COMPUTATIONAL MODELLING OF LOW STRENGTH BRICKWORK WALL/BEAM PANELS WITH RETRO-FITTED REINFORCEMENT

Computational modelling of low strength brickwork wall/beam panels with retro-fitted reinforcement

V.Sarhosis¹, S.W.Garrity² and Y.Sheng³

^1,2,3 Research Fellow, Hoffman Wood Professor of Architectural Engineering and Associate Professor, respectively,

School of Civil Engineering, University of Leeds, Leeds, West Yorkshire, England, UK. LS2 9JT.

v.sarhosis@leeds.ac.uk,  s.w.garrity@leeds.ac.uk, y.sheng@leeds.ac.uk

 

ABSTRACT

Retro-fitted stainless steel reinforcement is being used increasingly to strengthen the masonry cladding of low to medium rise buildings, particularly where cracking has occurred adjacent to a long-span window or similar opening. This paper describes the development of a computational model which was used to predict the behaviour of reinforced clay brick wall/beam panels subjected to vertical in-plane static loading. In practice, cracking in unreinforced walls of this type, particularly where low cement content mortar has been used, tends to occur along the brick/mortar interfaces and failure usually results from de-bonding of the bricks. As a result, software based on the Distinct Element Method (DEM) of analysis was used. The bricks were represented as an assemblage of stiff but deformable distinct blocks and the mortar joints were modelled as zero thickness interfaces. These interfaces could open or close depending on the magnitude and direction of the stresses applied to them. Reinforcement was modelled using spring connections attached to the masonry surface.

The masonry material parameters were obtained from the results of experimental tests carried out in the laboratory on full-scale unreinforced wall/beam panels. The computational model was then used to predict the behaviour of wall/beam panels containing bed joint reinforcement. Good correlation was achieved with the results obtained from the testing of full-scale reinforced panels in the laboratory, in particular, the load to cause first visible cracking, the propagation of cracks with increasing applied load, the mode of failure and the magnitude of the collapse load.

KEYWORDS: masonry, walls, reinforcement, distinct element modelling.

A COMPUTATIONAL MODELLING APPROACH FOR LOW BOND STRENGTH MASONRY

SAHC2012, Structural Analysis of Historic Construction, 15th to 17th of October 2012 in Wrocław, Poland

Historical masonry is typically characterised by its low bond strength. Cracking of such masonry is often as a result of the de-bonding of the masonry units from the mortar joints and the post-cracking response up to collapse is influenced by the characteristics of the masonry unit/mortar interface.

The development of a computational model that is applicable for low bond strength fired clay brick masonry is described. Using software based on the Distinct Element Method of analysis, bricks were represented as an assemblage of distinct blocks. The mortar joints were modelled as zero thickness interfaces which can open or close depending on the magnitude and direction of the stresses applied to them.

The material parameters for the masonry constitutive model were obtained from the results from the load testing of large wall/beam clay brick masonry panels in the laboratory. Initially the panels were modelled computationally using an assumed set of material parameters. The differences between the results obtained experimentally and numerically were then minimized by adjusting the parameters in the constitutive model using an optimization technique. The model was then used with the optimized parameters to predict the structural response of other unreinforced and reinforced wall/beams that had been tested previously in the laboratory. Good correlation was obtained with the experimental results.

Keywords: Bricks, Bond, Computational modelling, Masonry

Improving the resilience of masonry structures

Distinct element modelling of masonry wall panels with openings

CCP: 88

PROCEEDINGS OF THE NINTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL STRUCTURES TECHNOLOGY

Edited by: B.H.V. Topping and M. Papadrakakis

Paper 17

Cracking of the masonry walls of low-rise domestic and commercial buildings above windows and door openings often occurs as a result of a lack of support from a lintel or other similar structural member. Many thousands of low-rise domestic and commercial buildings in the UK are afflicted by these problems which are exacerbated by in-plane thermal and moisture movements as well as the effects of settlement and subsidence.
The cost, time and effort associated with repairing such damage is usually significant from the owner's perspective, particularly when the disruption to the normal use of the building is taken into account. Therefore, it is important to gain an improved understanding of the behaviour of masonry wall panels with openings so that cost-effective, minimum disruption repair or strengthening techniques can be developed. The development of a reliable computational model is particularly important to avoid the need for costly, repetitive laboratory testing of large-scale wall panels.

In the current research work, a two-dimensional distinct element modelling software package (UDEC) [1] has been used to predict the qualitative behaviour of a single leaf thick brick masonry wall panel with a 2m clear opening when subjected to a gradually increased vertical load. The results from UDEC were compared with the experimental results obtained from the full-scale testing of unreinforced brick masonry wall panels at the University of Bradford [2].

The experimental results showed that there were three notable features of the behaviour of the wall panels that occurred with increasing applied load. These consisted of initial flexural cracking followed by the propagation of diagonal stepped cracks from the corners of the opening leading, finally, to collapse. These three features were also predicted using the UDEC model. In particular, the model captured the tensile and shear failure of the brick/mortar interfaces as well as the collapse mode. Hence, it is concluded that a distinct element numerical model such as UDEC can be used successfully to predict the qualitative behaviour of a panel of brickwork containing a large opening when subjected to a gradually increased in-plane vertical load.

The next stage of the research will be focused on improving the performance of the model so that a quantitative analysis can be performed. Different forms of strengthening will then be incorporated into the model which will be further validated using the results from full-scale laboratory testing of strengthened wall panels

References
1 Itasca Consulting Group, "Theory and Background", Universal Distinct Element Code, ITASCA consulting group Inc., Minneapolis, Minnesota, USA, 2004.
2 S.W. Garrity, "Independent Testing of the Bersche-Rolt System for Strengthening a Brickwork Lintel Containing a Soldier Course", Test Report, University of Bradford, Bradford, United Kingdom, 2004.

Computational modelling of clay brickwork walls containing openings

Abstract

The use of the Distinct Element Method to simulate the response of single leaf clay brickwork walls with openings to vertical, in-plane, static loading is described. The walls were modelled as an assemblage of stiff yet deformable bricks with mortar joints as zero thickness interfaces. Conventionally, the results of tests on small specimens are used to determine the material or interface parameters. These values usually need to be adjusted to allow for inherent variations in the materials, workmanship effects and differences in the boundary conditions of the small-scale tests compared with those in the larger structure. In this research the material and interface parameters were determined by applying a manual optimisation to the results of a series of laboratory tests carried out on full-scale wall panels. The computational model was then used to predict successfully the behaviour of a longer span wall panel constructed from a similar brick and mortar combination.

Download:
http://eprints.whiterose.ac.uk/42880/7/8IMC_207-Garrity_front.pdf

See simulation video:
http://www.youtube.com/watch?v=BtUfB1nSS1o&feature=plcp