In the following article written for the Crimson, Professor C. L. Dawes of the Engineering School describes the work which the students are doing on the properties of electrical Insulation, as well as the Importance of such research.

The Electrical Engineering Department of the Harvard Engineering School, at the present time, is engaged in the study of the properties of electrical insulation. As is well known, electrical engineering involves the use of three types of materials: conducting materials such as copper and aluminum; magnetic materials such as iron and steel from which magnets are made; and insulating materials such as rubber paper, cotton, oils, porcelain, and other similar materials which have the property of insulating the electric circuit. It is only by employing insulating materials that electric currents can be made to flow in the electric conductors. Otherwise, they would leak away, and it would therefore be difficult to transmit power any considerable distance.

The properties of conducting materials are very well known and there has been no substantial improvement in these materials during the past thirty or forty years. Also, magnetic properties of iron and steel are very well known; although some special magnetic alloys have been developed recently which have unusually high magnetic qualities with low degrees of magnetism.

Insulation Improves

Great improvements have been made in certain classes of insulating materials during the past few years. As a result, it has been possible to design overhead transmission lines to operate at 220,000 volts. Transformers are built to operate up to 500,000 volts. Furthermore, when lines, transformers, and generators, or switches and similar apparatus are put into service, it is almost safe to predict that they will operate indefinitely without electrical break-down, unless the insulation is damaged by some abnormal electrical surge such as lightning strokes, or high voltages induced in the system by short circuits, or arcing grounds. With underground cables, however, the situation is entirely different. In the first place, it is not permissible to carry high-voltage lines into the centers of population because of the danger. Therefore, it becomes necessary to use underground cables. The ordinary high-voltage underground cable is insulated with paper tapes which are thoroughly dried, and then impregnated with insulating oils. This paper insulation is then surrounded with a lead sheath. Such cables behave most erratically. They may be tested in the factory at four or five times their working voltages, and yet in three or four months' service they may explode in several places.

Cable Weakest Link

Up to the present time, there does not seem to be any method for determining from preliminary test whether a cable will stand up in service or not. In other words, the cable is the weakest link in the modern power system. Moreover, cables have not been designed to operate on voltages commensurate with those employed for overhead lines.

The primary cause of cables breaking down in service is due to the fact that small voids or air spaces are formed within the paper insulation. With present methods of manufacture, these voids appear to be inevitable. When such voids form, a minute electric discharge takes place within the void; that is, small electric arcs form, and the molecules of air find themselves in an electric field which causes them to attain high velocities. This effect is called ionization. The ions then impinge on the paper and oil, and have the effect of a rapid and continual bombardment. They not only perforate the paper with holes, but also because of the formation of nitrates and ozone, cause adverse chemical actions in the impregnated oil. This insidious reaction is cumulative, and in the course of time, the papers and compounds become more and more carbonized and disintegrated which finally results in an explosion of the cable.

Tests Made

At this time the Engineering School is studying the foregoing phenomena. We are attempting to learn more of the electrical properties of the papers, oils, and of these electric discharges. To do this, we have already developed a precision high-voltage electric bridge which is capable of measuring very minute amounts of power at extremely high voltages, with very great precision. By means of the bridge, we are determining the electric properties of oils and papers at different temperatures and frequencies, and under different conditions of impregnation. Also, we are investigating the electrical characterisics of the electric discharges which take place in small air spaces.

We have already determined some of the essential electrical properties of these ionized gas films. With this information, we have been able to determine from simple power measurements on a cable just how much power is being lost as ionization in the destructive air films. Naturally, the life of the cable is associated with the amount of this loss. Experiments have shown that cables which have the highest value of this ionization loss also appear to be the poorest cables. A paper on this subject was presented at the last Midwinter Convention of the American Institute of Electrical Engineers in New York.

Conduct Cable Tests

This work is now being continued by conducting life tests on cable samples. The samples are approximately fifteen feet long, and they are subjected to high voltage for fifteen hours each day. Also, their electrical characteristics are measured every day so that we can determine the electrical characteristics of the cables over their entire life period.

Curves are being plotted showing the electrical characteristics of these cables taken every day. At this time we appear to be obtaining a very definite relation between these electrical characteristics, and the number of hours which the cable will operate. Although investigations of this general character have been conducted heretofore, none have ever made the detailed and accurate analysis which is being done by the Engineering School.