Thus we have an expression involving the number of watts EI, two temperatures 01 and 02, at the distances r1 and r2 from the axis of the heating coil, and the length of the heating coil. If the quantities are mesured in terms of watts, centimeters, and degrees Centigrade, then the conductivity is defined as the number of calories of heat transmitted in one second between two faces of a centimeter cube when the faces are at a temperature difference of 1° C.

METHOD OF MESUREMENTS

The cylinder of clay prepared as described as above, was enclosed in another cylinder with walls an inch thick. It was then placed in a vertical position upon thick magnesite bricks. The whole was surrounded with magnesite bricks with the exception of a small space thru which to insert the thermo-couples. To the terminals of the heating coil was connected a Weston alternating-current wattmeter, which gave the amount of electrical energy converted into heat. The temperature difference at different distances from the axis of the cylinder was mesured by means of two platinum, platinum 10 per cent rhodium thermocouples. The electro-motive forces of these couples was mesured by means of a Leeds and Northrup Type K potentiometer. These thermo-couples were carefully calibrated and curves drawn, relating electromotive force and temperature, from which the temperatures involved in the experiment were obtained.

The method of observation consisted in heating the cylinder for about six hours until a steady state was observed, then inserting the thermo-couples in the various holes at known distances from the axis of the cylinder, and reading their electro-motive forces. The temperatures corresponding to these electro-motive forces were obtained from the calibration curves mentioned above. In this manner a series of values of the conductivity at different temperatures was obtained, ranging from 150° C. to 850° C. for the couple nearest the axis of the cylinder. The object in view was to determine whether or not the conductivity was a function of the temperature.

RESULTS

Two samples of clay were tested which came from Hebron, North Dakota. These samples were a high-grade pottery clay. From a visual examination they appeared to be of the same texture. A sample of English ball clay of unknown origin was also tested. The results for the samples of North Dakota clay are contained in Tables I and II, and those for the English ball clay in Table III.

The samples of clay from the state show a decreasing conductivity between the temperatures at which it was observed, thus making the conductivity a function of the temperature. This is more clearly brought out by the fact that, for the same. sample and during the same steady state, the values of the conductivity are greater for a greater ratio of the radii which enter into the computation. For example, in Table II, the conductivities are computed using radii of 10, 15, 20, 25, and 30 millimeters and the temperatures corresponding to these radii. The conductivity corresponding to the radii 10 and 30 mm. is larger than that for the radii 10 and 15 mm., showing that for a higher average temperature which existed between the radii 10 and 15 mm. the conductivity was less than that corresponding to the lower average temperature between the radii 10 and 30 mm. This is in accord with the supposition that the conductivity decreases with the temperature. These samples were not initially fired, but were simply dried thoroly, hence it may be expected that as the temperature rises the conductivity will change owing to some change in the test sample. The change may be due to a driving off of gases due to combustible matter naturally included in the clay.

It may be argued that this is a fault inherent in the method, but it does not seem likely, owing to the fact that when the thermo-couples were inserted to various depths in the holes there was no temperature gradient except near the ends of the sample.

The English ball clay shows a much higher conductivity and not the remarkable change with a rise of temperature. It, therefore, seems apparent that North Dakota clay of texture and composition similar to the samples which have been tested exhibit admirable heat-insulating properties owing to the low thermal conductivity. However, the samples, as stated above, were not initially fired and after a thoro firing may show entirely different results. At the present time, experiments are in progress deal

ing with the conductivity of other samples of North Dakota and English ball clays, also mixtures of the two; Kaolin clay is also being attempted. Larger specimens are being used in order to obviate any possibility of the flow of heat not being approximately radial.

An investigation of similar character was published in a bulletin of the University of Illinois Experiment Station. There the investigators found that fire clay, obtained from the LecledeChristy Clay Products Company of St. Louis, Missouri, had a conductivity ranging from .0024 to .0035, a much higher conductivity.

Electrical Testing of Telephone
Cables*

TH

EDWARD BEATTIE STEPHEnson,

Instructor in Physics, University of North Dakota

HE cable has been one of the important features in the development of telephone service, especially where protection from the weather, the limitations of space, or the esthetic standards of the community do not permit the use of bare wires suspended on poles. It is the principal object of this article to consider the testing of the cables at the manufacturing plant, but for the proper understanding of the purpose of the tests it will be necessary at times to go briefly into general electrical and telephone theory without attempting rigorous mathematical treat

ment.

The dry-core, paper-insulated, lead-covered type of cable is in almost universal use. This cable is built up as follows: Each copper wire has a strip of special paper about an inch wide wound spirally around it, two wires with different colored paper are twisted together into a pair, two pairs are laid up into a quad, the quads are arranged in concentric layers wound in opposit directions, around the outside two layers of paper are wrapt in opposit directions, and finally, after baking in an oven at 212° F. until perfectly dry, the whole is covered with a continuous lead sheath which is molded on the cable at a very high pressure, but low temperature.

The electrical properties of a cable depend largely on its design. The ideal condition is to get as many wires as possible into a small space and yet preserve the insulation and talking qualities. A full consideration of the talking qualities of a cable would necessitate a complicated mathematical study1 but it may be

Most of the methods and apparatus described are those in use at the present time by the Western Electric Company, as learned by the writer during the past summer's work in the Inspection-investigation department at the Hawthorn plant, Chicago. The writer wishes to express his thanks to the department manager for explanations of many of the theoretical and practical points involved in the methods and apparatus.

For a mathematical study of the subject, see Oliver Heaviside, "Electromagnetic Theory," "Collected Papers"; Abbott, "Electrical Transmission of Energy." For a comprehensive study of the practical side, see Kemster B. Miller, "American Telephone Practice."