Masonry Failures

Neil R. Baer, P.E. When determining coverage for the failure of an older masonry wall, the cause of the failure should be investigated by an engineer. Most failures involving older masonry walls are caused by one, or a combination of, three factors; improper design/construction, long-term material deterioration and a change in the intended use of the structure.

Fig. 1 Cracked unreinforced masonry wall

Fig. 2 Deteriorated mortar

Even as recently as 20 years ago, masonry structures in some areas were built without the benefit of a professional engineers design or supervision. This resulted in structures that do not meet proper design standards and, therefore, are vulnerable to long-term structural failure due to their lack of durability. An underdesigned or poorly built masonry structure can exist for decades with no apparent problems, but small loading changes over time, that a properly designed masonry structure would withstand, can cause it to fail. These small loading changes can be brought about by weather changes such as El Nino, or changes adjacent to the masonry structure such as new construction. A typical example would be a masonry retaining wall that was not properly reinforced and lacked proper drainage, Fig. 1. During periods of unusually heavy rain, such as El Nino, this wall could fail due to the increased lateral pressure on the filled side of the wall. Structural fatigue, brought on by the masonry being overstressed over a long period of time, can also lead to failure in an improperly designed wall.

Some mortars from as recent as the 1960s have been known to deteriorate and loose their cohesiveness, Fig. 2. Masonry has never been recognized as having the ability to resist large tensile forces. As the mortar deteriorates, the masonry begins to lose its tensile strength, much of its shear and compressive strength also. When this occurs, small loading changes that would not affect sound mortar, could cause failure in a masonry structure. Again, the small loading changes could be weather related or changes around the structure that change the nature of the loading of the structure. The use of any structure can change with time and ownership. If a masonry wall that was intended for use as an interior wall becomes an exterior wall and it is exposed to wind loads, it could become overstressed and fail. If a load-bearing masonry wall has its loading condition changed by adding additional floors or floor area, the wall could fail in compression or buckling due to overstressing. Another example of a change in use would be an open field above a masonry retaining wall that was changed into a parking lot for automobiles. The surcharge from the automobile loading would induce unanticipated lateral loading in the masonry retaining wall and could cause it to fail. A recent storm or other event may appear to be the cause of failure in a masonry structure when actually the primary cause could be a design/construction defect, mortar deterioration or improper use.

Cracking of Buildings due to Shrink/Swell in Clay Soils

Purpose The purpose of this technical note is to provide basic guidance to the owners and custodians of heritage buildings which are suffering cracking due to shrink and swell in clay soils. There are many potential causes for cracking and approaches to mitigating the problem. Specialist engineering and geotechnical advice is usually necessary. 1.0 Cracking of masonry buildings built on reactive soils is a common issue throughout Australia wherever clay soils of significant depth underly buildings. Reactivity refers to the tendency for the clay soil beneath the footings to shrink and swell with changes in moisture content which can lift and lower the building. The major factors influencing this are: The ability of water to reach the clay material beneath the footings; The composition of the soil; The depth of the soil; The effect of trees. 2.0 Is the cracking in my building due to shrink/swell in the subsoil? There are many other potential causes of cracking to masonry buildings including: consolidation of poorly compacted soils (either natural soils or fill) when loaded with the weight of a building; vibration causing consolidation of soils (eg traffic, construction activities, earthquake, mining blasting); mine subsidence due to underground tunnelling; wash out or softening of the ground below the footings due to saturation by subsurface flow (eg stormwater or broken service pipes); corrosion of lintels or embedded steelwork or decay of embedded timber within walls; earthquake or wind loads; deterioration of materials (eg failing rubble walls)
CRACKING OF BUILDINGS DUE TO SHRINK/SWELL

2 3.0 Engineering advice: It is recommended that a structural engineers advice is sought to determine the probable cause of cracking and advise on further action. Qualified engineers are usually advertised in the yellow pages under Engineers Consulting. Names may also be obtained through the Institution of Engineers Australia and the Association of Consulting Engineers Australia. The structural engineer will also advise whether geotechnical advice and investigation is necessary. If your building is of heritage significance it is recommended that your engineer has heritage experience. 4.0 There are several options for treatment of problems due to shrink/swell in clay soils.

Control of water sources : All piped services in the ground should be checked by a plumber to ensure that there are no breaks or leaks leading to periodic local wetting of the subsoil. This includes sanitary drainage, stormwater and water supply services. External water taps and downpipes should discharge into drains not onto the ground. Irrigation systems that maintain a reasonably constant moisture content in the soil may be acceptable however those that are used only periodically may exacerbate wetting and drying cycles. Generally it is preferable not to have planting or irrigation systems immediately adjacent to walls of buildings in reactive clay soils. Flower beds should be separated from the building by paved areas in the order of 2 metres wide, if possible. Tree root control : the presence of trees near the building (eg within a distance of 2 x tree height from the walls) will exacerbate shrinkage in clay soils as the tree sucks moisture out of the ground. Trees can be a major contributor to shrinkage in reactive soils. Options include installation of cut off walls for root control with appropriate root pruning (an arborists advice is required to ensure tree health and stability is maintained) or removal of trees within the influence zone of the footings. The influence zone will vary with tree size and type, soil type and landform, consequently an arborists advice and possibly trial excavation to determine extent of roots is necessary. Works to trees generally require Council approval. Moisture control to the building perimeter : this is a strategy whereby the changes in moisture content below the footings are minimised by appropriate site grading, drainage and moisture control devices. The purpose of moisture control devices is to maintain a stable water content in the soil below the footings by creating an increased path length for evaporation to the surface. Typical systems include: - paving to the perimeter of the building; - cut off walls to the perimeter of the building; - buried impermeable membranes around the building; - a combination of the above;
CRACKING OF BUILDINGS DUE TO SHRINK/SWELL

3 - drip irrigation systems to maintain moisture content in the soil. These approaches can assist in minimising ongoing movement but will not eliminate it. These methods are howeve r generally considerably less expensive than underpinning. Moisture control aprons have potential problems in that if damp proof courses are inadequate the increased moisture levels can exacerbate rising damp in walls. They can also increase moisture content beneath houses leading to conditions which could exacerbate mildew, fungal rot and insect attack. Consequently, their use may need to be coupled with repair or installation of a damp course and maintenance or installation of appropriate underfloor

ventilation. The potential problems caused by increasing moisture content therefore have to be balanced with the advantages in reducing reactivity. Underpinning: this is the most complete solution whereby the footings of the building are underpinned with either concrete, masonry or piles to carry the load of the building down to a more stable stratum (eg rock or soils below the reactive zone). This solution is usually the most costly, particularly if there are access difficulties or if internal walls require underpinning, which may require lifting internal floors. 5.0 Subsoil drains : Subsoil (or agricultural) drains are often used to control stormwater in association with site grading. Subsoil drains are useful in preventing excessive saturation of the soil however should not be used immediately adjacent to external walls. Free drainage adjacent to the walls can have the effect of increasing the rate of wetting and drying. Consequently, subsoil drainage is only recommended at the perimeter of paving and generally at least 2 metres away from external walls. Site grading and paving should grade away from the walls towards such drains. The drains should not be too deep in case they dry out the soil excessively. 6.0 Repairs to cracked walls. Once a wall has cracked it can be difficult to prevent recurrent cracking at the same location. If underpinning has been used it may be possible to repair cracks without significant risk of re-cracking. If moisture control methods have been adopted, without underpinning, then continued, albeit minor, movement is likely to occur. Cracks can either be repaired and be allowed to re-crack (although to a lesser extent) or can be filled with flexible sealants to allow for minor movement. An alternative is to install articulation joints at selected locations so that movement is handled in a controlled manner in selected discrete locations (eg a vertical joint in the wall behind a downpipe). Crack repair can consist of raking out and refilling open joints (matching existing mortar in porosity and strength), replacing bricks, rebuilding sections of wall or can include stitching across cracks with stainless steel reinforcing bars. A structural engineers advice should be sought regarding appropriate methods of crack repair and advice on the risk of re-cracking.
CRACKING OF BUILDINGS DUE TO SHRINK/SWELL

4 References Australian Standard AS 2870 Residential slabs and footings. Stabilizing Heritage Buildings Founded on Reactive Clay, J W Jordan, B J Collins NSW Heritage Office Library Guide to Home Owners on Foundation Maintenance and Footing Performance by P F Walsh CSIRO Sheet No. 10-91, July 1986, CSIRO Information Service. Written by Simon Wiltshier for the Technical Advisory Group (TAG) of the Heritage Council of NSW May 2004

Advantages
The Cintec anchoring system is simple and easy to use, with fast installation and minimal cleanup. But its advantages go far beyond ease of use.

Masonry Repair Strategies

Finite/discrete element models for assessment and repair of masonry structures D. RJ. Owen, D. Peric and N. Petrinic Department of Civil Engineering, University of Wales Swansea, UK CL. Smokes Gifford and Partners, Southampton, UK P.J.. James CINTEC International Ltd., Newport, UK

ABSTRACT
The predictive modelling of masonry structures represents a challenge due to their semi-discrete and composite nature. This paper presents an approach to modelling which considers the composite action of individual masonry components as an alternative to macro-modelling based on a homogenised continuum. A primary aim of the current research is to develop computational procedures for assessing the remnant strength of damaged masonry structures and to examine the efficiency of alternative repair strategies. Technical issues considered in the paper include the coupling of polygonal discrete elements for simulating the behaviour of masonry structures with circular discrete elements for representing geotechnical fill material. The approach adopted for parallel solution is also summarised.

INTRODUCTION
The ultimate objective of the work described in this paper is to develop industrially applicable computational procedures to assess alternative repair strategies for damaged masonry structures. In many Stances, structures which have been damaged can be successfully repaired by the insertion of anchors, use of prestressing systems, etc. For relatively modern structures the decision to repair, rather than demolish, may be based on economic considerations, but for

historic structures the use of remedial techniques becomes a cultural necessity. Currently, appropriate stitching patterns are selected on the basis of previous experience and a major aim of the research is to provide a rational approach to determining an efficient and near-optimal anchor arrangement In this way, an industry standard simulation capability will be provided to aid decision making in quantifying the remnant structural integrity of masonry structures and to examine optional remedial actions. The predictive modelling of the behaviour of masonry structures, particularly in the non-linear range, remains a challenge, due predominantly to their semi-discrete and composite nature. An adequate computational model must include the fundamental mechanisms that characterise the composite action: (i) sliding along a bed or bead joint at low values of normal stress, (ii) cracking of the masonry units (bricks, blocks, etc.) in direct tension, (iii) diagonal tensile cracking of masonry units at values of normal stress sufficient to develop frictional behaviour in the joints and (iv) splitting of units in tension as a result of mortar dilatency at high values of normal stress. Further aspects which may require consideration include the treatment of reinforcement and/or prestressing in composite construction techniques or repaired structures. To date, most computational predictions have been based on a macro-modelling approach, in which attempts are made to incorporate some or all of the phenomena described above within a continuum description; employing homogenisation concepts to produce a smeared representation of the brick/joint action and classical plasticity concepts to model the tensile/compressive failure of the resulting composite. Such a modelling strategy has been dictated both by the limits of available computing power and the lack of maturity of other semi-discrete computational techniques. Whilst considerable fundamental information can be derived from solutions based on a continuum approach, a more natural treatment of this class of problem is offered by use of discrete element methods. The use of discrete elements originated in geotechnical and granular flow applications and are based on the concept that individual material elements are considered to be separate and are (possibly) connected only along their boundaries by frictional/adhesive contact. With present day computational power large scale discrete element models can be considered and for industrial applications in the field of rock blasting, etc. 10-50,000 elements are routinely employed. Earlier work on discrete element techniques was based on the assumption that each element was rigid, but later extension to include local deformation has permitted a more rigorous treatment of both the contact conditions and fracture requirements. The incorporation of deformation kinematics into the discrete element formulation has also led naturally to a combined finite/discrete element approach in which the problem is analysed by a combination of the two methods [5,14,16]. In a finite/discrete element analysis the main issues which require consideration, for both dynamic and quasi-static behaviour, are: (i) Appropriate element modelling of the continuum and discrete regions with a view to incorporating the deformation mechanisms necessary to model stress, stain and frictional contact/adhesive conditions, (ii) The inclusion of elastoplastic/viscoplastic behaviour in both the finite and discrete elements, (iii) Algorithms for treatment of the progressive fracturing of elements, (iv) Detection procedures for monitoring

contact between discrete elements and the continuum regions and (v) Representation of frictional/adhesive contact conditions for contacting elements. The first phase of this ongoing study has embraced recently developed numerical techniques for solving problems in applied mechanics which exhibit large discontinuous geometrical and nonlinear material behaviour. The technical developments considered in this paper concentrate on coupling of polygonal discrete elements for simulating the behaviour of masonry structures with circular discrete elements for representing geotechnical fill material for masonry arch bridges, etc.. The approach being currently pursued to enable the parallel solution of these CPU intensive problems is also summarised.