Rendered rainscreen walls - cavity ventilation, ventilation drying and moisture-induced cladding deformation

Sammanfattning: Over the past five years, moisture damages in external walls of wood frame structure and rendering on insulation as a cladding has been a widely debated subject which has received a lot of attention. In the light of this, the interest in an alternative construction method where rendering is applied on boards instead of insulation has increased. In this method, carrier boards are mounted on battens so that an air cavity is formed between the cladding and the backup wall structure. Provided that this cavity is open at the top and bottom, wind and thermal forces will create air circulation which means that moisture in the wall can be convectively removed. Under real conditions, ventilation rates in wall cavities are affected by many factors and there is a lack of knowledge regarding both its magnitude and significance for the drying process. Consequently, uncertainties arise during the analysis and moisture safety design of ventilated external walls. The main objective of this study was to increase knowledge of the significance of the cavity design for ventilation rates and convective moisture transport in a cavity behind rendering on a board. Special attention has been paid to the investigation of differences depending on two types of cladding support systems that are often used in practice – vertical wooden battens and perforated horizontal metal battens. Another objective of the study was to investigate how a cladding made up of rendering and boards is deformed on exposure to moisture changes. Understanding of the deformation behaviour is fundamental in assessing the risk of the occurrence of cracks and other defects that affect the appearance and durability of the rendered façade. In the theoretical part of the study, applicable equations for the modelling of air flow, heat balance and moisture balance in ventilated cavities were identified. The experimental part comprised an investigation of some important fluid mechanical parameters in a full scale setup in the laboratory, as well as field measurements of air velocity, temperature and relative humidity in experimental walls on a test house in Lund. The walls had four 2.15 m tall and 25 mm deep cavities, and the factor that was varied between the cavities was the design of the cladding support system. From the field measurements, it was possible to estimate that the average ventilation rate during the period October to February was in the interval 230-310 h-1 for the cavity with vertical wooden battens. With perforated metal battens, the ventilation rate was 60-70% lower. Relations between average ventilation rates in the different cavities were roughly in agreement with predictions based on modelled flow characteristics. The study sets out a number of conclusions regarding the way in which wind and thermal forces influenced ventilation rates. Calculations of average ventilation rates in the cavities of the experimental walls were performed by using tabular climate data for Lund and a simple driving force model. In a comparison of the calculation results and the experimental determination the deviation was less than 20% which indicated that the driving force model was reasonable. With the support of this result, calculations of ventilation rates were made for a number of notional variants of the actual experimental walls, such as different facade colour and different cavity depth. By applying the knowledge collected, drying rates in ventilated cavities of different designs were calculated during three phases of a drying process. In order to demonstrate the practical significance of the calculation results, drying times for a wet sheathing of gypsum, mounted in the direct vicinity of a cavity, were compared. An example of results from these studies is that the drying time is greatly prolonged if the depth of the cavity is reduced from 10 to 5 mm, but will be only marginally longer if the depth is reduced from 40 to 25 mm. In practical moisture safety design of ventilated external walls, different types of simulation programs are often used in which the user defines the ventilation rates. The current practice is that the ventilation rate is assumed to have a fixed value, with limited regard to the cavity geometry and the actual climatic conditions. On the basis of the models used in this study, an improved method is suggested for the estimation of realistic ventilation data for use in simulations. In the deformation study, the material characteristics of rendering and board material which are used in two commercial system solutions were determined. Strips of board material with and without render were then exposed to repeated wetting and drying in an experimental setup, and the free deformations in-plane and out-of-plane were continuously measured. The measurement results were compared with calculations based on a simple mechanical model. Because of limitations in the model and the fact that the influence of gravity had not been eliminated in the experimental setup, measurements and calculations exhibited large differences.