This paper presents a field study on the hygrothermal performance of hydrophobised masonry walls, provided with vapour tight or capillary active interior insulation. As a reference, also nonhydrophobised and non-insulated walls are analysed. To study the hygric performance, apart from traditional relative humidity sensors, in-house made moisture pins are embedded in the walls and are shown to yield valuable information in the high moisture range. Attention is given to the hygric performance of the wooden beam heads, the impact of wind-driven rain on the moisture conditions in the masonry wall, as well as to the impregnation depth of the water repellent agent. For the latter, the field study is supplemented with X-ray and liquid droplet measurements on a small semi-duplicate test wall. During the hydrophobisation process, an increased moisture level was induced. A drying period was needed to again reduce this level. After this period, a hydrophobic treatment showed a positive impact on the wall's hygric performance. The non-hydrophobised wall with a vapour tight interior insulation system showed the highest moisture level, especially deeper in the wall and thus where the wooden beam ends are located.

With the development of more efficient hygrothermal computer models, simulation studies have become increasingly important in the design of building components. To obtain trustworthy results from these studies, accurate hygric properties are required. The existing methods for moisture storage properties, however, are not very well suited to accurately measure moisture retention curves within a compact timeframe. To improve on this front, this paper introduces the steady state centrifuge technique, a common experiment in soil physics, for use on porous building materials. The laboratory centrifuge, used for the validation of this technique, is self-made to limit its cost and account for specific design choices. In the first part of the paper, the design of the laboratory centrifuge is described and all problems encountered during the development are explained and resolved. The two main problems are excessive heat generation by the motor and unwanted evaporation from the sample’s surfaces. The excessive heat generation is solved by extraction of heat both at the source, by using a ventilator, and at the rotor, by adding carefully positioned air extraction holes. The unwanted evaporation is eliminated by incorporating sample holders to shield the sample from the surrounding air. In the second part of the paper, the steady state centrifuge experiment is used to measure the desorption moisture retention curves of a ceramic brick starting from both saturated and capillary moisture content. The results are validated by their similarity to the curves obtained by mercury intrusion porosimetry. Besides providing accurate results, the determination of the full moisture retention curve requires only 1–2 weeks, which is significantly quicker than other common protocols, such as the pressure plates, which take about 2 months. Additionally, the ability to measure the desorption moisture retention curve from capillary moisture content as well as the limited cost of the centrifuge design (€6000) provide major advantages.