You are here

Research Papers

Issue date: 
Category: 
Engineering
Paper number: 
405

Sliding hinge joint connection with Belleville springs

Shahab Ramhormozian (supervised by Charles Clifton), University of Auckland (EQC funded project 14/U687)

Non-Technical Abstract

Following a severe earthquake, modern traditionally designed and well-built buildings are expected to suffer extensive damage, causing the building to be demolished (as has happened in Christchurch). A new approach to address this issue is low-damage design of structures. Building owners, occupant businesses and insurers find low-damage design more advantageous than traditional design. There have been a number of systems proposed by researchers to make the low damage design possible by dissipating part of the earthquake imposed energy, to minimize damaging seismic forces.

A unique low damage system was developed by Associate Professor Charles Clifton from 1998 to 2005. This system is named Sliding Hinge Joint (SHJ) and has been widely in New Zealand, making the SHJ a very successful low damage system which has been implemented in practice. This performance has also been tested in the field; in August 2013 a severe earthquake hit the Te Puni village buildings at Victoria University, Wellington, which use the SHJ system, were built in 2008 and which won the IStructE international engineering award in 2009. These buildings had no damage. Similarly in November 2016 the Kaikoura earthquake also caused no damage, despite being strong enough to require the demolition of several modern buildings in Wellington.

Despite this, the performance of the SHJ is not ideal. It permanently loses strength after a very severe earthquake, necessitating retightening of the bolts and may not always bring the building back to its pre-earthquake position. The focus of the current research is to eliminate those less than desirable characteristics. This will transfer a successful low damage system (SHJ) into an ideal no-damage resilient system.

SHJs with Belleville springs (BeSs) are energy dissipating components of the building which act similar to a self-correcting circuit breaker in an electrical wiring system. They remain rigid under minor earthquakes to provide building’s integrity, stably slide under severe earthquake to dissipate high energy and prevent other damage to the building’s members, return to their initial position to again provide the building’s integrity following the earthquake, and return the building to its pre-earthquake as-built condition.

By this means that, not only is the building collapse prevented, but immediate functionality and occupancy is achieved following a major earthquake. This will eliminate heavy economic losses due to post-disaster repair as well as eliminate the cost of the building’s closure downtime. The SHJ with BeSs 1) can dissipate high energy 2) is very cost effective 3) is easy to design and build 4) is practical 5) is architecturally versatile 6) is damage-free 7) Has a stable and repeatable seismic behaviour.

The SHJ with BeSs uses readily available components such as Belleville spring and high strength bolt which are designed and arranged in a smart way to add no additional cost or complexity to the current system but perfect the functionality. The research on this has been focused on detailed mathematical, numerical, and experimental aspects of the subject to solidly support the SHJ with BeSs.

Technical Abstract

The Sliding Hinge Joint (SHJ) is a low damage joint for moment resisting steel frame (MRSF) seismic-resisting systems. It is ideally intended to be rigid under serviceability limit state (SLS) conditions. For greater shaking, beyond the ultimate limit state (ULS), rotation between the column and beam is expected. At the end of the shaking the joint is expected ideally to seize up and become rigid again. A key component in the sliding hinge joint is the asymmetric friction connection (AFC) which allows large beam-column rotation with minimal damage through sliding. However, previous research has shown that, after a severe earthquake, the post sliding strength and stiffness of the SHJ connection as currently applied is reduced, such that re-tightening or replacement of the bolts is likely to be needed. This is because the AFC bolts lose part of their initial tension during joint sliding. Hence the joint falls short of meeting one of the key original low damage performance requirements of not requiring any structural intervention following a severe earthquake. In this research the remedy to this shortcoming was proposed as using Belleville springs in the SHJ’s AFC in the optimum way. An innovative form of using Belleville springs was proposed to, in addition to improving the post-earthquake SHJ’s strength, improve the self-centering ability of the SHJ and the building. The following points are the main outcomes of the project:

• Establishing the optimum use of the Belleville springs. This removes the concerns about the post-earthquake elastic strength loss, damaging prying effects, and the variations of the bolt tensions during sliding.

• Establishing the optimum level of installed bolt tension in the AFSHJ.

• Developing a methodology to tighten the HSFG bolts with BeSs in the bolt’s elastic range. This removes the concerns about the delivered installed clamping force in the friction sliders.

• Establishing the optimum surface preparation/roughness level for the AFC plies sliding surfaces. This also removes the CoF variability concerns about the friction sliders.

• Proposing required changes on current NZ/Australia bolting standards. This removes the concerns about the reliability of the HSFG bolts delivered installed tension.

• Developing a dynamic SDOF SHJ model to investigate the effect of dynamic loading frequency, mass, and wind down on the static and dynamic self-centering capability at component level.

• Experimentally investigating the shim-less AFC, AFC with TiN coated shims, and AFC with abrasion resistant cleat and shims.

• FEM modelling of the SHJ AFC with and without BeSs to numerically investigate the effect of BeSs, optimum bolt tension, effect of number of bolt rows, effect of prying actions, effect of plies thickness reduction.

• Developing the AFC bolt model to design the AFC. This is based on the first principals, will explain in details the behaviour of the AFC, and gives the modified design procedure.

• Developing a MDOF SHJ model to research the SHJ dynamic self-centring capability using SAP2000 considering everal parameters such as column base rotational stiffness, type of the friction damper, the additional linear elastic spring between the column and beam, and stepping column base.

• Pre and post-earthquake system identification of the Te Puni Village SHJ building using SHM data.

Page last updated: