John Forbes Olesen

Design of Mechanically Reinforced Glass Beams: Modelling and Experiments

The present paper is a study on how to obtain a ductile behaviour of a composite transparent structural element. The structural element is constructed by gluing a steel strip to the bottom face of a float glass beam using an epoxy adhesive. The composite beam is examined by four-point bending tests, and the mechanisms of the beam are discussed. Analogies to reinforced concrete beam theory are made; thus, four different design criteria, depending on the reinforcement ratio, are investigated. Analytical expressions are derived that are capable of describing the behaviour in an uncracked stage, a linear cracked stage and a yield stage. A finite element model, capable of handling the cracking of the glass by killing elements, is presented. Both analytical and numerical simulations are in fairly good agreement with the experimental observations. It appears that the reinforcement ratio is limited by the risk of anchorage failure and must be adjusted accordingly to obtain safe failure behaviour in a normal reinforced mode. Analysis of anchorage failure is made through a modified Volkersen stress analysis. Furthermore, different aspects of the design philosophy of reinforced glass beams are presented.

Accurate determination of asymptotic postbuckling stresses by the finite element method

Abstract Koiters method for initial postbuckling and imperfection sensitivity analysis of elastic structures is conveniently formulated in terms of finite elements. Special care must, however, be taken in the computation of the postbuckling stresses and the postbucklmg constant determining the initial curvature of the postbuckling path. The cause of the problem lies in the different degree of approximation of in-plane and lateral displacements and is inevitable in a standard compatible finite element formulation. It is shown that only a minor change in the computation of the postbuckling stresses is needed. The procedure is extended to cover Byskov and Hutchinsons method for cases with nearly sinultaneous buckling modes.

An Investigation on the Influence of Root Defects on the Fatigue Life of the Welded Structure of a Large Two-Stroke Diesel Engine

The crankshaft housings of large two stroke diesel engines are welded structures subjected to constant amplitude loading and designed for infinite life at full design load. A new design of the so-called frame box has been introduced in the engine using butt welded joints of thick plates, welded from one side only, with no access to the root side. Linear elastic fracture mechanics applied to three-dimensional finite element models has been used to assess this new design from the fatigue viewpoint. The methodologies used, from coarse models of the complete engine structure to refined sub-models of the welded joints, are described and results presented. In addition, large-scale test specimens with controlled “lack of fusion” weld root geometry were manufactured and fatigue tested to develop “S-N” curves and determine threshold stress intensity factor range values. These were established for opening mode loading both under the influence of residual stresses from production and in stress relieved specimens.

Cohesive cracked-hinge model for simulation of fracture in one-way slabs on grade

Numerical analysis of slab on grade structures subjected to mechanical loads is a complex matter often requiring computationally expensive models. In order to develop a simplified and general concept for non-linear analysis of slab on grade structures, this paper presents a cohesive cracked-hinge model aimed at the analysis of the bending fracture of the cemented material. The model is based on the fracture mechanics concepts of the fictitious crack model with a linear stress–crack opening relationship. Moreover, the paper presents a two-parameter spring foundation model applied to realistically capture the continuity in the supporting medium. The functionality of the proposed model is compared to numerical analysis with application of the more conventional cohesive zone model. The results obtained show that the methodology is a attractive and powerful one well-suited for practical use and further development.

Improvement of Fatigue Life of Welded Structural Components of a Large Two-Stroke Diesel Engine by Grinding

The crankshaft housings of large two-stroke diesel engines are welded structures subjected to constant amplitude loading and designed for infinite life at full design load. A new design of the so-called frame box has been introduced in the engine using butt weld joints of thick plates, welded from one side only, with no access to the root side. Various investigations on the fatigue life of the structural components of this new design have been carried out. The present investigation concentrates on the improvement in fatigue life which may be obtained by grinding of the weld toes. The tests performed showed a significant increase in fatigue life due to the grinding, ranging from a factor of approx. 2.8 to ∞, depending on the load level. Although the number of tests was limited, the results indicate a favourable change in slope of the S-N curve, from m ~ 3.0 for the test series without grinding to m ~ 6.0 for the test series with grinding. In one of the test series, it was observed that in most cases crack initiation moved from the weld toe to the non-ground surface between the ground areas at the weld toes. Tests were made on steel S 275, on centrally and eccentrically loaded test specimens.

Determination of Cohesive Fracture Parameters for Wood

The application of non-linear fracture mechanics to wood is a relatively new topic in the area of wood science, however, linear elastic fracture mechanics (LEFM) was first applied to wood in the nineteen sixties. The fracture propagation in wood is mainly governed by two aspects, namely the direction of the principal stresses, and the microstructure. There are six principal crack propagation systems in wood, which are illustrated in Fig 2, see eg. Smith et al. [1] or Reiterer and Tschegg [2]. The usual presumption is that wood is perfectly brittle-elastic (LEFM). When a crack propagates perpendicular to grain, the observed fracture behaviour is quasi-brittle, i.e. a tensile softening branch exists [2]. This indicates that a fracture process zone of some length exists. The LEFM approach is not an adequate approximation if the specimen size is of the same order of magnitude as the length of fracture process zone.

FRACTURE MECHANICS AND PLASTICITY MODELLING OF THE SPLIT CYLINDER TEST

ABSTRACT The split cylinder test is subjected to an analysis combining nonlinear fracture mechanics and plasticity. The fictitious crack model is applied for the analysis of splitting tensile fracture, and the Mohr-Coulomb yield criterion is adopted for modelling the compressive crushing/sliding failure. Two models are presented, a simple semi-analytical model based on analytical solutions for the crack propagation in a rectangular prismatic body, and a finite element model including plasticity in bulk material as well as crack propagation in interface elements. A numerical study applying these models demonstrates the influence of varying geometry or constitutive properties. For a split cylinder test in load control it is shown how the ultimate load is either plasticity dominated or fracture mechanics dominated. The transition between the two modes is related to changes in geometry or constitutive properties. This implies that the linear elastic interpretation of the ultimate splitting force in term of the uniaxial tensile strength of the material is only valid for special situations, e.g. for very large cylinders. Furthermore, the numerical analysis suggests that the split cylinder test is not well suited for determining the tensile strength of early age or fibre reinforced concrete.