The success of the low-tension system was followed by the introduction of the alternating system, using high potential primaries with the converters at each house, reducing, as a rule, from 1,000 down to either 50 or 100 volts. I am not familiar with the early alternating work, and had not at my disposal sufficient time in preparing my notes to go at any length into an investigation of this branch of the subject; nor do I think that any particular advantage could have been served by my doing so, as it has become generally recognized that the early alternating work with a house-to-house converter system, while it undoubtedly helped central station development at the time, proved very uneconomical in operation and expensive in investment, when the cost of converters is added to the cost of distribution. The large alternating stations in this country have so clearly demonstrated this that their responsible managers have, within the last few years, done everything possible, by the adoption of block converters and three-wire secondary circuits, to bring their system as close as they could in practice to the low-tension direct-current distribution system. I do not want to be understood as undervaluing the position of the alternating current in central station work. It has its place, but to my mind its position is a false one when it is used for house-to-house distribution with converters for each customer. The success of the oldest stations in this country, and the demonstration of the possibilities of covering areas of several miles in extent by the use of the three wire system, resulted in much capital going into the business. One of the earliest stations of a really modern type installed on either side of the Atlantic was built by the Berlin Electricity Works. The engineers of that station, while recognizing the high value of the distributing system, went back to Edison's original scheme of a compact direct-connected steam and electric generator, but with dynamos of the multipolar type designed and built by Siemens & Halske, of Berlin, the engines being of vertical marine type.

This was followed by the projecting in New York of the present Duane Street station, employing boilers of 200 pounds pressure, triple and quadruple expansion engines of the marine type, and direct-connected multipolar dynamos. Almost immediately thereafter, the station in Atlantic Avenue, Boston, somewhat on the same general design so far as contents is concerned, was erected. In 1891 a small station, but on the same lines, was projected for San Francisco, and in 1892 the present Harrison Street station of the Chicago Edison company was designed, and, benefiting by the experience of Berlin, New York and Boston, this station produces electric current for lighting purposes probably cheaper than any station of a similar size anywhere in this country.

It is not necessary for me to go into detail in explanation of the modern central station. You are all doubtless quite familiar with the general design, but if you will examine the detail drawings of the Harrison Street station, which I have brought with me, you will find that every effort has been made to provide for the economical production of steam, low cost of operating, good facilities for repairs and consequently low cost, and for permanency of service. You have but to go into any of the modern central stations in midwinter, to see them turning out anywhere from 10,000 to 80,000 amperes with a minimum of labor, to appreciate the fact that central station business is of a permanent and lucrative character.

To go back to the question of alternating currents, the work done in connection with the two-phase and three-phase currents and the perfection of the rotary transformer has resulted in introducing into central station practice a further means of economizing the cost of production—by concentration of power. According to present experience, it is (except in some extraordinary cases) uneconomical to distribute direct low-tension current over more than a radius of a mile and a half from the generating point. The possibility of transmitting it at a very high voltage, and consequently low investment in conductors, has resulted in the adoption of a scheme, in many of the large cities, of alternating transmission combined with low tension distribution. The limit to which this alternating transmission can be economically carried has not yet been definitely settled, but it is quite possible even now to transmit economically from the center of any of our large cities to the distant suburbs, by means of high potential alternating currents, distributing the current from the subcenter distribution by means either of the alternating current itself and large transformers for a block or district or else, if the territory is thickly settled, by means of a system of low-tension mains and feeders, the direct current for this purpose being obtained through the agency of rotary transformers.

There are various methods of producing the alternating current for transmission purposes. In some cases the generators are themselves wound for high potential; in others they are wound for 80 volts, and step-up transformers are used, carrying the current up to whatever pressure is desired, from 1,000 to 10,000 volts. In other cases dynamos are used having collector rings for alternating current on one side and a commutator for direct current on the other side of the armature, thus enabling you, when the peak in two districts of a city comes at two different times, to take care of this peak by means of the same original generating unit, furnishing direct low-tension current to the points near the central station and alternating current to the distant points. In other cases, where a small amount of alternating current is required on the transmission line, it has even been found economical to take direct current from a large unit, change it by means of a rotary transformer into alternating current, step up from 80 to, say, 2,000 volts, go to the distant point, and step down again to 80 volts alternating, and then convert again by means of a rotary transformer into low-potential direct current.

The introduction of alternating current for transmission purposes in large cities is probably best exemplified in the station recently erected in Brooklyn, where alternating current is produced and carried to distant points, and then used to operate series arc-light machines run by synchronous motors, the low-tension direct-current network being fed by rotary transformers, and alternating circuits arranged with block converters, and even in some cases separate converters for each individual customer in the scattered districts.

It would be very interesting to go at length into the details of cost in this, the latest development of central station transmission, but time will not permit; nor have I the time at my disposal to go at length into the central station business as developed by the electric street railways now so universally in use, or another phase of the business as exemplified by the large transmission plants, the two greatest examples of which, in this country, are probably those at Niagara Falls, N. Y., and Lachine Rapids, near Montreal. So far as street railways and power transmission are concerned, I would draw your attention to the fact that the same underlying principle of multiple-arc mains and feeders originally conceived by Mr. Edison is as much a necessity in their operation as it is in the electric lighting systems, whether those systems be operated on the old two-wire plan, the three-wire plan or by means of alternating currents.

Passing from a review of central station plants and distribution system naturally bring us to the operating cost and the factors governing profit and loss of the enterprise. In considering this branch of the subject, I will confine my remarks to the business as operated in Chicago by the company with which I am connected.

Our actual maximum last winter came on December 20, our load being approximately 12,000 horse power. A comparison of the figures of maximum capacity and maximum load of last winter shows that we had a margin in capacity over output of about 20 per cent. The load curves shown this evening represent the maximum output of last winter (December 20), an average summer load last year (June 4), and an average spring load of this year (May 2). For our purposes we will assume the maximum capacity of the plant and the maximum load of the system to be identical. The maximum load last winter occurred, as I have stated, on December 20, about 4:30 o'clock in the afternoon, and lasted less than half an hour. It should be borne in mind that the period of maximum load only lasts for from two to three months, and that the investment necessary to take care of that maximum load, has to be carried the whole year. It should not be assumed from this statement that the whole plant as an earning factor is in use 25 per cent. of the year. The fact is that, during the period of maximum load, the total plant is in operation only about 100 hours out of the 8,760 hours of the year; so that you are compelled, in order to get interest on your investment, to earn the interest for the whole of the year in about 1½ per cent. of that period, on about 50 per cent. of your plant.

This statement must bring home to you a realization of the fact that by far the most serious problem of central station management, and by far the greatest item of cost of your product, is interest on the investment. It may be that the use of storage batteries in connection with large installations will modify this interest charge, but even allowing the highest efficiency and the lowest cost of maintenance ever claimed for a storage battery installation, the fact of high interest cost must continue to be the most important factor in calculating profit and loss. This brings home to us the fact that in his efforts to show the greatest possible efficiency of his plant and distribution system, it is quite possible that the station manager may spend so much capital as to eat up many times over in interest charge the saving that he makes in direct operating expenses. It is a common mistake for the so-called expert to demonstrate to you that he has designed for you a plant of the highest possible efficiency, and at the same time for him to lose sight of the fact that he has saddled you with the highest possible amount of interest on account of excessive investment. Operating cost and interest cost should never be separated. One is as much a part of the cost of your current as the other. This is particularly illustrated in connection with the use of storage batteries. Those opposed to their use will point out to you that of the energy going into the storage battery only 70 per cent. is available for use on your distribution system. That statement in itself is correct; but in figuring the cost of energy for a class of business for which the storage battery is particularly adapted, the maximum load, that portion of your operating cost affected by the 30 per cent. loss of energy in the battery, forms under 4½ per cent. of your total cost, and it must be self-evident, in that case at least, that the 30 per cent. loss in the storage battery is hardly an appreciable factor in figuring the operating cost of your product. So far as I have been able to ascertain, it would appear to be economical to use storage batteries in connection with central station systems the peak of whose load does not exceed from two to two and one-half hours.

In order to illustrate the important bearing which interest has on cost, I have prepared graphical representations of the cost of current, including interest, under conditions of varying load factors. For the purpose of this chart I have assumed an average cost of current, so far as operating and repairs and renewals and general expense are concerned, extending over a period of a year, although of course these items are more or less attested by the character of the load factor. For the purpose of figuring interest, I have selected seven different classes of business commonly taken by electric light and power companies in any large city. Take, for instance, an office building. It has a load factor of about 3.7 per cent., that is, the average load for the whole year is 3.7 per cent. of the maximum demand on you for current at any one time during that period; or, to put it in another way, this load factor of 3.7 per cent. would show that your investment is in use the equivalent of a little over 323 hours a year on this class of business. This is by no means an extreme case. You can find in almost every large city customers whose load factors are not nearly as favorable to the operating company, their use of your investment being as low as the equivalent of 75 or 100 hours a year. Take another class of business, that of the haberdasher, or small fancy goods store. As a rule these stores are comparatively small, with facilities for getting a large amount of natural light and little use for artificial light. The load factor as shown by the chart is about 7 per cent., the use of your investment being not quite twice as long as that of the office building. Day saloons show an average of 16 per cent. load factor; cafetiers and small lunch counters about 20 per cent., while the large dry goods stores, in which there is comparatively little light, have a load factor of 25 per cent. and use your investment seven times as long per year as the office building. Power business naturally shows a still better load factor, say 35 per cent., and the all-night restaurant has a load factor of 48 per cent.

You will see from this that the great desideratum of the central station system is, from the investors' point of view, the necessity of getting customers for your product whose business is of such a character as to call for a low maximum and long average use. This question of load factor is by all means the most important one in central station economy. If your maximum is very high and your average consumption very low, heavy interest charges will necessarily follow. The nearer you can bring your average to your maximum load, the closer you approximate to the most economical conditions of production, and the lower you can afford to sell your current. Take, for instance, the summer and winter curves of the Chicago Edison company. The curve of December 20, 1897, shows a load factor of about 48 per cent.; the curve of May 2, 1898, shows a load factor of nearly 60 per cent. Now, if we were able in Chicago to get business of such a character as would give us a curve of the same characteristics in December as the curve we get in May; or, in other words, if we could improve our load factor, our interest cost would be reduced, an effect would be produced upon the other items going to make up the cost of current, and we probably could make more money out of our customers at a lower price per unit than we get from them now.

Many schemes are employed for improving the load factor, or, in other words, to encourage a long use of central station product. Some companies adopt a plan of allowing certain stated discounts, provided the income per month of each lamp connected exceeds a given sum. The objection to this is that it limits the number of lamps connected. Other companies have what is known as the two-rate scheme, charging one rate for electricity used during certain hours of the day and a lower rate for electricity used during the balance of the day, using a meter with two dials for this purpose. Other companies use an instrument which registers the maximum demand for the month, and the excess over the equivalent of a certain specified number of hours monthly in use of the maximum demand is sold at greatly reduced price. The last scheme would seem particularly equitable, as it results in what is practically an automatic scale of discounts based on the average load factor of the customers. It does not seem to be just that a man who only uses your investment say 100 hours a year should be able to buy your product at precisely the same price as the man who uses your investment say 3,000 hours a year, when the amount of money invested to take care of either customer is precisely the same. Surely the customer who uses the product on an average 30 times longer than the customer using it for only 100 hours is entitled to a much lower unit rate, in view of the fact that the expense for interest to the company is in one case but a fraction per unit of output of what it is in the other. This fact is illustrated by the interest columns on the graphic chart already referred to. Supposing that the central station manager desired to sell his product at cost, that is, an amount sufficient to cover his operating, repairs and renewals, general expense, and interest and depreciation, he would have to obtain from the customer having the poorest load factor, as shown on the load chart, over four times as much per unit of electricity as it would be necessary for him to collect from the customer having the largest load factor. No one would think of going to a bank to borrow money and expect to pay precisely the same total interest whether he required the money for one month or for twelve; and for the same reason it seems an absurdity to sell electricity to the customer who uses it but a comparatively few hours a year at the same price at which you would sell it to the customer using it ten hours a day and three hundred days a year, when it is remembered that interest is the largest factor in cost, and the total amount of interest is the same with the customer using it but a few hours a year as it is with the customer using it practically all the year around.

I have dwelt thus at length on the question of interest cost in operating a central station system, not alone for the purpose of pointing out to you its importance in connection with an electrical distribution system, but also to impress upon you its importance as a factor in cost; in fact, the most important factor in cost in any public service business which you may enter after leaving this institution. Most of the businesses presenting the greatest possibilities from the point of view of an engineering career are those requiring very large investment and having a comparatively small turnover or yearly income. Of necessity, in all enterprises of this character, the main factor of cost is interest, and if you intend following engineering as a profession, my advice to you would be to learn first the value of money, or, to put it another way, to learn the cost of money.

Before leaving this question of interest and its effect upon cost, I would draw your attention to the fact that while interest is by far the most important factor of cost, it is a constantly reducing amount per unit of maximum output in practically every central station system. When a system is first installed, it is the rule to make large enough investment in real estate and buildings to take care of many times the output obtained in the first year or so of operation. As a rule, the generating plant from the boilers to the switchboard is designed with only sufficient surplus to last a year or so. In the case of the distributing system the same course is followed as in the case of real estate and buildings, with a view to minimizing the ultimate investment. Mains are laid along each block facing, feeders are put in having a capacity far beyond the necessity of the moment; consequently interest cost is very high when a plant first starts, except, as I have stated, in the case of the machinery forming the generating plant itself. As the business increases from, year to year, the item of interest per unit of maximum output consequently will constantly decrease, owing to the fact that each additional unit of output following an increase of connected load increases the divisor by which the total interest is divided. The result is from year to year the interest cost of each additional unit of maximum output is a constantly reducing amount, and consequently the average interest cost of each unit of maximum output should, in a well regulated plant, grow less from year to year until the minimum interest cost per unit is reached. This minimum interest cost is reached when the capacity of the whole system and the total units of output at maximum load are identical, although of course it will always be necessary to have a certain margin of capacity over possible output, as a factor of safety.

This same rule, although to a less extent, applies to the operating and general expense cost, that is, the cost other than interest. To particularize, the manager's salary and other administrative expenses do not increase in proportion to maximum output of station; therefore, the cost of administration per unit of output, if the business is in a healthy condition, must be from year to year reduced. There are a great many other expenses that are not directly in proportion to output, and these follow the same rule. In a well-run plant the percentage of operating expenses to gross receipts will stand even year after year, while the income per unit of output will be constantly reduced. This is an excellent evidence of the fact that the cost per unit of output is constantly being reduced, as, if it were not, the percentage of expenses to gross receipts would be increased in direct proportion to the reduction in price. Moreover, it should be borne in mind that there are many difficulties in the way of universal use of electric energy from a central station system. It is the rare exception to find a house not piped for gas and water. In the case of the latter it is almost invariably the rule that owners are compelled to pipe for water, under the sanitary code of the municipality. On the other hand, in a large residential district, it is the exception to find a house wired for electric light; consequently the output of current per foot of conductor is at the present time very low as compared with the output of gas per foot of gas pipe in any of the large cities. The expense of wiring (which must of necessity be borne by the householder) is large, and it is often a barrier to the adoption of electric illumination, but as the rule to wire houses becomes more general, the output per foot of main will constantly increase, and therefore the interest per unit of output per foot of main will constantly decrease. This same rule will apply in the case of expenses of taking care of and repairing the distribution system, although to not so great an extent.

If you will take into account these various factors constantly operating toward a reduction of operating and general expense cost, and interest cost, the conclusion must necessarily be forced upon you that the price at which current can be sold at a profit to-day is in no sense a measure of the income per unit which it will be necessary for central station managers to obtain in the future. In 1881-82 it was difficult to make both ends meet with an income of 25 cents per kilowatt hour, to-day there are many stations showing a substantial return on their investment whose average income does not exceed 7 cents per kilowatt hour, showing 70 per cent. reduction in price in less than two decades. How far this constant reduction in cost, followed by a constant reduction in selling price, will go, it is difficult to determine; but if so much has been accomplished during the first 20 years of the existence of the industry, is it too much to predict that in a far less time than the succeeding 20 years electric current for all purposes will be within the reach of the smallest householder and the poorest citizen? But few industries can parallel the record already obtained. If you will trace the history of the introduction of gas as an illuminant, you will find that it took a much longer time to establish it on a commercial basis than it has taken to establish most firmly the electric lighting industry. All the great improvements in gas, the introduction of water gas, the economizing in consumption by the use of the Welsbach burner, have all been made within the time of those before me, and yet, notwithstanding that when these gas improvements started, the electric lighting business was hardly conceived, and certainly had not advanced to a point where you could claim that it had passed the experimental stage—notwithstanding this, the cost of electrical energy has decreased so rapidly that to-day there are many large central station plants making handsome returns on their investments at a far lower average income per unit of light than the income obtained by the gas company in the same community. In making my calculations which have led me to this conclusion, I have assumed that 10,000 watts are equal to 1,000 feet of gas. This comparison holds good, provided an incandescent lamp of high economy is used as against the ordinary gas burner. To make a comparison between electric illumination and incandescent gas burners, such as the Welsbach burner, you must figure on the use of an arc lamp in the electric circuit instead of an incandescent lamp, which is certainly fair when it is remembered that incandescent gas burners are, as a rule, used in places where arc lamps should be used if electric illumination is employed.

With such brilliant results obtained in the past, the prospects of the central station industry are certainly most dazzling. While the growth of the business has been phenomenal, more especially since 1890, I think it can be conservatively stated that we have scarcely entered upon the threshold of the development which may be expected in the future. In very few cities in the United States can you find that electric illumination exceeds more than 20 per cent. of the total artificial illumination for which the citizens pay. If this be the state of affairs in connection with the use of electricity for illuminating purposes, and if you will bear in mind the many other purposes to which electricity can be adapted throughout a city and supplied to customers in small quantities, you may get some faint conception of the possible consumption of electrical energy in the not far distant future. Methods of producing it may change, but these methods cannot possibly go into use unless their adoption is justified by saving in the cost of production—a saving which must be sufficient to show a profit above the interest and depreciation on the new plant employed. It is within the realms of possibility that the present form of generating station may be entirely dispensed with. It has already been demonstrated experimentally that electrical energy may be produced direct from the coal itself without the intervention of the boiler, engine and dynamo machine. Whether this can be done commercially remains to be proved. Whatever changes may take place in generating methods, I should, were I not engaged in a business which affords so many remarkable surprises, be inclined to question the possibility of any further material change in the distributing system. Improvements in the translating devices, such as lamps, may add enormously to the capacity of the distributing system per unit of light; but it does seem to me that the system itself, as originally conceived, is to a large extent a permanency. Should any great improvements take place in the medium employed for turning electrical energy into light, the possible effect on cost, and consequently selling price, would be enormous.

The proposal of Gov. Black, which has now become law, to depute to Cornell the care of a considerable tract of forest land, and the duty of demonstrating to Americans the theory, methods and profits of scientific forestry, has a curious appropriateness much commented on at the university, since two-thirds of the wealth of Cornell has been derived from the location and skillful management of forest lands, the net receipts from this source being to date $4,112,000. In the course of twenty years management the university has thrice sold the timber on some pieces of land which it still holds, and received a larger price at the third sale than at the first. The conduct of this land business is so systematized that the treasurer of the university knows to a dot the amount of pine, hemlock, birch, maple, basswood and oak timber, even to the number of potential railroad ties, telegraph poles and fence posts on each fourth part of a quarter section owned by Cornell. Certainly, Cornell is rich in experience for the business side of a forestry experiment such as Gov. Black proposes. The university forest lands from which its endowment has been realized are in Wisconsin.

Books may be called heavy when the qualifying term is not applied to their writers, but to the paper makers. It is falsifications in the paper that give it weight. Sulphate of baryta, the well known adulterate of white lead, does the work. A correspondent, writing to The London Saturday Review, gives the weight of certain books as: Miss Kingsley's "Travels in Africa." 3 pounds 5 ounces; "Tragedy of the CÆsars," 3 pounds; Mahan's "Nelson" (1 vol.), 2 pounds 10 ounces; "Tennyson" (1 vol.), 2 pounds 6 ounces; "Life and Letters of Jowett" (1 vol.), 2 pounds 1 ounce. To handle these dumb-bell books, The Saturday Review advises that readers take lessons in athletics.[1]

Before the Electrical Engineering Department of Purdue University, Lafayette, Ind., May 17, 1898.