New Zealand Society for Earthquake Engineering Conference
April 13-15, 2012, Christchurch, New Zealand
The Nelson Marlborough Institute of Technology Arts and Media building was completed in 2011 and consists of three seismically separate complexes. This research focussed on the Arts building as it showcases the use of coupled post-tensioned timber shear walls. These are part of the innovative Expan system. Full-scale, in-situ dynamic testing of the novel building was combined with finite element modelling and updating to obtain an understanding of the structural dynamic performance within the linear range. Ambient testing was performed at three stages during construction and was combined with forced vibration testing for the final stage. This forms part of a larger instrumentation program developed to investigate the wind and seismic response and long term deformations of the building. A finite element model of the building was formulated and updated using experimental modal characteristics. It was shown that the addition of non-structural elements, such as cladding and the staircase, increased the natural frequency of the first mode and the second mode by 19% and 24%, respectively. The addition of the concrete floor topping as a structural diaphragm significantly increased the natural frequency of the first mode but not the second mode, with an increase of 123% and 18%, respectively. The elastic damping of the NMIT building at low-level vibrations was identified as being between 1.6% and 2.4%
The ambient movement of three modern multi-storey timber buildings has been measured and used to determine modal properties. This information, obtained by a simple, unobtrusive series of tests, can give insights into the structural performance of these forms of building, as well as providing information for the design of future, taller timber buildings for dynamic loads. For two of the buildings, the natural frequency has been related to the lateral stiffness of the structure, and compared with that given by a simple calculation. In future tall timber buildings, a new design criterion is expected to become important: deflection and vibration serviceability under wind load. For multi-storey timber buildings there is currently no empirical basis to estimate damping for calculation of wind-induced vibration, and there is little information for stiffness under wind load. This study therefore presents a method to address those gaps in knowledge.
It has been shown that measurement of elastic constants of orthotropic wood-based panel products can be more efficiently measured by modal testing technique. Identification of vibration modes and corresponding natural frequencies is key to the application of modal testing technique. This process is generally tedious and requires a number of measurement locations for mode shape identification. In this study, a simplified method for frequency identification was developed which will facilitate the adoption of the vibration-based testing technique for laboratory and industrial application. In the method, the relationship between frequency order and mode order is first studied considering the boundary condition, elastic properties of the orthotropic panel. An algorithm is proposed to predict the frequency values and mode indices based on corresponding normalized sensitivity to elastic constants, initial estimates of orthotropic ratios and measured fundamental natural frequency. The output from the algorithm can be used for identification of sensitive natural frequencies from up to three frequency spectra. Then the algorithm is integrated with the elastic calculation algorithm to extract the elastic constants from the sensitive frequencies. The elastic constants of cross laminated timber panels were measured by the proposed method. The moduli of elasticity agree well with static testing results. The calculated in-plane shear modulus was found to be within the expected range.
This research paper deals with the evaluation of the dynamic modal vibration tests conducted on an innovative timber structure, the ETH House of Natural Resources. The building serves as a demonstrator of several innovative structural systems and technologies relating to timber. The main load-bearing structure comprises a posttensioned timber frame, which was subjected to modal vibration tests, firstly in the laboratory and, subsequently on the construction site. In this paper, the modal characteristics (eigenfrequencies, damping ratios and mode shapes), obtained from the laboratory testing campaign are presented. The modal vibration data is evaluated using polynomial and subspace identification techniques. The obtained results reveal that the structure exhibits pure translational, beam and column modes, as well as mixed beam-column modes. The bottom connection of the columns delivers significant influence on the modal characteristics, whereas the level of post-tensioning force yields no substantial influence in the modal characteristics obtained from low amplitude modal vibration tests.
Innovations in timber engineering have led to new slab systems built from engineered wood products like cross-laminated-timber (CLT). High stiffness of CLT can enable attainment of better vibration performances than is possible with traditional shallow profile-long span floors constructed from timber and other materials. However, realization of this depends on engineers being able to accurately predict effects various construction variables have on dynamic responses of CLT slabs. Past physical experiments have provided insights into those effects. However, testing is a very expensive and time consuming means of acquiring necessary knowledge. Discussion here addresses finite element (FE) simulations as a cost effective method allowing engineers to understand and assess relationships between design variables and dynamic responses of CLT floor slabs. Presented modelling techniques are verified by demonstrating close correlation between numerical predictions and experimental modal response characteristics of CLT slabs.
Ninth European Conference on Noise Control (Euronoise)
June 10-13, 2012, Prague, Czech Republic
In residential multi-storey buildings of timber it is of great importance to reduce the flanking transmission of noise. Some building systems do this by installing a vibration-damping elastic interlayer, Sylomer or Sylodyn , in the junction between the support and the floor structure. This interlayer also improves the floor vibration performance by adding damping to the structure. In the present work the vibration performance of a floor with such interlayers has been investigated both in laboratory and field tests. A prefabricated timber floor element was tested in laboratory on rigid supports and on supports with four different types of interlayers. The results are compared with in situ tests on a copy of the same floor element. The effect on vibration performance i.e. frequencies, damping ratio and mode shapes is studied. A comparison of the in situ test and the test with elastic interlayer in laboratory shows that the damping in situ is approximately three times higher than on a single floor element in the lab. This indicates that the damping in situ is affected be the surrounding building structure. The achieved damping ratio is highly dependent on the mode shapes. Mode shapes that have high mode shape coefficients along the edges where the interlayer material is located, result in higher modal damping ratios. The impulse velocity response, that is used to evaluate the vibration performance and rate experienced annoyance in the design of wooden joist floors, seems to be reduced when adding elastic layers at the supports.
This paper deals with a certain type of C.L.T. (Cross Laminated Timber) construction, in a residential building in Fristad, Sweden. The objective is to study impact noise transmission, at the lower frequency range (10-200 Hz), where wooden dwellings perform inefficiently, in terms of acoustic quality. The vibrational behavior of lightweight structures and specifically a multilayered floor separating two vertically adjacent bedrooms are investigated. A numerical model of the multilayered test plate, which includes sound insulation and vibration isolation layers, is developed using the Finite Element Method (F.E.M.) in commercial software. The design process, the analysis and improvement of the calculated outcome concerning accuracy and complexity are of interest. In situ vibration measurements were also performed so as to evaluate the structures dynamic behavior in reality and consequently the validity of the modelled results. The whole process from design to evaluation is discussed thoroughly, where uncertainties of the complex F.E.M. model and the approximations of the real structure are analyzed. Numerical comparisons are presented including mechanical mobility and impact noise transmission results. The overall aim is to set up a template of calculations that can be used as a prediction tool in the future by the industry and researchers.
Vibration serviceability of various types of timber floor systems has claimed much attention during past decades. Yet the definition of robustly reliable engineering design approaches has remained elusive, except in well-defined situations. Successful de...
During the last years the interest in multi-storey timber buildings has increased and several medium-to-high-rise buildings with light-weight timber structure have been designed and built. Examples of such are the 8-storey building Limnologen in Växjö, Sweden, the 9- storey Stadthaus in London, UK and being constructed at the moment, the 14-storey building Treet in Bergen, Norway. These are all light-weight and flexible structures which raise questions regarding the wind induced vibrations. For the building in Norway, the calculated vibration properties of the top floor are on the limit of being acceptable according to the ISO 101371 vibration criteria for human comfort. This paper will give a review of building systems for medium-to-high-rise timber buildings. Measured vibration properties for some medium-to-high-rise timber buildings will also be presented. These data have been used for calculating the peak acceleration values for two example buildings for comparison with the ISO standards. An analysis of the acceleration levels for a building with double the height has also been performed showing that designing for wind induced vibrations in higher timber buildings is going to be very important and that more research into this area is needed.