In-plane tension behaviour of masonry

Abstract: The main purpose of the present work is to increase knowledge of in-plane tension behav-iour of masonry and develop a model able to describe the impact of parameters influencing cracking caused by restrained shrinkage. In the process the mechanism of load transfer between units and mortar joints is described. The influence of cohesion between units and mortar joints, unit and joint geometry and external loads on the mechanical properties of masonry and the response to in-plane loading is studied. It is concluded that, in order to obtain a realistic in-plane tension behaviour, tension and shear softening of the mortar joints must be included in the masonry model. A discontinuum model, implying discrete model-ling of units and mortar joints, is employed to reproduce masonry behaviour in crack zones. Where cracking is not expected, masonry is modelled as an elastic continuum. The model has been implemented in a commercial finite element programme. To obtain realistic data for the simulations, two types of clay bricks and three types of mor-tars were tested. Measurement of natural frequencies by a non-destructive method and direct tension tests up to failure were carried out. The moduli of elasticity determined by the two methods showed good agreement. Deformation controlled shear tests, at three different levels of normal pre-compression, were carried out on joints, with cement respectively lime rich mortars. The joints with the lime rich mortar exhibited more deformation in the pre-peak stage and a slower diminishing of the load bearing capacity in the post-peak stage compared to the cement rich mortar. Three-point bending tests were carried out on beams consisting of glulam at the ends and masonry in the central parts. Perforated bricks with a cement respectively lime rich mortar were used – testing with massive bricks was not possi-ble due to pre-mature failure caused by low bending strength in the joints. The mode I fracture energy of the mortar joints manufactured with the lime rich mortar was 50 % com-pared to the joints with cement rich mortars. Simulation of retrained shrinkage on massive, one-storey masonry walls with one-sided restraint show that application of slip joints between the wall and obstructing foundation has the strongest crack delaying effect. A shift from the strong cement rich mortar to the lime rich mortar allows the spacing of vertical movement joints to be increased by 10 %. Cracking in walls with openings occur at approximately 40 % lower imposed shrinkage levels than in massive walls. Cracking at windows located at the end of walls longer than 7 m can not be avoided be application of vertical movement joints.