Summary
Highlights
The presentation begins with an overview of the topic: evaluating the load-bearing capacity of existing structures with a focus on seismic resistance. It delves into the legal basis in Austria, explaining how norms gain legal character through provincial building regulations and OIB guidelines, referencing Eurocode 0 to 9. For existing structures, a deviation from current technical standards is permissible, but a guideline for assessing load-bearing capacity and serviceability must be applied. This involves detailed condition assessments (Bestandserhebung) across three levels, determining the depth of investigation required for modifications like door openings, affecting multiple floors, or involving significant load increases. The new ÖNORM B4008-1 specifically addresses the load-bearing capacity of existing structures, with prerequisites including a condition assessment and bringing the building closer to current standards, particularly for seismic design. Basic load combinations (snow, wind, live load) must always be met, and an increase in occupancy is limited to 50%.
The concept of safety in construction is discussed as a matter of probability, specifically accepted failure probabilities (acceptable risk). The semi-probabilistic safety concept of Eurocode 0, where design values of actions are less than design values of resistances, is explained. Meeting Eurocode 0 and 1-9 achieves a minimum reliability level of Beta 3.8. This corresponds to reliability class 2 and consequence class 2, which are equated in Eurocode 8 Part 3 and B4008-1. For new buildings, an accepted failure probability per building per year is 10^-6 for CC2 and 10^-7 for CC3. For existing buildings, these probabilities can be increased to 10^-5 for CC2 and 10^-6 for CC3, aligning with accepted failure criteria.
Austria's building stock is predominantly residential, with a significant portion (over 75% in city centers) being over 100 years old. Sustainable urban development emphasizes renovation and densification, often through adding stories to existing structures. This necessitates assessing the load-bearing capacity of these existing parts. The knowledge about the construction and any prior damage is crucial, and inhabitants are invaluable sources of information regarding a building's history of damage. While new structural elements must comply with current Eurocodes, existing elements can be assessed using the new standard. Following Swiss precedents, Austria allows existing buildings to be reassessed with reduced reliability for seismic resistance. The seismic performance factor (Erfüllungsfaktor) is defined as the ratio of resistance to seismic action. Minimum performance factors are set to limit individual risk (probability of death for an individual in an earthquake). For a standard CC2 building, this factor is 0.25, meaning the structure can withstand 25% of the design earthquake. For schools, Switzerland opted for a minimum factor of 0.4. The relationship between the performance factor and failure probability is illustrated, though it's noted that building failure doesn't directly equate to individual fatalities, with roughly a factor of 10 difference.
Person occupancy, as defined in B 1990 Table B2, is relevant for risk assessment. For apartments, it's 2.35 persons. For seismic load cases, a time factor (e.g., 0.3 for schools) is applied. If occupancy increases, the overall risk must remain constant, meaning the failure probability must be reduced. A target failure probability and a required performance factor can be calculated based on this. The lecture then clarifies the definition of the seismic performance factor (Alpha), particularly noting a common misinterpretation in Eurocode 8 Part 3's national annex. It emphasizes that only the seismic action component should be scaled by Alpha, not the entire combined loading. An example calculation for a concrete frame illustrates this, showing a much lower Alpha when correctly applying the definition. Special attention is drawn to concrete frame corners, where older designs for static loads may inadequately reinforce for positive moments arising from seismic actions, potentially leading to small Alpha values.
Capacity verification according to EC8 Part 3 is a displacement-based method ensuring that demand (from the design earthquake) does not exceed the deformation capacity. For existing components, mean material properties are used, adjusted by 'confidence factors' based on the achieved 'knowledge level' about the structure. New components require nominal material properties. The knowledge level also determines the permissible calculation method. Acquiring knowledge involves reviewing geometry (plans, visual inspection, full inspection) and material properties. Eurocode 8 Part 3 and its national annex provide tables and flowcharts for determining knowledge levels. For example, a 'knowledge level 1' (limited knowledge) requires dividing mean material values by a confidence factor of 1.35, while 'knowledge level 3' (comprehensive knowledge) allows using mean values directly (confidence factor 1.0). For masonry, comprehensive knowledge requires at least two test series per 1000m² gross floor area, with each series comprising three individual tests or six rebound/penetration tests (for mortar and brick strength). This number can be reduced for lower knowledge levels.
For calculations, cracked cross-sections are generally assumed, pragmatically using 50% of the uncracked stiffness for masonry. Linear analysis methods (static or multimodal) require stiffening walls in both directions, continuous walls over all stories, rigid floor slabs acting as diaphragms, no level changes in slabs, and a stiffness ratio between the strongest and weakest seismic wall less than 2.5 to avoid excessive torsion. If these conditions are not met, non-linear methods, such as the static (pushover) method, must be used. Pushover analysis generates a capacity curve (base shear vs. roof displacement) by incrementally increasing horizontal forces, showing progressive damage and reduction in resistance. The capacity limit is often defined when the resistance drops to 80% of its peak. Older buildings, specifically those built before 1920 (Gründerzeithäuser), were typically not designed for earthquake loads, particularly in Vienna, which was considered outside hazardous zones. The first mention of seismic action in Austrian norms was in 1955, significantly lower than current standards. Earthquakes like the 1976 Friaul event led to the 1979 ÖNORM B4015, specifically for seismic forces on non-vibration-sensitive structures, and a revised hazard map with higher acceleration values.
The 1979 ÖNORM B4015 introduced a hazard map for Austria, showing significantly higher acceleration values compared to earlier guidelines, explicitly noting hot spots like Murau and Nassfeld. This map specified effective horizontal ground acceleration for a 100-year earthquake. With Eurocode 8, a reference return period of 475 years was established for the design earthquake, corresponding to a 10% probability of exceedance in 50 years. This led to a revised hazard map and its implementation in ÖNORM B4015 (from 1997) and subsequently Eurocode 8 (from 2009). Historical earthquake research provides crucial insights; for instance, a strong earthquake near Vienna in 1590 (intensity 8-9) suggests that a similar event could occur again by around 2065, based on a 475-year return period. The 1972 Sebenstein earthquake, affecting Vienna at about 60% of its design earthquake, showed that many older structures survived with minor damage, implying a seismic performance factor of at least 0.6 for these buildings. Current reference peak ground accelerations (aGR) for hotspots like Murau are around 1.17 m/s², while Vienna is 0.7-0.8 m/s², and Nassfeld is 1.34 m/s².
The differentiation in reliability is achieved through the importance factor (Gamma I), defined in Eurocode 8. This factor scales the reference peak ground acceleration based on the building's importance category (1 to 4). Category 4 buildings (e.g., hospitals, power plants), whose integrity is crucial during an earthquake, have a Gamma I of 1.4, increasing the return period to approximately 1303 years. Return periods depend on the slope of the hazard curve (exceedance probability vs. spectral acceleration). Different ground classes (A to E), characterized by the shear wave velocity (Vs30) in the top 30m, influence the shape of the elastic response spectrum, specifically the corner periods (TB, TC, TD) and the plateau height. The lecture then details the calculation of the elastic and design response spectra for different period ranges, incorporating the behavior factor Q for the design spectrum.
The relationship between acceleration response (Sa) and displacement response (Sd) is explained through basic structural dynamics: Sa = omega² * Sd. This allows for transforming the acceleration response spectrum into a displacement response spectrum, which shows constant displacements beyond a certain period (TD). Further transformation creates the Acceleration-Displacement Response Spectrum (ADRS) format, where displacement is on the x-axis and acceleration on the y-axis, with periods represented by radial lines from the origin. For designing seismic actions according to Eurocode 0, the design seismic action (AED) is determined by inertial effects of masses, which for Eurocode 8 consider permanent loads and a portion of variable loads (e.g., 30% for residential and office buildings). The equivalent static force method determines the seismic base shear from the design spectrum's acceleration, mass, and a correction factor (Lambda) for higher mode contributions.
The non-linear static (pushover) method, also known as the capacity spectrum method, involves applying monotonically increasing horizontal forces (with specific distributions, e.g., uniform or modal, as required by Eurocode 8) at floor levels' centers of mass. The resulting relationship between base shear and roof displacement is the pushover or capacity curve, which can be idealized (e.g., bi-linear). This curve is then transformed into an equivalent single-degree-of-freedom (SDOF) system. The 'target displacement' is the displacement of this SDOF system determined from the elastic response spectrum. The applicability of pushover analysis requires the first vibration mode to dominate and minimal torsional effects. By overlaying the SDOF capacity curve (resistance) with the ADRS (demand), the interaction point defines the system's performance. The seismic performance factor is then calculated as the ratio of the required displacement to the structure's actual displacement capacity. A critical Eurocode 8 requirement for displacement-based methods is that the capacity curve must extend to 1.5 times the target displacement, effectively adding a safety factor of 1.5 even after using mean material properties.
For masonry structures, Eurocode 8 Part 3 provides equations for flexural (rocking) and shear failure, specifying limits for inter-story drift (8‰ for flexural, 4‰ for shear). These allow generating bi-linear capacity curves for individual masonry walls. Assuming rigid diaphragms and no torsion, these curves can be combined to produce an overall building capacity curve, illustrating how stiffness diminishes as individual walls fail sequentially. For complex structures, specialized software like "Trimurti" (an Italian software) is used. While powerful, such software can act as a "black box," making results difficult to verify. The speaker advocates for performing a simple hand calculation (e.g., using the equivalent static force method) first to establish a basic understanding and plausibility check. This initial conservative estimate helps to validate complex 3D FEA results. An example from the Trimurti software for a specific building showed a minimum seismic performance factor of 0.42 for a ground-floor column, implying that its failure would lead to a partial collapse jeopardizing human lives. The presentation concludes by inviting questions and offering to share parts of the presentation.