RESEARCH PROJECTS

Brittle and flexible: the structural stabilisation of painted plaster on suspended wooden ceilings

[published as 'Brittle and flexible: the structural stabilisation of painted plaster on suspended wooden ceilings', Zeitschrift für Kunsttechnologie und Konservierung, 14 (2000)]
Heinz Leitner

Summary

The principal aim of this study was the assessment of methods of structural stabilisation of painted plaster on suspended wooden ceilings. This structural problem is well represented by the 18th-century ceiling painting of over 600 m2 of the Jesuit Theatre, Vienna, which exhibits both the typical properties and failures of this type of construction.

As a result of over 15 years of involvement with the conservation of this painting, an empirical method of stabilisation by means of stitching was developed, and many questions were raised concerning the nature of the structural failures and the adequacy of the various methods used to address them.

Therefore, the first part of the present study focuses on providing a basis÷historical examples and materials characterisation÷on which to clarify the original structural techniques of this type of plastered ceiling and its typical failures (Chapters 1-4). This is followed by a review of the previous methods of structural stabilisation (Chapter 5), and an extensive presentation of the principles and techniques of the newly developed stitching method as applied to the Jesuit Theatre in 1997 (Chapter 6). Finally, the principal stabilisation methods÷pinning, grouting and stitching÷were comparatively assessed by means of mechanical testing of analogue replicas (Chapters 7-8).

 


For the context of the historical development of the original techniques, primary sources were examined and examples of painting cited. Primary sources begin with Vitruvius, but in general are rather limited. Examples of plastered ceilings with reed reinforcement are cited from at least the Roman period, and become fairly common from the late 16th century forward. From the point of view of the technique of execution, the system is characterised by a combination of layers with very specific and differing qualities: wooden beams, wooden boards, reeds, wire and nails, rough plaster, and fine plaster. The key point of this technique is the combination of rigid and brittle render layers with the moving and flexible wooden support structure. The differing layers are characterised both by the behaviour of their component materials and by their relation to other layers. Therefore, the study aimed to clarify the techniques of execution and the properties of the components of the system.


Having clarified the structural role that each component plays in the overall system, previous methods of stabilisation were presented and analysed to assess their relation to typical failures and their visible phenomena. The technical problem of suspended wooden ceilings with plaster decoration has two main aspects.

First, how to attach a rigid, brittle and heavy plaster to the wooden support in such a way that the suspended load is distributed within the limits of the tensile and cohesive strength of the plaster.

 

Second, how to accommodate the difference in movement between the wooden support÷which is subject to dimensional change due to environmental factors, structural movement, wind and external stresses such as being walked over etc÷and that of the rigid plaster layer.


The main task of the research was the evaluation by means of mechanical testing of the proposed stitching system and of two other remedial systems÷pinning and grouting÷as well as of the original technology of the ceiling of the Jesuit Theatre. 



Two test series were undertaken: the first on four separate 1 m2 replicas of each of the four techniques; the second on a 1 x 4 m replica including all four techniques. The mechanical tests of the first series included static loading, dynamic loading and shearing, and for the second series, displacement, dynamic loading, and shearing. Results of the first series indicated that the stitching stabilisation was effective in reducing cracking and loss, and successfully imitated aspects of the original technique that allow sufficient flexibility in the system. 


Results of the second test series were less conclusive, primarily due to the use of a weaker plaster. Nonetheless, this series also provided useful information on potential modifications to the stitching method.

The results of the testing largely reinforced the conclusions drawn empirically from the behaviour of the original system, emphasising the importance of both understanding and respecting the original technology.

September 1997