Gripping Surfaces for Artificial Hands


A hand has to grip and hold objects: if one has to design a functional artificial hand should one look at the normal hand for inspiration, or should one look at gripping tools? Are hands and pliers really very different? It must be admitted that when some artificial hands are examined one gets the impression that they have been intended to look like hands, but to operate like tools, and that the product of combining these characteristics is unfortunate.

Of course, the skill and ability of the human hand to hold objects of widely differing shapes and sizes has long been a cause of wonder to students of many branches of science and philosophy, and the successful extension of man's power by the use of tools, has been as much due to his own ability to hold them as to his capability to design them. There is no need to emphasise that his hands are unique amongst the range of prehensile devices which nature supplies to the animal kingdom, a uniqueness only paralleled by his consequent enormous dependence on tools, of every sort, including pens.

The tool illustrated in Fig. 1 belongs to the subgroup of holding tools, as distinct from manipulative tools, and here the first difference between hand and tool appears. Tools are graded according to the size of object they are expected to hold and, except over a small range, are not smoothly adaptable to different ranges of size. The common characteristic of many tools, however, is the type of gripping surface they present; a ridged surface with the ridges running at right angles to the long axis of the jaws of the tool.

The purpose of the ridging is undoubtedly to give a firmer grip, and undoubtedly it does achieve a fairly considerable advantage over the grip given by, say, the round-nosed plier used for shaping, but it is worth looking at its action.

There are two ways in which the ridges may act; the first is dependent on irregularity, or curvature, of the surface of the object, so that a part, or parts, of it interdigitate with the ridging. The second way is that by having sharp edges to the ridges, the pressure achieved by the applied force is orders of magnitude higher than would be with a plane surface, and so the jaws produce by their own action a corresponding ridging on the object which precisely mates with the ridging on the jaw.

If one looks back at the normal hand, the ridging is certainly there also-but is its action the same? It certainly is not hard enough to impose its pattern on any but a few materials, and if it were important as a method for latching onto fine irregularities on objects then smooth objects (and here I would emphasise that I mean even surfaced but not slippery or with a low coefficient of friction) would be difficult to hold, and would not be in common use. The half-pint glass as we know it now would simply not be practical.

If the ridging is not to hold objects, what could be the reason for its presence? Dry hands do not hold well-could the grooving be simply a system of micro-guttering to distribute and retain and deliver sweat and maintain a big coefficient of friction? If this were so then it is obvious that the serrated jaws of the pliers are not a super finger surface; they may belong to an almost totally different system of gripping.

If we now look at the fingers to see how they behave in gripping we immediately find a major difference-in most cases it is the finger which deforms and not the object, and the deformation which takes place is complex. It is complex because this is not the sort of deformation which would occur if some sort of elastic material were fitted to the jaws of the pliers. With the finger, the object depresses or indents the skin, at first meeting little reaction. Quite sharply, however, this reaction rises and the pad of the finger becomes stiff, and the pressure under the skin rises and is demonstrated by blanching on the sides of the fingers if sufficient force is applied. On releasing the pressure the indentation remains for a time ( Fig. 2 and Fig. 3 ).

What sort of mechanism could give this effect? A possible one would be the flow of material away from under the object, the flow ceasing when the material has rearranged itself in the available volume. Any further depression or indentation of the skin by the object would then be impossible because their only action would be to decrease the volume further, unless the skin stretched.

The characteristics of the system therefore appear to be the following:

  1. A rigid framework to give stiffness to the finger.
  2. A flexible membrane, anchored to the framework, deformable but largely non-stretch under the action of the forces normally met with.
  3. A closed and finite volume bounded by the membrane and containing a material which can flow under pressure.

The construction of a model of this system is extremely simple, so simple in fact that I thought I must be mistaken and did not bother to make it for about a year. A strip of metal is taken, the width of the finger. An appropriate powder (I used ordinary table salt in my first model) is then put in a suitable bag-a short length of half-inch Paul's tubing is convenient. The bag of powder is then fixed to the strip by wrapping it round with a suitable non-stretch membrane-I used leather, but a reinforced plastic would probably be more suitable. With two such surfaces in opposition, any object in their grip displaces the powder until the non-stretch membrane is reduced to equal the volume of the powder, when the pressure in the sac then rises. This results in increased frictional forces between the particles of the powder, which consequently lock, giving what is effectively a rigid contour moulded to the object, and anchored to the frame ( Fig. 4 and Fig. 5 ). An extremely firm grip results, and if the powder is a conducting material a feedback of pressure can be available. The system is at present under development.