Abstract
Failure in wood is commonly considered as a sign of properly chosen adhesive for wood adhesive joints. It is argued that as the adhesive is stronger than the surrounding wood material, no concern has to be given to the adhesive properties in joint design. However, by means of theoretical arguments in combination with experimental experience in the present study, it is stated that local bond line properties are of great consequence to global joint load capacity, also in case of failure in wood. Furthermore, it may be concluded that for many joints, optimal strength can be expected if the failure is located in the adhesive material instead of the surrounding wood material.
A nonlinear fracture mechanics model, including gradual softening, for strength analysis of adhesive joints is presented. It is a unified theory which incorporates linear and nonlinear elastic analysis as well as conventional stress criteria and linear elastic fracture mechanics as special cases. Closed-form solutions are given for pure shear analysis. For mixed-mode a coupled deformation based model is developed which exhibits path-independence in stresses but path-dependence in fracture energy. The model is implemented in a finite element code and used for analysis of some of the proposed test specimens in-glued steel plates and for a parametric study of finger joints. A major outcome is that both local bond strength and fracture energy are of importance in strength evaluation of regular wooden joints.
The complete stress-deformation curve for bond lines is determined for pure shear, pure tension perpendicular to the bond line and in mixed-mode. To perform stable tests for brittle adhesives, large demands are placed on the test specimen, setup and testing machine. New small size material test methods have been developed with emphasis on stiffness and on uniform and univocal stress states. In the material test series resorcinol/phenol, PVAc and one-component polyurethane were used.
Larger specimens to verify and to study fracture propagation are developed in the study. For the adhesives used in the test programmes, failure in wood occurred more frequently for increased amount of peel stresses. This make the evaluation more complicated, as the fracture energy is a mixture of wood and adhesive properties and as the fracture area may be difficult to determine. In shear-dominated failure processes the correspondence between calculated and experimental results is good.
A nonlinear fracture mechanics model, including gradual softening, for strength analysis of adhesive joints is presented. It is a unified theory which incorporates linear and nonlinear elastic analysis as well as conventional stress criteria and linear elastic fracture mechanics as special cases. Closed-form solutions are given for pure shear analysis. For mixed-mode a coupled deformation based model is developed which exhibits path-independence in stresses but path-dependence in fracture energy. The model is implemented in a finite element code and used for analysis of some of the proposed test specimens in-glued steel plates and for a parametric study of finger joints. A major outcome is that both local bond strength and fracture energy are of importance in strength evaluation of regular wooden joints.
The complete stress-deformation curve for bond lines is determined for pure shear, pure tension perpendicular to the bond line and in mixed-mode. To perform stable tests for brittle adhesives, large demands are placed on the test specimen, setup and testing machine. New small size material test methods have been developed with emphasis on stiffness and on uniform and univocal stress states. In the material test series resorcinol/phenol, PVAc and one-component polyurethane were used.
Larger specimens to verify and to study fracture propagation are developed in the study. For the adhesives used in the test programmes, failure in wood occurred more frequently for increased amount of peel stresses. This make the evaluation more complicated, as the fracture energy is a mixture of wood and adhesive properties and as the fracture area may be difficult to determine. In shear-dominated failure processes the correspondence between calculated and experimental results is good.
Original language | English |
---|---|
Qualification | Doctor |
Supervisors/Advisors |
|
Place of Publication | Lund |
Publisher | |
Publication status | Published - 1994 Nov |
Subject classification (UKÄ)
- Reliability and Maintenance
- Wood Science
Free keywords
- Joints
- Adhesives
- Wood
- Strength
- Fracture energy
- Mixed-mode
- Softening
- Finite Element Method
- Testing
- Theoretical simulation