Method Statement
Literature review
We will conduct a comprehensive review of available literature (books, journals, conference papers etc.) relevant to the project, and particularly focussed on the structural design of Gothic cross vaults, their strength and collapse modes. We also hope to use the literature review to aid our understanding of the effect of certain parameters (such as spandrel fill height) and the testing processes. This will not only help us to form our research question, but also to assess whether any previous research can be used to enhance our own, and to potentially compare our own findings with those of other similar studies. Site visits We have visited various examples of both existing and collapsed Gothic-vaulted structures in order to enhance our understanding of these structures and the behaviour and interaction of their components. Observations from these visits will be used in conjunction with literature to assist our research, for example by enabling us to compare the varying heights of spandrel fill used in vaults at different locations. As part of a visit to Holyrood Abbey we also took measurements of one of the aisle vaults, in order to confirm that the dimensions of the quarter-scale model accurately reflect those of the actual vault. Material tests We will produce wallettes using the model construction material, building them in the same bond pattern as the vault. These will be tested under four-line loads in order to obtain the material’s stress-strain relationship and flexural strength. Two orthogonal orientations of the wallette material will be tested to find the properties in both directions. Three wallettes will be tested for each orientation, thus improving the reliability of the results. The wallettes will be simply supported and tested under line loads until failure. The test set-up is shown in the diagram to the right of this text (Theodossopoulos, 2001). Construction of model We will construct a quarter-scale model of one of the quadripartite vaults in the south aisle of Holyrood Abbey. This will be composed of a shell of oblong wooden units bonded with lime-based mortar joints, supported by ribs composed of the same materials. We will use existing wooden units from a previous study by Dr. Theodossopoulos. The vault model will be supported upon four concrete plinths, with one plinth supporting each springing. The plinths will be of sufficient height to allow access beneath the model to remove formwork and position the laser scanner if necessary. The model will be constructed in the Research Workshop of Minto House, in the University of Edinburgh’s School of Architecture. The ribs of the vault will be erected first. Once the ribs are in position, formwork will be assembled in order to support the shell during the construction phase. Once the shell has been constructed, the formwork will be removed. Although we already have the wooden components for the ribs and shell, we will have to construct the formwork ourselves, using the previously erected ribs as a template. Testing of model This will consist of two phases:
For the first phase of testing, the model will be subjected to a constant dead load provided by strategically positioned lead weights, in order to simulate the uniform self-weight of the real structure. The deformation of the model vault as a result of the dead load will be monitored, with varying heights of spandrel fill for each test. It is intended to carry out this part of the experiment so that all deformations are elastic, meaning that the structure will return to its original form when the weights are removed in between tests. When testing to failure, an outward movement will be imposed on the front two abutments, which will simulate the deformation caused by the outward thrust of the upper structure in real Gothic buildings. The abutments will be moved gradually using a turn buckle. We will use a spandrel fill of 40% of the vault height, as in the previous test by Dr. Theodossopoulos. Repairs will be carried out when cracks start to form, probably using FRP. Based on a comparison with Dr. Theodossopoulos’ results, a recommendation will be made as to whether such repairs are effective and should be implemented in actual existing structures. Monitoring of model In both the elastic and failure tests, deformations of the model will be monitored by either one or two laser scanners positioned strategically around the vault in order to maximise the amount of model that can be observed. The laser scanner(s) will be borrowed from AOC Archaeology. |
All photos are the property of Michael Scott or James Sanderson unless otherwise stated.
© 2013 James Sanderson and Michael Scott | All rights reserved.
School of Engineering, College of Science and Engineering, University of Edinburgh
© 2013 James Sanderson and Michael Scott | All rights reserved.
School of Engineering, College of Science and Engineering, University of Edinburgh