Page images
PDF
EPUB

returning over the other. Many such pairs of conductors may be bunched together into cables for as great distances as five miles without cross-talk. For in each pair of conductors there are equal and opposite currents which, either by leakage or induction, would tend to produce equal and opposite currents in either branch of any neighboring conductor, or, in other words, no current at all.

This device, therefore, prevents cross-talk, whether due to leakage or induction, and, with cables constructed in this way, the retardation is at a minimum, although experience shows that it is much easier to talk over five miles of single-conductor cable made on the Faraday plan than it is to talk over five miles of metallic gutta-percha cable, such as is used in Paris.

The retardation in Paris is, however, not large enough to be an obstacle for as great distances as it is ever necessary to use within the city. If it were desired to talk through such conductors, and then out across the country to neighboring cities, as we have occasion in America, it is an open question whether the retardation would not become a serious obstacle.

A great objection to the metallic circuit cable is that it requires two wires for each subscriber, which, to say the least, doubles the cost. Moreover, there has never yet been devised any real practicable method of connecting a metallic circuit system with a single circuit system so that conversation could not be carried on between parties in one city and parties in a neighboring city, unless both cities, as well as intervening trunk lines, were constructed with metallic circuits.

Thus far we have merely discussed the electrical difficulties which are met with when we take our wires down from the house-tops and poles, and bunch them into cables, to be laid under ground.

A bird's-eye view of the wires in any of our cities would show that there is an enormous engineering difficulty in the way of placing absolutely every wire under ground. The best solution of this difficulty has been found to be to radiate, by means of cables containing a hundred or several hundred wires, from the central office to a considerable number of points so located about the city that one or another of them could be conveniently reached by a short overhead line from any subscriber's station. This has been done to a considerable extent in many American cities by means of cables carried over

head. There is no reason other than that of increased cost why these same cables should not be put under ground; in fact, in many cities in Pittsburg and in Washington, for examples this has been done.

In this way by far the greater part of the wires may be placed under ground at an expense not greatly in excess of an overhead system. But it is easy to see that, if it were necessary to go a step further and place the overhead lines diverging from the points of radiation to the subscribers, also under ground, there would be an enormous additional expense, for it costs nearly as much to put one wire under ground as fifty, since the expense of excavating, laying the conduit, and refilling would be practically the same.

With regard to the method of laying the cables under ground, this must vary with the conditions present in the different cities.

In Washington, for example, where the streets may not be taken up for other purposes, it has been found sufficient to lay the cables in wooden troughs, filled with asphalt, about two feet below the surface of the street.

In some of the streets of Boston, which are continually being disturbed, it has been found necessary to construct under ground chambers at crossings, connected by means of wrought-iron pipes, through which the cables may be drawn. These, however, are engineering difficulties, and are easily settled for each particular case. While there exists no technical obstacles to the placing of all wires of a telephone exchange under ground, I do not, by any means, consider that it is economically practicable. It is practicable to extend wires from the central office under ground out to a considerable number of points, some one of which shall be easily accessible by a short overhead line from any subscriber's station, and when we consider that such a system is more durable than an overhead system, requiring practically no expenditure for repairs, and being always in good order, it is, in the long run, perhaps more economical to place the wires in this way than entirely overhead.

At the conclusion of Mr. Jacques's paper, Prof. CROSS said, among other things, that he had had occasion recently to compute the size of wire, and the thickness of insulation needed for an Atlantic cable for telephonic purposes, according to the figures given by Mr. Jacques in his paper read before the Society last winter.

The method of computation used was the one which would most naturally suggest itself, there being a variety of methods possible. He found the diameter of the wire itself would have to be three inches, and the diameter of the completed cable thirty-nine inches.

MEETING 338.

Improvements in Steam-Heating.

BY MR. FREDERIC TUDOR.

Application of Solar Heat to the Warming of Buildings.

BY MR. S. H. WOODBRIDGE.

The 338th meeting of the SOCIETY OF ARTS was held at the Institute on Thursday, Nov. 12th, at 8 P. M., Prof. C. R. Cross in the chair.

After the reading of the minutes of the previous meeting, and the election of new members, the chairman introduced Mr. F. Tudor, of Boston, who read a paper on "Improvements in Steam-Heating."

Mr. TUDOR said: The two objections to steam-heating are waterhammering and absence of control of temperature. The first is the result of bad design or workmanship, or of mismanagement of the valves. I shall try to show how both can be obviated. The circulation depends upon gravity, and upon very slight differences of pressure in the several parts of the apparatus, this difference of pressure having no necessary relation to the boiler pressure. Whether the boiler pressure be ten pounds or one hundred pounds, the circulation is not affected, except that difficulties may arise from a difference in the rate of condensation, which is greatly increased by greater pressures, since high-pressure steam has a higher temperature than lowpressure steam. The flow of steam through the great length of pipes, necessary in most cases, is attended by friction and a consequent reduction of pressure at remote points. In a well-proportioned appa

ratus these reductions of pressure will be very slight, and, when they are sensible, suitable arrangements must be made to overcome their effects. [The speaker then explained some diagrams on the board, bringing out very clearly the points mentioned above.] If the pressure were the same throughout the apparatus the water would be level, but when the circulation is going on, with a diminished pressure in the radiator, it is clear that the level of the water in the returnpipe will rise to a hight above that in the boiler equal to the hight of a water-column which balances the difference of pressure. For example: if the difference of pressure between the boiler and radiator is one pound, the difference of level will be two feet three inches. Where this difference becomes sufficient to sustain a water-column higher than the difference of level between the water-line of the boiler and the lowest part of the radiator, water enters the latter from the return-pipe, and water-hammering begins. In such apparatus this defect can be obviated by limiting the boiler-pressure so as to reduce the difference in pressure on which the hight of the water-column depends. In a well-proportioned apparatus this difference will not exceed one pound, while the difference of level provided is usually four feet.

Up to this point we have only supposed one radiator. Let us consider a number of them, horizontally disposed, with a returnpipe common to all, and for most of its length above the water-line. In such a case there would be no appreciable loss of pressure in the nearest radiator, and the steam, passing down through it, would establish the same pressure on the main return-pipe. At the most distant radiator there would be a lowered pressure, consequently the movement in the return-pipe would be toward it rather than toward the boiler. The result would be that the condensation-water, being transferred to the point of lowest pressure, would accumulate there, obstructing the circulation and causing incessant noise.

If we change this horizontal system into an inclined one, a degree of inclination will be reached where the movement of the water, due to gravity, will have sufficient force to overcome the friction of the steam moving past it in the return-pipe on its way to restore the pressure in the more distant radiators, and the water will then reach the lowest point rather than the point of least pressure. If, now, we continue to increase the inclination of our horizontal sys

tem until we reach the vertical, we shall have the common type of radiators disposed vertically, with upright rising mains, and the reason circulation is good, notwithstanding a largely reduced pressure is sufficiently clear.

If, in the horizontal system, the main-return is placed below the water-line, the loss of pressure in the radiators will be balanced in the return-pipe by an elevation of the water-columns in the upright branches, and the circulation will be perfect, notwithstanding the differences of pressure.

The details of the apparatus I have described are commonly supposed to be especially suitable to a low pressure,that is, of two or three pounds per square inch; but, since the circulation does not depend upon pressure, the system is suitable for any pressure. In passing, I will say that a high-pressure apparatus, so called, is one so badly proportioned that there can be no return to the boiler of the water of condensation which accumulates at the remotest point, where there is the greatest loss of pressure, whence it must be removed by special apparatus.

It has appeared that, in an apparatus of good design, slight differences of pressure are unavoidable, but that there must be a limit beyond which they must not go. Suppose, now, we limit the boilerpressure, so that it shall not exceed that of a water-column whose hight is equal to the difference of level between the water-line of the boiler and the bottom of the radiator. We can then impose an artificial obstruction in the steam-pipe and graduate its flow, even shutting it off altogether, without deranging the circulation in other radiators. This obstruction is the steam-valve, which, under these conditions, we can open more or less, and obtain more or less heat. cannot usually do this, because the pressure is too high in the case of horizontal systems having the returns sealed by water-columns; and in the vertical systems, the returns not being sealed, there would be a reversed current in the returns if the supply were throttled, steam would flow in from the return-end and the condensation-water would be driven back and retained in the radiator. Consequently, there is no control of the supply of steam and of the heat emitted; the valves must be wide open or tightly closed.

We

The method of regulating the heat by limiting the pressure, and thus affording a control over the steam-current in horizontal water

« PreviousContinue »