174 THE COLLIERY GUARDIAN. January 25, 1918. Fig. 10 shows another system of supply which enables a very considerable saving in copper to be effected. This is a three-phase balanced wiring, which may be of very small section, compared with a single-phase supply. The lamps are connected from the three phases to the neutral, the circuit being balanced as closely as possible, which prevents the neutral wire carrying any current. The lamps are connected much the same as in a three- wire direct-current scheme, i.e., the lamps are two in series across the lines, but the centre points are all connected to the neutral wire, which is usually allowed to be one-fourth the cross section of one of the outers. Since the lamps work from line to neutral, each outer wire will carry one-third of the current, 3° = 10 o amperes. The resistance per line for equal total loss will be 102 x a* = 30 watts .*. x = 0-30 ohm, or six times the resistance of one single-phase line, and conse- quently only one-sixth of its weight. As one wire is one-sixth one single-phase line, it will be one-twelfth the total single-phase copper, as there are two lines; and therefore, with three wires one-twelfth the size and weight we have three-twelfths = one-quarter, i.e., 25 per cent, of the copper used in the single-phase line. Fig. 7. ___________C X r/ 3 3 A. R ~ *|oo oh Fig. 8. The neutral being one-quarter the weight of one of the three-phase lines, it will be one-48th the total single-phase copper, i.e., approximately 2 per cent., making the total weight of copper, required for a cir- cuit as shown in fig. 10, only 27 per cent, that necessary for direct-current or single-phase working. The drawbacks to this system of supply are, roughly, two : the use of four wires, and the voltage which would exist between lines in the tubing carrying the wires. This voltage will be equal to 50 x ^3 = 86'5 volts. However, if this system is used in tubing of ample size, with plenty of draw-in boxes, no difficulty should be experienced; neither should there be any danger from the extra voltage existing between the lines. Voltage Drop in Transmission. It has already been noticed, from the curves for low- voltage supply, that a very slight drop in voltage seri- ously affects the candle-power; and, assuming the source of supply gave an electro-motive force of the value required at the lamps, it would be necessary to put in very large cables in order to reduce the drop in volts to a minimum. However, a characteristic common to underground lighting circuits assists in solving this difficulty due to voltage drop. Usually, most of the pit lights are controlled by one main switch, and therefore the load on the transmission line, be it long or short, is constant. This being so, it may be designed to give a definite drop in volts between the point of supply, and that where the lamp circuits are connected, so that a reasonable size cable may be used. For instance, assume a simple single-phase supply, to give 50 volts at the lamps, and the supply has to be transmitted 100 yds. before coming to the lamps. There is no reason why, after calculating the current which will be required, the cable should not be of a resistance high enough to give a drop in voltage along its length of five or more volts, the pressure of supply being 55 volts, which would be reduced to 50 volts by the time the lamps were reached. In this way, by careful calculation, quite a reasonable size cable would suffice where otherwise a much larger one would be required. Voltage Drop in Lighting Circuits. Having reached the lighting circuits, the question of voltage drop must again be considered, in order that lamps connected to the extreme ends of the circuits shall have sufficient pressure to give their rated candle- power as near as possible. Most underground lighting consists of lighting points placed at intervals along the roadway, and may extend for a considerable distance. To calculate the size of a cable necessary to supplv such a circuit to give a definite drop at the end of the line appears rather difficult, owing to the fact that the amount of current passing along the line diminishes as each lamp is passed, until, from the second lamp from the end, to the last lamp, the wires only carry the current of a single lamp. The ideal cable to meet such a case would be one tapering down in the same propor- tion as the current; but such cables are not made. If it is definitely known that no extensions beyond that contemplated are likely to take place, only the,lamps about to be installed would be considered. Take the case of a straightforward run of 20 lights, each taking 0*5 ampere, i.e., a total of 10 amperes, the lights being spaced 30 ft. apart over a distance of 570 ft. An ordinary two-wire circuit will be considered for simplicity, but it can easily be applied to any other. A drop of two volts will be assumed at the last lamp. As the current is gradually getting smaller as the lamps are passed, it would not be right to base the resistance of the required cable upon the total current supplied, i.e., 10 amperes. The easiest method is to calculate the size of cable by halving the number of lights, and finding the size of cable required to give the pre-determined drop in volts In the present case, instead of the drop being based upon 10 amperes, it would be based on 5 amperes. E total resistance of the line would be R = - C The 2 _ 5 0-4 ohm, or 0-2 ohm per line. The resistance per line no from lamp to lamp would be ^-0-0105 ohm, and for both lines, i.e., lead and return, 0*0105 x 2 = 0-210 ohm. The various drops along the line are approximately as follow:—0-199, 0*189, 0-178, 0-168, 0-157, 0*147, 0-136, 0-126, 0-135, 0*105, 0*095, 0-084, 0*077, 0-063, 0-052, 0-042, 0*031, 0-021, 0*010—total, 2-005 volts. It will be noticed that a heavy drop occurs in the beginning of the line, but is compensated by the very small drop towards the end of the line. In the present case the cable suitable to give the required drop would be approximately ?/18 S.W.G. It might be mentioned, in passing, that if a line of lamps can be fed at a point mid-way between the ends, the section of cable for the same drop to the extreme end lights would be only half that required to enable the lights to be fed from one end. It is therefore advisable, in low-voltage lighting, not to miss such an opportunity, should it present itself. Transformers. Transformers designed for low-voltage supply should have their secondaries provided with tapping of not more than three volts between each. In the case of a 50-volt transformer, there should be at least one tapping above, and one below, the normal voltage, so that, in the event of an error in calculation of the con- ductors of the circuit, either an increase or a reduc- tion can be effected to make an adjustment. When a very low voltage is required for upcast light- ing, it may sometimes be convenient to transform down from the lighting voltage of the downcast pit bottom, instead of transforming from a high voltage to a very low one. This method is certainly more desir- able, as it would not be so serious if the primary side happened to break down into the secondary. In any case, provision should be made to safeguard any such occurrence. In the Memorandum attached to the Rules for the Use of Electricity in Mines, it is suggested that small lighting transformers should be run in solid like a joint box. If this suggestion is meant to convey the idea that small transformers for lighting a roadway may be placed on a power cable and run in solid, it hardly commends itself, as, in the first place, the pro- Fig. 9. C X i c JO . A . "R - -30. oh-m > Fig. 10. tection afforded in case of breakdown of the trans- formers is not sufficient, especially where gas is likely to be present; and, in the second place, should a short- circuit occur between the turns of the transformer windings, and the whole become hot, there would be a grave possibility of the transformer being blown to pieces, as has occurred to many joint boxes, which have literally exploded under certain conditions. This would not only extinguish the lights, but would pro- bably so damage the cable as to put it out of commis- sion, causing much inconvenience if the cable be one supplying power to motors. Oil-immersed transformers which allow of their tanks being completely filled with oil, are the best type to use ; whilst if no oil be used, the casing should be well ventilated and thoroughly flame-proof. Switchgear. All underground lighting circuits should be adequately protected. Fuses should be prohibited altogether in return airways, whether in flame-proof boxes or otherwise. Oil-immersed automatic switches are the only apparatus which meet the needs of safe and efficient control of electrical circuits underground. For low-pressure lighting, the switches employed should be of ample capacity, and the trip coils of very low resistance, so that the drop in volts across the switch contacts and through the trip coils will be a minimum. Since a lighting load is constant, the switch may be set quite close to the operating load, so that very little extra current will be sufficient to open the circuit. The first prize of £500 in the Glasgow tramway tombola, which realised over £2,000, was won by a Stirlingshire miner, who visited the city on the first day on which the tickets were sold. COKE AS BLAST FURNACE FUEL. Mr. Henry Peile, of the Priestman Collieries Limited, presided over a meeting of members of the Newcastle Section of the Society of Chemical Industry, held on Wednesday, January 16, at which Mr. G. W. Hewson, chief chemist to Messrs. Palmer’s Ship- building and Iron Company Limited, Jarrow, spoke on “ Coke as a Fuel for the Blastfurnace.” Mr. Hewson stated that without coke many of the metallurgical advances of the past century would have been unattainable. The substitution of coke for char- coal or coal as a blast furnace fuel introduced an era of increasing productiveness from the blast furnace, resulting from the greater height of furnace that could be used, and the increased pressure of blast permis- sible. The increase in height was largely due to the ability of the coke to stand the crushing action of the descending materials much better than coal or char- coal, and thus to reach the tuyeres before which there was the greatest combustion. The chief object of the use of fuel in a blast furnace was the production of intense local heat in the vicinity of the hearth, the heat acting as a reducing agent. The materials in the hearth melted at a rate proportional to the rate of heat development, and that, in turn, was dependent upon the rate of combustion between the carbon of the fuel and the oxygen of the blast. Forsythe said the characteristics which made a fuel desirable for blast furnace use were: (1) A well- developed cell structure, since porous fuel would pre- sent more surface to the action of the blast than a dense fuel, and, therefore, facilitate and hasten com- busion. (2) Firmness. A fuel which changed its shape in the furnace, either through being crushed by the weight of the accompanying materials or through softening under the action of heat, was undesirable, as the filling of the interstices of the charge with fine or pasty material impeded the current of gases, and hampered combustion. (3) Purity. Other conditions being equal, it was evident that the higher the fixed carbon the more efficient the fuel. The non-carbon- aceous material developed no heat, but formed a slag, which absorbed heat in melting. Whilst charcoal was more desirable than any other fuel for its readiness to combine with the oxygen of the blast, and for its purity, the impossibility of the supply meeting the demand ruled it out of consideration, except in a limited way, as a blast furnace fuel. In a general way, coke came next in efficiency and value, and was more desirable than coal. The demand for coke became so great, that thousands of beehive ovens must have been built, with the sole object of supplying coke for blast furnaces, and, so good was that coke, that, at a later stage of coke making, many managers preferred beehive coke to by-product. Beehive ovens, however, had their limitations. Coking small coal in such ovens was only possible as long as the coal pos- sessed sufficient agglutinative power to produce coke at temperatures within the limits of what beehive ovens were capable of withstanding; so that beehive ovens began to be superseded. Retort ovens were first used to produce coke from coals low in volatile matter and agglutinative power, to obtain a greater yield of coke, and to increase coke production by shortening the coking period. Improvements effected to obtain such results included: (1) Improvements in the arrange- ments for controlling the admission and circulation of gas and air in the flues in such a way as to produce higher temperatures and to distribute the heat uniformly along the whole length and height of the oven walls; (2) adapting the dimensions of the ovens and the coking time to suit the nature of the coal; and (3) improving the firebrick work so as to allow the side walls to be diminished in thickness, to give a more i apid conduction of heat to the ovens. Later, modi- fications had from time to time been made to? collect various by-products from the so-called waste gases expelled from the ovens, and variations in the rate of coking and temperature had been made to improve the yield of by-products. There was a limit to which that could safely be carried. Working Factors. The blast furnace, although unchanged in principle for several centuries, remained the most complex in principle and most difficult to operate of any metal- lurgical apparatus, and it was very necessary that all factors affecting its operation which could be con- trolled and regulated should receive very serious atten- tion. Among the chief of these was the physical and chemical composition of the coke used, lack of regu- larity and uniformity in coke quality being a verv serious handicap to the producer of pig iron. The fundamental principle underlying successful opera- tions at the blast furnace was that the composition and quantity of the slag, together with the temperature of the smelting zone and the hearth, determined the quality of the iron produced. These, in turn, were decided principally by the quality of the coke used and the distribution of the burden. In the past, production of pig iron from the blast furnace had been largely a question of putting in the materials in the quantities found by experience to give a good marketable product. Fortunately, the range of composition of saleable pig iron had been somewhat wide, and it had been possible to grade the iron made, and to send supplies to customers according to frac- ture or analysis of the finished product. Modern requirements, however, demanded a product within narrower limits of composition and of greater unifor- mity than had hitherto been obtainable. These demands were such that a slight modification in the furnace working might have a very detrimental effect upon the operations. The principal source of heat was the coke charged. With reservations, the available carbon of the coke was a measure of its heat producing capacity. Steel furnaces had, in the past, taken pig iron varying from 2 to 3*5 per cent, silicon, largely because thereby they were more assured of an iron sufficiently low in sulphur to ensure a steel within the limits allowed. With the introduction and develop-