Reparatii Zidarie Cu Ancore Interne

C L Brookes and R J R Swift: Numerical Modelling of Masonry to Explore the Performance of Anchor BasedRepair Systems and the Repair of Monuments in Cairo

Numerical Modelling of Masonry to Explore the Performance of AnchorBased Repair Systems and the Repair of Monuments in Cairo

C L Brookes and R J R SwiftGifford and Partners, UK

ABSTRACT: This paper is in two parts; firstly a method of numerical modelling isdescribed that is being used to investigate the performance of masonry strengthening;secondly how more pragmatic approaches are being used for localised masonry repairs withexamples taken from the Mosque of al Ghuri, Cairo.

Numerical modelling based on a discrete element formulation has been employed tosimulate the response of masonry to seismic loading. Using dynamic non-linear numericalanalysis the performance of shear walls with and without retrofitted strengthening has beencompared. Models representative of ashlar masonry laid with a weak lime-based mortar havebeen investigated. The benefits of strengthening by the introduction of passively stressedreinforcement are predicted for various arrangements when subjected to seismic loading. Thereinforcement is represented explicitly in the analysis allowing direct assessment of damageand potential failure mechanisms. It is concluded that, although more work is required toverify simulations against tests and field experiences, the discrete element technique isideally suited to dynamic masonry simulation and overcomes many difficulties experiencedwith traditional finite element analysis. The overall performance of masonry actingcompositely with retrofitted reinforcement has been predicted and comparisons madebetween different reinforcement dispositions. It has also been shown that unless carefullyplaced additional reinforcement may actually reduce seismic resistivity.

Gifford and Partners have been assisting with the conservation of the monuments of Cairosince the earthquake (5.9 on the Richter scale) in 1992. Recently repairs have beencompleted to the mosque of Al Ghuri. These works are described.

INTRODUCTION

The use of continuum based numerical modelsto simulate discontinuous structures, such asmasonry, is fraught with difficulty. Theintroduction of discontinuities such as cracksduring the loading event, or as a result of loadinghistory, has to be wholly or partly predetermined.The use of gap elements allows cracks to open andmaintain normal and shear force connection whenclosed but the crack locations have to be known inadvance. Another approach is to avoid explicitrepresentation of discontinuities but instead smeartheir effect by using a brittle non-linear materialmodel. However, these models fail to predictmechanisms where, for example, initially isolated

parts react dynamically together. Continuummethods can give satisfactory results but generallyfail to provide a practical method of analysis formasonry.

As an alternative to the traditional finite elementcontinuum approach, a discrete element (DE)formulation has been employed to simulatemasonry with and without strengthening. So far theresults of the analyses have been applied to parts ofbuildings and used to help develop remedial designphilosophies by providing simulations underspecific ground excitations. A separate projectwhere the engineering analysis has been based onthe DE technique has involved the successfulstrengthening of over thirty masonry arches.

C L Brookes and R J R Swift: Numerical Modelling of Masonry to Explore the Performance of Anchor Based Repair Systems andthe Repair of Monuments in Cairo

Predictive verification of full-scale tests underliesthis work and has involved calculated collapseloads of masonry arch bridges as well assupplementary load tests on in-service bridges.Results have been shown to correlate very closelywith tests (Brookes, Tilly 1999). As the techniqueis developed it is hoped that the performance ofwhole buildings can be checked beforestrengthening systems have been installed.

1 ANALYTICAL REQUIREMENTS

In order to represent masonry with or withoutretrofitted reinforcement, particularly in seismicengineering where non-linear structuralperformance defines how ductility and energyabsorption characteristics are exhibited, thefollowing types of fundamental behaviour need tobe included in the model.

i) Material and geometric properties of themasonry blocks themselves.

ii) Contact-gap-friction effects along jointsbetween the masonry blocks.

iii) Depending on block and joint properties, theability to evolve further joints by fracturingwhich in turn depends on limiting tensilestrength and fracture energy.

iv) Full account of stiffness and derived inertialoads which may occur over very short timeintervals.

v) The capability to model post-failure behaviourto help verify simulations against the evidencecollected after observed seismic damage andcollapse.

vi) To allow stress and initial damage fromprevious seismic events to be included.

vii) The ability to represent retrofittedreinforcement including materially non-linearbehaviour of the steel and the non-linear shearcoupling behaviour of the bond with thesurrounding masonry.

To date, most numerical simulations of masonryhave been based on finite element continuummethods in which sophisticated and often-complexmaterial models in conjunction with arrays of gapelements are used to account for the requirementslisted above. A more intrinsically satisfactoryapproach for masonry is to base the analysis on aseries of discrete elements. This more naturalapproach can be used to represent ranges ofmasonry from completely intact buildings to pilesof random rubble.

2 DISCRETE ELEMENT TECHNIQUE

The technique used to perform all the analysisin this study is the discrete element (DE) method.This is a development of the distinct elementmethod (Cundall, 1971) in which the concept ofindividual elements being separate and reactingwith their neighbours by contact throughfriction/adhesion was first successfully applied togeotechnical and granular flow problems. Hereelements were considered rigid but laterdevelopments (Munjiza et al, 1995) included theaddition of element deformations and fracturing,with some overlap with traditional finite elementtheory.

In the current investigation the DE formulationavailable in the explicit dynamic version ofELFEN (Rockfield Software Limited, 1998) hasbeen used. Essentially three different approacheshave been used for the non-linear analysis ofmasonry each requiring different modellingapproaches.

Macro Blocks. The category where the jointsbetween blocks have predominantly no strengthand models the construction generally found inhistoric structures.


Brittle Material. This is where the masonryblocks and joints have predominantly similarstrengths, as is more likely in modern forms ofconstruction or where masonry is weak andrandom.

Brittle Macro Blocks. Here the macro blockapproach is used with brittle materials thuspermitting a blocky representation to fracture intofurther parts.

The first two approaches have been investigatedfor shear wall applications to investigate thesensitivity of seismic resistivity to mortarproperties (Brookes, Mehrkar-Asl, 1998).

2.1 Macro block

The macro block approach has been achievedby separately modelling each block or group ofblocks in the structure and applying permanentstatic loads and seismic excitation to the base.Individual blocks of elements have defined elasticand plastic materials and are arranged to therequired bond. All joints and therefore potentialdiscontinuities are predefined and have frictionparameters assigned. It is assumed that failure atjoints always develops before blocks fail.However, the introduction of a von-Mises non-linear material model without hardening has beenused to approximately represent block crushingthus giving a compressive stress cap. Materialproperties have been based on characteristic valuesdetermined for the masonry as a whole.

2.2 Brittle material

Where masonry includes high strength mortaror where the strength of blocks is low, a brittlenon-linear material model has previously beenused. Here the continuum becomes discretised dueto evolving fractures in the blocks and possiblythrough joints. This is achieved in the analysisautomatically using adaptive mesh algorithms.Using a Rankine material model, including fractureenergy, newly generated cracks become contact

surfaces requiring friction parameters to beassigned as for the macro block approach.However, for ancient ashlar walls with little or nomortar the macro block approach is preferred.

3 REINFORCEMENT REPRESENTATION

The finite element technique is used to modelthe reinforcement independently of the masonryusing a partially constrained spar formulation(Roberts, 1999). Connection between thereinforcement and masonry models is achievedthrough non-linear bond elements. Modelling ofreinforcement arrangements is completelyautomated without the need for topologicallyconsistent element meshes thus accelerating themodelling process and permitting rapidcomparison of designs. Currently, the capacity forreinforcement elements crossing masonry joints togenerate transverse shearing strength or doweleffects is ignored. This is believed to beconservative.

4 SHEAR WALL INVESTIGATION

As part of the continuing development of Cintecanchor applications and the expansion of jointventure historic structure remedial projects,Gifford and Partners with Cintec InternationalLimited are undertaking limited studies toinvestigate how the seismic resistivity of low-risemasonry buildings might be improved (CintecInternational Limited, 2000).

Cintec anchors are comprised of stainless steelbar(s), a grouting sock and an engineered grout.Installation is by precisely drilled holes using wetor dry diamond coring technology. The sockconsists of a specially woven polyester fabricshaped into a tubular sleeve to fit the required holediameter. The sock controls the volume of groutused and ensures good contact is achieved with thesurrounding masonry. Presstec grout is usedhaving similar characteristics to Portland Cementbased products, contains graded aggregates andother constituents which, when mixed with water,


produce a pumpable grout that exhibits goodstrength with no shrinkage. The size of the steelanchor, strength of grout and diameter of hole canbe varied to provide the required design parametersand to provide good stiffness compatibility withthe masonry. Design parameters such as the bondstrength between the grout and the masonry, whichis often critical, are normally derived from staticpullout tests. Figure 1 shows a diagrammaticCintec anchor.

To date effort has concentrated on the detailedanalysis of masonry shear walls, the primarystructural element in masonry buildings, it beingrecognised that the out of plane behaviour ofmasonry panels has been the subject of muchprevious work (Key, 1998). These shear walls aredescribed below. The main objectives of theanalyses were to provide some comparisonbetween the performance of the walls with andwithout strengthening and to continue to explorethe potential of the DE technique includingmodelled reinforcement applied to masonry underdynamic loading. An on-going programme ofstrengthening projects using the DE method topredict the ultimate strength of masonry arches,including comparisons with full scale tests, hasshown the technique to be very accurate and betterthan alternative analyses for the case of staticloading. Further work involving the out of planeprediction of masonry wall behaviour under high-speed dynamic loads arising from blast is alsoencouraging.

Figure 1 Typical Installed Cintec Anchor

4.1 Description

The masonry shear wall under investigation isshown in Figure 2. The wall is 6.15m wide, 6.76mhigh, has shallow foundations over rock and hasbeen considered with and without a large openingin each storey. It forms the shorter side of therectangular building shown in Figure 3 andsupports two floors capable of behaving asdiaphragms. The longer side walls (not explicitlymodelled) partly support the floors, and have littleout-of-plane shear resistance. Vertical body forcesand imposed loads are supported uniformly by allof the external walls. Loads developed byhorizontal seismic ground accelerations in thetransverse direction of the building are resisted byin-plane forces in the side walls. One of these wallsis the subject of the current investigation.

4.2 DE Model

Several plane stress DE models of a single sidewall were developed incorporating the masonryblocks and slabs. The vertical loads and massesattributed to the slabs were modified to reflectmass and load transfer from the rest of thebuilding. The wall is constructed from ashlarblocks, bedded on narrow and relatively weaklime-mortar.


Figure 2 Stone Masonry Shear Wall Details

(all dimensions in mm)

Without full scale testing of part of the wall,uncertainties exist for most material parametersand indeed the modelling. Previously both macroblock and brittle material masonry models wereused to help characterise the walls behaviour(Brookes, Mehrkar-Asl, 1998). The focus of thework was the sensitivity of structural behaviour toequivalent friction of the mortar joints rather thanthe relative performance of different strengtheningarrangements. In the current investigation masonrymaterial parameters have been fixed and based onthose most representative of the form ofconstruction. Similarly macro block models havebeen used throughout so that the focus of theinvestigation is on the effectiveness ofstrengthening.

Although it is feasible to include all of theblocks in the macro block representation, previouswork on masonry arches had shown that there waslittle advantage in terms of accuracy andcomputational efficiency. Hence, each block mayin reality include several squared stones. The floorslabs and foundation were defined as separatecontinuums with similar perimeter frictionalproperties to the blocks. It has been assumed thatthe floors and foundation are constructed such thattheir global behaviour is linear elastic. Forexample reinforced concrete slabs.

All models were developed within a solidmodelling environment using DISPLAY3 (EMRC,1999), translated using in-house software andimported from within the ELFEN pre-processor.

Shear wall Other walls not shown for clarity

Ground

4.3 Material Properties

Masonry material properties were based onthose typical of well-built ashlar construction andusing a weak calcareous sandstone laid with a limebased mortar. The strength and stiffness of themodelled blocks have been based on averagecomposite values for the masonry treated as awhole. The contact and frictional behaviour of themortar is modelled explicitly at the joints. Hence, itis not necessary to individually include thestiffness and strength of the joints or the stiffnessand strength of the stone blocks. Table 1summarises the material parameters used.

Table 1 Masonry Material Parameters

Description Parameter Value

Sandstone/limemasonry

Density 2200 kg/m3

Young’s modulus 3.5 kN/mm2

Strength 5 N/mm2

Mortar joints Coefficient offriction (µ)

0.6

Cohesion 0.15 N/mm2

Adhesion 0

Figure 3 General Arrangement of Idealised Building


4.4 Loading

Hypothetical horizontal seismic loading basedon a circular frequency of approximately 0.6 Hzand containing six shocks was derived and appliedto all of the models as displacement functions atfoundation level (Figure 4). Two magnitudes ofthis simplified motion have been used with peakaccelerations of 0.15g and 0.3g.

Vertical accelerations were not considered dueto the inherent inconsistencies in the distribution ofmass that were required to simplify the problem toone of two dimensions. Horizontal motion resultsin a greater proportion of the effective mass beingdistributed to the two shear walls than thatcorresponding to vertical motion. All load-bearingwalls resist vertical motion. However, whilstconcurrent vertical motion has an influence on theoverall behaviour of masonry shear walls, it isgenerally accepted that horizontal motion iscritical.

Figure 4 Time History of Input Motions

4.5 Strengthening

Both the plain wall and wall with openings weremodified to include various arrangements ofstrengthening. Proposed dispositions ofreinforcement included horizontal throughindividual block courses, vertical at the ends of thewall and end diagonal. Combinations of thesepatterns have also been considered. Thereinforcement is fully bonded along its length.

The reinforcement is introduced into the wallusing the Cintec anchor system. All dispositions ofreinforcement investigated used single 20mmdiameter ribbed bars installed in 50mm diameterholes. The anchors are designed not to bedeliberately stressed but attract load during aseismic event. The modelled anchors permitrecovery of bond stresses, axial stresses andslippage along the length of the anchor at any timeduring loading.

Table 2 shows the various arrangements of walland reinforcement that were the subject of theinvestigation. Figure 5 shows a typical modelincluding DE boundaries (bold), finite elementsubdivisions (fine) and modelled reinforcement(bold).


Table 2 Wall and Strengthening Arrangements

Arrangement No. Description(friction with cohesion and 0.15gloading used unless otherwiseindicated)

2,1214

Plain wall. Also plain friction and0.3g loading investigated.

3,11Horizontal reinforcement at thecentre of the first storey. Also plainfriction investigated.

4 Horizontal reinforcement at thecentre of the second storey

5Horizontal reinforcement at thecentre of the first storey and thecentre of the second storey

7 Diagonal reinforcement at the endsof the first storey

8 Diagonal reinforcement at the endsof the second storey

6 Vertical reinforcement at the ends ofthe wall

13

Combination of diagonal andhorizontal reinforcement. Bottomreinforcement in second layer ofblocks

19Combination of diagonal andhorizontal reinforcement. Bottomreinforcement in first layer of blocks

2021

Combination of diagonal, verticaland horizontal reinforcement.Bottom reinforcement in first layer ofblocks. Also 0.3g loadinginvestigated

22 Wall with openings in each storey

2324

Combination of diagonal, verticaland horizontal reinforcement.Bottom reinforcement in first layer ofblocks. Also 0.3g loadinginvestigated

5 DISCUSSION OF RESULTS

Chiefly, DE simulations were carried out toshow how the strength and ductility of the wallsvaried with reinforcement arrangement. However,for the plain wall and the wall with reinforcementarrangement No. 3 mortar with cohesion, which isconsidered to be most probable, and withoutcohesion has been considered. This limitedinvestigation into the influence of mortarproperties was carried out to allow comparisons tobe made with simpler earlier work. Where wallshave exhibited a high degree of seismic resistivityan additional ground motion with peakaccelerations of 0.3g have been applied. Theresults have been illustrated by arrays of contourdiagrams in which, the friction behaviour in thejoints, the magnitude of ground motion and theposition of reinforcement have been varied.

5.1 Unstrengthened simulations

Figure 6 illustrates compressive stresses anddeformed geometry half way through the seismic

Figure 5 Typical Wall DE and Reinforcement Model


event (time 5.2s) and after ground motion hasceased (time 12.5s). The shaded contours rangebetween 2.2 N/mm2 (dark) and -0.2 N/mm2.

It has been theoretically shown that thepredicted ductility of the walls is highly sensitiveto the properties of the joints and the duration ofthe event (Brookes, Mehrkar-Asl, 1998).Furthermore, the inherited damage history frompreceding shocks increases the seismicvulnerability of the wall. As part of the currentinvestigation the influence of openings is alsoconsidered with the aim of developing anarrangement of reinforcement that works equallywell for plain walls as well as those with openings.At the level of the idealised openings, representingsmeared windows and doors, approximately 36%of the wall is removed. A lintel supports the wallremaining above each opening.

Halfway through event After event

Plain wall ( No.2) – Maximum acceleration 0.15g

Plain wall (No.14) – Maximum acceleration 0.3g

Wall with openings (No.22) – Maximum acceleration 0.15g

Figure 6 Unstrengthened Walls

(Contours of principal compressive stresses)

Figure 6 shows deformed principal compressivestress contours halfway through the event and afterthe event for the plain wall with (No.22) andwithout openings (No.2). The plain wall is alsoloaded with a 0.3g ground motion. After the 0.15gevents the plain wall remains relatively undamagedwith cracking in both storeys. The cracks in thesecond storey result in significant dilation acrossthe wall whereas the cracks in the first storey areless dilated and are associated with locked instresses. Doubling the acceleration results inmassive damage. With much less masonry capableof resisting shear, the wall with openings is


severely damaged after three shocks and collapsesbefore the end of the event.

The results show the general behaviour as wellas the initiation of cracking. Cracking is marked bysudden stress discontinuities as well as the relativemovement of blocks. This movement developsrapidly into local failure mechanisms whensubjected to continued shocks in the hypotheticalseismic event. The predicted failure and localcollapse, reflecting modelled ductility and energyabsorption, is similar to damage frequentlysustained in seismic regions. Hence, these threemodels have been used as the benchmarks tocompare the performance of various retrofittedreinforcement schemes.

5.2 Strengthened simulations

Dispositions of reinforcement that have beeninvestigated including horizontal, vertical, diagonalat the ends of the wall and combined patterns.

Figure 7 shows stress and reinforcementslippage results obtained for horizontalreinforcement with (No.3) and without (No.11)cohesion effects included in the mortar. In themore realistic case with cohesion the introductionof reinforcement has resulted in more damage.Without cohesion but with µ=0.6 remaining, lessdamage occurs and good correlation is obtainedwith earlier simulations (Brookes, Mehrkar-Asl,1998). The following reasons are likely for thisbehaviour.

i) With cohesion, less energy is dissipated acrossthe joints resulting in less structural damping.

ii) Increased shear capacity across joints providedby cohesion helps a rocking mechanismdevelop in the wall. This mechanism causessudden vertical oscillations of the wall abovethe reinforcement level resulting in additionaldamage.

Axial slippage of reinforcement in excess of5mm indicates that bond with the masonry hasfailed. The graphs show progressive bond failure at

both ends of the reinforcement. As with themasonry, less damage occurs to the reinforcementwhen no cohesion is considered.

Plain wall ( No.3) – Friction with cohesion

Halfway through event After event

Axial slippage of reinforcement

(Ordinate – slippage [mm], Abscissa – relative position)

fail

failfail

Plain wall ( No.11) – Simple friction

Axial slippage of reinforcement

(Ordinate – slippage [mm], Abscissa – relative position)

fail

Figure 7 Horizontal reinforcement



In Figure 8 the results at the end of the event aregiven for different reinforcement arrangements. Allschemes except for No.19 and No.8 (not shown)cause either more damage to the masonry orproduce higher locked in stresses compared withthe similar unstrengthened wall. ArrangementNo.19 has marginally improved resistance andprevented any blocks from falling away from thewall. In the case of No.13 an oscillatorymechanism develops above the first storeyhorizontal reinforcement causing increased damageby vertical movement.

Maximum acceleration 0.15g

Diagonal (No.7) Vertical (No 6)

Diagonal and horizontal(No.13)

Diagonal and horizontal(No.19)

Figure 8 Diagonal, Vertical, Horizontal Reinforcement


Results obtained from schemes with the mostdeveloped dispositions of reinforcement combiningdiagonal, vertical and horizontal bars are shown inFigure 9. Here similar schemes have been appliedto both the plain wall and the wall with openings.It is similar to No.19 except that the bottomhorizontal reinforcement lies in the bottom courseof blocks. This helps eliminate vertical motioncaused by rocking of the blocks below thereinforcement level.

The plain wall remains intact throughout theseismic event with the reinforcement controllingthe development of cracking. Even under 0.3gloading, although higher locked in stresses arepredicted little damage is evident. Withoutreinforcement, blocks left unarrested rapidlypropagate failure mechanisms leading to collapse.

The wall with openings is similarly improvedexcept that considerable cracking occurs to the firststorey with appreciable dilation on one side.Although the improvement compared with theunstrengthened wall is dramatic debonding ofsome reinforcement has marked the beginning of afailure mechanism and continued shocks would bevery damaging.

In both walls the reinforcement, apart fromdirectly supporting the ends of the walls, hasencouraged shearing at the foundation levelcreating an energy absorbing process as well asoffering some degree of base isolation.

Maximum acceleration 0.15g Maximum acceleration 0.3g

Diagonal, horizontal and vertical (No.20 and 21)

Diagonal, horizontal and vertical (No.23 and 24)

Figure 9 Combined reinforcement arrangements



6 CONCLUSIONS

By combining the discrete element techniquewith a finite element formulation for reinforcementnumerical models have been developed that haveallowed rapid evaluation of the relativeperformance of reinforcement based retrofittedstrengthening. This process has been illustratedusing a plane shear wall subjected to simplifiedhypothetical ground movements. The followingconclusions have been drawn.

i) The overall performance of masonry actingcompositely with retrofitted reinforcement canbe predicted.

ii) The sensitivity of seismic resistivity to thedisposition of reinforcement has beendetermined thereby allowing the comparisonof practically viable schemes.

iii) A disposition of reinforcement has beeninvestigated that improves seismic resistanceand that works equally well for walls with andwithout openings.

iv) The results have also shown that retrofittedreinforcement unless carefully placed mayactually reduce seismic resistivity.

The development of strengthening schemeswould have been extremely difficult and costlyusing conventional analysis or testing. However, tobe completely confident that the results obtainedby these numerical simulations are correct furtherwork is required to verify the simulations againstobserved behaviour of masonry structuressubjected to seismic loading.

7 MONUMENTS IN CAIRO

Although numerical modelling of masonry isbeing used to develop retrofitted strengtheningsystems often the representation of these systemsand respective building parts are necessarily

idealised. In practice a degree of pragmatism andengineering judgment is required especially wherelocalised repairs are involved. The recentexperiences in Egypt described below are a case inpoint.

7.1 Background

Egypt has a long and relatively well-documented earthquake history. The country washome to some of the earliest known civilisations,as a consequence of which records of earthquakedamage extend back at least as far as 2800BC. Inrecent years significant earthquake events haveoccurred once every 30-50 years.

There are over six hundred monuments in Cairowhich have earned the city its title of a “City ofHuman Heritage” by UNESCO. Many of thesemonuments have endured at least sevenearthquakes exceeding five on the Richter scale.

Cairo has not only suffered from the effects ofan earthquake but has also suffered from theproblems of a leaking drainage system and thecontaminated perched water table has caused therapid degradation of the masonry structures. Thusbefore the 1992 earthquake, the monuments ofCairo had been suffering from serious neglect anddecay.

According to initial evaluations, the 1992earthquake occurred approximately 30km to thesouth-west of Cairo, was of only mediummagnitude 5.5-5.9 on the Richter scale and was ofshallow focal depth. The city stands on very deepalluvial sands, silts and clays forming either side ofthe Nile delta. The peak ground acceleration wasestimated at 10% gravity. The nature of the quakewith its relatively high frequency motion meantthat damage to low structures of up to 5 storeyswas more likely.

The Egyptian Antiquities Organisation reportedthat 150 of the 600 monuments had been damagedby the earthquake. Typical of the damage incurredwas the damage to the Mosque of al Ghuri.


7.2 The Mosque of al Ghuri

The Mosque of al Ghuri, monument number189 dates from 909AH or1504AD and is shown inFigure 10.

The funerary complex of Sultan al Ghuri issituated in the Fahhamin quarter of old Cairo in alMuizz Street. On the west side of the street thereis a kanqah and mausoleum as well as a SabilKuttab. The minaret is a four storied rectangularstructure approximately 50 metres high.

An inspection of the madrasa reveals some verysevere long-standing problems. The floor of themosque undulates dramatically, evidence of verysignificant foundation problems of the masonryvaults supporting the floor. Attempts have beenmade in the past to underpin the sleeper wallssupporting the vaults. Those have failed. All ofthe walls of the mosque exhibit very severefractures. These are a function of the problemsbrought about by the rising contaminated groundwater table as well as the shaking due to theearthquake.

The next result of the above is that the mosqueof al Ghuri is now in a very delicate state ofequilibrium. Despite having survived for nearly500 years, the toll of a rising water table,earthquakes and neglect had brought this structureto the point of collapse. Urgent measures wererequired to reintroduce some structural strengthsand stiffness into the building.

In summary, typical damage evident at al Ghuriincluded:-

• Lack of adequate anchorage between the wallsresulting in vertical cracks and separation inthe corners at the connection between walls atright angles – ‘Out of plane bending’.

• Spreading of the sahn arches of the court anddropping of voussoirs.

• Settlement of the floors.

• “Pounding” adjacent to the minaret.

• Failure of roof to wall connections.

• Loss of integrity of the wall construction.

7.3 Repair

Several monuments did not survive theearthquake because of their decayed conditions. Itwas important therefore to use appropriate repairtechniques to ensure continued survival.

The Cintec masonry anchor, described above,was introduced extensively at al Ghuri and used torestore structural integrity and provide additionalductility to the building. Figure 11 illustrates atypical anchor application in an arched facade. Theanchors, which were installed, were up to 12metres long and served to stiffen each individualwall. The walls of al Ghuri are generally of twofacing skins infilled with a core of rubble. Thelarge arched openings in the mosque are particularpoints of weakness in the structure. Longitudinalties in each of the stone facings of the wall abovethe arch would serve to resist the thrusts naturallyproduced by the arch as well as serving to assistthe walls to resist the next earthquake. In additionto longitudinal ties, transverse ties of length equalto the thickness of the wall acted as consolidationanchors. The Cintec ties were also intended to beused to tie the roof structure to the perimeter wallsand create a diaphragm action.

The intention of the designs for the anchor is:-

• Restore integrity of construction of wallthickness.

Figure 10 The Mosque of al Ghuri


• Resist “lateral toppling” out of plane bendingby tying walls at right angles to each other.

• Connection of roofs and floors to walls.

• Repair of arches.

• Pinning of individual decorative voussoirstones.

Of particular concern to the EgyptianAntiquities Organisation was the possible damagewhich might be caused to the decoration of thesoffits of the voussoirs stones of the arches to thecourt. This concern was overcome by using a thinwalled core drill to extract a cover pellet first andthen the anchor was installed and the pellet re-fixed.

The repair works were recently completed andthe mosque re-opened.

8 REFERENCES

Brookes, C.L. Mehrkar-Asl, S. 1998. Numericalmodelling of masonry using discrete elements.Proceedings of the 6th SECED Conference onSeismic Design Practice into the Next Century,Oxford.

Brookes, C.L. Tilly G P. 1999. Novell method ofstrengthening masonry arch bridges. StructuralFaults and Repair – 99, 8th InternationalConference, London.

Cintec International Limited. 2000. The Cintecanchor system. Newport, Wales, UK.

Cundall, P.A. 1971. A computer model forsimulating progressive, large scale movement inblocky systems. Proceedings: Symp. ISRM,Nancy, France, Vol. 2, 129-136.

Engineering and Mechanics Research Corporation1999. DISPLAY3 version 9.0 production. Troy,Michigan, USA.

Key, D. E. 1998. Parameter influences on the outof plane seismic collapse of unreinforcedmasonry panels. Proceedings of the 6th SECEDConference on Seismic Design Practice into theNext Century, Oxford.

Munjiza A., Owen, D.R.J., Bicanic, N. 1995. Acombined finite/discrete element method intransient dynamics of fracturing solids.Engineering computations, 12, 145-174.

Rockfield Software Limited 1998. ELFEN version2.8.0a_MT Archtec version. University ofWales Swansea.

Roberts, D.P. 1999. Finite element modelling ofrockbolts and reinforcing elements. PhD Thesis,University of Wales Swansea.

Figure 11 Anchor Arrangement in Arched Facade

Documents

Reparatii Zidarie Cu Ancore Interne