ÁñÁ«ÊÓƵ

It is a natural phenomenon that things that grow tend to divide. This is certainly true of plants, colonies of social insects and, of course, knowledge. As science progressed it divided into physics, chemistry and biology. Physics was then further divided into heat, light, sound, magnetism and electricity, plus an additional group known in prewar days as "properties of matter" - topics such as surface tension, viscosity, elasticity and others that did not seem to "fit" into the more specific classifications.

Once such subdivisions have been accepted and become "traditional" it is difficult to imagine the original subject being divided up in a different way. Finally, however, it became possible to read for a degree in "materials science" that included such topics as ceramics, semiconduction and metallurgy, each of which was clearly a mixture of physics and chemistry. Interdisciplinary sciences have multiplied almost as rapidly as fundamental particles, and justifiably so.

Both of these books are on relatively new classifications of science, and both began as university lecture courses. Both are about engineering, but of course both contain a great deal of traditional physics and chemistry. Hiromu Nakazawa justifies the reason for creating new disciplines beautifully: "In thermodynamics, and indeed in any field, by transforming previously fragmented information and knowhow into a systematic body of knowledge, it becomes universally available worldwide and part of our human heritage." And how well he then goes on to do it. Principles of Precision Engineering is a textbook, and yet it is in part a history book. It is a collection of a wide variety of subjects connected with accuracy in manufacturing, yet some are so unusual that at first it seems unlikely that they would yield to attempts to quantise them - subjects such as "surface roughness", "out of roundness" and "smoothness". But these are clearly defined, to the extent that the expression "micro-smoothness" is seen to have a meaning.

It could be argued that friction plays such an important part in all our lives that it is surprising it took so long to establish it as a formal discipline. Yet the name tribology - the study of friction and lubrication - is a mere 25 years old, as J. A. Williams points out in his excellent book on the subject. His opening chapter contains a wonderful picture, copied from an ancient Egyptian stone carving c.1880bc, of more than 100 men pulling along a huge stone colossus on a sledge. The caption reads: "Perhaps an early tribologist can be seen at the front of the sledge lubricating its passage from a jug of liquid." There, to be sure, is a figure with a vessel remarkably like a modern oil can clearly doing just that. The author goes on to explain that no less a person than Leonardo da Vinci expended a great deal of effort in extensive investigations of the subject.

Nakazawa claims to keep his book "readily understandable by making liberal use of illustrations and avoiding the use of mathematical equations as much as possible I so that it can be read anywhere, even on a train on the way to work". He certainly keeps his promise. It is not only a good reference book but also contains the kind of collected information any practising engineer would want to read.

Williams's book is certainly not as devoid of mathematical analysis, but is none the less easy to read. Above all, he retains a physical insight that is not lost, even when the text extends to numerical solutions. This is rare in modern books, many of which give the impression that they wish to appear learned for the sake of learning.

He sets out to write primarily for final-year undergraduates and postgraduates in mechanical engineering but ends up achieving much more than that, for he has written a reference book for practising engineers, almost a Bible of tribology.

Of course, it is not light reading, but only an elementary knowledge of basic science is needed before tackling it. When I undertook to review it, I feared that as an electrical engineer I would be out of my depth, but it was not so. Not only was it full of interest but is a valuable reference book for any electrical engineer involved in developing moving machinery. At the same time it is by no means superficial. The analytical parts are backed up by numerous set problems, with answers.

Both books involve a lot of surface physics and they complement each other to a considerable extent. One has only to consider that the "roughness" of a surface, as manufactured, is bound to play a large part in determining the amount of friction it will encounter when in contact with another surface.

Nakazawa's book deals with cutting, grinding and polishing, with the different techniques used and the resulting machinery that has been designed. As one progresses through the book one becomes ever more amazed at the accuracies now attainable. The author fires our imagination quite early by noting that a magnetic reading head in a computer is positioned above the disc, rotating at 3600rpm at a clearance of less than 0.15 of a thousandth of a millimetre. The length of the magnetic head is 3.2mm. He points out that if these figures are magnified to the scale of a jumbo jet, they correspond to the plane flying 3.3mm above the ground.

Diffraction gratings call for 1200 grooves/mm to be cut to an accurate shape. The error allowed in machinery is 8 x 10-9 mm, which is only 22 times the interatomic distance in copper. The result of machining affecting the remaining surface, which might subsequently change shape, is such that "a change of 0.85m in the next 26 years cannot be ignored". In terms of angular precision 30 millionths of a degree is not impossible. Holes 1 5m in diameter can be cut using a laser at a power density of 100 megawatts/sq. cm.

Using "mechanochemical" polishing a scratch-free surface on a sapphire removes individual particles whose diameter is only a "few Angstrom units". The phrase "optically flat" takes on a new meaning. The distance between the highest peak and lowest trough in the roughness of this surface is now less than 0.015m (=10 nanometres) Processing methods include elastic emission, ion beam, electric discharge, laser and electron beam. Etching techniques include reaction ion, photoetching and anisotropic etching. There exist both hydrodynamic and hydrochemical polishing.

In pointing out that James Watt's first steam engine became workable only as the result of improved precision, this book makes the uninitiated more aware of the value of precision in the modern world.

I recommend both books very highly indeed.

Eric Laithwaite is emeritus professor, Imperial College, London, and visiting professor, University of Sussex.