THE COLLIERY GUARDIAN AND JOURNAL OF THE COAL AND IRON TRADES. Vol. CXII. FRIDAY, JULY 28, 1916. No. 2900. Electric Signalling in Mines. The Home Office has issued a report on electric signalling with bare wires so far as regards the danger of ignition of inflammable gaseous mixtures by the break- flash at the signal wires, by Dr. R. V. Wheeler, of the Home Office Experimental Station, Eskmeals, and Prof. W. M. Thornton, of Armstrong College, Newcastle-upon- Tyne. The previous report,* dealing with battery-bell signalling systems, showed that the break-flash that occurs at uncovered signal wires when a signal is given can, in the majority of cases, ignite inflammable gas, and must, therefore, be regarded as “ open sparking.” At the same time, it was shown that the break-flash can be rendered innocuous. The object of the present report is to define the pre- cautions necessary in order that the bare-wire signalling system shall be rendered safe. Given certain restrictions as to battery power, there are several ways in which safe and efficient bells and relays can be constructed. Battery Bells. The conclusions reached respecting battery-bell systems of signalling in the report already referred to have been amply confirmed by further experiments. Further examples of bells have been received at the experimental station from different makers and tested, and it is clear that the ordinary type of bell causes a dangerous break-flash at the signal wires when a battery of either dry or wet Leclanche cells, of voltage in some cases as low as 4-5, is employed. On the other hand, examples of bells were received which could be used safely with a battery of dry cells at 25 volts. Following the method of testing previously adopted, the “ battery efficiencies ” of all the bells received were determined, the measure adopted being the resistance (in ohms) that had to be included in the circuit in order to render the bell only just capable of ringing. This was Table I. Bell number. Resistance of magnet coils. Ohms. Minimum ringing current. Amperes. 25 volts dry cells. 15 volts dry cells. 26 volts wet cells. 15 volts wet cells. Minimum igniting current at 25 volts. Amperes. Coefficient of self- induction. Measured at the igniting current. Henries. ! Maximum current. Amperes. Resistance to reduce to minimum. Ohms. Maximum current. Amperes. Resistance to re luce to minimum. Ohms. Maximum current. Amperes. Resistance to reduce to minimum. Ohms. Maximum current. Amperes. Resistance to reduce to minimum. Ohms. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 (L) 9*0 11*0 11 5 12*7 12*7 14*2 14*8 21*2 22*0 25*0 25*3 28*9 28*9 500 52*0 106*8 120*5 '(2.) 0*13 0*29 0T3 008 0*12 012 0*08 0*06 0*21 0*20 0*11 0*10 0*05’ 0*15 0*05 007 0*06 (3.) 1*80 1*68 1*48 1*41 1*41 1*26 1*26 1*00 0*95 0*88 0*86 0*75 0*75 0*46 0*45 0*21 0*19 (4-) 170 72 170 275 175 170 270 340 90 95 188 195 400 no 395 210 230 i (5-.) ! 1*28 1*18 1*04 0 96 0*96 0 90 0*90 0*66 0*61 0*58 0*55 0*49 0*49 0*29 0 28 0’13 0*12 (6.) 100 40 98 160 ■ 100 ! 100 160 i 200 ! 45 1 50 105 no ; 230 | 47 220 85 100 (7.) 0*75 0*74 0'72 0*69 0*69 0*65 0*66 0*57 0*55 0*54 0*53 0*49 0*49 0*33 0*34 0*18 0*16 (8.) 160 51 152 250 155 150 260 320 70 80 172 186 360 92 375 190 210 (9.) 0*60 0*57 , 0*56 0’54 0*53 0*50 0*49 0 42 0*40 0*39 0*37 0*34 0*34 0 22 0*22 0*11 0*10 (io.) 75 25 81 145 85 80 145 180 30 35 90 90 210 30 1% 67 70 (11.) 0*22 0*22 0 14 0*18 0*23 0T2 0*20 0*20 0*18 t 0*20 0*11 0*10 0*12 0*08 + § (12.) 0’37 0*46 0*46 0*45 0*43 0*64 0*46 0*70 0*46 0*57 1*09 0*95 1*10 2*43 + This bell had parallel short-circuited winding. Ignition was obtained with 1T4 amperes at 41 volts. X This bell had a copper sleeve. Ignition was obtained with 0*35 ampere at 41 volts. § This bell was wound with brass wire. Ignition was obtained with 0*17 ampere at 22 volts. determined with batteries of 25 and 15 volts dry cells (” Dania,” size C), and 25 and 15 volts wet Leclanche cells (quart size). The results are recorded in Table I., which also shows the maximum currents obtained in each case when the resistance of the bell only was included in the circuit; the currents required to cause the ignition of an 8 per cent, methane-air mixture by the break-flash at the signal wires and the inductances of the bells at these igniting currents. As was shown in the previous report, the usual types of bells are all capable of giving a dangerous break-flash at bare signal wires, and nearly every flash that* occurs when the bare wires are separated after a signal has been given would ignite a firedamp-air mixture containing 7-5 to 9-5 per cent, of methane if present. The only break-flashes that cannot ignite such mixtures are those that occur so far along the signal wires that the resis- tance of the wires reduces the current available at the flash below the minimum igniting current. Comparison between columns 2 and 11 in Table I. shows that with * Colliery Guardian, April! 23, 1915, p. 855. many bells the minimum igniting current is but little greater or is even less than the minimum ringing current. The maximum battery power permissible under the Coal Mines Act is 25 volts, and column 4 shows that the battery efficiency at this pressure when dry cells are used varies from 400 bo 72 ohms, the mean value for all the bells being 200 ohms; that is to say, 200 ohms resistance can be included in the circuit before the bell ceases to ring. If a battery of dry cells at 15 volts is used, the mean efficiency drops to about 120 ohms, with a minimum of 40 ohms; while with wet Leclanche cells at 15 volts it drops to about 90 ohms with a minimum of 25 ohms. Enquiries as to present practice as regards the type of battery usually employed show that wet Leclanche cells are generally favoured. Reports from his Majesty’s inspectors of mines are agreed in stating that, except in North Staffordshire, where 10 per cent, of the battery bells used are actuated by dry cells, the use of wet Leclanche cells is practically universal. There should, therefore, be no great inconvenience caused if the use of wet Leclanche cells were made compulsory. Compari- son of columns 3 and 7 or 5 and 9 of Table I. shows the desirability of such a restriction. It is questionable, however, whether it would be wise to reduce the voltage allowable from 25 to 15 or less. Even with 3 or 4 volts some types of bells at present in use are unsafe, whereas the introduction of “ anti- sparking ” devices, as illustrated by bells Nos. 10 and 16, would allow a battery at 25 volts safely to be used. While, therefore, no gain in safety would be experienced by stipulating a maximum voltage of, say, 15, the ringing efficiency of most of the bells would be seriously affected thereby. On the other hand, a battery of wet Leclanche cells at 25 volts has little greater danger than any battery necessary to give a reasonable working current. Relays. As a considerable amount of mechanical work is expended in ringing a bell, a corresponding amount of additional battery power is required beyond that neces- sary to overcome the total resistance of the circuit. A relay introduced into the main signalling circuit reduces the battery power required to give- a signal by reducing the mechanical work that has to be performed on that circuit, a usual practice being to cause a subsidiary circuit, containing the bell and a separate battery, to be closed by the action of the relay. The only mechanical work required 'of the relay circuit is the attraction of the armature of its electro magnet against the pull of a light spring. Since the distance through which this arma- ture need move (in order to close the bell circuit) can be made quite small, and the spring control can be very light, only a small battery is required to actuate a well- designed relay over long distances. Since relays contain electro-magnet coils, often highly self-inductive, the same considerations respecting the possible danger of the break-flash at the signal wires apply as with bells. A number of relays obtained from different makers have been examined, the mode of pro- cedure being similar to that adopted for bells. In Table II. are given the main results of the tests made, which were measurements of “ battery efficiencies ” with different batteries of wet Leclanche cells, determi- nations of ” minimum working currents,” “ minimum igniting currents,” and inductances at the igniting currents. In a separate column are recorded the num- bers of wet Leclanche cells which caused the break-flash at the signal wires to ignite an 8 per cent, methane-air mixture, only the resistance due to the magnet coils of the relay being in circuit. The commonest types of relays are those represented by Nos. 1, 2, 3, 6, and 7. The battery efficiencies are greatly superior to those of most bells, a battery power of 10 volts in most cases allowing ample margin for the resistance of bad connections on the signal wires. This superior efficiency is discounted by the higher induct- ances of relays compared with bells, due to the larger number of turns of wire with which the relay magnets are wound; so that though relays can be worked with a smaller battery than most bells, they have, in general, a lower “ igniting current ” at which the break-flash on the signal wires becomes dangerous. As regards present practice in respect of the use of relays, more particularly as regards the battery power usually employed, it would appear from reports obtained from H.M. inspectors of mines that although in the majority of mines smaller batteries are employed to actuate relays than would be used for bells over the same length of line, yet, with few exceptions, the battery power is in excess of what is sufficient to give a dan- gerous break-flash at the signal wires. The commonest types of relays are dangerous when used with a battery of more than four wet Leclanche cells, while the reports of H.M. inspectors of mines show that a battery of less than six such cells is used comparatively rarely. The same methods can be applied to render relays safe as for bells. Relays Nos. 10 and 11 (Table II.) are examples of what can be done in this direction. Details regarding the application of safety devices to relays are given in a later section. Growth of Current. The current which flows through the circuit when a signal is given is acted upon by two pressures : the electromotive force of the battery urging it forward, and the electromotive force of self-induction holding it back, the combined effect being to make the growth of the current obey a lagarithmic law. This growth can be represented by a curve which rises rapidly at first, and approaches its maximum value, E/r, more and more slowly. The character of the curve indicates the amount of inductance of the circuit; since a rapid rate of growth must be associated with a low inductance, and vice versa. Tmif <»>"! lid Fig. 1. In fig. 1 are given curves illustrating the manner of growth of current in a trembler bell, observed in a revolving mirror by means of an oscillograph. Current is measured along the vertical lines and time along the horizontal, and a succession of “ makes ” and “ breaks ” by the trembler contacts is shown. In studying such curves, it is convenient to adopt as the unit of time the ratio L/r for the circuit, the ■so-called “ time constant.” During the growth of the current the effective voltage is increasing also, until it reaches the battery voltage, while the resistance of the circuit remains constant. When, however, the metallic circuit is broken, either at the trembler of the bell or at the signal wires, the resist- ance is suddenly increased. The electromotive force of self-induction, tending to maintain the decreasing current, therefore rises, and may become many times greater than the original (battery) voltage of the circuit, and is sufficient to cause a discharge through the small air space which at the first instant separates the ends of the metallic circuit. The incandescent particles of metal and air which constitute the flash act for an instant as a conductor, and a momentary electric arc is established. It will be apparent that a flash can only occur on breaking circuit at the signal wires if the trembler con- tacts at the bell are closed at that moment. The greatest flash that can occur at the signal wires is, however, no