FRAMED-MASONRY RESEARCH PROJECT
FRAmed-MAsonry composites for modelling and standardization (FRAMA)
In r-c frames infilled with masonry (“framed-masonry”) the infill walls stiffen the frame and reduce the first-mode period leading to a reduction in drift response to strong ground motion. At the same time, the addition of the masonry w all to the frame tends to increase the base-shear response and reduce the drift capacity of the structure. The increase of lateral force and reduction of drift capacity leads to serious vulnerabilities unless proper proportioning is exercised. For frames with competent walls, the challenge for safe and economical design is to be able to take advantage of the stiffening but to make certain that the increase in lateral forces and reduction in drift capacity do not handicap performance. Available evidence has pointed out that shear strength of the confining column is the ”Achilles’s heel” of the system. Solution of the problem requires understanding the behaviour of masonry and concrete subjected to dynamic and random loading reversals, a challenge that demands testing under reasonably realistic conditions for confident analysis of the problem and its generalization.
This is a proposal to investigate the safety and behaviour of buildings with masonry infilled r-c frames through near full-scale dynamic earthquake simulation tests accompanied by supporting pseudo-dynamic tests of structural assemblies and components and by calibrated analytical solutions. The overall goal is to develop pragmatic methods, by cooperative efforts of team members and co-working international body, for design, safety evaluation and standardization. Because framed-masonry serve both architectural and structural demands efficiently, people in seismic regions live and will continue to live in buildings of this type. An organized solution of the safety of such construction is essential. This proposal intends to put the ”framed-masonry” composite up as a full-fledged building type, it is ”transformative” and will change design practice.
The vulnerability of unreinforced masonry to earthquake became evident to industrial society as early as in Naples, 1857, (Mallet, 1862) and Messina, 1908, (Morris 1909). Specific flaws in unintentional frame-wall systems were identified in the aftermath of the 1963 Skopje, Macedonia, earthquake (Berg, 1963; Muto, Okamoto & Hisada, 1963; Leeds, 1964; Sozen, 1964; Unesco, 1968):
- weakness introduced by openings in the wall,
- captive columns,
- out-of-plane collapse of walls, and
- column failures under reversals of combinations of shear and tensile or compressive forces.
These flaws have continued to cause tragic consequences in subsequent urban earthquakes (Aschheim, 2000; Sezen, 2003), while most recent examples occurred in profusion in Wenchuan, China (2008) and L’Aquila, Italy (2008).
The Kocaeli (1999) and Duzce (1999) earthquakes and the development of major earthquake research centres within the European Union led to a wealth of experimental work and brilliant new perspectives (see references by Calvi, Carvalho, Dolšek, Fajfar, Fardis, Negro, Žarnić, among others). A review of the literature shows that consensus on the effects of the frames and in-plane masonry walls interaction is still in lacking. Some researchers have suggested that infill walls have led to collapse of many buildings (Aschheim, 2000; Sezen et al., 2003, Kyriakides, 2008) and that infill walls may affect the response of frames detrimentally (Murty et al., 2006). Some others have suggested that masonry infill panels may be beneficial (Akin, 2006; Hassan, 1996; Fardis and Panagiotakos, 1997; Henderson, 2002; Mehrabi et al., 1997). Dolsek and Fajfar (2008) captured the essence of the problem stating: “The infill walls can have a beneficial effect on the structural response, provided that they are placed regularly throughout the structure, and that they do not cause shear failures of columns.” The existence of contradictions in the views of the research community have led to the deconstruction of the frame-wall system by many regional building codes that contain warnings about the interaction of frames and walls but are mostly silent on providing recommendations and bounds on their proper proportioning. This is the driving reason for the proposed project.
In the new buildings designed in accordance with the European Standard, EN 1998, the masonry infill is threat as a source of structural additional strength and so called “second line defence”. However, the reduction of input seismic action, as a result of possible favourable infill effects, is not allowed. Considering this, design of reinforced-concrete (r-c) buildings with masonry infill according to EN 1998 is on the safety side but this design is not rational because it leads to significant increase of reinforcement in the structural r-c elements when compared to the design of bare r-c frame (where the infill acts only as a dead weight on structure). The problem is even more complex in the case of seismic performance evaluation of the existing r- c buildings with masonry infill. It was previously stated that the influence of the infill is most significant when the structural system itself doesn’t possess adequate seismic resistance, which is often case in large number of substandard r-c buildings constructed before 40-50 years in ex-Yugoslavia countries, (before implementation of seismic codes), as well as in the case of newly designed buildings without respecting capacity design approach. In such buildings the explicitly consideration of infill in analytical model and their verification are necessary.
Solutions to problems related to captive columns are self-evident (non-intentional structural masonry infills) and are controllable by architectural decisions. Problems related to openings, out-of-plane collapse, and column strength under shear and axial load demand special attention by structural engineers. Of the three, the last is by far the most critical problem leading often to building collapses causing heavy casualties. Because the failure of the reinforced concrete column in combined shear and axial load leads immediately to visually impressive damage to the filler wall and because the modelling of the masonry wall appears to be a more challenging task than that of the column, the criticality of the column strength has not received sufficient attention.
In multi-story construction, the most important attribute of the structure is its capability to retain its integrity at story drift ratios on the order of 1.5%-2%. A recent test of a full-scale three-story structure by Pujol and Fick at Purdue university has demonstrated that drift ratios of that magnitude can be achieved by a r-c frame with solid filler walls provided the columns have the ability to sustain the required shear force under reversals of shear and axial forces. Arguments stated above are strengthened by the results of neural-network analysis (Kalman Šipoš and Sigmund, 2013) based on EDIF Database formed by test data of 113 tests of one- bay one-story infilled frames available in literature:
- failure mode depends on the relative strength ratio between the r-c frame and masonry infill and occurs in weaker component,
- failure mode of the r-c frame depends on both the relative stiffness ratio of beam to column and column longitudinal reinforcement ratio,
- failure mode in masonry infill depended on the masonry thickness,
- yield drift depends on both the height to length ratio of the system and magnitude of the vertical axial forces acting on columns,
- ultimate drift depends on the height to length ratio of the system,
- yield force depends on both the longitudinal reinforcement ratio of columns and compressive strength of masonry infill and
- ultimate force depends on the height to length ratio of the system, compressive strength of both concrete and masonry infill wall and also on magnitude of the vertical loading.
Explicit recognition of the pivotal role of column shear strength introduces a new and important factor in the evaluation and design of reinforced concrete frames with infill walls.
The aim of this research proposal is to investigate, through dynamic earthquake-simulation tests, the safety and behaviour r-c frame system containing infill masonry walls since these systems serve both architectural and structural demands efficiently. Moreover, the majority of population in earthquake-prone zones live and will continue to live in such buildings, of which safety rely on frame-wall composites. The overall goal is to develop both pragmatic procedures to assess the safety of existing framed-masonry systems and design methods for new buildings. The proposed project is going to be initiated by tests of full-scale buildings at one of the main EU laboratories at the Institute of Earthquake Engineering and Engineering Seismology, University Kiril and Metodij at Skopje, Macedonia, FYR, continue with supporting tests and studies at GFOS, UNIOS and be completed by production of appropriately vetted assessment and design methods by the Croatian and International Advisory Group team members.
The specific goals of initial phase are to determine:
- the relationship between drift capacity and properties of the framed-masonry system controlling drift capacity,
- the stability of the masonry wall subjected to out-of-plane inertia forces,
- the effect of openings in masonry walls on response of the frame-wall system,
- the development and calibration of a new sensor to detect crack development and enable remote sensing of safety state of a building after an earthquake.
The definition of the test structures is governed with the aim to determine the failure mechanism of the column in shear. Studies of existing laboratory and field evidence (References) highlighted five important constraints in development of proper experimental structure. It is essential:
- to test the specimens in a dynamic (earthquake simulation) environment,
- to have multiple stories in order to approach the fluctuations of axial force that a structural system which involves the interactions of a reinforced concrete frame with masonry infill walls would experience,
- to use near full-scale materials and dimensions (masonry units, concrete aggregates, reinforcement, and mortar joints),
- to include an intermediate column (a column between two walls) in the test structure,
- to limit the building footprint to approximately 5 by 5 m and its weight to 45 tons because of the limits of available earthquake-simulation systems.
The constraints have resulted in the choice of a three-story test structure with two bays in the assumed N-S direction and one bay in the E-W direction. The height of the building is planned to be 5 m above the test platform. It is planned to use 150 mm square columns reinforced with approx. 2% of the longitudinal reinforcement. The beams will measure 150 mm wide by 200 mm deep and will have reinforcement ratios of approximately 1.5% top and 1% bottom. The masonry wall will be assembled using hollow clay masonry units of Type 2a with compressive strength of 10 MPa. They will be laid using M5 mortar. Floor diaphragms will be 90mm thick reinforced concrete slabs monolithic with the girder. The wall in the mid-frame on E-W direction will have large openings. Target concrete strength is 30 MPa (at 28 days). Reinforcement will be cut from deformed-bar stock with a nominal yield stress of 500 MPa.
After the first series of tests, the test structure will be repaired and tested again until collapse. The object of this second test is to investigate whether the framed-masonry, consisting of repaired r-c frame and masonry infill, has enough displacement capacity to sustain displacement cycles in the non-linear range. The main aim in seismic rehabilitation is to upgrade the strength, ductility and the lateral stiffness. Strengthening and/or repair of individual structural members (beams and columns) become feasible when the number of members to be rehabilitated is limited and the lateral stiffness of the building is adequate. If a great majority of the frame members have ductility problems and/or if the frames do not have adequate lateral stiffness, the rehabilitation by introducing infilled frames becomes an attractive solution and this option will be tested experimentally.
A synergistic by-product of the proposed project will be the opportunity to use the test structures as platforms for developing a much-needed technology, remote sensing of the state of safety after earthquakes of masonry and reinforced concrete buildings. Inspired by the work by Wood (2001) and Morita (2006), we are developing a low-cost (can be manufactured using inexpensive off-the-shelf components), low-data-rate sensing system to provide a categorical “up or down” indication of critical damage in a reinforced concrete or masonry structures. The driving criteria in the design of the sensor are simplicity of its manufacture, its ruggedness, and its ability to discriminate crack width.
The well-known complexity and randomness of critical crack trajectories in masonry poses both a challenge and an opportunity. The decision process for sensor length and locations will have to start with “blind predictions” followed by those made on the basis of analytical modelling based on observations. Undoubtedly, a positive outcome of the study will transform the remote-sensing technology for post-quake safety of the urban environment.
Before initiation of the experimental research two competitions will be announced to the research community:
- Development of a new sensor designed to gather critical information on the state of safety of masonry walls.
- Blind prediction of tests outcomes.
It is anticipated that those competitions, popular in engineering community (e.g. Hyogo Earthquake Engineering Research Centre, 2009; PEER and NEES, 2010; CUEE, 2011; 15WCEE, 2012), will result in innovation and creativity. Nevertheless, the competition would act as an additional tool for spreading the word about the project.
Phase II follows the tests on the earthquake simulator and is focused on static-cyclic component tests at the GFOS, UNIOS .Certain elements of this phase are known from the previous synergetic project in Croatia (“Seismic design of infilled frames”, Vladimir Sigmund) that covered experimental quasi-static tests of strong (steel- and seismically designed frame) and weak (confined-masonry frame) with strong, medium and weak masonry infill. This research was done within the Ph.D. theses finished at the GFOS, University of Osijek. Other elements of this phase will be agreed after the earthquake simulation tests and after finishing the tests on the non-ductile infilled frames and damaged r-c frames strengthened by masonry infill (partial fulfilment of Ph.D. theses within the mentioned previous project). Up to five different models are planned to be tested in this phase. Extensive test data will be acquired, crack trajectories and widths on concrete and masonry within this phase will be monitored visually accompanied by a 3D optical measuring system.
A number of distinct approaches in the field of analysis of infilled frames since the mid-1950s have yielded several analytical micro- meso- and macro-models (Asteris, P.D. et. al, 2011 and 2013). These studies stressed that the numerical simulation of infilled frames is difficult and generally unreliable because of the very large number of parameters to be taken into account and the magnitude of the uncertainties associated with most of them. Improvements of this situation will be sought within the Phase III of the project. Within the Croatian research project titled “Seismic design of infilled frames” (Sigmund, 2007-2013), an information data base of available experimental data arranged in a protocol for presenting the existing data in a performance format has been created along with the Artificial neural networks that gave reliable estimates of the FRAmed-MAsonry behaviour under earthquakes. Analytical investigations considering the influence of infill on seismic performance of r-c frame structures are under way providing a wealth of information and pointing out the evident lack of knowledge. An overview of different analytical models proposed for the analysis of inﬁlled frames has been done in the theses done at the GFOS, UNIOS (2007-2013), based on the previous and original experimental results and observed stiffness and strength degradation of FRAmed-MAsonry structures. Conclusions will be re-checked with the new experimental data and practical recommendations for implementation of different numerical models will be presented based on the calibration from results of the previous phases. Work in this phase will be supported with the ongoing bilateral Croatia-Macedonia research project: “Nonlinear analysis of r-c frames with types of infill characteristic for Macedonia and Croatia” (2012-2014).
Expected results of this Phase are refinements of design procedure, including explicit consideration of the infill with appropriate stiffness values that ensure adequate performance of the newly designed infilled frame structures in seismic conditions. Final result will lend a fully calibrated methodology that leads, with local site hazard input, to a procedure that is sensitive to variations in seismic intensity and construction tradition. This methodology should make calculation simpler, establish the minimum material requests more corresponding to reality and enable building of safer and cheaper infilled-frames buildings, as some of the unknown regions will be cleared. The desired performance levels will be corresponding to demand and ideally will involve a minimum of design effort.
This project will be fully coordinated by the International Advisory Group (IAG) formed from the well-known researchers in the field (joint “Council”) and team members. A consensus of the joint “Council”, and its advice, will be sought at each of the projects Milestones and the final Phase IV of the project shall represent a cooperative effort by the joint “Council” (co-working international body from Croatia, Macedonia, Serbia, Bosnia Herzegovina, Slovenia, Germany and USA). Meetings of the joint Council will be mostly through internet and final result of the project will be consensus procedures for design and safety evaluation of the FRAmed-MAsonry building structures.
The special and distinguishing features of the proposed research project are:
- This proposal intends to put the “FRAmed-MAsonry” composite up as a full-fledged building type.
- Commitment of the researchers to the development of a set of design recommendations for safety assessment and design.
- Testing of a building complex in a large scale under real conditions to gather physical data on a detail.
- Testing of a set of model frames in a scale 1:2 to get an in-depth understanding of the importance of various parameters that influence the response.
- Development of the fully calibrated methodology that leads, with local site hazard input, to a procedure that is sensitive to variations in seismic intensity and construction tradition. It will bring improvements of the “infills provisions” of EN 1998-1 (new design methodology and stricter quality assurance requirements in the infills construction). This methodology should make calculation simpler, establish the minimum material requests more corresponding to reality and enable building of safer and cheaper FRAmed-MAsonry buildings.
- Development of “infills provisions” for EN 1998-3 will be proposed as it is urgent and cost effective in the context of assessment and retrofitting of existing buildings since, in many cases, infills represent a significant proportion of its weak framed structures.
- Development and calibration of a new sensor to detect crack development and enable remote sensing of safety state of a building after an earthquake.
The knowledge to be generated by this research project will contribute to the safety of urban populations in seismic regions. It will lead to elimination of massive economic and human losses in future earthquakes. In addition, the necessary video conferencing and international participation as team members of the well-known experts, will involve Croatian researchers in a rich educational and cultural activity.