Observations placeholder

THE METAL-BENDERS” by JOHN B. HASTED

A child who could rotate surfaces could, without touch, twist a single metal strip about its own axis; metal strips were exposed singly rather than in crossed pairs.

An important question to be answered is how does the twisting depend upon the dimensions of the exposed specimen? It was answered by allowing Willie G., Andrew G. and Stephen North to twist aluminium strips of different widths, identical in other respects.

The pitches were then measured, and analysed in terms of the torque G necessary to produce twisting, through an angle of a strip of cross-section dimensions a and b, and linear modulus of elasticity n. This torque is given by the equation:
G = n*pi*theta*(a^2 + b^2)ab/(12*l)
The dimensions of the metal strips were as follows: 10 <= l <= 40 cm, b = 0.75 mm, 1.5 <= a <= l3 mm.

It follows that if pitch is proportional to a^2 + b^2, then the torque per unit strip width a is constant. The data displayed in Figure 7.2 show that this proportionality holds over more than an order of magnitude. Since the torque is force multiplied by strip width a, it follows that the quasi-force exerted by the surface of action is independent of the strip width. If the width were sufficiently small, these quasi-forces would be capable of doing serious damage to the metal; perhaps bringing about structural change. However, extrapolation through six orders of magnitude down to atomic dimensions would be too much of a liberty to take!

If the axis about which the surface of action rotates were not in the plane of the surface itself, strips of metal would not be twisted in the same way. If it were parallel to the surface but separated from it, as if the surface in rotation formed a tube, the strip would be bent into an Archimedean spiral.

If it were inclined to the surface and passed through it, the strip would be formed into a helix.

All these types of action have been found, but without leading the metal-bender towards a desired result.

Usually the supposed continuous rotation about a fixed axis does not continue for more than part of a single cycle. Non-uniform rotations and translational movements are the general rule. These result in the decorative shapes that some children, in particular Andrew G., claim to produce. They vary widely in size, from as large as 50 cm to as small as l mm. Andrew at one time must have achieved a considerable measure of control over his action so as to be able to produce the profusion of abstract and representational designs which have been seen by many people at a London exhibition and elsewhere (Plate 3.2). Julie Knowles has also exhibited art-work.

Difficulties about the conservation of angular momentum must be faced in interpreting these events. For a twisted strip to be produced by the quasi-force of a rotating surface, the strip must be held at one end, for example in the subject’s hand. But some subjects insist that this is not always the case, and that all sorts of twirled patterns can be formed on their own.

Although this presents difficulties of credibility, I have come at length to believe that it could sometimes be so. The solid surface on which the event takes place, a table, carpet or bed, can contribute forces; and one must also consider the possibility that two surfaces of action, or at least two parts of the same surface, could exert opposing quasi-forces. It will be recalled that suspended metal specimens receiving strain gauge signals hardly swing on their suspension wires, even though quite large strains are involved. The strains arise from within the metal rather than from an external interaction.