Post-tensioned Timber Frames with Supplemental Damping Devices
Thesis DisciplineCivil Engineering
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy (PhD)
In recent years the public expectation of what is acceptable in seismic resisting construction has changed significantly. Engineers today live under demands which are far more intensive than their historical counterparts and recent seismic events have shown that preserving life is no longer sufficient, and a preservation of livelihood is now the minimum. This means that after a major seismic event a building should not only be intact but be usable with no or minimal post-quake intervention. In addition to this already high expectation these demands must be met in a green and sustainable fashion with minimal (or even negative) environmental impact. This doctoral project looks to further advance the research into a new and innovative method of timber construction which satisfies (and exceeds) these demands.
In response to these higher expectations recent developments in the field of seismic design have led to the development of damage control design philosophies and innovative seismic resistant systems. Jointed ductile connections for precast concrete structures have been implemented and successfully validated. One of these systems, referred to as the hybrid system, combines the use of unbonded post-tensioned tendons with grouted longitudinal mild steel bars or any other form of dissipation reinforcing device. During the controlled rocking of the system under seismic loading the post-tensioning provides desirable recentering properties, while the devices allow adequate energy dissipation from the system as well as increased moment resistance at column bases and beam-column connections.
The hybrid concept is material independent and in 2004 an extensive campaign was begun to investigate the performance of the hybrid system when applied to large engineered timber members. Numerous small and large scale tests on both subassemblies and full buildings were performed showing that post-tensioned timber meets the seismic resilience demands now imposed by society. Recently this technology has also been applied in practice with over ten structures now using post-tensioned timber walls or frames, or a combination of the two, in New Zealand.
In-spite of the extensive research effort and the acceptance and adoption in practice of post-tensioned timber as a structural system, significant work was still required in the review and refinement of both the system itself and the analytical and numerical methods used to predict and analyse structural performance. The objectives of this research were to review and refine comprehension of the static and dynamic response, analytical and numerical modelling, and design of post-tensioned timber frames under lateral loading. In order to do this a three phase experimental testing campaign was devised and performed including quasi-static testing of an angle dissipative reinforcing device, quasi-static testing of a full-scale beam-column joint and the mono-directional dynamic testing of a 2/3rd scale three storey frame. All testing used glue laminated timber, which had not been previously used in post-tensioned timber structures.
Insight gained from the experimental testing was used to analyse and refine existing analytical modelling techniques. These techniques were split into two categories: 1) modelling of the local behaviour of a post-tensioned timber beam-column joint, with particular focus on stiffness and energy dissipation capacity, and 2) evaluation of the seismic demand (in the form of design base shear) on post-tensioned timber frames looking at current Force Based (FBD) and Displacement Based (DBD) design methods.
This analysis led to the development of recommended alterations in the existing beam-column joint analytical procedure enabling the procedure to provide better prediction of initial and post-yield stiffness. Analysis of the FBD and DBD procedures showed that both methods are capable of providing accurate prediction of seismic demand provided correct assumptions are made regarding system ductility and damping characteristics. Recommendations have been made on how designers can ensure that assumptions are either sufficiently accurate in the beginning of a design or require minimal iteration to be performed. Current numerical modelling techniques have also been compared against the quasi-static and dynamic testing results providing confidence in their accuracy when applied to post-tensioned timber frames. Modelling techniques were also extended to the widely used SAP2000 modelling programme which had not been previously used in post-tensioned timber research.
Although many observations and conclusions were made, a common theme continued throughout this research. This was the importance of the deep understanding of displacements within a post-tensioned timber frame and the impact of these displacements on frame performance. Displacements occur throughout a frame in dissipative reinforcing devices, in the connection of these devices, in beams, columns and joint panels as well as at the interfaces between members. When these displacements are allowed for through proper design excellent seismic performance, possible using this innovative system, is obtained.