Resistance-curves and wood variability: Application of glued-in-rod

Published Date
October 2016, Vol.70:1–9, doi:10.1016/j.ijadhadh.2016.04.015

Author 

J.L. Coureau ^a,,

P. Galimard ^a

A. Cointe ^a

J. Lartigau ^b

S. Morel ^a

aI2M – Department of Environmental and Civil Engineering, UMR 5295 (Université de Bordeaux/CNRS), 351 cours de la Libération, Bâtiment A11bis, 33405 Talence Cedex, France

bESTIA Recherche, Technopôle Izarbel, 97 allée Théodore Monod, 64210 Bidart, France

Accepted 15 April 2016. Available online 3 May 2016.

Abstract

The current study investigates the withdrawal strength of glued-in rods as part of linear elastic fracture mechanics. An experimental campaign was performed in order to observe the effect of the specie (spruce and oak) on the axial strength of glued-in rods for given geometrical configurations. Finite elements modelling was presented in order to consider the progressive damage and the crack propagation located at the wood-adhesive interface (failures obtained during experiments). The approach aims at separating the progressive failure due to mode I and mode II. For this, Resistance-Curves, regarded as material properties, were used to characterize the peeling and the shear effects at the ultimate state. The study reveals that the mode I initiates the damage in the glued interface. Using several finite element runs, the predicted pull-out strengths were estimated from the elastic properties of substrates (wood, adhesive and steel) and the fracture properties of wood. Numerical results show the dependence of the strength according to the stiffness of the materials. Moreover, the scattering of the results is also affected by the variability of the fracture energies of the wooden substrates. The investigation leads to propose a robust approach which is able to predict the axial strength of glued-in-rods, considering the variability of each material and combining damage and crack propagation of the wooden substrate. It reveals that the prediction of the ultimate load cannot be performed considering only the failure mode II.

Keywords

Glued-in rods

Destructive test

Finite element analysis

Linear elastic fracture mechanics

---

Nomenclature

a

length of the crack along the interface as part of LEFM (mm)

a_c

length of the crack producing a crack propagation at a constant critical energy released rate (mm)

a_u

length of the crack at ultimate load (mm)

A_s

area of the rod (mm²)

COV

coefficient of variation (%)

G^*(a)

critical energy released rate induced by damage and crack growth (J/m²)

G_I(a)

energy released rate in mode I for a load equal to 1 N (J/m²)

G_II(a)

energy released rate in mode II for a load equal to 1 N (J/m²)

G^*_I(a)

critical energy released rate in mode I when G*(a) is reached (J/m²)

G^*_II(a)

critical energy released rate in mode II when G*(a) is reached (J/m²)

G_RC,I (a)

critical energy released rate in pure mode I (J/m²)

G_RI (a)

R-curve in mode I according to a (J/m²)

K

ratio between critical energy release rates in mode I and mode II

l_a

anchorage length (mm)

P_u

failure load of the single bonded-in rods (N)

S

initial stiffness of the connection (N/mm)

r

radius of the rod (mm)

S(a)

stiffness of cracked specimen (N/mm)

S′(a)

derivative function of S(a)

R_I(a)

part of mode I in function of the global energy released rate

Fig. 1.

Fig. 2.

Fig. 3.

 Table 1

Table 1.

Fig. 4.

Fig. 5.

 Table 2

Table 2.

Fig. 6.

 Table 3

Table 3.

Fig. 7.

Fig. 8.

Fig. 9.

Fig. 10.

• ⁎ 
  Corresponding author. Tel.: +33 5 400 065 14; fax: +33 5 400 035 95.
  
  

For further details log on website :
http://www.sciencedirect.com/science/article/pii/S0143749615002377