| Abstract (english) | Existing unreinforced masonry structures are sensitive to seismic excitation and pose a potential risk in the event of new strong shaking. The large number and age of such structures exacerbate the problem. Therefore, the renovation and regular maintenance of such structures is extremely important. In order to make the renovation process more efficient, it is necessary to assess the current condition of unreinforced masonry structures as accurately as possible. For this purpose, it is necessary to determine their mechanical properties, which have certainly deteriorated over time under the influence of weathering and are often even unknown because there is no project documentation from the construction period. For this reason, in-situ tests on such structures are an essential step in the restoration process. Normally, when testing masonry structures, various methods are used that are more or less invasive to the structure itself. In the context of this work, attention has been focused on several of them. The focus is primarily on the flat-jack tests, but the shear and the sonic pulse velocity tests are also used. As part of the research, more than 30 buildings were tested at over 60 test locations.
The main hypotheses are as follows:
H1: The recommended value of the modulus of elasticity for existing masonry
structures given in the new proposal of the EN 1998-3 standard is overestimated.
H2: The recommended value of the initial shear strength for existing masonry
structures given in the new proposal of the EN 1998-3 standard is underestimated.
2. Literature review and experimental procedure
There are various non-destructive, semi-destructive and destructive test methods for existing masonry structures. The Eurocode HRN EN 1998-3 recommends several of them, including the flat-jack test, the shear test, and the sonic pulse velocity test. The flat-jack system is a semi-destructive test method for testing existing masonry structures [50]. It was originally used in geotechnical engineering to measure the stress state in rocks and to determine the deformability of the rock mass. At the end of the 20th century, it began to be used for structural testing, so that it is also used in the assessment of masonry and concrete dams [51], [52]. The first application of the method in masonry structures dates back to 1981 [55]. Initially, it was used to determine the state of vertical stress in masonry and then to determine the deformation properties of the masonry [56]. Since then, the method has gradually evolved, so that today there are several types of flat-jack. The flat-jack consists of two metal plates that are welded together at the edges. It is connected to a hydraulic jack via a valve and a hose, which is used to generate pressure that is controlled by an analog pressure gauge. There are also various shapes of flat-jacks, such as rectangular, semi-rectangular, circular and semi-circular. Depending on the shape of the flat-jack, there are different ways to create openings in the masonry. A drill can be used for rectangular openings, while a drill with a suitable guide or an eccentric circular saw can be used for semi-rectangular, circular and semi-circular openings. As for the accuracy of the method, it can underestimate and overestimate the actual value when determining the state of stress in masonry, and the coefficient of variation can be up to 20 % [49]. When determining the deformation properties of masonry, the coefficient of variation can be up to 24 % [68]. The flat-jack method enables the mechanical characterization of masonry [70]. Three phases of the test can be distinguished. The first phase uses one flat-jack and gives the values for the state of vertical stress in the masonry. In the second phase, two flat-jacks are used to determine the behavior of the masonry under compression, i.e. the relationship between stress and strain. In the third and final phase, two flat-jacks and an additional third jack are used to determine the shear properties of the masonry. The test methodology itself is specified for all three phases in the form of American [49], [68], [71] and international [48], [72], [73] guidelines. Before the actual test, it is necessary to calibrate the flat-jacks. Shear properties of masonry are one of the most important properties, considering that they significantly affect the shear capacity of masonry walls in plane, which is often the dominant mode of failure. Determining these properties is not an easy task. There are various methods that are carried out directly on site, such as the diagonal compression test, the shear compression test and the shear test (shove test). There are also laboratory tests such as the couplet, triplet and Van der Plujim test [98], as well as cyclic tests on large wall samples. The shear properties are determined in the laboratory in accordance with the HRN EN 1052-3 standard [99], while the field test is carried out in accordance with the ASTM C1531 [71] and RILEM MS-D6 [73] standards. The field shear test is a partially destructive test, but certainly less destructive than the other tests listed, which are carried out in the field. For this reason, it is very often used in the assessment process of a masonry structure after an earthquake to determine the seismic capacity required for reconstruction. During the implementation of this research, on site shear tests were performed. As for the equipment itself, the same equipment was used as in the shear test with vertical stress control. In other words, a hydraulic pump in combination with a standard jack with a capacity of 200 kN and a power tool to remove the last layers of plaster and make openings for the jack. The test determines the local shear resistance to sliding parallel to the horizontal joint of the mortar in the load-bearing wall. The local properties tested are also assumed for the global properties of the walls in further analyses and calculations. Determining the speed of sonic elastic waves through masonry using a sonic test is a well-known and widely used method. The beginnings of research into the sonic method for testing masonry structures date back to the eighties of the twentieth century [107], [108]. The largest application is in the field of determining the elastic properties of concrete structures, but also for the qualitative assessment and detection of cracks and voids in concrete and masonry structures [109], [110]. Considering the fact that poor connection of masonry layers, voids and various damages reduce resistance of masonry elements, it is of great importance to assess the condition of such masonry [111]. The method of measuring the speed of a sound wave belongs to the group of non-destructive testing, which uses the mechanism of propagation of elastic waves through the material and their repulsion and reflection [112]. Low-frequency waves are generated by applying mechanical energy by hitting the masonry with a hammer. By measuring the velocity of the wave, it is possible to correlate it with the mechanical properties of the masonry, such as the modulus of elasticity. The MQI method or Masonry Quality Index method is a method for determining the mechanical properties of existing unreinforced masonry structures by visual analysis. The method belongs to the category of non-destructive methods as only the plaster has to be removed without further destroying the structure. The entire analysis is carried out on the basis of a visual inspection and a simple calculation. The result is given at the end in the form of a range of values for the mechanical properties of the masonry. These include the modulus of elasticity E, the compressive strength fm and the shear strength of the masonry τ0. The range of values given is wide, and additional destructive tests are required to determine the mechanical properties more precisely. The results of the analysis can also be used to compare and verify the results obtained with other methods.
3. Case studies
This chapter gives a brief overview of the case studies on which tests were carried out during the experimental part of the research. The selected case studies represent characteristic unreinforced masonry buildings made of solid clay bricks in lime mortar with timber floor and roof structures in the city of Zagreb and the surrounding areas. A large number of the observed buildings serve public purposes and are protected cultural heritage, which emphasizes the importance of assessment and restoration of such buildings. The construction methods of the buildings observed range from the 13th century to the end of the 20th century and include various architectural styles. In addition to the presented facades of the observed buildings, the chapter also presents detailed floor plans of the buildings, which clearly outline the load-bearing structure. A total of twenty-two case studies are presented with the corresponding forty-five locations that were tested using the flat-jack method. All tests on the buildings mentioned were carried out as part of the scientific ARES project (Assessment and rehabilitation of existing structures – development of contemporary methods for masonry and timber structures) under the direction of assist. prof. Mislav Stepinac.
4. Experimental results
This chapter presents the results obtained when testing a regular unreinforced masonry made of solid baked clay bricks in lime mortar. The tests were carried out on a total of 22 case studies. The case studies are mainly located in Zagreb and in the area of Sisak and Glina. Several test methods were used, with a focus on the flat-jack method. Other methods used are the sonic method, the shear (shove) test and the visual method MQI (Masonry Quality Index. They were used to compare and evaluate the obtained results.
The test results showed a certain tendency towards asymmetric scattering, which was to be expected given that the tests were carried out on existing masonry structures. The results are presented in various ways, including tables, bar charts, box and whisker plots, and histograms overlaid with probability density functions. Suitable models to describe the obtained results of the mechanical properties of masonry were mainly gamma and Weibull distributions.
5. Comparison of experimental results
The use of several different test methods is certainly desirable, especially to increase knowledge of the observed structure, but also to compare and confirm the quality of the obtained results. With the intention of exploring specific methods, but also looking for a possible correlation between individual methods, several levels of comparison of the results obtained with different methods were carried out in this chapter. The methods whose results were compared were the flat-jack method, the sonic method, the shear (shove) test and the masonry quality index method. First, a comparison of the methods for determining the state of vertical stress in masonry was carried out. The test with a single flat-jack was primarily used for this purpose. Additional methods were also used to determine the state of vertical stress. These were the analytical method for assessing the state of stress by means of load analysis and the use of numerical models. The 3muri software package was used to create the numerical models. Based on the results, it can be concluded that the stress results obtained by testing with a flat-jack in most cases yield significantly higher values than values obtained through load analysis and numerical modelling. On the other hand, the stress results obtained from the load analysis in most cases give lower values than other two methods. A side-by-side comparison shows good agreement between the results of the load analysis and the results of the numerical model. This can be explained by the same load values and similar distribution assumptions. The difference between the results obtained by tests and additional methods can be partly explained by the limited knowledge of the structure itself, its layers, exact dimensions, load distribution and the state of the structure. Of course, the unreliability that may occur during the test should also be taken into account.
The use of a faster and less destructive test, such as a shear (shove) test, can be useful in assessing the condition of existing masonry structures due to the greater test coverage and the representativeness of the results for the whole structure. The disadvantage of the shear test is the lack of knowledge about the state of vertical stress at the test location, which must be evaluated analytically by load analysis or numerical modelling. For this reason, a comparison of the results obtained with the semi-destructive methods of the flat-jack test and the shear test was carried out. From the presented results, it can be concluded that the flat-jack method generally yields lower values for the initial shear strength than the values determined by the shear test. The possible cause of the different values lies in the method used to determine the shear properties, i.e. the initial shear strength of the walls. The shear test provides a shear strength value that must be further converted into the initial shear strength, taking into account the friction coefficient and the state of vertical stress. The mentioned two parameters must be assumed, i.e. the recommended values must be taken or determined by a load analysis. In this way, it is possible to misjudge their actual value. On the other hand, when preparing the measuring point for the flat-jack method, mechanical tools are increasingly used, which can damage the masonry through vibrations and thus negatively influence its mechanical properties. The use of non-destructive testing is of great importance in assessing the condition of existing masonry structures, especially those that are part of the cultural heritage. For this reason, a comparison was made between the results of the semi-destructive flat-jack method and the non-destructive sonic method. The conclusion is that the flat-jack method and the sonic method are good methods for determining the elastic properties, i.e. the modulus of elasticity of masonry. It is also strongly recommended to use both methods in the assessment of existing masonry structures in order to optimize the destruction of the structure and the tesing costs. The use of more than one method is also useful to confirm the results obtained with a particular method. Depending on whether the structure is part of the cultural heritage or not, the ratio can be optimized and a larger number of sonic tests or flat-jacks tests can be performed. By combining these two methods, the testing process itself is also accelerated and a larger amount of data is collected due to the non-destructive nature and speed of the sonic test. This gives a better insight into the condition of the existing wall and the results are representative of the entire structure due to the larger number of tests. Next, the results obtained by testing with flat-jacks were compared with the results obtained by analysing the masonry quality index. The values of compressive strength, modulus of elasticity and initial shear strength of the masonry were compared. The results show relatively good agreement, but in some case studies there are still considerable discrepancies. The results of the MQI method for the modulus of elasticity generally overestimate the values obtained with flat-jacks. For the initial shear strength, the results may be over- or underestimated depending on the case. As for the compressive strength of the masonry, the results of the MQI method in each case study significantly overestimate the value of the compressive strength extrapolated from the flat-jack test data. Finally, all methods were compared in terms of the value of compressive strength, modulus of elasticity, coefficient of friction and initial shear strength of the masonry.
6. Discussion
The existing masonry building fund varies considerably in terms of construction period, condition of materials, construction quality, maintenance and other local conditions. In addition, the presence of moisture, low temperatures, lack of power supply, organizational problems, non-functioning equipment and unforeseen obstacles or gaps in the walls can complicate, slow down and impede the execution of the test. For this reason, one's own experiences as well as the experiences of others are a great help in further research, especially for new, less experienced researchers. This chapter therefore describes the common difficulties that can arise when testing with flat-jacks, based on numerous tests with flat-jacks. The tests themselves are challenging in a number of ways and are by no means quick or easy. In addition, comparisons are presented between the results of the tests with flat-jacks and the recommended values of the new proposal of Eurocode 8 (EN 1998-3) and the Italian NTC standard for existing regular unreinforced solid clay brick masonry in lime mortar. The values of the modulus of elasticity, the coefficient of friction and the initial shear strength of the masonry obtained in the tests are very close to the recommended values.
The other test methods used, such as the shear test, the sonic test and the MQI test, are also discussed in terms of their advantages and disadvantages. In addition to the quantitative comparison of the results obtained with the methods listed above, a qualitative analysis of the same methods was also carried out. This type of comparison says more about the applicability and quality of the method than about the results themselves. The methods used were compared on the basis of thirteen criteria, which were grouped into categories relating to the performance of the test and the properties of the test results themselves. The MQI method is certainly the most favourable method in terms of performing the tests. The sonic and shear tests are similar, but the sonic test has more favourable performance criteria. The flat-jack system has by far the most demanding test procedure but is also the most favourable in terms of results. The other three methods are similar and have lower rated criteria in terms of results.
7. Conclusion
Based on the comparison of the test results with the flat-jack method and the recommended values given in the new version of the European standard EN1998-3, it was found that the test results of the existing masonry structures made of solid bricks in lime mortar indicate that the recommended value of the modulus of elasticity of the masonry in the standard slightly overestimates the actual value. The results of the coefficient of friction overlap very well with the recommended value from the standard and scatter evenly above and below the recommended value. The results of the initial shear strength are generally above the recommended values given in the standards. It can therefore be concluded that the recommended value for the initial shear strength specified in the standard underestimates the actual value. For the preparation of this thesis, all experiments were carried out on site on existing masonry buildings. The tests were therefore not carried out under laboratory conditions and certain conditions were not controlled, i.e. they were not constant. Future research could therefore preferably have a smaller scale of field tests, but with additional tests under controlled laboratory conditions. Based on experience of conducting the tests so far, some of the following ideas emerge that would be worth exploring in more detail:
• Investigate the effects of using an electric impact tool to remove plaster and bricks on the deterioration of the mechanical properties of walls;
• Investigate the effect of moisture used in making openings for flat-jacks on the mechanical properties of walls;
• Investigate the influence of unexpected vertical openings (old chimneys) in the wall at the test location on the results;
• Investigate the influence of boundary conditions when testing an isolated specimen in the laboratory and when testing a real structure; |
