and measurements indicate that less moisture was absorbed in test specimens with end face sealing consisting of three coats of paint than in test specimens with no end face sealing.
• There is no difference between surface treatment with wood oil and linseed oil on the test specimens of spruce/heartwood, oak and thuja. Surface treated specimens absorbed less moisture than specimens with untreated surfaces.
• For pine/sapwood there was no significant difference between the surface structures: sawn, planed and cutter planed.
• Based on cracking and moisture absorption, it was not possible to ascertain clear differences between the fastening methods: glue, dowels, screws or nails.
• For technical reasons it was not possible to assess if there was a difference between vertically and horizontally mounted test specimens.

In "The Driving Rain Chamber" the protective properties of drip caps on the wood facade were tested with focus on the following test variables:

• Depth
• Slope
• Design of drip notches

The test results from The Driving Rain Chamber show:

• Protection immediately below the drip cap increased with the depth of the drip cap, but the protective effect decreased as the distance below the drip cap increased. Drip caps with a depth of 70 and 95 mm provided better protection than those with a depth of 45 mm.
• Drip caps with slopes of 1:5 and 1:4 provided better protection than drip caps with slopes of 1:10 and 1:3. However, drip caps with a 1:10 slope also provided good protection, but in practice it is cause for concern that the drip cap may be fixed in a way that causes water to run backwards.
• Drip caps without notches provided poorer protection than drip caps with circular or sharp-edged notches.

A facade element was tested in the Climate Simulator with focus on the following test variables:

• Wood species
• Orientation
• Surface treatment
• Board length
• Fastening distance from end wood
• Fastening method
• Board joint
• Drip edge (window)
• Window details

The test results from the Climate Simulator show:

• All four wood species: larch (European), larch (Siberian), thuja (Danish) and thuja (WRC) seem suitable for facade cladding as all of them have few cracks, little cupping and narrow gaps between boards.
• The wood species have almost the same cupping, number of cracks and cracking pattern regardless of the inside/outside orientation. The moisture content in thuja (Danish) with outside orientation is somewhat higher than the moisture content in other wood species.
• After exposure, boards surface treated with Rentolin absorbed less water at a drip test than boards with no surface treatment.
• Long boards have somewhat larger gaps between the boards than short boards. And long boards of thuja (WRC) have somewhat more cracks than short boards.
• The number of cracks decreases with increased fastening distance from the end wood.
• Fastening by means of gun nails causes fewer cracks than fastening nails manually.
• Based on the measurements performed, it is impossible to see whether the boards were joined by different types of joints.
• The moisture content in the drip edge covered with aluminium was decidedly less than the moisture content in the drip edge not covered with aluminium where it was higher than 20 weight-%.
• It was impossible to note any difference whether a 50 × 50 mm lining or a 25 × 50 mm lining was used as window detail. The moisture content below the lining was very high in both cases and water entered the gypsum board behind, also in both cases.

Tilbage til publikationen 'Fugtforhold i træbeklædninger'