Historiska träbärverk : Strukturmekaniska metoder och visualisering

Licentiate thesis, 2003

In designing and erecting buildings in historical times, creating the loadcarrying structure was perhaps the greatest challenge with which the architect and the builder were faced. We encounter structures in which there was a strive to reach vertical heights in the pillars and walls and to extend the horizontal span of the beams and vaults that were built. Yet the roof trusses, working structures that were often hidden in an attic behind a large assortment of decorated vaults and ceilings, was the part of the building that represented the greatest challenge of all. In discovering these, we encounter impressive structures of highly differing form reflecting a desire to increase the span and to expand the space. The aim of the thesis is to explore and develop methods of facilitating an understanding of the static behaviour of historic load carrying structures of this kind generally and more specifically of roof trusses, endeavouring to provide a sound basis for decisions concerning the maintenance and preservation of such structures. An essential element in the approach described is use of the Finite Element method (FEM) combined with visualisation. To provide an overview of different types of structures and problems to which the methods proposed are applicable, six specific examples are taken up. These highlight the importance of a conceptual understanding of the structural behaviour involved, both before and during the design process or any repair of an existing structure, as well as the advantages that visualisation can provide. Paper 1 examines the roof truss of the Swedish castle of Glimmingehus by use of FEM and visualization in order to determine the probable reasons for certain damages observed in the structure. Highly predictable causes were found and visualizations of their influence on the behaviour of this structure were presented. Paper 2 introduces a new approach, so-called static eigenvalue analysis, for investigating how effectively different truss types are able to carry both dead and wind load. The importance of different structural elements and their placing were assessed and a measure of this in terms of relative deformations was provided. The deformation modes emerging from the method and the comparison of deformation patterns in terms both of shape and of size were found to be intuitive and graspable.