Author: Finn, E V;Needham, F H
N/A
Standard: £10 + VATMembers/Subscribers: Free
Members/Subscribers, log in to access
Finn, E V;Needham, F H
The Structural Engineer, Volume 42, Issue 1, 1964
The stresses and deflexions in a castellated girder are calculated by the simple theory of bending and by two methods in which it is assumed that the girder behaves as a Vierendeel truss. The calculated values are compared with experimental data and it is shown that the simple theory of bending usually underestimates deflexions by 20 to 30 per cent and that the measured stresses at some points in the girder deviate appreciably from the calculated values. However, the Vierendeel truss analogy is shown to provide more reliable values. J. KOLOSOWSKI
Mr. A. Jennings, Lecturer in the Civil Engineering Department of the University of Manchester, writes: ‘The standard methods of solution of linear redundant structures are all formulated in terms either of the unknown forces or of the unknown displacements and use the flexibility or stiffness properties of the structure to effect a solution. The method of solution of free cable problems advocated in this paper is peculiar in that it does not formulate the analysis directly in terms of either the unknown displacements or the unknown forces, and no concept of stiffness or flexibility is introduced.
Mr. R. Park, Lecturer in Civil Engineering at the University of Bristol, writes:- ‘In Mr. Christiansen’s paper an expression for the arching couple Ca is developed as a function of the central deflexion of the beam and the magnitude of the deflexion which makes Ca a maximum is determined, This maximum value of Ca is taken to be the arching couple at the ultimate (maximum) load of the beam. It appears that in some cases the central deflexion of the beam at the ultimate load so found is extremely small and Mr. Christiansen points out that in the tests conducted the ultimate load was reached at central deflexions which were larger than those predicted by the theory. This discrepancy in deflexions may result from the assumption that the plastic hinges of the beams are fully developed at all stages. When a beam is loaded to failure it will only develop full plasticity at the critical sections when the deflexion is large enough to cause the required strains. At small deflexions the stresses at the critical sections are less than the plastic values. Hence if the maximum arching couple given by the theory occurs at too small a deflexion it may be outside the range for which the theory is applicable. The theory should overestimate the ultimate load in this case since the beam will reach ultimate load at a higher deflexion and the arching couple will be reduced. It is evident that a method for determining the central deflexion at which full plasticity at the hinges has just developed is required, since only when this deflexion is reached does the theory become applicable. For laterally restrained beams the tension steel reaches yield stress before the concrete reaches its ultimate value. Hence the central plastic deflexion when the last plastic hinge fully develops could be written in terms of the ultimate strain at the compressed edge of the concrete, the length of the region of the hinge, the depth to neutral axis and the position of the hinge in the beam. It is evident, however, that some other features of the theory are conservative since the failing loads found by Mr. Christiansen’s tests exceed the theoretical ultimates in spite of the difference in central deflexions. ‘ It is doubtful whether arching action can be