Response of primary and secondary systems under dynamic excitation

Engineers have gained much knowledge on how a primary system behaves under harmonic excitation and earthquakes, but they do not know as much about how a secondary system affects the behavior. In seismic areas, buildings (primary system) are built to withstand the impact of the earthquake, but the effects on secondary systems are often not taken into consideration when the structure is designed. In the past 40 years a lot of research has been done on developing methods to analyze the nonstructural elements during seismic excitation. However, these methods were mainly focused on the safety of critical equipment in, for example, power plants. In the later years it has been shown that the non-structural elements in conventional buildings also should be taken into consideration in the earthquake design process. A floor response spectrum from the primary structure is often used to estimate the dynamic response of the secondary system, but one problem by doing so is that it does not take into consideration the effect of the interaction between the two systems. The so-called Cardington Building, the primary system used in this thesis, was a seven storey in-situ concrete building erected inside an old Zeppelin hanger which housed the Cardington Laboratory near Bedford, UK, owned and operated by the BRE (Building Research Establishment Ltd). The building was built in 1998, and was a part of the European Concrete Building Project (ECBP). It was constructed like an office block, and its goal was to provide improved design codes, especially for the dynamic properties of the concrete structure. The analyses in this thesis are based on a numerical model of the Cardington Building made in a finite element program called Ruaumoko. The model is calibrated against experimental results from full scale tests performed by Jónas Thór Snæbjörnsson, Ódinn Thórarinsson and Símon Ólafsson in 2000 [2]. The FE model in this thesis is exposed to harmonic excitation in the first tests and excited by earthquakes in the second tests. Also, the seismic coefficients obtained from the earthquake excitations have been compared with the seismic coefficients calculated by equations from the Eurocode 8 standard. From the harmonic excitation tests it is seen that the secondary systems with their natural frequencies inside the resonant frequency range of the primary system get a much higher displacement than the ones outside the resonant frequency range, but in exchange the displacement of the primary system is reduced. For secondary system 6 the displacement of the primary system drops 34 % for the fourth floor and 35 % for the top floor (Figure 5.10 and Figure 5.11). This is true for secondary systems with both 0.8 % damping and 5.0 % damping. When the earthquake tests are performed it is seen that the floor response spectra method gives values that are generally higher than the values computed from the numerical model when the secondary system is attached to the primary system, see Table 5.10. It is seen that the natural frequency and the damping ratio of the secondary system affects the response generated by the two methods of measuring in different ways