Experimental and finite element investigations of the serviceability behaviour of CLT floors
Uí Chúláin, Caitríona
MetadataShow full item record
This item's downloads: 98 (view details)
Global environmental concerns with regard to sustaining our growing population is prompting the building community worldwide to increase urban density, while using optimised and sustainable resources. The traditional materials used for high density and high rise development are carbon intensive, in contrast timber construction sequesters carbon and is renewable. Therefore, there is a concerted effort in timber engineering research to offer credible solutions with respect to sustainable construction in contemporary design. There are numerous prefabricated timber engineered building elements that are suitable for modern buildings. A product which is emerging as the most viable alternative to concrete or steel construction for multi-storey development is cross laminated timber (CLT). This study aims to characterise the serviceability performance of CLT floor systems used in mid-rise buildings, focusing on floor deflection and excitation due to human activity within the range of people’s perception. The research includes experimental investigations exploring the serviceability performance of floors using particular connection details in keeping with the prevailing CLT floor-to-wall junctions that are suitable for mid-rise construction of residential, office, and educational buildings. The influence of non-structural elastomeric interlayers and the effect of non-structural added mass is also explored. The effect of CLT panels connected in parallel and the influence of the integration of the floor in the building are examined in field studies. The dynamic influence of floor voids to accommodate vertical circulation and building services are examined using a numerical modelling approach. This approach was also used to characterise the effect of non-structural intermediate supports and irregular mass distributions, representative of typical floor loadings. The floor assemblies tested were all substantially compliant with current European timber design frequency criteria. However, the field tests on multi-panel floors indicated a marked reduction in the natural mode separation. This is attributed to the floor widths and the number of panels connected in parallel. The mode separation and the number of first order modes below a defined range can prove significant with regard to compliance with velocity impulse criteria. The experimental testing showed that alternative assembly configurations did not substantially influence the rotational stiffness of the floor support. All tested floor experimental results indicate that the responses to pedestrian traffic will be transient. The results suggest that optimum CLT floor design, in non-industrial buildings, with respect to serviceability performance will more significantly rely on the floor’s mass and flexural stiffness, which are dictated by the floor panel geometry and intermediate supports. The numerical models showed that the natural mode shapes of the floor are significantly influenced by floor void positions and that the natural frequency and mode shape values are strongly effected by intermediate supports. All investigations showed that a CLT floor’s dynamic performance is substantially affected by added mass and the distribution of the mass over the floor area. Conclusions from this research will serve to advise building designers with respect to floor-to-wall connection assembly, floor geometry, floor voids, and void support locations, and the positioning of intermediate supports with respect to mitigating disturbing structural vibration in CLT floors due to pedestrian traffic and other human induced floor excitation.