This project was initiated to resolve uncertainty regarding the long-term performance of epoxied rod connections in glue laminated timber (glulam), specifically the long term-deformations associated with beam-column moment joints. The excellent short-term performance of this type of connection is well known. However there is some scepticism in industry over the long-term performance. The behaviour of this connection when subjected to long term loading is thought to be primarily affected by creep in the timber and heterogeneity of the cross section. It has been hypothesised that the stress distribution at the beam-column interface may change over time leading to a localisation of stresses, rotations within the joint and ultimately, excessive deflections of the frame at serviceability limit states. Furthermore, crushing of the timber perpendicular to the grain may occur in the long-term as a result of the stress redistribution over time which could cause failure at ultimate limit state. In order to investigate the long term performance, two types of experimental testing were carried out in this project: (i) long term testing of several frames and lever specimens; and (ii) short term testing of several smaller glulam specimens. Four frames with a “C” shaped layout (i.e. one column and two beams) were tested. Two of these frames used the “fully-epoxied” connection, where the steel rods were epoxied along the full length, whilst the other two frames used the “tensioned half epoxied” connection, where the steel rods were epoxied into one member and tensioned to the other. Two different geometries were tested (the column running past the end of the beam and the beam extending over the top of the column). The lever specimens were designed to individually measure the different components of creep. These measured the creep of the timber itself (both parallel and perpendicular to the grain), the creep of a timber specimen with epoxied rods at both ends, and the creep of the glue alone. Short-term bending, compression and pullout tests were carried out on glulam according to ISO recommendations to characterise the timber used in this project. The results of the compressive strength tests perpendicular to the grain gave compressive stresses that were approximately 50% of that specified in the New Zealand Timber Standard. This is of concern and needs attention given the importance of this property when designing any joints where structural performance depends on bearing perpendicular to grain. Some important conclusions have been reached. The results from the Glulam Lever Specimens showed that creep perpendicular to the grain may be up to 8 times greater than that parallel to the grain. The remaining lever specimens showed that long term pullout deformations are likely to be negligible in dry conditions. The “fully epoxied” connection performed significantly better than the “tensioned half epoxied” connection. This is due to better transfer of forces in the “fully epoxied” connection and hence very little reliance on the compressive strength of timber perpendicular to the grain. Creep of timber was found to mainly affect the deflection of the frame specimens, as no significant stress redistribution was noticed in the long-term between timber and rods. It was also pointed out that a significant contribution to the total deflection is given by the flexibility semi-rigidity of the connection itself, which is usually designed as a rigid joint but should more realistically be considered as a semi-rigid connection. Design equations have been proposed and are shown to correlate well with the results obtained. Lastly, a method has been described to estimate the long-term stresses in the glulam and steel elements.

"Long term deformation of epoxy glued glulam portal frame knee joints" May 2009.

FRAGIACOMO, Massimo;
2009-01-01

Abstract

This project was initiated to resolve uncertainty regarding the long-term performance of epoxied rod connections in glue laminated timber (glulam), specifically the long term-deformations associated with beam-column moment joints. The excellent short-term performance of this type of connection is well known. However there is some scepticism in industry over the long-term performance. The behaviour of this connection when subjected to long term loading is thought to be primarily affected by creep in the timber and heterogeneity of the cross section. It has been hypothesised that the stress distribution at the beam-column interface may change over time leading to a localisation of stresses, rotations within the joint and ultimately, excessive deflections of the frame at serviceability limit states. Furthermore, crushing of the timber perpendicular to the grain may occur in the long-term as a result of the stress redistribution over time which could cause failure at ultimate limit state. In order to investigate the long term performance, two types of experimental testing were carried out in this project: (i) long term testing of several frames and lever specimens; and (ii) short term testing of several smaller glulam specimens. Four frames with a “C” shaped layout (i.e. one column and two beams) were tested. Two of these frames used the “fully-epoxied” connection, where the steel rods were epoxied along the full length, whilst the other two frames used the “tensioned half epoxied” connection, where the steel rods were epoxied into one member and tensioned to the other. Two different geometries were tested (the column running past the end of the beam and the beam extending over the top of the column). The lever specimens were designed to individually measure the different components of creep. These measured the creep of the timber itself (both parallel and perpendicular to the grain), the creep of a timber specimen with epoxied rods at both ends, and the creep of the glue alone. Short-term bending, compression and pullout tests were carried out on glulam according to ISO recommendations to characterise the timber used in this project. The results of the compressive strength tests perpendicular to the grain gave compressive stresses that were approximately 50% of that specified in the New Zealand Timber Standard. This is of concern and needs attention given the importance of this property when designing any joints where structural performance depends on bearing perpendicular to grain. Some important conclusions have been reached. The results from the Glulam Lever Specimens showed that creep perpendicular to the grain may be up to 8 times greater than that parallel to the grain. The remaining lever specimens showed that long term pullout deformations are likely to be negligible in dry conditions. The “fully epoxied” connection performed significantly better than the “tensioned half epoxied” connection. This is due to better transfer of forces in the “fully epoxied” connection and hence very little reliance on the compressive strength of timber perpendicular to the grain. Creep of timber was found to mainly affect the deflection of the frame specimens, as no significant stress redistribution was noticed in the long-term between timber and rods. It was also pointed out that a significant contribution to the total deflection is given by the flexibility semi-rigidity of the connection itself, which is usually designed as a rigid joint but should more realistically be considered as a semi-rigid connection. Design equations have been proposed and are shown to correlate well with the results obtained. Lastly, a method has been described to estimate the long-term stresses in the glulam and steel elements.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11697/27384
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact