The Journal of Strain Analysis for Engineering Design

Mild-steel vessels with initially flat ends, having four different thicknesses and various knuckle shapes, have been formed without dies. Similar shapes were obtained by forming at room temperature and at a high temperature with saturated steam as the pressurizing medium.

The elastic stresses due to internal pressure in 24 pressure-formed vessels have been analysed using frozen stress photoelastic models. The peak stresses in these shapes are generally lower than in the best torispherical heads of the same head height.

The mechanics of thermal self-straining in fibre-reinforced materials are explained. Stress distributions obtained from finite-element calculations are compared with results from photothermoelastic analysis of models.

Pearlitic flake-graphite grey cast irons have long been widely used for the manufacture of diesel-engine combustion-chamber components. Nonetheless, very few pertinent high-temperature data are to be found in published work. The investigation described attempts to enlarge current knowledge on the high-temperature behaviour of grey cast irons in view of requirements highlighted by high thermal ratings. A general consideration of the problem showed that the often quoted Eichelberg material-quality factor is largely irrelevant and an alternative assessment formula is proposed. This is broadly confirmed by a systematic examination of the mechanical and physical properties of 166 plain and alloyed cast irons and a limited number of engine components.

Different engine components impose different demands upon the material properties but for economic reasons a single cast-iron melt capable of meeting most of the requirements of all components is needed. It is shown that this situation is best met with an alloyed iron containing 1.3 per cent of copper-nickel and 0.4 per cent of molybdenum. The scatter in properties obtained from production casts was found to be typical for such irons. For components having safety factors of 2 or less, a prerequisite of economic engineering design, the service reliability is crucially dependent upon the scatter of material properties. Thus, there are advantages in reducing this scatter and it is shown that this can be achieved by attention to the cleanliness of the matrix and tight control of the chemistry of the melt.

In general, the absolute values of mechanical and physical properties of flake-graphite cast irons appear to be governed mainly by the form and size of the graphite flakes and the chemistry of the matrix (i.e. CEV and alloying additions). Regression analysis of the results permitted the derivation of empirical formulae for the prediction of the iron properties. The general findings are used to suggest possible ways of improving the thermal resistance.

A printing technique to replace the usual marking practice of punching dots or scribing lines on a round tensile specimen is described. With this method the specimen surface will not be influenced by marking, hence premature failure can be avoided in high-speed testing. The printing technique is simple to apply. The printed gauge mark can be controlled to an accuracy of ± 0.002 in for a 1.4 in gauge length. The width of the printed lines is about 0.003 in.

A simple formula is derived for predicting the flexure of uniformly loaded point-supported circular plates. The classical solution of Michell for a clamped plate under a single point load is extended for any number of point loads regularly spaced around a circle concentric with the plate edge. The resulting series for the edge moments and shears are summed and are shown to be very similar to a simple sine wave. Replacing the exact expressions by single sine waves enables the clamped edges to be set free by a simple superposition of solutions. The point reactions are equilibrated by a uniform load and the resulting deflection surface for a free uniformly loaded point-supported plate is obtained immediately. Deflection curves for the particular case of a plate supported at three points are given in the form of contours of equal deflection. This particular case is compared with some experimental results which were obtained by optical methods.

For three supports, maximum deviation from the flat is least when the supports are equispaced around a circle of radius approximately two-thirds that of the plate. The contours for this case show that the central area is remarkably flat and that there are three diameters along which the deflections are almost constant.

Stress-concentration factors are analytically determined for relatively deep U-shaped edge notches of constant radii in a semi-infinite sheet with uniaxial tension applied parallel to the sheet edge. Because a strong geometric similarity exists between a U-notch and a semi-elliptical notch of the same tip radius, a comparison of such analytical data is effected. Stress-concentration factors for U-notches in a semi-infinite sheet are also compared with other comparable analytical and experimental data available in the literature.

Finite-width correction factors are applied to the results of the semi-infinite-sheet analysis to provide practical design data. These data are compared with experimental results provided by a number of investigators.

A series of combined stress tests in torsion-tension space is carried out on thin-walled tubes of 1100-F commercially pure aluminium. One initial and four subsequent yield loci are established to a maximum prestrain level of 14 per cent tensile plastic strain.

The results are analysed in terms of a proposed, rationally based, yield criterion constructed according to statistical reasoning.

It is shown that during prestrain a significant degree of geometrical distortion is undergone by the yield loci, accompanied by a strong Bauschinger effect and a flattening of part of the locus opposite to the loading point.

It is found that the yield locus does not rotate during prestrain and, contrary to the case for torsion prestrain, exhibits evidence of a sharply rounded corner developing at the loading point.

The proposed criterion is shown to fit the experimental results extremely well and the mechanism of distortion is explained in terms of a statistical model for work-hardening materials.

A previous paper (1)∗described an analysis for plates that made use of non-linear large-deflection theory. The results of the analysis were compared with measurements of deflections and stresses in simply supported rectangular plates.

In this paper the analysis has been used to calculate the stresses and deflections for clamped-edge plates and these have been compared with measurements made on plates of various aspect ratios. Good agreement has been obtained for the maximum values of these stresses and deflections.

These maximum values have been plotted in such a form as to be easily usable by the designer of pressure-loaded clamped-edge rectangular plates.

A creep laboratory for variable-load and variable-temperature testing in uniaxial tension has been established at the University of Newcastle upon Tyne. The philosophy behind the specification and design of the equipment is described, considerable attention having been paid to accurate control of the test environment. The method of processing experimental data, obtained with a data acquisition system, is outlined.