library ELECTRIC IGNITION FOR COMBUSTION MOTORS BY FORREST R. JONES, M.E. Member of the A merican Society of Mechanical Engineers, and of the Society for the Promotion of Engineering Education. Formerly Professor of Mechanic Arts in the University of Tennessee, and of Machine Design successively in the University of Wisconsin, the Worcester Polytechnic Institute and Cornell University FIRST EDITION FIRST THOUSAND NEW YORK JOHN WILEY & SONS LONDON: CHAPMAN & HALL, LIMITED 1912 COPYRIGHT, 1912, BY FORREST R. JONES Stanhope ipms F. H. GILSON COMPANY BOSTON. U.S. A / ^'773 Engineering Library PREFACE. THE plan of this work is based upon the supposition that some of its readers may possibly be unfamiliar with electricity and electrical devices. To meet this condition, the fundamen- tal principles involved in each case are given before commercial forms are described. The range of the subject matter is from the small ignition apparatus used on motor cycles to that used on the largest gas engines. More than half of the illustrations have been prepared espe- cially for the book, largely from sketches, drawings, photographs, and information kindly furnished by those making or dealing in ignition appliances. Electric connections, both internal and external, are shown, by wiring diagrams and other means, for complete ignition sys- tems and the parts of which they are composed. Considerable attention is given to the operation, care, adjust- ment, and testing of ignition systems and their various parts. Lack of knowledge in this respect is the chief cause of ignition troubles in connection with the excellent equipments now ob- tainable. FORREST R. JONES. KNOXVILLE, TENNESSEE, January, 1912. 257830 CONTENTS. CHAPTER I. INTRODUCTORY. ART. PAGE 1. Electric ignition in general i 2. High-tension and low-tension electricity i 3. Sources of electricity 3 4. Permanent magnets. Forms and action 3 5. Poles of a magnet 5 6. Magnetic field 6 7. Magnet keeper and its effect 6 8. Magnetic and non-magnetic materials 7 9. Compound or composite magnets 8 10. Principle of electric generators & 11. Armature of electric generator 9 CHAPTER II. LOW-TENSION ALTERNATING-CURRENT MAGNETOS WITH SINGLE-WOUND SHUTTLE ARMATURES. 12. Field magnets of a magneto 10 13. Abutted magnets n 14. Armature of magneto n 15. Electric arc from a shuttle-wound armature 14 16. Effect of speed of armature on position for maximum arc 16 17. Positions of armature for strong electric arc 16 18. Laminated armature core 18 19. Magnetic flux in a rotating I-shaped armature core 19 20. Electromotive force and current induced in an armature 24 21. Armature lag 24 22. Alternating current generated 25 23. Graphical representation of current in a shuttle-wound armature 28 24. Cycle of current 29 25. Form of current curve as affected by shape of pole-pieces 29 26. Position of armature for maximum arc 30 27. Low-tension alternating-current magneto with shuttle- wound armature 31 28. Stationary armature and rotary magnetic sleeve 35 29. Action of the magnetic sleeve 37 30. Note regarding low- tension magnetos for high-tension ignition 37 CHAPTER III. DIRECT-CURRENT MAGNETOS. 31. General 38 32. Elementary form of drum armature , 39 33. Generation of current 39 34. Commutation of current in a direct-current generator 41 35. Continuous-current electric generator 43 36. Laminated drum armature core 43 v VI CONTENTS ART. PAGE 37. Complete drum armature 44 38. Commutator for direct-current generator 44 39. Armature connections 45 40. Complete direct-current magneto 48 CHAPTER IV. TESTING FOR DIRECTION OF CURRENT. 41. Water test. Bubbles at submerged wire-end 50 42. Color test 51 43. Magnetic compass test 52 44. Extemporized compass needle 53 45. Test with measuring instruments v 53 CHAPTER V. ELECTRIC MEASURING INSTRUMENTS. 46. General 54 47. Ammeters 54 48. Voltmeters 56 49. Volt-ammeters 57 50. Dead-beat indicating needle 59 CHAPTER VI. ELECTROMAGNETS. 51. Plain bar electromagnet 60 52. Plunger-core electromagnet 61 53. U-shaped electromagnet 62 54. Ring-shaped electromagnet with consequent poles 64 55. Bipolar ring-shaped electromagnet 64 56. Four-pole ring-shaped electromagnet 64 CHAPTER VII. DIRECT-CURRENT GENERATORS WITH ELECTROMAGNETS. 57. General 65 58. Bipolar direct-current generator with U-shaped shunt-wound electro- magnets 65 59. Bipolar direct-current generator with ring-shaped shunt-wound electro- magnets 67 60. Four-pole direct-current generator with shunt- wound electromagnets. . . 71 61. Series-and-shunt field winding - 74 62. Field rheostat for regulating voltage 75 63. Reversing the rotation of the armature 76 CHAPTER VIII. PRIMARY BATTERIES. 64. Carbon-zinc battery 78 65. Elementary Leclanche carbon-zinc wet cell 78 66. Polarization of primary electric cell 80 67. Dry cell with carbon and zinc electrodes 80 CONTENTS Vll ART. * AGE 68. Deterioration of dry batteries while idle 82 69. New type of carbon-zinc dry cell 82 70. Exhaustion of dry batteries in service 83 71. Recuperation of dry cells 83 72. Lalande and Chaperon wet cell 84 73. BSCO wet cell 84 74. Edison primary battery 86 CHAPTER IX. BATTERY CONNECTIONS. 75. General 88 76. Series-connected battery of four cells . . . 88 77. Reversed cell in a series battery 89 78. Parallel-connected battery of four cells 90 79. Reversed cell in a parallel-connected battery 91 80. Parallel-series batteries 92 81. Wrong arrangement of' a battery 93 82. Connection to external circuit 94 83. Screw-top battery cells . 95 CHAPTER X. STORAGE BATTERIES, ALSO CALLED ACCUMULATORS AND SECONDARY BATTERIES. 84. Storage battery defined 97 85. Electrodes, or plates, of a lead storage cell 98 86. Complete storage cell 99 87. Voltage of a storage cell having lead plates 101 88. Maximum rate of discharge of storage cell 102 89. General description of storage battery .^ 102 90. Exide storage battery 104 91. Charging the storage battery 107 92. Chemical action in a lead storage battery 108 93. Capacity of a storage cell 109 CHAPTER XI. FLOATING THE STORAGE BATTERY ON THE LINE OF A DIRECT-CURRENT GENERATOR. 94. Method of floating the battery on the line no 95. Automatic cut-out 113 96. Two- voltage system with two storage batteries floated on the line 117 97. Two-voltage system with lamps and ignition apparatus 118 98. Switchboard for two- voltage system with battery floated on the line . . 119 CHAPTER XII. MECHANICALLY OPERATED MAKE-AND-BREAK IGNITERS AND KICK-COILS FOR LOW-TENSION IGNITION. 99. Mechanically operated igniter : 123 100. Duration of contact between the electrodes 126 101. Bosch mechanically operated igniter 126 102. Truscott Boat Manufacturing Company's igniter 128 viii CONTENTS ART. PAGE 103. Fay & Bowen low-tension igniter 131 104. Westinghouse make-and-break igniters 132 105. Snow Steam Pump Works mechanically operated igniter 133 106. Mechanical make-and-break operating mechanism of the Snow Steam Pump Works igniter 133 107. Four-unit low-tension mechanism 137 108. Allis-Chalmers gas engine with mechanical make-and-break igniters.. . 138 109. Kick-coils 138 1 10. Screw-top kick-coil . 142 in. Tell-tale kick-coil 142 CHAPTER XIII. MECHANICAL MAKE-AND-BREAK LOW-TENSION IGNITION SYSTEMS. 112. Battery, reactance coil, and make-and-break igniter 144 113. Duration of contact 145 114. Magnet and make-and-break igniter 146 115. Direct-current generator, kick-coil, and mechanical make-and-break ig- niter 148 116. System for four combustion chambers. Mechanical make-and-break igniter, storage battery, primary battery, and direct-current generator 149 117. Alternating-current magneto, primary battery, and four make-and- break igniters 150 118. Storage battery floated on the line of a shunt-wound generator, and mechanical make-and-break igniters 151 119. no- volt generator and 6- volt primary battery mechanical make-and- break system 152 1 20. Multiple system with switchboards, primary batteries, and no- volt direct-current generator 155 121. Storage battery and no- volt generator mechanical make-and-break system 159 122. Storage battery, primary battery, and no- volt generator make-and- break system 160 123. Multiple system with storage batteries, no-volt generator, primary battery, and switchboards 161 124. System using current from no-volt direct-current service without ground connection 162 125. Multiple system using iio-volt to 125-volt direct current from general service circuit 165 126. Triple low- tension ignition system for large engine with four double- acting cylinders 165 CHAPTER XIV. ELECTROMAGNETIC IGNITERS AND IGNITION SYSTEMS FOR LOW-TENSION CURRENT. 127. Principle of operation 169 128. Elementary ignition system with a timer and an igniter having an elec- tromagnet with a plunger core 169 129. Dual ignition system with plunger-core electromagnets in the igniter. . 171 130. Wisconsin Engine Company's electromagnetic igniter with plunger cores. Details 175 131. Details of one-ring timer for large engine with four combustion cham- bers 1 78 CONTENTS ix ART. PAGE 132. Allis-Chalmers four-ring timer for large engine 181 133. Details of Allis-Chalmers electromagnetic igniter for a large engine 184 134. Wiring diagram for a four-ring timer and igniters actuated by a rotary magnet armature 191 135. Bosch igniter with vibratory magnet armature 191 136. Magneto and wiring diagram for magnetic-plug ignition system 194 CHAPTER XV. TRANSFORMER SPARK-COILS AND SYNCHRONIZER, OR MASTER, TREMBLER-COILS. 137. General 200 138. Elementary transformer spark-coils 200 139. Operation of elementary transformer spark-coil without trembler 202 140. Trembler transformer spark-coil 204 141. Safety spark-gap 213 142. Lag of spark-coils 214 143. Tremblers, or vibrators: Bow-spring, hammer-break, and plain types. . . 214 144. Complete trembler transformer spark-coils 220 145. Synchronized spark-coils with master-trembler 222 146. Trembler spark-coil for use with high-tension distributor 223 147. Plain transformer spark-coils without a trembler 224 148. Connections to trembler spark-coils 225 CHAPTER XVI. TIMERS AND SPARK-PLUGS FOR HIGH-TENSION IGNITION. 149. Elementary form of timer 227 150. Roller-contact timer 227 151. Sliding-contact timer 230 152. Timer with normal-pressure contacts 230 153. Spark-plugs. General description 231 154. Single-gap jump-spark plugs 231 155. Spark-plugs with two or more spark-gaps 234 156. Separable spark-plugs 236 157. Width of spark-gap 238 CHAPTER XVII. IGNITION SYSTEMS WITH MAGNETIC TREMBLER INTERRUPTERS AND INDIVIDUAL TRANSFORMERS. 158. Introductory 240 159. System with one spark-plug and trembler spark-coil 240 160. Auxiliary condenser in an ignition system 242 161. Grounded spark-coil condenser ; . 243 162. Individual trembler-coil systems 244 163. Synchronized system with master trembler-coil 245 164. Synchronized system with master trembler-coil and auxiliary condensers 247 165. Ammeter and voltmeter permanently in the circuit of a high-tension ignition system 249 166. Speed of the timer 251 CONTENTS CHAPTER XVIII. HIGH-TENSION DISTRIBUTOR SYSTEMS WITH BATTERY CURRENT. ART. PAGE 167. General 252 168. High-tension distributor system with trembler spark-coil 252 169. Mechanically operated contact-maker and high-tension distributor sys- tem with battery current 254 1 70. Unisparker 255 171. Combined contact-maker, spark-coil, and distributor 258 172. Comparison of unisparker and ordinary timer 260 CHAPTER XIX. SPARK-PLUGS IN SERIES AND IN SERIES-SHUNT. 173. Method of operating 261 174. Series-shunt connection of spark-plugs 262 175. Constructive form of series-shunt spark-plug ignition system 265 CHAPTER XX. INTERRUPTER MAGNETOS AND JUMP-SPARK IGNITION SYSTEMS WITH MAGNETO CURRENT ONLY. 176. Introductory 268 177. Interrupted primary-current magneto ignition system 268 178. Interrupted shunt-current magneto ignition system 271 179. Shunted-current magneto ignition system 272 180. High-tension magneto with single-wound armature 272 181. Double-wound high-tension magneto 273 182. Bosch high-tension magneto with double- wound rotary armature 275 183. U. & H. magneto with rotary double-wound shuttle armature 291 184. Remy magneto with stationary armature and rotary inductor 295 185. Effect of advance and retard on the strength of the ignition spark 301 186. Charged-and-discharged condenser ignition system 302 187. Shaft couplings for advancing and retarding the ignition 303 187.1 Eisemann high-tension magneto with automatic spark-advance mechanism 305 188. Interrupted short-circuit magneto 307 189. Movable extensions of magnet poles for constant strength of spark. . . . 309 190. Pittsfield magneto with stationary armature and rocking pole-extensions 311 191. Mea magneto with rocking magnets 316 CHAPTER XXI. HIGH-TENSION DUAL AND COMBINED IGNITION SYSTEMS. 192. Introductory 326 193. Remy ignition system with separate transformer 327 194. Splitdorf ignition system with separate transformer 330 195. Eisemann ignition system with separate transformer 334 196. Eisemann-Carpentier ignition system 337 197. Bosch dual ignition system 340 197. i Duplex high-tension ignition system having a battery in series with the primary of the magneto 345 198. Magneto with two high-tension windings for dual ignition 352 CONTENTS XI CHAPTER XXII. HIGH-FREQUENCY ALTERNATING-CURRENT MAGNETOS. ART. PAGE 199. Introductory 353 200. W. & S. magneto 353 201. K-W high-frequency magneto . 356 202. Ford high-frequency magneto 358 CHAPTER XXIII. VARYING THE TIME OF IGNITION. MULTIPLE IGNITION. 203. Advancing the timer on account of lag in the ignition apparatus 361 204. Varying the time of ignition relative to the rate of combustion of the charge in the motor 362 205. Varying the time of ignition with variable speed 364 206. Reduction of the variation of ignition by the use of two simultaneous ignition sparks 365 207. Advancing and retarding the spark in a variable-speed motor 366 CHAPTER XXIV. CARE AND ADJUSTMENT OF IGNITION SYSTEMS. 208. Introductory 368 209. Cleaning the spark-plug 369 210. Adjusting the width of spark-gap in jump-spark igniters 369 211. Repairing the spark-points of contact igniters 370 212. Adjusting the trembler 370 213. Lubricating and cleaning the timer 371 214. Care of the magneto or dynamo 372 215. Filing or dressing the contact-points 374 216. Care of batteries in general 374 217. Keeping electric connections tight 375 218. Testing a primary battery 376 CHAPTER XXV. TESTING OF STORAGE BATTERIES. 219. Voltage-and-current test of a storage battery 378 220. Ammeter in series with resistance 379 221. Lamp for testing 381 222., Testing cells individually 381 223. Voltmeter test of a storage battery 381 224. Voltage-drop test 381 225. Lowest safe voltages of storage batteries while discharging 382 226. Curve of voltage drop while a storage battery is discharging 383 227. Testing the density of the electrolyte 384 CHAPTER XXVI. CHARGING AND CARE OF STORAGE BATTERIES. 228. Precautions 387 229. Connections for charging a storage battery 387 230. Connections for charging two batteries at the same time 389 231. Rate of charging a storage battery 390 xii CONTENTS ART. PAGE 232. Charging and care of lead-plate storage batteries 390 233. Removing the sediment from lead-plate cells 393 234. Taking a lead-plate cell out of commission 394 235. Charging and care of nickel-iron storage batteries 394 236. Taking a nickel-iron battery out of commission 396 CHAPTER XXVII. TIMING THE IGNITION. 237. General features 397 238. Timing, or setting, the timer 398 239. Timing a rotary magneto of the interrupter type 400 240. Relative positions of the crank-shaft and piston of a motor 402 CHAPTER XXVIII. IGNITION SYSTEM FAULTS AND REMEDIES. 241. Defects and conditions in the ignition system which cause faulty ignition 405 242. In the igniter 405 243. In the spark-coil 407 244. In the reactance coil, or kick-coil 408 245. In the timer 408 246. In the magneto 409 247. In the dynamo 413 248. In the battery 413 249. In the connections 415 CHAPTER XXIX. OPERATING TROUBLES POSSIBLY DUE TO THE IGNITION SYSTEM. 250. Introductory 416 251. Motor will not start 416 252. Preignition 416 253. Back-firing into the intake of the motor 417 254. Overheating of the motor 418 255. Misfiring and exhaust explosions without other serious troubles 418 256. Knocking or pounding 419 257. Sudden stoppage of motor 419 258. Motor does not develop full power 419 259. Spark control must be advanced more than usual, and motor behaves erratically 420 ELECTRIC IGNITION. CHAPTER I. INTRODUCTORY. 1. Electric ignition, as applied to motors, or engines, which burn a combustible mixture of gas inside of the cylinder of the motor, is accomplished, in modern practice, either by producing an electric spark, or by drawing an electric arc, inside of the cylinder in the inclosed space which is filled with the combustible mixture. The spark, or arc, as the case may be, ignites the com- bustible mixture and causes it to burn. High-tension and Low-tension Electricity. 2. There are two classes of electric ignition as applied to combustion motors. One class is known as high-tension, or jump-spark, ignition; the other class as low-tension, make-and- break, contact, or touch-spark ignition. The terms " high-tension " and " low-tension " are used in accordance with the intensity of the tension, also called pressure, of the electricity that is used to produce the spark, or arc, for igniting the combustible charge. An idea of the distinction between high-tension and low-tension electricity as used for ignition can be obtained from its action on the animal, or human, body. The high-tension electricity used for ignition is capable of giving a very severe shock to one who touches the metal of a wire or apparatus which is charged with the high-tension electricity. The shock is not dangerous, however, unless continued for a considerable time. The low-tension electricity for ignition in motors used on automobiles, traction engines, and other similar appliances is 2 ELECTRIC IGNITION hardly capable of making its presence known by giving a shock to one touching the bare portions of electric conductors charged with it, when ordinary conditions exist. The same is in general true qf the electricity used for low-tension ignition in stationary engines when the electricity is supplied by apparatus especially adapted to ignition usage. The pressure of the electricity sup- plied by such means for low-tension ignition varies considerably in different systems. Some systems operate on about 4 volts, while others operate at about 50 volts. The latter pressure is about the same as that used in some commercial lighting systems, and gives only a slight shock under ordinary conditions. Some stationary engines take electricity from commercial service mains at a pressure of about no or even 220 volts. In such cases the electric current is generally passed through incandescent lamps which utilize a portion of the electric pressure and prevent, by their electric resistance, too much current from passing through the ignition apparatus. The reason that the high-tension electricity used for ignition is not dangerous, although its pressure is several thousand volts, is that the apparatus used does not have capacity to furnish enough electricity to do serious harm. The pressure is enor- mously high, as compared with that in service wires for incan- descent lighting, but the quantity of electricity is minutely small. The condition of the skin where it comes into contact with an electrically charged metal wire or piece of apparatus has much to do with the extent of the shock that is received, especially when the pressure is as low as that generally used for incan- descent lighting and for low-tension ignition. When the skin is dry or oily, the shock is very much less than when it is moist or wet. If the skin is cut or deeply scratched so that the raw flesh comes into contact with the charged metal, then the shock is still more severe than when the skin is intact but wet. The animal skin, or cuticle, offers more resistance to the passage of electricity than the other parts of the body, at least the soft parts. When the skin is removed, more electricity will pass through the body than when the skin is intact, even though moist or wet with water. In other words, a wet skin allows more INTRODUCTORY 3 electric current to pass through it, on account of its lesser resist- ance, than will pass through when the skin is dry or oily and consequently offers greater resistance to the passage of electricity. Sources of Electricity. 3. The electricity used for ignition is obtained either from a power-driven machine, called an electric generator, or from an electric battery in which the electricity is produced by chemical action. When an electric generator has permanent magnets, it is gen- erally called a magneto. An electric generator which does not have permanent magnets is ordinarily called either a dynamo or an electromagnetic generator. Magnetos and electromag- netic generators are both further classified according to the nature of the electricity they produce. This will be taken up in connection with the discussion of electric generators. Electric batteries are of two distinctive kinds, known as primary and secondary. A secondary battery is also called an electric accumulator and a storage battery. A primary battery is one which is ready to give out electric current as soon as it is constructed. A storage battery cannot give out electric current as soon as it is mechanically constructed, but must have electricity charged into it before it is ready to deliver electric current. It must also be frequently recharged with electricity during its life. Each unit of which an electric battery is made up is called an electric cell. A primary electric cell is also called a galvanic cell and a voltaic cell, after the names of its inventors. It is quite common practice to call an electric cell a battery, or a battery cell. Permanent Magnets. 4. Forms and Action. Doubtless the most familiar forms of permanent magnets are the horseshoe magnet and the magnetic needle. The latter is part of the magnetic compass for deter- mining the directions of north and south. Small horseshoe magnets are sold in toy stores and hardware stores. The mag- ELECTRIC IGNITION netic compass is regularly used on board ocean-going vessels, in surveying instruments, and in pocket compasses. A common form of horseshoe magnet is shown in Fig. i . It is made of a piece of bar steel bent to bring the ends of the bar near together. After bending, the steel is first hardened and then magnetized. The magnet will attract pieces of iron and FIGS, i, 2, and 3. Horseshoe Magnets. Permanent. steel placed near the ends of the bent bar, and, if the pieces are free to move, they will be drawn up against the magnet and held there. It is immaterial what forms the pieces of iron and steel to be attracted have. They may be in the form of wire nails, tacks, balls such as used in ball bearings, rings, or any other form. There is little or no magnetic attraction in the region of the crown, or curve, of the bent bar. Nails or tacks of comparatively mild steel or soft iron do not remain magnetic to any great extent after they are removed from the magnet. But hardened or tempered pieces of steel, especially those of an elongated form, such as a knife blade, sewing needle, and writing pen, retain considerable magnetism for some time after removal from the magnet, and will pick up INTRODUCTORY 5 tacks and other small pieces of steel. Still harder pieces of steel, such as a file, will retain magnetism for a longer time than the articles just mentioned. 6. Poles of a Magnet. The usefulness of the magnetic com- pass depends on the fact that one end of the magnetic needle, and that always the same end, points approximately in the direc- tion of the north pole of the earth when the needle is allowed to swing freely. The end of the needle which takes its position toward the earth's north pole is called the north pole of the needle; the opposite end, which points toward the south, is called the south pole of the needle. If a magnetic compass is placed near one of the ends of a horseshoe magnet, the compass needle, if left free to swing, will take a position with one end pointing more or less directly toward the nearest end of the bent bar. For convenience, it will be assumed that the north pole of the compass needle points toward the nearest end of the bent bar. Then, if the compass is moved to a new position so as to bring it near the other end of the bent bar, the needle will swing on its pivot so that the south pole of the needle will point more or less directly toward the nearest end of the bent bar. If the compass is repeatedly moved away from the bent-bar magnet and brought back near it as just described, the north pole of the needle will always be attracted by and point toward the same end of the bent bar, and the south pole will always behave in the same manner relative to the other pole of the bent bar. In the horseshoe magnet, the bar-end which attracts the north pole of the compass needle is called the south pole of the horseshoe magnet ; and the bar-end which always attracts the south pole of the needle is called the north pole of the horseshoe magnet. The letters N and S are customarily used to designate the north and south poles respectively of a magnet. If the compass is placed immediately between the bar-ends of the horseshoe magnet, the needle will take a position straight across between the poles, with the north pole pointing toward the south pole of the horseshoe magnet, and its south pole pomting toward the north pole of the horseshoe magnet. 6 ELECTRIC IGNITION 6. Magnetic Field. The region throughout which a magnet acts to attract pieces of iron and steel is called the " magnetic field " of the magnet. The magnetic field has its greatest strength near the ends, or poles, of the magnet, and is especially strong in the space between the ends of the horseshoe magnet. The magnetic field is said to be permeated with " magnetic lines of force." The position which a very small compass needle takes when placed in the magnetic field of a relatively large magnet indicates with fair accuracy the direction of the lines of force in the locality occupied by the compass needle. The length of the needle approximately coincides with the direction of the lines of force. The needle must of course be free to swing. If the north pole of a long, thin magnetic needle is placed between the poles of a horseshoe magnet, the magnetic force tends to move the needle pole in the direction from the north pole of the horseshoe magnet toward the south pole of the horse- shoe magnet, which is the direction in which the lines of force act in that locality. If the north pole of the needle is placed in any other part of the magnetic field of the horseshoe magnet, the magnetic force of the latter also tends to move the needle pole in the direction of the lines of force at the needle pole, and to carry it along the same lines of force in the direction from the north pole to the south pole of the horseshoe magnet. The path which the needle pole will follow may be a very indirect one of a circuitous nature. It is customary to assume that there is a flow of magnetism, called magnetic flux, in the magnetic field from the north pole to the south pole of a magnet, the flux at any point being in the direction in which the magnetic force tends to move a magnetic north pole. The complete magnetic circuit through which the flux occurs includes the length of the steel bar from end to end. 7. Magnet Keeper and Its Effect. A magnet keeper of soft iron or soft (mild) steel is generally provided with a horseshoe magnet. In Fig. 2 such a keeper A is shown in place across the ends of the magnet bar. The purpose of the keeper is to prevent the magnet from losing its magnetism. When the space between the poles of the magnet is bridged by INTRODUCTORY 7 the keeper as shown in the figure, the magnet will not attract pieces of iron and steel with nearly as much force as when the' keeper is not in place. The keeper has a similar effect, and to practically the same extent, if laid against the sides of the bent bar so as to bridge the space between the poles of the magnet. Or the keeper may be made so as to fit between the ends of the bar as shown in Fig. 3, and will then also have the effect of weak- ening the magnetic field in the manner described. The weakening of the magnetic field by the keeper is more complete when the keeper fits accurately against the magnet so that there is a large area of metallic contact between them, than when the surfaces that touch each other are rough and make but imperfect contact. The keeper offers an easier path for the magnetic flux than is offered by air, therefore nearly all of the flux is through the keeper, only a small proportion passing through the air. The magnetic flux through the bent bar of the magnet is not decreased by bridging the space between the poles with the keeper, but, on the contrary, a large increase of flux through the magnet bar is caused by putting the keeper in place. Proof of the last statement will appear in connection with the method of operating one of the various types of magnetos for generating electricity. 8. Magnetic and Non-magnetic Materials. Iron and alloys containing a large proportion of iron are the only materials that are magnetic to an appreciable extent. Steel is a combination, or an alloy, of chemically pure iron (ferrum) with other chemical elements. Except iron and steel, all of the materials ordinarily used in machinery and electrical apparatus are either non-magnetic, or magnetic to only such a slight extent, compared with commercial iron and steel, that they can be considered non-magnetic for the present purpose. These non-magnetic materials include brass, bronze, aluminum, aluminum alloys generally, zinc, porcelain, steatite (soapstone), glass, mica, rubber, pitch, dry wood, wood fiber, cotton, silk, hemp, flax, and paper. Strictly non-magnetic materials are not attracted by a magnet. 8 ELECTRIC IGNITION The magnetic field is not affected by their presence. If a piece of non-magnetic material is placed so as to bridge the air gap between the poles of a magnet after the manner of using a magnet keeper, no appreciable change will be produced in the magnetic field. 9. Compound or Composite Magnets. The permanent mag- nets used in magnetos for generating electricity for ignition pur- poses are generally made up of several bar magnets bent into the shape of the letter U and grouped closely together. Fig. 4 shows a method of grouping the individual magnets together that is very commonly used for forming a composite magnet. The north poles of all the indi- vidual magnets are placed together, and likewise the south poles. Rectangular bars bent flatwise as shown are very generally used, but other forms of steel bars, including round ones, are also used to some extent. The reason for using composite magnets is that it is dif- ficult, in fact practically impossible, to make magnets of one piece sufficiently large for magnetos to be used for ignition purposes. FIG. 4. Compound or Composite Magnet. Principle of Electric Generators. 10. In all power-driven electric generators the generation of electricity is due to directly and repeatedly varying the number of lines of magnetic force that pass through the opening of a coil of wire (not through the length of the wire itself). This varia- tion generally includes reversing the direction of magnetic flux through the coil opening, although in some unusual designs the magnetic flux is not reversed in direction through the coil open- ing. Varying the number of lines of magnetism through the coil of wire induces an electromotive force in the wire, which tends to cause a flow of electricity through the wire and also INTRODUCTORY 9 through whatever apparatus may be suitably connected to the wire. The electromotive force is induced in the coil only during the time of variation in the amount of magnetic flux through the coil. A constant flux of magnetism through a coil, without change in the number of lines of magnetic force passing through the coil opening, does not induce an electromotive force in the coil. The methods by which the number of lines of force passing through a coil are made to vary are numerous. They come under two general methods, however. In one of these general methods the wires of the coil are caused to cut through the lines of magnetic force in such a manner as to vary the number of lines of force passing through the coil. In the other general method the magnetic flux through a bar which the coil encircles is caused to vary in intensity, also generally reverse, without the wires of the coil cutting through any of the lines offeree. The more important ways in which the variation of magnetic flux through a coil is accomplished will appear in the descriptions of various types of generators used in connection with electric ignition. 11. The armature of an electric generator consists of a coil, or coils, of wire over or around a core, or cores, of magnetic material which is not a permanent magnet. In the more usual forms of generators for electric purposes, the armature rotates relative to the magnets. In less usual forms, the armature remains stationary with regard to the magnets. In the latter form there is a rotor (rotating part) of soft iron or mild steel, called an inductor. In both cases the rotation, or oscillation, of the rotor (armature or inductor) causes a variation in the num- ber of the lines of force passing through the opening of the arm- ature coil, and thus induces an electromotive force in the coil of the armature, so that the coil will deliver current when the elec- tric circuit is properly closed, as through the other apparatus of an ignition system. CHAPTER II. LOW-TENSION ALTERNATING-CURRENT MAGNETOS WITH SINGLE- WOUND SHUTTLE ARMATURES. 12. The field-magnets of a magneto are shown in Fig. 5. The field magnet is composite, being made up of six individual magnets in pairs, each pair consisting of a large magnet fitted over a small one. Two pole-pieces, or pole-shoes, of mild steel or cast-iron, are fast- ened opposite each other and against the inner surfaces of the inside individual magnets by means of screws which pass through the magnets into threaded holes in the pole -pieces. The space between the extreme ends of the magnets is bridged by a piece of non-magnetic metal, such as brass or aluminum alloy, which forms a base for the mag- netic field and is fastened to the Field-Magnets and Pole-Pieces of a po le-pieces by screws, one of which is shown partly removed beneath the base. The opposite faces of the pole-pieces are bored out cylindrical, so as to fit close to the armature, or. inductor, which rotates, or oscillates, between them when the machine is operating. The ends of the pole -pieces are slightly counter- bored to receive an end-plate (not shown) and hold it con- centric with the bore of the pole -pieces. The use of the end-plate is to support the armature and other parts, as will appear later. FIG. 5. 10 ALTERNATING-CURRENT MAGNETOS II 13. Abutted Magnets. Another arrangement of the field- magnets is shown in Fig. 6. The magnets are placed on opposite sides of the armature space with their ends against each other so that the north poles are together and have one of the pole- N | N s I s FIG. 6. Abutted Field-Magnets of a Magneto. pieces fastened to them. The south poles are also together and have the other pole-piece fastened to them. Composite magnets can be used in this arrangement as well as that shown in the preceding figure. Rotary Armature Types. 14. An armature suitable for rotating between the pole-pieces of the magnets just described is shown in Fig. 7. The general 82- FIG. 7. Armature of an Alternating-current Electric Generator. Shuttle-wound Type. nature of the construction of the armature can be seen by referring to Figs. 8, 9, and 10. The core of the armature has the general form shown in Fig. 8. It is made of mild steel or very pure and soft iron machined so that the crowned surfaces have a cylindrical form to fit between the pole-pieces. The crowned surfaces should fit as close as 12 ELECTRIC IGNITION possible to the pole-pieces without touching them, in order to make the air-gap between the armature and pole-pieces as small as possible. The armature winding is a coil of wire wound around the neck which lies between the crowned sides of the core. Fig. 9 shows the core with part of one layer of winding in place. The neck and sides of the space in which the coil is wound are first cov- FIG. 8. Core of Shuttle-wound Armature. FIG. 9. Partly Wound Armature of the Shuttle Type. FIG. 10. Elements of Shuttle-wound Armature. ered with some insulating material, such as silk or paper that has been oiled or varnished, mica, or wood fiber cut from sheets or molded to form. The insulating material does not allow electricity to flow through it in appreciable quantity. A side piece of insulation is shown at i in Fig. 9. Copper wire, covered with cotton or silk thread wound around it so as to form an insulating covering, is used ordinarily. In the better class of work, the wire is covered with two windings of silk or cotton, one winding on top of the other. The wire used in magneto ALTERNATING-CURRENT MAGNETOS 13 armatures is commercially known as armature wire or magnet wire, single-covered or double-covered, as the case may be with regard to whether there is one or two layers of thread wound on it. One end of the wire is bare and fastened to the metal of the armature core at 2, so that the copper of the wire and the metal of the core are in metallic (electric) connection. The rest of the wire is carefully insulated from the core, and the different layers of the coil are also carefully insulated from each other. The wire-end 3 is intended to represent where the wire has been cut off before the first layer of winding was completed, in order to show the nature of the winding. After the first layer of the winding is complete, the second layer is wound over it, and so on, layer upon layer, till the wind- ing is complete. There is one continuous winding throughout all of the layers. Some insulating material, such as heavy paper or cloth, is generally placed between the layers, especially at the bends of the wire, as a protection to the silk or cotton wrap- ping of the wire, and to make the insulation more perfect between the layers. The coil is then wrapped with insulating tape and bound around circumferentially of the core with bare non-magnetic wire, as shown in Fig. 7. The tape is generally of cotton or linen web saturated with insulating varnish. Liquid varnish is also generally applied to the tape while wrapping it over the coil. The varnish is waterproof if the magneto is to be used where there is any likelihood of water reaching it. It is better for it to be waterproof so as to exclude atmospheric moisture even if it is intended to be used only in places free from water. Brass or bronze wire is generally used for the circumferential bands. A disk-shaped head of non-magnetic material is fastened to each end of the armature core after the winding is completed. Brass, bronze, or aluminum alloy is generally used for these heads. Each head carries a spindle which projects outward from the winding. The spindles are usually made of steel. They run in suitable bearings during the rotation, or oscillation, of the arma- ture while the magneto is operating. ELECTRIC IGNITION One of the spindles shown in Fig. 10 is hollow. The outer end of the armature wire passes through the hole so that its end projects beyond the spindle at i . Carrying an end of the arma- ture winding out through a hollow spindle is very commonly used in magnetos intended for ignition purposes, although the wire itself is not generally carried through the spindle. It is more usual to connect the armature wire to a rod or screw which extends through the hole and is insulated from the metal of the spindle either by a tube of hard rubber or vulcanized fiber, a wrapping of sheet mica, or some other suitable means of insulation. An armature with a core of the shape shown in Fig. 8, and wound as just described, is called either an I-armature, an H- armature, or a shuttle- wound armature. The latter name is on account of the resemblance of the armature in general appearance to the shuttle of a weaving loom. The names I-armature and H-armature come from the resemblance of the core, when looked at endwise, to either the letter I or the letter H, ac- cording to whether the core is held, or lies, with the crowned surfaces at the top and bottom, or at the sides. 15. Electric Arc from a Shuttle- wound Armature. Fig. u shows a shuttle- wound armature in place be- tween the pole -pieces of permanent field-magnets. The armature is of the Field-Magnets and Armature of general form of that shown in Figs. 7 Magneto with Device for an( j IO< T^ bare end i of the copper Showing Positions of Arma- r ,-, -,. ture for Maximum Electric Wlfe f the armature wmdln g P r J ects Arc or Spark. from the end of the hollow spindle, part of the insulation 2 being removed to expose the wire. A cam-shaped projection 3 is shown on the forward end of the front spindle. A cam of this particular form is not usual in commercial magnetos, but is added here to facili- tate the explanation of the principle of operation of the magneto. FIG. ii. ALTERNATING-CURRENT MAGNETOS 15 A bent wire 4 is shown in contact with the bare end i of the armature wire and also in contact with the convex surface of the cam or lug 3. If the armature is rotated in the direction indicated by the arrow on the armature head, and the bent wire 4 is held stationary in the position shown, so as to have rubbing or slipping contact with the wire-end i and cam 3, then when the bent wire snaps off the edge of the cam as the latter rotates with the armature, an electric arc will be drawn between the end of the wire and the edge of the cam at the point of separation of the two, provided the speed of rotation is sufficiently high. A rotative speed of 30 to 40 revolutions per minute is sufficient to draw quite a large arc in some magnetos when the parts separate. A magneto of the size commonly used on portable combustion motors will give a large arc when the armature is twirled around through one or two revolutions by grasping the spindle in the fingers ; half a rev- olution caused in this manner will give a good-sized arc in some low- tension magnetos for ignition. If while the armature is rotating at a speed more or less uni- form, the bent wire 4 is swung around to different positions but held so that it is always in contact with both the wire-end i and the cam 3 just before the edge of the latter breaks contact with the end of the bent wire, it will be found that while a good-sized arc will be drawn for some positions of the bent wire, at other positions no arc whatever will appear when the contact is broken. After one position of the bent wire for the largest electric arc corresponding to the speed of rotation is determined, it can be seen, by swinging the bent wire slowly around the axis of the rotating spindle, that the other position of the bent wire for maximum arc is diametrically opposite the position first deter- mined. In other words, the largest arc is obtained at two positions of the cam-edge diametrically opposite each other, which may be expressed by saying that the two positions of the cam-edge are half a revolution apart for maximum electric arcs. The position of the cam at any instant corresponds of course to definite positions of the armature, since the cam and armature are rigidly fastened together. 1 6 ELECTRIC IGNITION It can also be seen, by the same process as just described, that the two positions of the bent wire at which no arc is obtained lie midway, or approximately midway, between the positions for maximum arcs. The two positions for no arc are also diametri- cally opposite each other, and therefore half a revolution apart. A larger, or hotter, arc is obtained at a high speed of rotation than at slow speed. 16. Effect of Speed of Armature on Position for Maximum Arc. If the positions of the armature for maximum arc and no arc are first determined as above while the armature is rotat- ing at slow speed, say 50 revolutions per minute, and then the armature speed is increased to say 1 200 revolutions per minute, and the experiment repeated, it will be found that the bent wire must be held farther around in the direction of rotation of the armature in the latter case in order to obtain the maximum arc and no arc. This means that the maximum arc occurs later in the revolution of the armature at high speed than at low speed, each revolution being assumed to begin at the same position of the armature, as when the neck of its core is in a vertical position. The cause of this lag in the time of production of the maximum arc will be explained later. 17. Positions of Armature for Strong Electric Arc. The posi- tions of the armature at which the strongest arc can be obtained vary with variation in the forms of the pole-pieces and armature core, as well as with the speed of rotation, as has been mentioned. Most magnetos that are to run at a variable speed when in service are generally so constructed that a strong arc can be produced throughout a considerable range of position of the bent wire applied as stated above. This is done on account of the advance and retard of ignition relative to the position of the pistons of the motor, as well as on account of the lag in the mag- neto with regard to the position of the armature at the instant of maximum arc. The lag is not so great for the speed usual for combustion motors but that the position of the armature for maximum arc can be pointed out in a general way, as in the following paragraphs. Fig. 1 2 shows conventionally the pole-pieces and shuttle-wound ALTERNATING-CURRENT MAGNETOS armature of a magneto. A double, or two-lobed, cam is shown fastened to the armature spindle. A bent wire for making electric connection between the cam and the bare projecting end of the armature wire is also shown in place. The two lobes of the cam are diametrically opposite each other and are in such positions relative to the core of the armature that if a line were drawn FIG. 12. Positions of Armature for Maximum Electric Arc or Spark. Approximate. through the two cam-edges at which the contact with the bent wire is broken as the armature rotates, the line would be parallel to the neck which connects the two crowned parts of the core. FIG. 13. Positions of Armature for No Electric Arc or Spark. Approximate. The bent wire is held so as to break connection with the cam- edge which is uppermost just as the armature core passes through its upright position in which its end view resembles the letter I. The arrow indicating the direction of rotation of the armature may be taken as cut into the metal of the core so that it rotates with the core. If the armature is rotating at a very slow speed, the maximum electric arc for that speed will be obtained by breaking the cir- cuit while the armature is passing through the vertical positions i8 ELECTRIC IGNITION shown in Fig. 12, or while it is passing through positions very near the vertical. The maximum spark will occur twice during each revolution. The positions of no spark are shown in Fig. 13. The neck of the armature core is horizontal, or nearly so, in these positions, so that the end view of the core resembles the letter H. FIG. 14. Maximum-spark Positions of Armature. If the armature is rotated at a speed as high as 1200 revolu- tions per minute, then its positions of maximum arc will be some- what as indicated in Fig. 14, which positions are somewhat later FIG. 15. Sparkless Positions of Armature. in the revolution than for slow speed of rotation. The positions of no spark at high speed will be somewhat as shown in Fig. 15. 18. Laminated Armature Core. The rotation of the arma- ture in the magnetic field induces electric currents, called fou- cault currents, in the armature, as well as current in the winding of the armature. If the core is made of one solid piece of steel or iron, the foucault currents in it cause it to heat and are other- wise objectionable. In order to keep this objectionable action as small as possible, it is common practice to build up the core ALTERNATING-CURRENT MAGNETOS 19 from thin sheet steel cut into I-shaped pieces. The steel used for these pieces is commercially known as armature steel. The pieces are generally cut out by a stamping press. Fig. 1 6 shows a laminated armature core built up in this manner. The " disks," or laminations, are sometimes separated from each other by some such material as silk fabric or thin sheet paper properly prepared by oiling or varnishing; in other cases, FIG. 16. Laminated Core of Shuttle Armature. varnish or the black scale on the surface of the steel is depended on as sufficient insulation. The sheet steel from which the disks are cut is of about the thickness of that used for stovepipes. The armature disks are pressed together under heavy pressure, as that of a hydraulic press, after they have been grouped to form the core. They are then fastened together by rivets or other suitable means. 19. Magnetic Flux in a Rotating I-shaped Armature Core. It has been stated that the electromotive force and current in the winding of an armature are induced by changing the amount of magnetic flux through the space surrounded by the coil. In the shuttle-wound armature the variation of magnetic flux occurs in the steel neck of the core on which the insulated wire is wound. Fig. 17 shows the general nature of the magnetic flux through an I-shaped magnetic core during the time it is rotating between the pole-pieces of the magnets. Only the end views of the core and pole -pieces are shown in the figure. The pole-pieces are marked N and S to indicate the north and south poles respec- tively. The letter and the feathered arrow indicating the direc- 20 ELECTRIC IGNITION (a) FIG. 17. Magnetic Flux through Shuttle Armature Core in Different Positions. ALTERNATING-CURRENT MAGNETOS 21 FIG. 17 (continued). Magnetic Flux through Shuttle Armature Core in Different Positions. tion of rotation of the core may be considered as cut into the metal of the core. In (a) the core is in the H position and the magnetic flux through it is from N to 5, as indicated by the lines with arrow heads along them. This is one of the positions of the core in which the greatest amount of magnetic flux occurs through the neck that connects the crowned ends of the core. When the core has been rotated to the position (b) there is less flux through it, because less of the crowned ends, or sides, is opposite the pole- pieces. The direction of flux for this position is in general as in- dicated by the arrow-headed lines. When the core is in position (c) there is very little flux through it, because the crowned surfaces have almost entirely moved away from opposite the pole-pieces. In the vertical position of the core, as shown in (d), there is no longer any magnetic flux through the core -neck around which the coil is wound in the complete armature. In other words, the magnetic flux through the coil is of zero value when the core is in the I position. There is some flux through the crowned 22 ELECTRIC IGNITION ends of the core, however, from pole to pole, as indicated by the arrow-headed lines; but this has no effect to produce electro- motive force and current in the armature winding. In position (e) the magnetic flux through the core -neck is in the opposite direction from that in the first three positions, but the flux is from the N pole to the S pole, as it always is. In the first three positions the flux is through the core from the arrow-marked side toward the blank side, but in position (e) the flux is from the blank side toward the arrow-marked side of the core, as is also the case in positions (/) and (g). In the latter position the flux is again a maximum of the same value as for position (a), but in the opposite direction through the core. In positions (g), (ti), (i), (&), and (/) the paths of flux are similar respectively to those in (a), (>), (c), (e), and (/), but the flux is in the opposite direction through the core -neck on account of the core being half a revolution further around in the positions (h) to (/) than in (b) to (/). In position (j) there is no flux through the core-neck, but the flux through the crowned sides of the core is in the opposite direction from what it was in (d), on account of a difference of half a revolution between the two positions. Fig. 1 8 is a diagram representing the relative amounts of magnetic flux through the core -neck of an H-armature for all of its positions during one-half a revolution. The distance from O vertically up to the curve is the amount of flux when the core is stationary in the position shown in Fig. 17 at (a). The vertical distance B, Fig. 18, is the flux when the core has been rotated 45 degrees to the position in Fig. 17 at (&). At Z>, where the curve crosses the zero line, there is no flux through the core -neck. This corresponds to position (d) in Fig. 17. When the armature is in position (/), three-quarters of a revolution from the starting position, the flux is equal to that for position (b) but in the opposite direction. This is indicated in the diagram by taking the distance F below the zero line. When the core is rotating at a very slow but uniform speed, the rate of change in the magnetic flux through the core -neck is more rapid while the core is passing from position (c) to position (e), and from position (i) to position (&), than during the other ALTERNATING-CURRENT MAGNETOS portions of the revolution. (The reason for limiting this and the following statement to slow speed will appear later.) While the core is passing through positions at and near those shown in (a) and (g), the rate of change in magnetic flux through the core is very low compared with the rate for the movements just mentioned. In one position at or near (a), and another at or Zero Line J Revolution- FIG. i 8. Graph Showing Magnetic Flux in Shuttle Armature. FIG. 19. Graph Showing Rate of Variation of Magnetic Flux in Shuttle Armature. near (g), the rate of change falls to zero. This is when the flux stops decreasing and just before it begins increasing. As the armature core moves from position (c) to position (e), the decrease first occurs in the flux through the core -neck, fol- lowed by reversal and increase of flux in the opposite direction through the core-neck. These are together equivalent to a continuous decrease of magnetic flux. The same is true for the movement from position (i) to position (&).* * This may possibly be more readily understood by considering a somewhat analogous case in the flow of water, as follows: If two pipes are delivering water into a reservoir at the same time, a large pipe at the rate of 50 gallons per minute, 24 ELECTRIC IGNITION Fig. 19 shows the rate of change of magnetic flux through the core-neck for all positions of the core during one-half a revolu- tion, starting from position (a), Fig. 17. The vertical distance C, Fig. 19, represents the rate of change of flux while the arma- ture is passing through the position shown at (c), Fig. 17. The rate of change for position (e) in the latter figure is shown as the vertical distance E, Fig. 19. If the curve were given for the second half -re volution, it would be below the zero line, since the increase and decrease take place in opposite directions through the core-neck from what they do in the first half -revolution. 20. Electromotive Force and Current Induced. If a coil of insulated wire is wound around the core-neck, as in Fig. 20, so as to form an armature, an electromotive force is induced in the coil while the armature is rotating in the magnetic field between the pole-pieces of the magnets. This electromotive force is pro- portional, or nearly so, to the rate of change of the magnetic flux through the core-neck. (A steady flux of magnetism of constant amount does not induce an electromotive force.) If the electric circuit is closed, a current will flow through the winding whenever there is an electromotive force. In the figure a complete closed electric circuit is obtained by connecting both ends of the insulated wire to the metal of the core. These con- nections are indicated by black spots and are numbered i and 2. 21. Armature Lag. It has been stated that the maximum arc is obtained later in the revolution of the armature when the speed of rotation is high than when it is low, and that this is due to armature lag. The armature lag is due to both the magnetic lag of the core and a small pipe at the rate of 10 gallons per minute, then the rate of increase in the amount of water in the reservoir is 50 + 10 = 60 gallons per minute. If the flow of the small pipe is stopped and another pipe opened to draw water from the reservoir at the rate of 10 gallons per minute, the rate of increase of water in the reservoir will be reduced to 50 10 = 40 gallons per minute. The difference between the two rates of increase is 60 40 = 20 gallons per minute. An analogous case is that of first flowing water into a tank at the rate of 10 gallons per minute, then stopping the inflow and drawing out water at the same rate. Drawing out water may be considered as a negative filling of the tank. The difference between the two rates of filling, one positive and the other negative, is 10 + 10 = 20 gallons per minute. ALTERNATING-CURRENT MAGNETOS 25 and the lag of the current behind the induced electromotive force. It requires an appreciable amount of time, in comparison with the speed at which the magneto rotates on a high-speed multi- cylinder motor, to change the rate of magnetic flux in a piece of steel or iron. And the electric current lags slightly behind the electromotive force that is induced by the change of magnetic flux. The current lag is due chiefly to the action of the current* in each turn (or single wrap) of the coil winding upon the cur- rent in the other turns of the winding, and to the reaction of the current upon the magnetism of the core. These and other causes together produce armature reactance and lag. Referring to Fig. 17, if (a) is one of the two positions of the core for maximum magnetic flux through the core -neck when the core is standing still or rotating at very slow speed, then at high speed of rotation the maximum flux will occur slightly later in the revolution of the core; that is, after the core has passed slightly beyond the position shown in (a). And if (d) is one of the positions for no flux through the core-neck when the core is not rotating, then the position of no flux through the core -neck will be somewhat further around in the direction of rotation at high speed. Thus, position (e) may be the position of no flux through the core -neck at excessively high speed of rotation. The same applies to positions (g), (/'), and (k). The reactions in the armature cause the maximum current to occur somewhat later than the maximum magnetic flux, as has been stated.* 22. Alternating Current Generated. For convenience in dis- cussing the nature of the current generated in an armature wind- ing, it will first be assumed that there is no lag in the armature. Referring to Fig. 20, (A) is one of the positions of maximum magnetic flux through the core-neck when the armature is stand- ing still, and also when it is rotating, the latter in accordance with the assumed condition of no lag. As the armature rotates * It is not thought desirable to give any further discussion of armature react- ance and lag in a work of this nature, especially as it is very probable that the largest, or hottest, arc is obtained by breaking the electric circuit slightly before the armature reaches the position which would give maximum current if the circuit were left closed. 26 ELECTRIC IGNITION (A) zzQn / CD) FIG. 20. Direction of Current Flow in Winding of Shuttle Armature. ALTERNATING-CURRENT MAGNETOS (K) (L) FIG. 20 (continued). Direction of Current Flow in Winding of Shuttle Armature. through the first quarter-revolution from position (A), the mag- netic flux through the core-neck decreases, slowly at first, and at an increasing rate till the armature has reached position (D) at the completion of the quarter-revolution, in which position there is no magnetic flux through the core-neck. The decrease of magnetic flux through the core-neck causes an electric current to flow through the insulated wire of the winding. The direc- tion of flow of the current is as indicated by the arrows on the wire. The path of the current is from 2 through the length of the wire to i, and thence through the metal of the core from i to 2. The current, beginning at zero value, keeps increasing during the first quarter-revolution and reaches its maximum value in position (D). From position (D) to position (G) the current decreases until it drops to zero at the completion of the first half -re volution, corresponding to position (G). During the second half -revolution, from (G) to (L), a similar 28 ELECTRIC IGNITION action takes place; but, since the coil has been turned over, the direction of current flow through the wire is opposite that during the first half -revolution. The direction of current flow during the second half-revolution is indicated by the arrows on the wire in (H)j (7), (/), and (K). It flows through the wire from i to 2. Briefly, under the assumed condition of no lag, starting from position (A), the current increases from zero to its maximum value during the first quarter-revolution, and decreases to zero dur- ing the second quarter-revolution; then increases to a maximum in the opposite direction during the third quarter-revolution, and decreases to zero again during the last quarter-revolution. This action is repeated during each revolution. The speed of rotation has been assumed to be constant. An electric current of the nature just described is called an alternating current. The lag causes the maximum current to occur later in the rotation of the armature than just stated. At a high speed of rotation positions (E) and (K) may be those for maximum cur- rent. Although the lag is very small as measured in fractions of a second of time, it may be very appreciable when measured in parts of a revolution of the magneto. Thus, a shuttle-wound magneto that is igniting a six-cylinder four-cycle motor runs at 1800 revolutions per minute when the speed of the motor is 1200 revolutions per minute. The current must rise from zero to its maximum value and drop back to zero again 3600 times per min- ute, which is 60 times per second. The current has ^o of a second to rise to its maximum and drop back to zero again. A lag of yAo of a second corresponds to 3*0 of a revolution, which is 9 degrees of angle. 23. Graphical Representation of Current in a Shuttle-wound Armature. Fig. 21 shows graphically the general nature of the current generated in a shuttle-wound armature. The revolu- tions of the armature are measured horizontally, and the amount of current is measured vertically. The rotation is measured from the position in which the magnetic flux through the core- neck is a maximum when the armature is not rotating; this is position (a) in Fig. 17, and position (-4) in Fig, 20. In the dia- ALTERNATING-CURRENT MAGNETOS 29 gram, Fig. 21, the point of zero rotation is indicated by O. It is assumed that the armature rotates at a uniform speed. The current flow is represented by the curved line ABODE. The current has zero value at A after the armature has rotated through a small angle. The current increases and reaches its maximum value at B slightly after the completion of the first quarter-revolution. The maximum value of the current at this point is BF. Decrease of current begins at B and continues till zero value is reached again at C, which is half a revolution from A. The flow of current then begins in the opposite direction -One Cycle of Current B 1 Revolution FIGS. 21 and 22. Current in Armature Winding as Affected by Different Forms of Pole-pieces. and increases till it reaches maximum value again at D, slightly after three-quarters of a revolution. This may be called a negative maximum. Its value is DG. Decrease then begins and the current falls to zero again at E, just after the completion of the revolution. 24. Cycle of Current. The series of changes through which the current repeatedly passes is called the cycle of the current. In this case the complete cycle is passed through during one revolution of the armature. 25. Form of Current Curve is Affected by Shape of Pole- Pieces. The shape of the pole-pieces, especially at the edges, ELECTRIC IGNITION or lips, determines to some extent the form of the current curve. The lips of the pole-pieces are the edges i, 2, 3, and 4 in Fig. 17, view (b). If the current curve in Fig. 21 is obtained with the pole-piece lips rounded as in Fig. 17, then for sharp-edged lips like those in Fig. 20 the current curve will be flatter and broader at the top and bottom, somewhat as shown in Fig. 22. The maximum current is not so great in the latter figure, but the current remains large during a greater part of a revolution than in Fig. 21. Two forms of pole-pieces, which give broad peaks of the nature of those at b and d in Fig. 22, are shown in Fig. 23. In (A) the (A) (B) FIG. 23. Pole-pieces with Rounded Lips and with Tooth-shaped Lips. middle portions of the lips extend out farther than the ends. In (B) one lip of each pole-piece is in the form of teeth which resemble, in a measure, those of a comb. The lips with teeth extend out farther from the magnet poles than those which have smooth edges. The pole-pieces in (B) are for an armature that rotates clockwise, so that the surface of the core that is next to the pole-pieces moves away from the pole-piece lips with teeth toward the smooth-lipped pole-pieces. Pole-pieces of the forms of those shown in Fig. 23 give a more gradual rate of change in the magnetic flux as the armature, or inductor, rotates than occurs when the lips are straight as in Fig. 17. 26. Position of Armature for Maximum Arc. In accordance with what has been stated, it may be seen that in a magneto with a shuttle-wound rotating armature, the largest, or hottest, arc is obtained by breaking the electric circuit while the armature is passing through a position near that in which the crowned ALTERNATING-CURRENT MAGNETOS 3 1 sides of the core span the space between the lips of the pole- pieces, as in Fig. 20 at () and (K). 27. A low-tension alternating-current magneto with shuttle- wound armature is shown in Fig. 24. The illustration is partly a longitudinal section and partly full view, the latter being mostly of interior parts. The beginning of the armature winding is connected to the armature core by means of a screw so as to make metallic (elec- FIG. 24. Bosch Low-tension Alternating-current Armature Mounted on Plain Journal per Revolution. 1. Field magnets, composite. 2 . Armature plain on journal bearings. 3. Insulated bolt to which one end of armature winding is connected. 4. Terminal with binding nut. 5. Metal mounting for carbon brush which presses against end of 3. 6. Front bearing-plate with plain journal-bearing and felt wick lubricator. Magneto with Rotary Shuttle-wound Bearings. Sectional View. Two Sparks 7. Rear bearing-plate with plain journal bearing and felt wick lubricator. 8. Felt wick with coiled spring under it. 9. Leather washer. 10. Wick holder. 11. Carbon brush with coiled spring. 16. Dust cover over armature. 20. Steatite insulating washer on ter- minal 4. trie) connection. The end of the winding is metallically con- nected to the insulated bolt 3 which passes through the hollow rear spindle and projects beyond the end of the spindle. The black around the bolt 3 indicates insulating material. A carbon 32 ELECTRIC IGNITION brush* mounted in a metallic holder 5 is pressed against the end of the insulated bolt 3 by a coiled compression spring. The three latter parts are carried in an insulated terminal 4, which is provided with a thumb-nut for holding the end of the wire through which current can be carried to other apparatus. A steatite washer 20, to which the terminal 4 is firmly attached, holds the terminal in place and insulates it. The spindles of the armature are of the plain cylindrical journal type and rotate in corresponding bearings in the plates 6 and 7. The surface of each journal slides over the surface of its supporting bearing. The carbon brush n is pressed against the rear head of the armature by a coiled compression spring so as to make electric connection between the metal of the armature and the body of the magneto. The path of the current that is generated in the armature winding, assuming a direction of flow, is from the insulated end of the armature winding through the insulated rod 3, carbon brush and mounting 5, terminal 4, wire leading to the external apparatus and through the latter, then back to the body of the magneto, through the brush 1 1 to the armature head, from which it flows to and through the core to the end of the winding which is connected to the core. The current also passes through the armature winding, of course. If means, such as brush n, were not provided for flow of * In the earlier forms of electric generators, or dynamo-electric machines, a "brush" was a brushlike bundle of copper wires used to make sliding contact between electric conductors for the purpose of allowing current to flow from one to the other. By common usage "brush" has come to mean any form of electric conductor that has sliding contact with another part for the purpose just stated. Ordinarily the brush slides continuously over the part against which it bears. The latter may be either all electric conductor, or it may be part conductor and part insulator. In some cases the electric contact is continuous; in others it is broken by insulation, on which the brush rubs part of the time. Either the brush or the part on which it rubs may be stationary, or both may move so as to have motion relative to each other. A carbon brush may be made of pulverized charcoal or graphite mixed with suitable binding material and compressed to the desired shape. Frequently fine- woven copper or brass wire (wire gauze) is embedded in the carbon to allow the current to flow more freely through the brush. ALTERNATING-CURRENT MAGNETOS 33 current between the armature core and the body of the magneto, the current would have to flow through the journal bearings. This is objectionable, since there is a thin film of oil between the rubbing metal surfaces of the journal and its bearing when they are properly lubricated with oil. Oil is an insulator, and there- fore prevents to some extent the flow of current even when the film is as thin as in bearings of this sort. The electric resistance of the oil film also causes heating of the rubbing surfaces and FIG. Bosch Low-tension Alternating-current Magneto with Armature Mounted on Ball Bearings. Sectional View. lution. i. Field magnets, composite. 21. 2a. Armature on ball bearing. 4. Terminal with binding nut. 22. 5. Metal mounting for carbon brush which presses against end of 21. 23. 6a. Front end-plate for ball race. ya. Rear end-plate for ball race. 24. i6a. Dust cover over armature. 20. Steatite insulating washer on ter- 25. minal 4. tends to burn the oil, thus injuring the effectiveness of lubrica- tion. All of this is objectionable. The journal bearings are each lubricated by means of a pencil- shaped felt wick, one of which is shown at 8. It is pressed up Rotary Shuttle-wound Two Sparks per Revo- Insulated bolt to which one end of armature winding is connected. Inside steatite insulator on arma- ture. Outside steatite insulator on arma- ture. Carbon brush with mounting and coiled spring. Screw cover. 34 ELECTRIC IGNITION against the bottom of the journal by a coiled spring. The felt wick and coiled spring are held in place by a wick-holder 10, which is a hollow screw. A leather washer 9 is placed between the screw head and the end-plate into which the wick -holder screws, in order to make an oil-tight joint. The wick-holder has a small hole, or holes, through its sides to let oil in to the wick from the oil reservoir which surrounds the body of the wick- holder. One of these small holes is shown in the wick-holder which is screwed into the rear bearing plate 7. Fig. 25 shows a low-tension magneto with ball bearings upon which the armature rotates. While this magneto is the same in its general nature as the one just described, it differs in the form and location of some of its parts. There are no oil reservoirs FIG. 26. FIG. 27. External View of Magneto shown Splitdorf Low-tension Magneto. Two Sectionally in Fig. 24. Sparks per Revolution. and no wick-oilers, since the ball bearings require only a few drops of oil once a month or so. The brush 24 for making elec- tric connection between the armature head and the body of the magneto is at the head end of the armature instead of at the rear end as in the other machine, and is perpendicular to the spindles instead of parallel to them. The caption immediately beneath the illustration describes the parts briefly. Fig. 26 is a full view of the magneto shown in section in Fig. 24. ALTERNATING-CURRENT MAGNETOS 35 The driving end of the armature spindle is shown projecting forward from the front end of the machine. Fig. 27 is a full view of a similar magneto showing the rear end and the projecting terminal to which a wire leading to other apparatus can be connected. Stationary Shuttle-wound Armature Types. 28. Stationary Armature and Rotary Magnetic Sleeve. In- stead of rotating the armature, it may be held stationary and the required magnetic flux through the core-neck obtained by rotating an inductor, having the form of a slotted soft steel sleeve, between the armature and the pole-pieces, the latter being bored considerably larger in diameter than the armature, so as to leave space in which the sleeve can rotate. Fig. 28 shows a magnetic sleeve for rotating between a sta- tionary armature and the pole-pieces. The armature is shown 82 FIG. 28. Rotary Magnetic Sleeve Inductor for a Magneto with a Stationary Armature of the Shuttle Type. FIG. 29. Stationary Shuttle-wound Armature to go inside of the Magnetic Sleeve shown in Fig. 28. in Fig. 29. It is of the same general form as a rotary shuttle- wound armature. When the parts are assembled the armature lies inside the sleeve. The two sides of the magnetic sleeve have the form of pieces ELECTRIC IGNITION cut from a tube by slotting it lengthwise. They are of mild steel as already stated, and are held together by disk-shaped heads (G) (H) FIG. 30. Magnetic Flux through Stationary Armature Core and Rotary Sleeve. of non-magnetic material, such as brass, bronze, or aluminum alloy, attached to them by suitable fastenings (screws in this case). One of the heads has a driving spindle. ALTERNATING-CURRENT MAGNETOS 37 29. The action of the magnetic sleeve in causing a variation of magnetic flux through the core-neck of the armature can be understood by reference to Fig. 30, in which the sleeve is shown in all of its positions for maximum magnetic flux and for no flux through the core-neck, no allowance being made for armature lag. The direction of rotation of the sleeve is indicated by the feathered arrow, which may be taken as stamped on the end of the sleeve. The general direction of magnetic flux is indicated by the lines with arrowheads along them. In (A) there is no magnetic flux through the core-neck, but a slight amount occurs through the crowned sides of the core. In (B) the flux has a maximum value through the core-neck from top to bottom; this position is about one-eighth of a revolution later than (A). In (C) the sleeve bridges the gap between the pole-pieces, and there is no flux through the core-neck. In (D) the flux again has a maximum value through the core-neck, from bottom to top, which is in the opposite direction from the flux in (B). In (E) there is no flux through the core-neck. This completes the first half -re volution, starting from position (A). It may be noted that the flux, and consequently the current, reaches maximum value twice during half a revolution, and since the flux is in opposite directions in (B) and (D), the current flows in opposite directions in these two cases. An alternating current is therefore generated. During the latter half of the revolution the variation of flux through the core-neck is the same as that during the first half- revolution. The current therefore passes through two complete cycles during one revolution. An arc can be drawn four times per revolution by breaking the circuit at or about the time maxi- mum current occurs. The field-magnets and pole-pieces for a magneto with a sta- tionary shuttle-wound armature and rotary magnetic sleeve can be of the same form as those for a rotary shuttle armature. 30. Note. There are several types of magnetos designed especially to deliver low-tension alternating current for use in high-tension ignition systems. The more important of these will be described in connection with high-tension ignition. CHAPTER III. DIRECT-CURRENT MAGNETOS. 31. General. By the use of a suitable form of armature between the pole-pieces of permanent magnets, a direct current can be obtained. The same magnets and pole-pieces can be used as for the shuttle-wound armature, but the armature core FIG. 31. Elementary Form of Drum Armature between Pole-Pieces of Magnets. is different. Several forms of cores are used, among which one in the form of a cylinder, and another having the form of a ring, are most common. The cylinder generally has lengthwise slots to receive the winding. An armature with a cylindrical core is 38 DIRECT-CURRENT MAGNETOS 39 generally known as a drum armature. This applies whether the core has a smooth cylindrical surface or is slotted as just stated. 32. Elementary Form of Drum Armature. A cylindrical core A with one turn of insulated wire W around it is shown in Fig. 31 in the magnetic field between the pole-pieces of a set of permanent magnets. The wire is continuous (without ends). The mag- netic flux is from the N pole across the air-gap between the N pole and the core to the core, through the core and across the air-gap between the core and the S pole to the latter. The same number of lines of magnetic force pass through the smooth cylin- drical core whatever its position with regard to rotation about its axis. This is also approximately true of a slotted cylindrical core of the usual form. 33. Generation of Current. When this elementary armature is standing in the position shown in Fig. 31, all of the magnetic lines of force in the core pass through the space inclosed by the wire. Other positions of the armature are shown in Fig. 32. When in position (A) part of the magnetic flux through the core passes through the space inclosed by the loop of wire, and part passes outside of the loop. In (B) none of the flux is through the space inclosed by the loop. In (C) part of the flux in the core is through the coil space, and in (D), half a revolution from the position in Fig. 31, all of the flux through the core passes through the coil. The direction of the flux through the coil space is in the opposite direction in (D) t relative to the coil, from its direction in Fig. 31, where it enters the coil space from the side next to the feathered arrow cut into the core. In (D) the flux enters the coil space from the side opposite the feathered arrow, which rotates with the core and coil. When the armature is rotating, the positions at which all of the flux in the core passes through the coil, and those at which none of the flux- passes through the coil, occur later in the revo- lution on account of magnetic lag and the reactions which occur on account of the electric current generated in the coil and other parts. For convenience in discussing the manner in which a direct ELECTRIC IGNITION FIG. 32. Current Flow in Winding of Elementary Drum Armature. current is obtained, the effect of magnetic lag and armature reactions will be neglected.* As the armature passes through the position shown in Fig. 32 at (X) while rotating in the direction indicated by the feathered arrow, electromotive force is generated in the wire of the coil and current flows through the wire in the direction indicated by * It is not thought necessary to discuss more fully the effect of cross-magnet- ization and armature reactions. DIRECT-CURRENT MAGNETOS 41 the arrow on the wire. The direction of current flow is similarly indicated in (), (C), (), and (F). There is no current flow in position (D) and in the position shown in Fig. 31. The elec- tromotive force at any instant is proportional to the rate at which the stretches of wire along the cylindrical surface of the core are cutting through the lines of force. This rate corre- sponds to the rate of change in the amount of magnetic flux through the space inclosed by the coil. The current is approxi- mately proportional to the electromotive force at any instant. In the positions shown in Fig. 31, and at (D) in Fig. 32, the wire is not cutting through lines of magnetic force, hence there is neither electromotive force nor current. In (E) and (F) the direction of current flow through the wire is opposite that in (A), (B), and (C). In (E) and (F) the arrow indicating the direction of flow through the stretch of wire across the front end of the core points in the same direction as the feathered arrow engraved in the end of the core, while in positions (A), (B), and (C) the arrows point in opposite direc- tions. The reason why the current changes its direction of flow has been discussed in 22. It should be noted that the direction of current flow through the portion of the wire that lies across the end of the armature core is always from the S pole toward the N pole when the armature is rotating clockwise. In other words, the current flow through the wire next to the south pole is always toward the observer when the rotation is clockwise.* 34. Commutation of Current in a Direct-current Generator. In Fig. 33 the coil of wire is cut in two at the front end and the ends fastened to two parts, i and 2, which are approximately half-rings of metal. Brushes, 3 and 4, of metal, carbon, or some other conductor of electricity, bear on the rings at points (really areas) diametrically opposite each other. From these brushes wires connect to an external circuit 5. While the armature is rotating clockwise through the position shown, the current flows from brush 3 (next to the S pole) to the external circuit 5, through the external circuit and then to the brush 4. This continues as * If the rotation were in the opposite direction, the current flow would also always be in the opposite direction from that indicated. ELECTRIC IGNITION long as brush 3 is in contact with segment i and while current is generated in the armature coil. When the armature reaches a position similar to that in Fig. 31, which may be called the " dead " position of the coil, the open spaces between the two segments have come under the brushes. The brushes therefore change from one segment to the other of the ring while no current is flowing, and consequently there is no spark formed during this VWWIAAJ FIG. 33 . Two-segment Commutator and Brushes of Elementary Drum Armature. change from one segment to the other. When segment 2 is alone in contact with brush 3 and current is again generated, the flow is, as before, from brush 3 through the external circuit 5 to brush 4. By the use of this two-segment commutator the flow of current through the external circuit is caused to be always in the same direction. The flow of current in the armature wire alternates as before, however. Since the flow of current is always from brush 3 to the external circuit, this brush is called the positive brush and is usually indi- cated by the sign +. The other brush, 4, toward which the current flows, is indicated by the sign , and is called the negative brush. DIRECT-CURRENT MAGNETOS 43 A current which flows in one direction only is called a direct current. As produced by the elementary generator shown in Fig. 33, it is intermittent, or, more specifically, pulsating. 35. Continuous-current Electric Generator. In order to ob- tain a continuous current it is necessary to use more than one armature coil and more than two commutator segments. To operate successfully for the usual requirements, the coils are spaced uniformly around the core, and the commutator segments are all of the same width circumferentially. In the more usual constructions there is the same number of commutator segments as there are coils, but not infrequently twice as many commutator segments as coils are used. Each coil may have only one turn, as in Fig. 33, or each may have several turns, or wraps. It is probably that in all direct-current generators intended for igni- tion purposes, each armature coil has several turns. FIG. 34. Thin Disk of a Laminated and Slotted Drum Armature. 36. Laminated Drum Armature Core. In the better genera- tors for direct current the armature core is built up of a number of thin disks cut from sheet metal and placed side by side in the same manner as has been described for shuttle-wound armatures and for the same reason. One of the disks for a direct-current generator is shown in Fig. 34. The metal is cut out at regular 44 ELECTRIC IGNITION intervals around the periphery to leave openings which, when the disks are grouped together in the armature, form the slots in which the wire is wound. The central opening is for the armature spindle, which is usually all in one piece and passes through the core. It is good practice to place a brass sleeve, or quill, between the steel spindle and the core-disks. 37. A complete drum armature for direct current is shown in Fig. 35. This armature has 12 coils, each of several turns of FIG. 35. Drum Armature of Direct-current Electric Generator. wire wound in the slots of the core, and 12 segments in the commutator. 38. A commutator similar in general form to that on the arma- ture in the preceding figure is shown in Fig. 36, in which (A) is a /- x .. .. (A) (B) FIG. 36. Commutator for Direct-current Electric Generator. Twelve Segments. view of the complete commutator, and (B) is a longitudinal sec- tion. The segments are dove-tailed and held in place by the DIRECT-CURRENT MAGNETOS 45 correspondingly dove-tailed inner sleeve and ring. Insulation i is placed between the adjacent copper segments, and at 2, 3, 4, and 5 between the segments and the metal sleeve. The annular space 6 may or may not have insulation in it, according to the will of the designer. There is a possibility of moisture collecting in this space if it is not filled with insulation. The insulation between adjacent segments must be of some material that will withstand heat and is not readily burned by the sparks that form as the brushes pass from one segment to the next, especially when the brushes are not properly set. Mica or some composi- tion composed chiefly of mica is used for this insulation. Various methods of fastening the ends of the armature coils to the segments are used. A common one is to notch or slit the end of the segment and solder the wire into the notch. A hard solder (one that does not melt at a low temperature) should be used, so that it will not melt and fly out in case the commutator becomes hot. It is well to swedge the segment down on the wire to prevent the latter from flying out in case the solder melts. Screws are sometimes used to fasten the wire to the segments, but they are apt to become loose, unless soldered, on account of the expansion and contraction due to heating while in service and cooling while at rest. 39. Armature Connections. Fig. 37 is a diagram showing conventionally how the ends of the armature coils are brought to the segments of the commutator. This diagram is for an arma- ture with 12 coils and the same number of commutator segments, intended for use in a bipolar generator. Only one turn of wire for each coil is represented, but each coil may have several turns. Starting at segment i, connection is made to one side A of a coil lying in a slot of the core. Side A is connected, across the back end of the core, to the side A' of the same coil, and A' is connected to the segment 2. Segment 2 is also connected to B, which is connected across the back end of the core to B', and the latter is connected to segment 3. The same method of connec- tion is followed out for all of the coils, thus: 3, C, C', 4; 4, D, D', 5; and so on to 12, L, L', i. The coils are not shown connected in the successive order in 46 ELECTRIC IGNITION which they have to be wound on the core. The successive order of winding is A, H, C, J, E, L, G, B, 7, D, K, F. If the coils were connected to the commutator in the order of their winding, as just given, the lengths of the different circuits through the FIG. 37. Winding Diagram of Direct-current Armature with Twelve Coils and Twelve Commutator Segments. armature would not be uniform, and the armature would not operate as satisfactorily as when the connections are made as has been shown. The direction of current flow through the connections to the commutator is indicated by the arrowheads on the lines repre- senting the connections. The brushes are shown in contact with the commutator, and are indicated as positive and negative by the signs + and . It can be seen that the -f brush electrically connects the seg- ments i and 2, to which the ends of the coil A are attached, when DIRECT-CURRENT MAGNETOS 47 the armature is in the position of its rotation shown in the dia- gram. The brush bridges the insulation between i and 2. This corresponds to the connection that occurs between the two half- Direct-current Magneto. 1. Field magnets. 2. Pole-pieces. 3. Armature. 4. Brass tube enclosing armature. 5. Commutator. 6. Brush that bears on commutator. 7. Brush-holder. Insulated. 8. Insulation between brush-holder and tube 4. 9. Insulation around screw that holds brush-holder in place. 10. Coil-spring for pressing brush 6 against commutator. FIG. 38. Hercules Electric Co., Indianapolis, Ind. 11. Terminal for external wire. 12. Bearing for armature spindle. 13. Bell-shaped friction pulley. 14. Friction facing of pulley 13. 1 5 . Collar for speed governor. 16. Governor balls and arms. 17. Governor spring. 18. Setscrew. 19. Base of magneto. 20. Clamps for holding magnets against pole-pieces. 21. Clamp bolt. 22. Name plate. rings in Fig. 33 when the armature in the latter figure has rotated one-quarter revolution from the position shown. But in Fig. 3 7 the coil A is in its dead position when thus short-circuited by 48 ELECTRIC IGNITION the brush, just as the coil is in Fig. 33 when the ends of the half- rings are under the brushes, therefore no electromotive force is induced in the coil to cause current flow in that coil which would cause sparking at the brush when segment 2 passes from contact with the brush and thus breaks the short-circuit of coil A . The same applies to coil G, which is shown short-circuited by the negative ( ) brush. It also applies to all the other coils as their segments pass successively under the brushes. 40. A complete direct-current magneto for giving a continuous current is shown in Fig. 38. The armature is surrounded by a brass tube 4 through which the pole-pieces project so as to come FIG. 39. Photographic View of Direct-current Magneto shown in Fig. 38. close to the armature. This tube, together with the heads which carry the spindle bearings, form a dust- and water-proof protec- tion for the armature. The brushes are made of phosphor-bronze wire gauze pressed in dies to a suitable form around a carbon core. The carbon acts as a lubricator for the rubbing surfaces of the brushes and commutator. A bell-shaped friction pulley 13 is carried on the driving end of the armature spindle. This pulley is faced with a suitable friction material, such as leather, paper composition, or rawhide, which is pressed by spring action axially against a rotating part, such as a flywheel or pulley that drives the friction pulley. The speed of rotation of the armature is controlled by a shaft governor of the fly-ball type. The balls DIRECT-CURRENT MAGNETOS 49 are drawn toward each other radially by a pair of coiled tension springs, one of which, 17, is shown. As the speed increases the balls move out radially and reduce the pressure between the FIG. 40. Direct-current Magneto with Friction Pulley and Speed Governor. Tritt Elec- tric Company, Union City, Indiana. facing of the friction pulley and the flywheel. This action allows the friction pulley to slip enough on the flywheel to keep the speed of the armature down to the required rate. Fig. 39 is a photographic view of a direct-current magneto of the general type shown in the preceding figure. Fig. 40 illus- trates another machine that operates in a similar manner. CHAPTER IV. TESTING FOR DIRECTION OF CURRENT. Chemical Tests. 41. Water Test. Bubbles Form at Submerged Wire-End. - To about half a pint of water in a glass tumbler or other vessel that is not a conductor of electricity add any one of the following : Common salt (NaCl), one teaspoonful; Common washing soda (sal soda, Na 2 CO3), one teaspoonful; Sulphuric acid (H 2 SO 4 ), half teaspoonful of the strength sold at drug stores; Hydrochloric acid (HC1), half teaspoonful of the strength sold at drug stores. Connect a small wire with each of the two brushes of a direct- current generator, or with the positive and negative terminals of any source of direct-current supply, and dip the free bare-metal ends of the wires into the impure water, first keeping the ends as far apart as possible and then gradually bringing them toward each other, without allowing them to touch. Bubbles of gas will form on and rise from the submerged end of the wire that is connected to the negative ( ) side of the source of direct-current supply. (This test does not apply to alternating current.) With the small currents and pressures commonly used for ignition purposes, there is no appreciable formation of bubbles at the positive wire-end. There are numerous other substances that can be used for adding to the water to make it impure for the direction-of -current test. In fact, water from city mains often contains enough matter in solution to make it impure enough (chemically speak- ing) for this test. The passage of the direct current through the impure water decomposes it into its chemical elements, hydrogen and oxygen, each of which is a gas. This action is more or less indirect so 50 TESTING FOR DIRECTION OF CURRENT $1 far as the water itself is concerned. The hydrogen is liberated at the negative terminal. When water is decomposed it gives two volumes of hydrogen for each volume of oxygen. The oxygen tends to collect at the positive terminal, but at least part of it is absorbed by the water and thus disappears so far as its being a gas is concerned. Each of the submerged ends of the wire is called an electrode. The liquid, in this case impure water, is called the electrolyte. The positive electrode (the one connected to the positive side of current supply) is called the anode, and the negative electrode is called the cathode. The decomposition of the water sets up a counter-electromotive force of about 1.48 volts. This is about the maximum electro- motive force of one cell of ordinary dry electric batteries such as are in common use for ignition. One cell of such a battery will not therefore generally give bubbles at the electrode in this test. At least two cells must be connected in series to make certain of producing bubbles. Different methods of connecting cells to form a battery are given later. 42. Color Test. If a tablespoonful of sal ammoniac (ammo- nium chloride, NH 4 C1) is dissolved in half a pint of water, and the bare ends of two wires placed in the liquid as described in the preceding article, the liquid around the positive terminal, or anode, will turn blue, and bubbles will form at the negative terminal, when current flows. There are several substances that will give color tests of this nature. Different colors are obtained according to the sub- stances used. Convenient devices for making color tests are found on the market. A small glass tube some two inches in length, set in a mounting and having suitable terminals, is a convenient form. The instrument should be marked so that there can be no mistake in determining which of the two wires connected to it to test them is positive or negative. ELECTRIC IGNITION Test with Magnetic Needle. 43. Magnetic-compass Test. Place the wire which carries the current immediately above the case in which the magnetic needle is mounted, as in Fig. 41, so that the direction of the wire is the same in general as the length of the magnetic needle. If the wire is not insulated it is best not to let it touch the metal of the case. It is immaterial whether it touches the glass. The Positive Negative Negative (A) FIG. 41. Magnetic Compass Indicating Direction of Flow of Electric Current by the Deflection of the Needle. needle will be deflected from its north-and-south position during the time a direct current flows through the wire placed in this position. If the current flows from the north toward the south above the needle, it will be deflected in a clockwise direction as indicated at (A). In other words, the needle will be turned through part of a revolution in the same direction that the hands of a clock rotate. A current from the north above the needle turns it clockwise is a convenient expression by which to remember the action of a direct current on a magnetic needle. TESTING FOR DIRECTION OF CURRENT 53 If the current flows in the opposite direction (toward the north) above the needle, it will be rotated in the opposite direc- tion as shown in (B). By placing the wire beneath the needle, the action on the needle will be the reverse of that when the wire is above the needle. 44. Extemporized Compass Needle. An ordinary sewing needle magnetized and floated on water in a non-magnetic vessel can be conveniently used in the absence of a compass. The needle can be magnetized by bringing it in contact with a magnet, or by means of an electric current. The latter method is ex- plained in Chapter VI. If the sewing needle is highly polished, as when new, it can be floated on water by first drying it and then rubbing it with a slightly oily cloth or one's fingers, slightly oily. If then laid, or dropped lightly, on the water it will float and quickly assume a north-and-south direction. A needle that is bright and prop- erly oiled will float a day or more. The action of the electric current upon this floating needle is the same as on the pivotally supported needle in a compass. Another method of floating the needle is to lay it on, or stick it through, a flat piece of cork or paraffin. A cup or saucer is convenient for holding the water. A brass, copper, or aluminum vessel will answer, but the wire, if bare of insulation, should not be allowed to touch the vessel in two places at the same time. A steel or iron vessel is not so satis- factory on account of the tendency of the needle to float up against the side. 45. Test with Measuring Instruments. Many of the amme- ters and voltmeters for measuring current and pressure are made so that the direction of the current flowing through them can be told. The terminals of the instrument are marked + and , or P-\- and N . The indicating needle, or pointer, of the instrument moves so as to give a reading on the graduated scale only when the wires are connected to the instrument terminals in accordance with the signs; the positive wire to the terminal with the + sign, and the negative wire to the terminal with the sign. CHAPTER V. ELECTRIC MEASURING INSTRUMENTS.* 46. General. It is often desirable to test a battery to deter- mine its condition by measuring its voltage and the amount of current that it will give; also to measure the amount of current that an ignition system is using. For this purpose numerous types of small portable instruments have been developed, and a lesser number of instruments intended to be fixed in place. The smaller portable instruments frequently resemble a watch or pocket compass in general appearance, and are about the size of a large watch. These instruments ordinarily operate on the principle that an electric current flowing through a coil of wire attracts or repels a permanent magnet or a piece of magnetic material and causes it to move when it is mounted so as to allow movement, or upon the principle that two coils of wire with current flowing through them attract or repel each other so that one coil is moved when mounted for such movement. The chief difference between the am- meter, for measuring current, and the voltmeter, for measuring pressure, is that the ammeter has a coil, or coils, of comparatively thick wire of short length which has a very low resistance, and the voltmeter has a coil, or coils, Portable Ammeter for Measur- of very thin wire of great length and ing Electric Current. Small verv frjgh resistance. 47. Ammeters. A small portable ammeter for measuring current up to 30 amperes is illustrated in * This chapter is intended to deal with only such measuring instruments as are used in connection with ignition systems, and only to an extent sufficient to give a general idea of their nature and use. It is not thought desirable to go into details of measuring instruments in a work of this nature. 54 FIG. 42. ELECTRIC MEASURING INSTRUMENTS 55 Fig. 42. The indicating needle (pointer, hand) which indicates the amount of current flowing through the ammeter is pivoted at the center of the instrument and shown pointing to zero of the graduated scale. When current is flowing through the in- strument, the needle is deflected and points to the reading that FIG. 43. Stationary Ammeter. Weston Electrical Instrument Company, Newark, New Jersey. corresponds to the number of amperes of current flowing. One of the terminals is the projection at the bottom of the case, and the other is at the free end of the attached wire. When testing a primary battery, one terminal of the ammeter is electrically connected to one terminal of the battery, and the 56 ELECTRIC IGNITION other terminal of the ammeter is electrically connected to the other terminal of the battery. Current from the battery then flows through the ammeter. (The ammeter should not be used in this manner for testing a storage battery, unless the instru- ment is especially designed for such use, and current should not be allowed to flow longer than necessary to obtain a reading not longer than two or three seconds.) TO measure the current flowing through any circuit, the am- meter must be interposed in the circuit (cut into the circuit) so that all of the current of the circuit will flow through the am- meter. Thus, the ammeter may be cut into a circuit that has a wire held by a binding screw, by disconnecting the wire from the binding screw and connecting one terminal of the ammeter to the binding screw, then connecting the other terminal of the ammeter to the disconnected end of the wire. Fig. 43 is a high-grade ammeter in- tended to be used in a fixed position, as on a switchboard. Its range of current is from zero to 50 amperes. Similar instruments are made with ranges re- spectively up to i, 5, 10, 15, 25, and 75 amperes. 48. Voltmeters. Fig. 44 is a small TIG. 44. portable voltmeter reading up to 10 Portable Voltmeter for Measur- volts. It resembles, in general appear- ing Electric Pressure. Small ance? ^ sma n ammeter just described. One terminal is at the bottom of the case, and the other at the free end of the attached wire. For measuring voltage, the terminals of the voltmeter are electrically connected to the two points between which the pressure is to be measured, one terminal of the voltmeter to each point. It is immaterial whether the circuit is otherwise open or closed between the points of connection, so far as the action of the voltmeter is concerned. Thus, the voltage of a battery can be measured by connecting the terminals of the voltmeter to the terminals of the battery, either while the battery ELECTRIC MEASURING INSTRUMENTS 57 has no other connection to it (on open circuit), or while the battery is connected in circuit and delivering current for its regular service. The reading of the voltmeter will not generally be the same under the two conditions, but this is because the difference of pressure between the battery terminals is not the same while it is delivering current as when it is on open circuit. The amount of current that flows through the voltmeter while measuring pressure is so small as to have no appreciable effect on the action of the battery or the circuit to which the voltmeter is connected. 49. Volt-ammeters. It is quite usual to combine a volt- meter and an ammeter in one instrument, called a volt-ammeter. In some of these only one indicating needle (hand, pointer) is used, and the reading scale has two graduations, one for amperes and the other for volts. Such an instrument cannot be used for measuring both current and pressure at the same instant. Other volt-ammeters are made up of two complete instruments, a voltmeter and an ammeter, and can be used for measuring both current and pressure at the same time. Fig. 45 is a small portable volt-ammeter of the type having only one indicating needle. The reading scale is graduated to '10 volts and 30 amperes. The instrument has three terminals. The one at the top is for both volts and amperes. The left- hand one at the bottom is for volts, and the other for amperes. For measuring purposes, the top terminal and the left-hand one at the bottom are connected respective to the points between which the pressure is to be measured. For measuring current, the connections are made to the top terminal and the right-hand lower terminal. A comparatively small volt-ammeter for use in connection with storage batteries (see Fig. 102) is shown in Fig. 46. Only one indicator needle is used. The needle is shown in its zero position, which is not at the end of the graduated scale. The lower scale is for amperes, and is graduated in both directions from its zero. When the instrument is used as an ammeter, the needle is deflected either to the right or the left, according to the direction in which the current is flowing through the ELECTRIC IGNITION instrument. To obtain a voltage reading, the push-button V below the scale is pressed in and held while taking the reading. FIG. 45. Volt-ammeter for Measuring Both Pressure and Current. Small Pocket Form. FIG. 46. Stationary Volt-ammeter Which Indi- cates the Direction of Current Flow. Apple Electric Company, Dayton, Ohio. The voltage scale is graduated in only one direction from zero. In Fig. 47 two complete instruments, a voltmeter and an FIG. 47- Ammeter and Voltmeter Mounted Together Permanently. ammeter, are mounted on the same base. They can both be used at the same time, and used continuously. The instrument ELECTRIC MEASURING INSTRUMENTS 59 is so constructed that it can be used on vehicles. The upper (middle) terminal is connected to both instruments. The right- hand terminal is for the voltmeter only, and the left-hand terminal is for the ammeter only. 50. " Dead-beat " Indicating Needle. Unless some means is provided for quickly bringing to rest the indicating needle of a voltmeter or an ammeter, the needle will continue vibrating for a considerable time after the current is first sent through the instrument. When the instrument is moved, as on a vehicle or in one's hand, the needle may never come to rest. This vibration of the needle makes it impossible to take an accurate reading. The better class of instruments are constructed so that the needle comes to rest quickly and stands almost without vibration even when the entire instrument is subjected to a reasonable amount of motion, yet is sensitive in its movement to indicate variation in the current or pressure. Such an indicating needle is said to be " dead-beat." This dead-beat effect is generally obtained by constructing the instrument so that the movement of the needle and its attached parts set up foucault currents that in turn resist the movement of the parts to which the needle is attached. This damping action is of much the same nature in its effect as that which can be obtained in a stationary instru- ment by attaching a vane, or wing, to the spindle on which the needle is mounted, and submerging the vane in a liquid. CHAPTER VI. ELECTROMAGNETS. 51. Plain Bar Electromagnet. If an insulated wire is wrapped around a bar of iron or steel as shown in Fig. 48, and a direct current of electricity sent through the wire, the bar will become a magnet and remain so as long as the electricity con- tinues flowing. When the current stops, the bar will lose nearly c f c (B) FIG. 48. Electromagnets of Straight Bar Type. all of its magnetism if it is of commercially pure soft iron or steel. A small amount of magnetism will remain. This is called residual magnetism. If the bar is hardened, or tempered like a sewing needle, knitting needle, or a file for working steel, it will retain a considerable amount of magnetism and be, for a while at least, a permanent magnet. If of the quality and con- dition of steel used for permanent magnets, it will remain a per- manent magnet after the current stops and the bar is removed from the coil. If the current flows in the direction indicated by the arrow- heads on the wire, then the upper, or top, end of the bar will be 60 ELECTROMAGNETS 61 a north pole, and the lower end, or bottom, of the bar a south pole. This applies to both (A) and (B) in the figure. If the current is made to flow in the opposite direction from that indi- cated, then the lower end of the bar will be a north pole, and upper end a south pole. The magnetic flux in the bar is from the south pole to the north pole. If the bar is held before the dial of a clock with one end point- ing toward the dial, and current is flowing through the wire in the direction of rotation of the hands of the clock, then the lines of magnetic force will flow through the bar toward the clock. The north pole will be next to the clock. Another method of determining which is the north pole is as follows: If the bar is vertical and the current in the portion of the coil between the observer and the bar flows east while the observer is looking north, then the north pole is at the top. If the bar is horizontal at the level of the observer's eyes, and the current in the por- tion of the wire between the observer and the bar flows downward, then the north pole is at the right-hand end. 52. Plunger-core Electromagnet. Fig. 49 shows a non-magnetic spool i with a coil 2 of insulated wire wound around it. It may be assumed that the coil is supported in a vertical position. An iron or steel bar 3 is shown with its upper end projecting a slight distance into the opening through the spool. It may also be assumed that no current is passing through the coil, and that the bar is resting on some support, such as a table. If an electric current, sufficiently large, is passed through the coil, the core will be drawn up into the spool and will remain suspended there as long as the current continues flowing through the coil. The bar will remain in the central part of the spool opening without touching any part. Its middle will be somewhat below the middle of the coil, on account of the weight of the bar. FIG. 49. Electromagnet with Plunger Core. 62 ELECTRIC IGNITION As soon as the current is stopped the bar will fall. The current must be direct (not alternating). As indicating the force with which the bar is drawn upward, it can be shot up completely through and above the spool if the current is stopped while the bar is still moving rapidly upward just after the circuit has been closed. 53. U-shaped Electromagnet. - Fig. 50 shows an electromagnet shaped like the letter U in an in- verted position. When a direct current flows through the winding as indicated by the arrowheads, the poles are N and S, according F IG> 5a to the marking on the bar in U-shaped Electromagnet. the % ure * If the current flows in the opposite direction, the po- larity of the poles is changed. It can be seen that, when looking at the ends of the bar, the FIG. 51. Ring Electromagnet with Consequent Poles. current flows clockwise around the south-pole leg of the bar, and counter-clockwise around the north-pole leg of the bar. The ELECTROMAGNETS winding is as if the whole coil had been wound on a straight bar, and the bar then bent to the U-form. FIG. 52. Ring Electromagnet with Two Projecting Poles. FIG. 53- Four-pole Ring-shaped Electromagnet. By attaching suitably formed pole-pieces to the ends of the bent bar, it can be used for furnishing the magnetic field of an electric generator. 64 ELECTRIC IGNITION 54. Ring-shaped Electromagnet with Consequent Poles. A plain ring of iron or steel can be magnetized electrically so as to make a north pole and south pole as indicated in Fig. 51. The manner of winding the coil and the direction of current are indicated in the figure. Poles located in this manner are called consequent poles. By attaching suitable pole-pieces to the ring, one at N and another at S, it can be used for the field-magnet of an electric generator. 55. A bipolar ring-shaped electromagnet with winding on pole-pieces is shown in Fig. 52. The pole-pieces may be either an integral part of the ring, or separate parts attached to the ring by bolts or other suitable fastenings. 56. A four-pole ring-shaped electromagnet is shown in Fig. 53. A magnetizing coil is wound on each pole-piece. The direction of the current through each coil is indicated by the arrowheads. CHAPTER VII. DIRECT-CURRENT GENERATORS WITH ELECTROMAGNETS. 57. General. Electromagnets can be used in conjunction with either an armature that delivers an alternating current, one that delivers a direct current, or one that delivers both direct and alternating current. When the armature delivers only alternating current, some auxiliary source of direct current must be provided for supplying electricity to magnetize the field- magnets; in other words, for exciting the field. But when the armature delivers direct current, all or part of the current can be used to excite the field-magnets, thus eliminating the necessity of the auxiliary source of current supply. It is believed that the only type of electromagnetic generators that are used for ignition purposes is that in which the armature delivers direct current, therefore only this type will be described. Shunt-wound Direct-current Generators. 58. A bipolar direct-current generator with U-shaped shunt- wound electromagnets is shown in elementary form in Fig. 54. Two circuits are connected to the brushes, one through the external circuit i, and the other through the field-coils 2 and 3. The latter is called a shunt circuit, or simply a shunt. The shunt is a comparatively small wire of considerable length and a great number of turns around the magnet core, so that only a small proportion of the current that the armature is able to deliver passes through the field-coils.* * The amount of direct continuous current that flows steadily through a wire is inversely proportional to the length, and directly proportional to the sectional area of the wire. A thin wire offers more resistance to the flow of current through it than a thick wire of the same material. This is analogous to the greater resist- ance offered to the flow of a liquid through a small pipe than through a large one. The resistance offered to flow in both the wire and the pipe is proportional to the length of the wire and the pipe. 65 66 ELECTRIC IGNITION A generator of this nature must first have its field-magnets magnetized by current from an exterior source, or by another magnet. After being once magnetized, the field-magnets retain enough residual magnetism to start the generation of a current in the armature when it is rotated, unless the magnets are sub- jected to some unusual demagnetizing influence. When the armature is started to rotate, the slight electro- motive force generated by the residual magnetism in the field- FIG. 54. Bipolar Direct-current Electric Generator with Shunt-wound U-shaped Electro- magnets. magnets sends a correspondingly small current through the field- coils. This current strengthens the magnets, which in turn induce a greater electromotive force in the armature, and more current flows through the field-coils. By this progressive action, the generator " picks up " or " builds up " its magnetism until a condition is reached where the increase of magnetism becomes slow in relation to the increase of current in the field-coils, and a constant electromotive force is then maintained as long as the external circuit remains the same. DIRECT-CURRENT GENERATORS WITH ELECTROMAGNETS 67 An increase of current through the external circuit, such as may be caused by removing a piece of apparatus from it, still leaving the circuit closed, causes a reduction of voltage at the brushes of the generator. The generator is usually so constructed for ignition purposes that the variation of pressure at the brushes is not excessive for variations of current within the range through which the machine is designed to operate. A method of preventing this FIG. 55. Shunt-wound Direct-current Electric Generator with Ring-shaped Bipolar Field- magnets. drop of pressure at the brushes, by using a " compound winding " on the magnets, is given later. 59. A bipolar direct-current generator with ring-shaped shunt- wound electromagnets is shown conventionally in Fig. 55. The principle of operation is the same as for the generator shown in the preceding figure. It may be noted, however, that the brushes do not stand in the same position relative to the pole- pieces as they do in the former figure. In Fig. 54 the position of the brushes relative to the pole-pieces corresponds to that in Fig. 37. In order to obtain the relative positions shown in Fig. 55, the commutator in Fig. 37 may be twisted around a quarter-turn counter-clockwise relative to the armature core 68 ELECTRIC IGNITION without changing the connections of the wires to the commutator. In any case, the brushes can be made to stand in any position relative to the pole-pieces, by making the connections to the commutator segments accordingly. A complete commercial machine of the type shown conven- tionally in Fig. 55 is illustrated in Fig. 56. The protective cap FIG. 56. (See also Figs. 57, 58, 59, and 60.) Bipolar Direct-current Electromagnetic Generator. The Dayton Electric Manu- facturing Co. Dimensions in inches: io| long; 5! wide; 5! high. r 3 amperes continuously. Capacity < 8 volts at 1000 r.p.m. 1 10 volts at 1 200 r.p.m. 1. Commutator. 2. Brush, insulated. 3. Brush spring, insulated at end that presses against brush. 4. Terminal. 5. Connection between brush and terminal 4. 6. Field Coil. 7. Steel tube around armature. (Discarded in later designs.) 8. Oiler with felt wick. 9. Spider with bearing for armature spindle and with brush-holders. is opened out on a hinge at the commutator end to show the working parts. The pole-pieces are above and below the arma- ture. The brushes are perpendicular to the commutator, so the armature can rotate in either direction. Each brush has a short ribbon spring, somewhat like a short clock-spring, for DIRECT-CURRENT GENERATORS WITH ELECTROMAGNETS 69 pressing it against the commutator. The magnetizing coil 6 of the top pole-piece is partly visible. At 7 is a steel tube which fits against the pole-pieces, and inside of which the armature runs without touching it. The use of this tube has been discon- tinued in later designs. The armature for this machine is shown separately in Fig. 35. The oiling device is shown in Fig. 57. It consists of an oil res- ervoir into the top of which is fitted a round felt wick that is pressed up against the journal of the armature shaft by a coiled compression spring. The capillary action of FIG. 57. ., . , ,, ., ,, , Oiling Device with Felt the wick carries the oil up to the bearing wkk? for Fig s6 gradually. The field-coils are shown in Fig. 58. Each coil is made up of insulated wire which, after being wound to form, has air and moisture removed by placing it in a vacuum, and is then insulated by impregnating it with FIG. 58. Field-Coils, for Fig. 56. liquid insulating compound that hardens like varnish upon dry- ing. The coil is then wound with insulating tape. The brushes, Fig. 59, are of graphite with a bronze-gauze core. The terminal wires are soldered to the gauze core. ELECTRIC IGNITION Fig. 60 shows the spider 10 which supports the brushes and the commutator end of the armature spindle. The capacity of a machine like that in Fig. 56, having the dimensions: i of inches long, 5! inches wide, and 5! inches high, FIG. 59. Commutator Brushes, for Fig. 56. as rated by the makers, is 3 amperes of steady current at a pres- sure of 8 volts when running at 1000 r.p.m., or a little more than 12 volts at 1200 r.p.m. FIG. 60. Bearing, Brushes, Oiler, etc., for Fig. 56. The only means of regulating the pressure is by variation of the speed. The current is also varied in practically the same proportion as the pressure when the circuit remains unchanged. A friction pulley or a belt pulley combined with a speed governor is provided with the machine. DIRECT-CURRENT GENERATORS WITH ELECTROMAGNETS 71 60. A four-pole direct-current generator with shunt-wound electromagnets is shown diagrammatically in Fig. 61. Only two brushes are used. They are placed at an angle of 90 degrees with each other of necessity. The path and direction of mag- netic flux is indicated by the broken lines with double arrow- heads on them. The connections for an armature of a four-pole machine with two brushes are shown in Fig. 62. This armature has twenty- FIG. 61. Elementary Four-pole Direct-current Electric Generator with Shunt-wound Magnets. one coils and the same number of commutator segments, or strips. The broken lines indicate the part of the winding that is in the rear of the armature core. The core is not shown, since it would detract from the clearness of the diagram. When the direction of rotation is clockwise, as indicated by the feathered arrow, the flow of current is as indicated by the arrowheads on the lines. Figs. 63 and 64 show a direct-current shunt-wound four-pole generator with two brushes. The former figure is partly in sec- tion. A governor spring bears against the commutator end of the armature shaft and presses the friction pulley against the flywheel that drives it. A speed governor is located on the 72 ELECTRIC IGNITION shaft between the friction pulley and the armature. This gover- nor moves the entire armature and the friction pulley endwise as the speed increases, so as to reduce the pressure of the friction pulley against the flywheel, thus allowing the friction pulley FIG. 62. Armature Connections for Drum Armature with Twenty-one Coils and the Same Number of Commutator Segments. For Four-pole Field-Magnets. to slip on the flywheel and limiting the speed to the desired rate. An adjusting screw is provided for varying the pressure of the governor spring against the end of the shaft so as to obtain the desired speed limit. There are two brushes set at 90 degrees with each other. They are a combination of wire gauze and graphite. The armature DIRECT-CURRENT GENERATORS WITH ELECTROMAGNETS 73 has twenty-one coils and the same number of segments in the commutator. The frame of the generator, which is also the GOVERNOR FIBRE THRUSTWASHER BEARING FRICTION PUUIY COHMUTATOR FIG. 63. (See also Fig. 64.) Four-pole Direct-current Electric Generator. Sectional View. Apple Electric Company, Dayton, Ohio. magnet ring, is cast from semi-steel. This is a material between soft steel and cast-iron in its physical and magnetic properties. The generator is provided with an automatic cut-out for open- ing and closing the circuit when used in connection with storage COVERNORjSPRING ADJUSTER BRUSH SPRING 4KIU3TER BRUSHES GOVERNOR- SPRING PRESSED STEEL LID AUTOMATIC'CUTOUT ' OIL CL)P LINE.WlRt.f BINDING. POST FIG. 64. (See also Fig. 63.) Commutator End of Four-pole Direct-current Electric Generator. batteries. This method of using is described later in connection with a complete ignition system (see Fig. 95 and several follow- ing figures). The machine is also equipped with a device for keeping the current nearly constant when the speed of the 74 ELECTRIC IGNITION armature is variable. This device automatically changes the amount of resistance in the field circuit. Compound-wound Direct-current Generators. 61. Series-and-shunt Field Winding. It is often desirable to have a generator that will keep the voltage at the brushes practically constant when the rotative speed of the generator is constant, so that the pressure will be practically constant whether FIG. 65. Compound-wound Direct-current Electric Generator. the amount of current delivered is variable or constant. This is accomplished by means of a double winding on the field-mag- nets. One of the windings is the regular shunt winding, and the other winding carries the current that flows through the external circuit. The latter is called the series winding. Fig. 65 shows diagrammatically a generator with compound field winding of this nature. The currents in both windings flow in the same direction around the magnet cores. It has been explained that the voltage at the brushes drops as the external current increases when only DIRECT-CURRENT GENERATORS WITH ELECTROMAGNETS 75 a shunt winding is used. This tendency is counteracted by the magnetizing effect of the current which flows through the series coil. The magnetizing effect of the series coil increases as the current through the series coil and external circuit increases. By giving the series coil a suitable number of turns around the magnet core, the voltage at the brushes can be kept almost constant in a properly designed machine. By giving the series FIG. 66. Rheostat in Shunt-coil Circuit of Electric Generator. coil more than this number of turns, the voltage can be made to rise as the current in the external circuit increases. This over-compounding is often desirable. The voltage of a compound-wound machine such as shown in Fig. 65 is approximately proportional to the speed of the arma- ture, within the range of speed at which the machine is designed to operate. 62. Field Rheostat for Regulating Voltage. When the rota- tive speed of the armature of an electromagnetic generator is constant, the voltage at the brushes can be regulated by varying 76 ELECTRIC IGNITION the electrical resistance of the shunt-coil circuit. The usual means of doing this is a rheostat. Fig. 66 shows a compound-wound generator with an elemen- tary form of rheostat in the shunt circuit. The principal parts of the rheostat are shown at A and R. R is a series of coils of wire. The ends of the wire are connected to metal contact- points i, 2, 3, 4, and 5. These points are arranged in an arc of a circle. A switch-arm A is pivoted at the center of the arc and is of such a form that it can be moved into contact with any of the contact-points just enumerated. This rheostat is inter- posed in the shunt circuit by cutting the shunt wire at any con- venient point and connecting the wire-ends thus obtained to the rheostat as shown. One end is connected to the contact-point i, and the other end to the pivot of the arm A. When the rheostat arm A is set in contact with point 4 as shown, the shunt current must pass through the rheostat coils that lie between i and 4, and the resistance of these coils is added to that of the field-coils. The current that flows through the field-coils is therefore less than would flow if the rheostat were not in the circuit, and the voltage at the brushes is in con- sequence less than it would be without the rheostat. By moving the arm into contact with either 3, 2, or i, the voltage can be increased; or by moving it to 5, the voltage can be decreased. The swinging end of the contact arm A is wide enough to touch two of the contact-points at the same time when moving from one to another. This is necessary in order to prevent breaking the circuit while varying the resistance in the circuit. The rheostat is generally a separate piece of apparatus. The resistance wire used in it is ordinarily of a material that has high electric resistance compared with that of copper. A rheostat can be used in the field circuit of a plain shunt- wound machine as well as in one that is compound- wound. Reversing the Rotation of the Armature. 63. Most of the generators for ignition usage have the brushes perpendicular to the commutator so that the armature can be rotated in either direction without injury. DIRECT-CURRENT GENERATORS WITH ELECTROMAGNETS 77 The field-coil connections to the brushes must be interchanged when the direction of rotation of the armature is to be reversed. If the armature is run in the new direction without making this change of connections, it will not pick up its magnetism and generate a current. This is because the different direction of rotation changes the polarity of the brushes, so that the one formerly positive becomes negative, and the former negative one becomes positive. The current which is generated by the residual magnetism therefore flows through the field-coils in the direction to demagnetize them instead of strengthen their magnetism. The result is that no appreciable pressure is gen- erated, and of course no appreciable current can be obtained without corresponding pressure. CHAPTER VIII. PRIMARY BATTERIES. Carbon-zinc Battery. 64. When an electric battery is subject to considerable motion, as in automobiles, railway motor cars, and motor boats, a " dry battery " is almost exclusively used if ignition current is supplied by a primary battery. The dry primary battery is also much used for ignition in stationary motors. Only one of almost innumerable types of dry batteries, as distinguished by the substances used in them, is used to a noticeable extent. In this commonly used battery the substances _^ which designate it are carbon, zinc, and sal ammoniac (also called ammonium chloride, NH 4 C1) . Other substances are used in connection with these and are essential to its operation for supplying current for motor ignition. In order to make clear the nature of this battery cell and its operation, the elementary form using only carbon, zinc, and sal ammoniac will be first described. 65. Elementary Leclanche Car- bon-zinc Wet Cell. A bar of zinc and a slab of carbon immersed in a FIG. 67. Elementary Wet Electric Battery Solution f Sal ammoniac in a glass Cell. vessel are shown in Fig. 67. The solution is made by dissolving sal ammoniac (a white salt) in water.* The carbon and zinc are connected together by a wire. * One-quarter pound of sal ammoniac to a quart of water is the proportion generally used in a cell. 78 PRIMARY BATTERIES 79 A current of electricity begins to flow from the carbon to the zinc through the wire as soon as the carbon and zinc are immersed in the solution. The current also flows through the solution from the zinc to the carbon inside of the cell. The amount of current decreases very rapidly immediately, and then continues to decrease at a slower rate till, after considerable time, there is scarcely a perceptible flow. The cause of the decrease of current is called polarization, and is described below. Decrease of current on account of polarization also occurs in this elementary form of cell when it is used in the manner required for motor ignition. This action makes it unsuitable for motor ignition purposes. The electric current through the wire is generated by chemical action between the zinc and the solution. The solution attacks the zinc and combines with it. This action dissolves, corrodes, or eats away the zinc. The above combination is called a primary electric cell. In earlier days it was commonly called a galvanic cell or a voltaic cell. Two or more such cells properly connected together form a battery of cells, called an electric battery. In commercial usage a single cell is generally also called a battery. The carbon and zinc are called the electrodes, and the solu- tion is called the electrolyte. All three together are called the active elements of the cell. The point at which the wire is attached to the carbon is the positive (+) terminal of the cell; and the point of attachment of the wire to the zinc is the negative ( ) terminal of the cell.* In commercial forms of cells, binding screws and nuts, or other suitable fastenings, are usually provided for attaching wires at the terminals. If the wire is cut in two and the ends separated, the flow of current through it is stopped. The chemical action in the cell * On account of the confusion which arises when one of the elements of the cell is referred to as the positive electrode, and the other as the negative electrode, the terms positive electrode and negative electrode are not used herein in con- nection with electric cells and batteries. The terms anode and cathode are omitted in connection with batteries for the same reason. 8o ELECTRIC IGNITION also stops with the stoppage of current through the wire, except that there is generally some slight amount of local action on the zinc. Impurities in the zinc, such as iron and copper, increase this local action. But even if the zinc is very pure, local action will still occur on account of a difference in the strength, or quality, of the portion of the solution at the top and that at the bottom. The local action due to this latter cause eats away the zinc at and near the surface of the solution. Local currents through the zinc are caused by this action. The electromotive force of a primary cell with carbon and zinc electrodes in sal-ammoniac solution is slightly less than 1.5 volts between the terminals after the cell has not been de- livering current for some time. The voltage drops as soon as the circuit is closed and current begins to flow. It slowly rises again to its full value after the current is stopped by opening the circuit. 66. Polarization of Primary Electric Cell. The decrease of current that occurs while the circuit of a cell is kept closed, in the case of a cell having only the elements carbon, zinc, and sal ammoniac, is due to the formation of hydrogen gas by the chemical action. The gas collects on the carbon and retards chemical action. This retardation is apparently chiefly due to a counter-electromotive force which the hydrogen sets up, and also partly due, but to a less extent, to the formation of bubbles on the carbon so as to prevent the electrolyte from having as good contact with the carbon as it has before any bubbles are formed. If the carbon is molded very dense and has a very smooth surface, polarization can be at least largely prevented by constantly brushing off the bubbles of hydrogen. This is not practicable, however. The usual method is to prevent polariza- tion by chemical means. This is ordinarily called depolariza- tion. It is successfully applied to both wet cells and dry cells. 67. A dry cell with carbon and zinc electrodes and chemical depolarizer is shown sectionally in Fig. 68. This is the type of dry cell that is almost universally used for ignition where the battery is subjected to much motion. The containing vessel, cup, or can i is made of sheet zinc and PRIMARY BATTERIES 8l is one of the electrodes. The can is lined with absorbent paper 2 (blotting paper) that is saturated with water in which sal ammo- niac and zinc chloride have been dissolved. In the center is a molded bar of carbon 3 around which is packed a mixture 4 of manganese dioxide (MnO 2 ) and car- bon dust. The manganese dioxide is in granular form (powder) . The paper is turned in over the top of the mixture so as to cover it nearly or completely. On top of the paper is a little sawdust or sand, and above this a sealing compound 5, composed chiefly of pitch, to make the cell water-tight. The absorbent paper and the mixture around the bar are saturated with the electrolyte before the cell is sealed. The carbon in the mixture is generally coke-dust, and the carbon bar is made of coke- dust mixed with a binder such as pitch. The plastic mixture for the bar is molded to form and then baked to give it strength and at the same time convert the binder into carbon. Carbon is a better con- ductor of electricity than manganese dioxide, hence mixing it with the manganese dioxide gives the cell less resistance to the flow of electricity than if manganese dioxide alone were used around the bar. A low resistance is desirable in a dry cell, especially one that is to be used for gas-engine ignition. The carbon bar is capped with a tight-fitting brass piece which has a binding screw and nut. Another binding screw is soldered to the top of the zinc can. The cap on the carbon bar at the center of the cell is the positive (+) terminal, and the binding screw fastened to the zinc can is the negative ( ) terminal. The manganese dioxide is a depolarizer. It gives up oxygen, FIG. 68. Dry Cell of an Electric Battery. 82 ELECTRIC IGNITION which combines with the hydrogen gas that is liberated by the action described in connection with the elementary carbon-zinc cell. The hydrogen and oxygen combine in the proportion to form water, which is a liquid and remains in the cell. The electromotive force of a dry cell of the kind just described is about 1.5 volts on open circuit when new and in good condition, irrespective of the size of the cell. A large cell will give more current than a small one. The size of dry cell which has become standard is 2\ by 6 inches long. It is cylindrical in form. The cell is usually covered with some insulating material, such as paper or straw- board, except the terminals. This insulating covering prevents the metal of one cell from coming into contact with that of another when the cells are grouped together to form a battery. A cell of this size will give from 15 to 20 amperes of current through a low-resistance ammeter when the circuit is first closed. The cell is practically short-circuited when its terminals are connected through a low-resistance ammeter, the resistance of the latter being about the same as that of a short, heavy copper wire. The cell will deliver this maximum amount of current during only a few seconds. The current rapidly decreases when the cell is short-circuited, but continues with constantly de- creasing value till the battery is exhausted. The exhaustion of a dry cell of the usual construction is due to weakening of the liquid electrolyte with which the absorbent paper in the cell is saturated. This weakening is on account of the chemical action necessary to produce electric current. 68. Deterioration of new permanently sealed dry batteries often occurs to a marked extent before they are put into use. They sometimes deteriorate in a few months or less of storage so as to become useless. Such rapid deterioration is not apt to occur in well-made cells whose materials are suitably pure. 69. New Type of Carbon-zinc Dry Cell, not Sealed. A type of dry cell which is actually and thoroughly dry until put into use was first exhibited at the Atlanta automobile show in the latter part of the year 1909. The cell resembles in general PRIMARY BATTERIES 83 appearance and construction the one shown in Fig. 68, except the carbon rod and the terminal on it. The carbon rod of the new cell is made hollow and is provided with a wooden stopper at the open end. The terminal is fastened to one side of the top of the carbon. When the cell is manu- factured it is left entirely dry, but all of the necessary chemical elements are put in it. It is chemically inactive and does not deteriorate in storage before putting into use. To prepare the cell for use, it is only required to fill the hollow carbon rod with water, after removing the stopper, which is replaced after the water is poured in. The water dissolves the chemical elements which with the water form the electrolyte. After being thus put into operation, the cell is subject to ex- haustion and deterioration the same as a permanently sealed cell of the same quality. 70. Exhaustion and Running Down of Dry Batteries in Ser- vice. The chemical action in the cell by which electric current is generated of course consumes the active materials and pro- duces new chemicals. The result is a dropping off of the activity of the cell and finally its exhaustion to such an extent that it becomes useless. In a properly constructed cell, the chemical elements are so proportioned that they become exhausted at about the same time. The zinc cup is not much thicker than necessary to furnish the requisite metal for the amount of chemicals present. The fact that the zinc is sometimes eaten through before the cell is nearly exhausted is an indication of local action in the zinc, probably on account of impurities in the metal. Local action is more apt to make itself known when the battery is allowed to stand idle during a considerable portion of its life. 71. Recuperation of dry cells can generally be effected by adding sal-ammoniac solution to the inside of the cell. The solution can be added to a permanently sealed cell by making an opening through the sealing compound at the top so as to expose the blotting paper, and pouring the liquid into the open- ing. The sealing compound can be readily dug out with a pointed instrument. It is not necessary to replace it, since the 84 ELECTRIC IGNITION cell will never be of much use after becoming exhausted the first time. This expedient of recuperation is hardly worth while except in case of emergency. It is probable that the new type of unsealed cell described above can be recuperated by pouring in a solution of sal ammoniac after the cell has been run down in service. The addition of water alone will sometimes recuperate a cell slightly. Copper Oxide and Zinc Wet Cell. 72. The Lalande and Chaperon wet cell, as brought out in 1881, has zinc amalgamated with mercury for one electrode,. and either iron or copper for the other. The electrolyte is a solution of either caustic soda (concentrated lye, sodium hydrate, NaOH) or caustic potash (potassium hy- drate, KOH). A depolarizer of copper oxide (cupric oxide, CuO) is used. Modi- fied forms of this cell, one known as the Edison-Lalande and more recently as the Edison primary battery, and the latest one , v . as the BSCO battery, are used to a consid- FIG. 69. (See also Figs. . ... 70 and 71.) erable extent for gas-engine ignition, espe- Copper Oxide and Zinc cially for stationary engines. Wet Cell. Edison Man- 73. A BSCO wet cell is shown in Fig. ufacturing Company, , Part of ^ cont aming vessel is broken Orange, New Jersey. . away to show the interior construction. The electrodes, depolarizer, and the cover of the cell are shown removed from the cell in Figs. 70 and 71. The copper electrode has the form of an inverted U-shaped, grooved, or channeled, frame A which hangs from the cover F of the cell, the connection being made by a bolt which passes through the cover and has thumbscrews above it. The depolar- izer slab C is clamped between the legs of the copper frame, which are drawn together against the beveled edges of the slab by means of a copper cross-bar, or bridge, D, so as to make good electric contact between the copper and the depolarizer slab. PRIMARY BATTERIES The lower ends of the copper frame are bent in under the slab to support it. The zinc plates B are suspended from the cross- bar by means of a steel bolt which passes through a porcelain insulator E and holds the zincs firmly in place. The porcelain insulates the zincs from the copper. An insulated wire is con- nected to the zincs, and its outer free end is the negative ( ) FIGS. 70 and 71. Elements and Top of Fig. 69. terminal of the cell. The bolt from which the copper frame is suspended is the positive (+) terminal. The depolarizer slab is a mixture of copper oxide and magne- sium chloride compressed to form in molds and then heated to make it a firm mass. The magnesium chloride acts only as a binder to hold the mass together. The zinc plates are amalgamated by incorporating about two per cent of mercury with them when they are cast. The liquid electrolyte covers the zinc plates completely to a depth of an inch or so above them. A layer of heavy mineral oil is poured on the electrolyte and floats at the top to protect the electrolyte from atmospheric action. If air is allowed to come into contact with the electrolyte, it oxidizes it and de- creases the length of life of the cell. 86 ELECTRIC IGNITION Since the rigid parts of the cell are firmly fastened together, the cell can be used where there is motion without danger of the electrodes coming into contact with each other or with other parts so as to short-circuit the cell. It can therefore be used on vehicles if the top is sealed on, for which provision is made in one type of the cell intended for ignition use. Cells for portable use are made with enameled steel containing vessels, or jars. The voltage of one of these cells on open circuit is slightly less than one volt. The pressure does not drop much below one volt during use until the battery is nearly exhausted. The renewals for an exhausted battery consist of the parts suspended from the cover, which are sent out as a unit fastened together, and the electrolyte of dry caustic potash to be dissolved in water. The oil for covering the electrolyte may also be in- cluded as one of the renewal items. Renewal of the parts is made by taking out the bolt which passes through the cover, discard- ing the copper frame and parts attached to it, and fastening the new frame and its attached parts in place with the old cover- bolt; also discarding the exhausted electro- lyte and dissolving the new in water poured into the jar. The capacity of the portable cell intended for ignition use is 200 ampere-hours. 74. The Edison primary battery, already referred to as an earlier form of the one FlG - 72. just described, differs from the newer form Early form of Edison only in mec hanical construction. One of Primary Battery shown . ... ., . . . in Fig. 69. these earlier cells is shown in Fig. 72, with part of the jar broken away to show the interior. The zinc plates are suspended from the porcelain cover, instead of from the copper frame, as in the later form, and the copper frame is of different form, with a bolt connecting the two sides under the depolarizer slab. The zincs are not so firmly and accurately held in place as in the later type of cell. The cell is therefore not so well adapted to portable use as the latter type. PRIMARY BATTERIES The capacities and dimensions of some of these cells are given in the following table according to the Manufacturer's rating: Diameter and Height over All. Com- plete Cell with Capacity in Ampere-Hours. Porcelain Jar. Enameled Steel Jar. Inches. 4^X 7l SIX 8f 7iXio| Inches. 4*X 6f 5lX 8 7iXio IOO 150 300 CHAPTER IX. BATTERY CONNECTIONS. 75. General. In order to obtain suitable electromotive force and current, cells are connected together to form a battery. The best arrangement of the cell with regard to the order in which their terminals are connected together depends on the nature of the service to be performed and on the resistance of the external circuit. In the following discussion it is assumed that all of the cells in a battery are alike. This assumption is in accordance with the best practice. Moreover, the discussion relative to different kinds and capacities of cells grouped together in a battery is more complicated than is thought should be presented in a work of this nature. For convenience of discussion, it will be assumed that either carbon-zinc cells or copper-zinc cells are used. The carbon ter- minal, or the copper terminal, as the case may be, is the positive one, and the zinc terminal is the negative one. Series-connected Batteries. 76. A series-connected battery of four cells, all alike, is shown in Fig. 73. -The carbon terminal of each cell is connected to the Negative Terminal- ^.3^ of Battery. Zinc. (jg) \^6^ \ ^4^ \ ^-P \-PositiveTerminal of Battery. Carbon or Copper FIG. 73. Electric Battery of Series-connected Cells. zinc terminal of another cell, except that the carbon terminal of the right-hand cell and the zinc terminal of the left-hand cell 88 BATTERY CONNECTIONS 8 9 are left free. These two free terminals are the terminals of the battery. The voltage of a series-connected battery, measured between its terminals, is equal to the sum of the voltages of all of the cells. In this case the battery voltage is four times that of one cell, since there are four cells, all assumed to be alike. If the electromotive force of one cell is 1.5 volts on open circuit, then the electromotive force of the battery of four cells is 4 X 1.5 = 6 volts. If the electromotive force of each cell drops to 1.25 volts when the battery is delivering current in regular service, then the working voltage of the battery is 4X1. 25 = 5 volts. FIG. 74. Reversed Cell in a Battery Intended to be Series-connected. The current that the battery will give is greater than a single cell will give. The increase of current obtainable by connecting the cells together is not so great in proportion as the increase in the number of cells in the battery. With the four cells, it is not possible to get four times as much current as from one cell. The current will be very nearly four times as great with the four cells, however, if the resistance of the external circuit is very great in comparison with the internal resistance of the battery. The internal resistance of the series-connected battery is the sum of the internal resistances of all of the cells. If the external resistance is very low compared with the internal resistance of the battery, the series-connected battery will give only very little more current than one cell alone will give. 77. Reversed Cell in a Series Battery. If one of the cells is 90 ELECTRIC IGNITION wrongly connected to its neighbors in a series battery, so that its carbon is connected to the carbon of one adjacent cell, and its zinc to the zinc of the other adjacent cell, the effect on the voltage of the battery is equivalent to removing two cells from the battery. It requires the electromotive force of one of the properly connected cells to counteract that of the reversed cell. The current that the battery will give is less than that of a properly connected series battery with two less cells. Fig. 74 is a six-cell battery intended to be series-connected, but one cell A is reversed. The battery is therefore somewhat less effective and efficient than a properly connected four-cell series battery. While this error of making connections appears plain on paper, it is one that frequently occurs and is not so easy to notice in practice. Multiple- or Parallel-connected Batteries. 78. A parallel-connected battery of four cells whose positive (carbon) terminals are all connected together by one wire, or other conductor, and whose negative (zinc) terminals are all connected together by another conductor, is shown in Fig. 75. FIG. 75- Parallel or Multiple Connection of Cells in a Battery. Any place on the wire connected to the carbons can be taken as the positive terminal, and any place on the wire connected to the zincs as the negative terminal. The voltage of the battery is the same as that of one cell. When connected to an external circuit of very high resistance, the battery will deliver only slightly more than the amount of current in amperes that one cell will deliver to the same circuit. But when the positive and negative wires of the parallel-con- nected battery are connected together by a conductor of very BATTERY CONNECTIONS low resistance, such as a thick, short copper rod, the four cells will give nearly four times as much current as one cell with its terminals similarly connected together. In general, the current is but slightly increased by putting cells in parallel if the resistance of the external circuit is high, but if the external resistance is low compared with that of the battery the current is materially increased. 79. Reversed Cell in a Parallel-connected Battery. In Fig. 76 three cells, i, 2, and 3, are connected together. The zinc of cell i is connected to the carbons of cells 2 and 3, and the carbon of cell i is connected to the zincs of 2 and 3. When the cells are connected together in this manner, current flows from the car- bon of cell i to the zincs of 2 and 3, and from the carbons of cells 2 and 3 to the zinc. of i. The resistances of the circuits are low, being only that of the connecting wires and of the cells. The internal re- sistance in circuit is one and one-half times that of one cell when they are connected in this manner. Unless the connecting wires are very unusually thin and long, FIG. 76. Wrong Connection of Cells in a Battery. FIG. 77- Reversed Cell in a Battery Intended to be Parallel-connected. the total resistance of the circuit, internal plus external, is not more than that of two cells added together. The result is that a large amount of current flows and the cells become exhausted in a short time. 92 ELECTRIC IGNITION Fig. 77 shows nine cells, eight of which are connected in parallel, but the remaining one A is reversed from the position proper for parallel connection with the others. The result is of the same nature as that just stated for three cells. Current flows as indicated by the arrowheads on the wires. The current is not of the same amount in all parts of the wires, however. It is greater in the wires which are between the cells near A than in those between the cells more remote from A. All of the cells will be rapidly exhausted. Parallel-series Batteries. 80. Fig. 78 shows two sets of series-connected cells with four cells in each set. One set is made up of cells i, 2, 3, and 4; the FIG. 78. Parallel-series Battery Connection of Two Sets of Series-connected Cells. other set is made up of cells 5, 6, 7, and 8. The two sets are connected together in multiple, or parallel. Any point on the wire connecting the two carbons can be taken as the positive (+) terminal of the battery, and any point on the wire connect- ing the two zincs can be taken as the negative ( ) terminal. The voltage of the parallel-series battery is the same as that of each series of cells. In this case the voltage is four times that of one cell, since there are four cells in each series. More current will be sent through the external circuit by the two sets of series-connected cells than by one series alone. The increase of current will be greater when the external resistance is low than when it is high. It is sometimes convenient to con- sider each series as a unit whose terminals are those of the series. BATTERY CONNECTIONS 93 When this is done the series can be dealt with as a single cell so far as regards its relations to other units of a similar nature. In Fig. 79 five sets, each of four series-connected cells, are connected in parallel, or multiple, with each other. The cells FIG. 79. Parallel Connection of Five Sets of Series-connected Cells. The of each series are in a vertical row in the illustration, voltage of the battery is four times that of one cell. 81. Wrong Arrangement of a Battery. In Fig. 80 five series- connected cells are shown in the upper row, and four series- FIG. 80. Wrong Arrangement of a Battery. connected cells in the lower row. The two series are connected in parallel with each other. The result is that current flows through the battery while the external circuit is open, on account of the electromotive force of the upper row of five cells 94 ELECTRIC IGNITION being greater than that of the lower row of four cells. The direction of the current is indicated by the arrowheads on the wires. The current will continue until the electromotive force of the five cells in series drops to that of the four series-connected cells. This action means exhaustion of the five cells. A battery should not be made up in this manner. 82. Connection to External Circuit. The method of con- necting two series batteries to the same external circuit is shown in Fig. 81. The negative terminal of each series of cells is connected to one of the contact-points, or poles, of a two-point switch. The contact-points are insulated from each other and from other parts of the system. The positive terminal of each FIG. 81. Correct Method of Connecting a Battery to the External Circuit. set of series-connected cells is connected to the external circuit R, from which connection is made to the switch-blade. When the switch-blade is in the position shown, the upper row of cells is the only one that delivers current. The other row of cells is on open circuit. If the blade is moved into contact with the lower switch-point, as indicated by the dotted lines, then the lower row of cells is brought into operation alone. By placing the switch-blade in mid-position, so that it has contact with both switch-points, the complete battery is cut into circuit and it all acts to furnish current to the external circuit. By throw- ing the blade over so that it has no contact with either point, the current is completely cut off from the external circuit and there can be no local current in the battery of the nature of that in Fig. 80, because there is no connection between the negative BATTERY CONNECTIONS 95 terminals when the switch-blade has no contact with either point of the switch. It may be noted that, even when two series, both of the same number of cells, are in parallel with each other, and one series is run down or exhausted, while the other is in good condition, there will be local current in the battery of the same nature as that in Fig. 80. 83. Screw-top Battery Cells. An exceedingly convenient way of connecting cells together in a battery is by means of a FIG. 82. Screw-top Cells and Battery Box. Stanley & Patterson, 23 Murray Street and 27 Warren Street, New York City. screw top on each cell and a plate provided with suitable contact pieces and connections. Such cells and a plate are shown in Fig. 82. It is only necessary to screw the cells into the plate in order to make the proper connections. The making of wrong connections, which is not an unusual happening with the ordinary cells, each having two terminal nuts, is thus entirely eliminated. The cell is connected into the battery in a manner similar to that in which an incandescent electric lamp is connected into the circuit of the service wires. The carbon terminal of the battery cell makes contact with a spring, and the zinc shell makes contact with the threaded metallic ring into which it screws. 9 6 ELECTRIC IGNITION \ The spring keeps the contacts pressed together and prevents their jarring loose. The cells are also made with combination screw-top and bind- ing-post terminals. By means of the latter the cells can be connected together with wires when desired. By the use of " emergency spring clips " the ordinary type of cell with two binding posts can be used in the 4L A m cap or plate. Fig. 83 shows five dry cells and a spark-coil (spark-coils are described later) , each screwed into its receptacle in the top-plate of the battery box. The spark-coil occupies practically the same amount of space as one of the cells. Battery boxes with this form of connection are also made up with two sets of cells and a switch which, when moved to its different positions, connects the two sets of cells either in parallel or in series, or connects either of the sets into the system, leaving the other idle. The boxes are made water-tight for marine use or for any place where there is much water or moisture. m FIG. 83. (See also Fig. 82.) Screw-top Spark-Coil and Cells in Battery Box. CHAPTER X, STORAGE BATTERIES, ALSO CALLED ACCUMULATORS AND SECONDARY BATTERIES. 84. A storage battery is one that must be charged by passing an electric current from some exterior source through it in order to bring it into a condition in which it can deliver an electric current. Before charging by passing an electric current through it, the storage battery is inert and cannot deliver current. After it has been once charged and then discharged by delivering current, it can be recharged and will again furnish current till it is discharged a second time. Charging and discharging can be repeated numerous times. The charging current while passing through the battery effects chemical changes in the elements of the battery. These chemical changes take place in both of the electrodes and in the electro- lyte. During discharge the reverse chemical actions occur and bring the battery elements at least partly back to their condi- tions before charging. The name " storage battery " is a misnomer, strictly speaking. Electricity is not actually stored in the battery. The charging current passes through and out of the battery, effecting chemical changes in the elements of the battery during its passage. The distinction between this action and the actual storage of elec- tricity will appear later in connection with electric condensers. While a great variety of storage batteries have been constructed and tried out, those in commercial use are limited almost exclu- sively to two kinds as classified by the materials used in them. Of these two types, the one which is used by far the more gener- ally is known as a " lead accumulator " or " lead storage bat- tery," on account of its electrodes being made of the metal lead and oxides of lead. The other type is known as the " Edison 97 9 8 ELECTRIC IGNITION storage battery." Its electrodes are composed chiefly of nickel and iron. 85. The electrodes, or plates, of a lead storage cell are usually made up of plate-shaped pieces of lead, or lead alloy, which have perforations, pockets, or other forms of receptacles filled with the electrically active material. This active material is usually put into the pockets in the form of paste during the construction of the cell. It is composed chiefly, when first put in, of an FIG. 84. Electrode, Plate, or Grid of a Lead Storage Battery. oxide, or oxides, of lead. The plates are then treated chemically and electrically so as to cause the paste to set firm and hard, and to change its chemical composition. The plate, or plates, to form the electrode to which the positive terminal of the cell is connected are treated so that the final condition of the paste is dioxide of lead (PbC^), and the plate has a brown color, including the paste. The plate, or plates, to form the other electrode are treated so as to remove the oxygen from the compound and leave metallic lead in a porous, or spongy, condition with the charac- teristic color of metallic lead. STORAGE OR SECONDARY BATTERIES 99 One form of lead grid for a storage battery is shown in Fig. 84. It has rectangular perforations for receiving and holding the paste. The lug projecting upward is for making connection to other plates or to the external circuit. 86. Complete Storage Cell. Fig. 85 shows several complete plates grouped together to form the electrodes of one cell. Three FIG. 85. Group of Plates for a Lead Storage Cell. of the plates are fastened together by a bar, or strap, which carries one of the terminals of the cell at the top. The other two plates are fastened together in a similar manner and have the other terminal of the cell at the top. The pockets of the plates, or grids, are shown filled with paste. Separators are used between the plates to keep them from coming into contact with each other. One of the separators IOO ELECTRIC IGNITION FIG. 86. Separator for Placing between the Plates of a Storage Cell. FIG. 87. Storage Cell Containing the Parts shown in Figs. 84, 85, 86 and 88. Dayton Electrical Manufacturing Company, Dayton, Ohio. STORAGE OR SECONDARY BATTERIES IOI used with these particular plates is shown in Fig. 86. It is made of hard rubber in one piece and has the general form of a number of bars laid across each other at right angles. The complete cell is made up of the plates and separators immersed in an electrolyte of dilute acid solution contained in a jar. The electrolyte should completely cover the main portion of the plates, but the terminals and upper portion of the lugs extend above the liquid. The electrolyte is generally dilute sulphuric acid. The water used with the acid should be very pure, as distilled water. A complete storage cell having plates and separators like those just described is shown in Fig. 87. The jar is of hard rubber suitable for portable work. It is provided with a cover which is sealed on to make the cell tight, except a small opening, or vent, which is shown between the terminals. The cover of the cell is shown in section in Fig. 88. The vent is so made that the bubbles of gas, escaping during the charging FIG. 88. , Cover of Cell shown in Fig. 87. of the cell and carrying the .liquid electrolyte up with them, can escape through the upper part of the vent-plug without carrying the liquid outside of the cell. The conical terminal of one set of plates is shown in the opening through the left-hand end of the cover. This cone is surrounded by a rubber washer and is drawn up tight so as to make a liquid-proof joint. 87. The voltage of a storage cell having lead plates and dilute sulphuric acid electrolyte is about 2.2 volts on open circuit. It drops below the open-circuit value as soon as the external cir- cuit is closed and current flows. The pressure drop is approxi- 102 ELECTRIC IGNITION mately proportional to the amount of current flowing, for a given cell, when the cell is in good condition. The pressure decreases very slowly during discharge until the battery is almost completely discharged, and then begins to drop very rapidly as the discharge continues. The voltage is independent of the number of plates in the cell. It is the same when there is only one positive plate and one negative plate, as when there are several of each. In one cell the plates are generally all immersed in the same lot of electro- lyte contained in one jar with no partitions to separate one portion of the liquid from another. Exceptions to the last state- ment are some unusual types of cells with porous jars. 88. The maximum rate of discharge of a storage cell, without injury to the cell, is approximately proportional to the amount of surface area of active material of the electrodes in contact with the liquid electrolyte. If the same size and make of plates are used to make up cells having different numbers of plates in them, then the safe maximum amperage of the cells will be approximately proportional to the number of plates in them. The maximum safe rate of discharge of a cell having 15 plates is about three times as great as that of one having only 5 plates, the plates in both being of the same size and make, as already stated. The rate of discharge is measured in amperes. An excessive rate of discharge is injurious to the cell. It causes the paste to swell and even to drop from the grids. 89. A storage battery is composed of storage cells connected together. Any of the methods of connection that have been given for primary cells can be used for storage cells. The storage batteries used for ignition purposes are generally made up of two or three cells connected together and inclosed in a case. A battery of this nature does not differ much, in appear- ance from a single cell. The plates of three storage cells are shown connected together in series in Fig. 89. The two positive plates of the right-hand set are Connected by a lead strap to the three negative plates of the middle set. The two positive plates of the middle set are similarly connected to the three negative plates of the left- STORAGE OR SECONDARY BATTERIES 103 FIG. 89. Connected Plates of a Three-cell Storage Battery. FIG. 90. Three-cell Storage Battery with Cover Removed. 104 ELECTRIC IGNITION hand set. The two threaded bolts at the opposite corners of the entire group are the terminals of the battery. In Fig. 90 a case containing the three sets of plates and their corresponding jars is shown before the sealing compound is put on the top. The complete battery is shown in Fig. 91. The FIG. 91. Complete Three-cell Storage Battery containing Plates shown in Fig. 89. knurled nuts on the terminals are shown just above the marks "P+" and " N-." The voltage of the three cells connected in series is three times that of one cell alone. While the three cells will send more amperes of current through a given external resistance than one cell will, the maximum allowable amperage is the same for the three cells as for one cell alone. 90. " Exide " Storage Battery. Fig. 92 shows another form of battery that used lead plates and dilute sulphuric acid electro- lyte. It is commercially known as the " Exide " battery. One side of both the case and the jar is removed in the illustration, STORAGE OR SECONDARY BATTERIES 105 H Exide Storage Battery. A. Terminals of battery. B. Inverted petticoat. C. Pillar post. D. Plate strap. E. Lug on plate. F. Positive plate. G. Acid-resisting paint. H. Handle. I. Vent plugs. J. Plastic asphaltum. K. Beveled edge at top of wood case. FIG. 92. Electric Storage Battery Company, Philadelphia, Pa. L. Connector. M. Hard rubber cover, sealed in with asphaltum J. N. Apron. Part of plate strap. O. Negative plate. P. Wooden separator. Q. Acid-resisting compound. R. Hard rubber cell or jar. S. Hard rubber ribs. T. Expansion joint. 106 ELECTRIC IGNITION and some of the interior members partly broken away to show the construction. 'The separator between adjacent positive and negative plates is made of wood chemically treated before using. The chemical treatment of the wood is to remove any substance that might be deleterious to the cell. There are deep vertical grooves in the separator on the side that goes next to the positive plate. These grooves are to allow free circulation of the electrolyte and escape of gas while the cell is being charged. In some forms of this battery a thin sheet of hard rubber with numerous perforations is placed between the positive plate and the grooved side of the separator. The plates rest on high rubber ribs at the bottom of the jar, so that there is ample space left for the sediment which collects at the bottom of the jar. This is important, since the battery is short-circuited internally if the sediment rises high enough to touch the electrodes. An apron N on each strap which connects the plates prevents the wooden separators from rising on account of the buoyant action of the liquid. Each vent plug / is a hollow cone with a hole near the top to allow the escape of gas from the cell while it is being charged, and a drainage hole at the bottom through the apex of the cone to let the electrolyte flow back into the cell, in case any of the liquid is carried up with the escaping gas. The binding posts are formed so as to prevent the acid electro- lyte from creeping up and spreading over the top of the cell. This prevention is accomplished by forming the lead alloy into the shape of an " inverted petticoat " which is below the binding screw far enough to be covered with the sealing compound of " plastic asphaltum " that covers the top of the battery except the terminals and vents. The edges of the wood case are beveled at the top so that the sealing compound covers them and thus prevents the acid from soaking down into the wood if any of the electrolyte is spilled over the top. The acid is injurious to the wood. Each grid is cast in one piece and has the form of numerous small horizontal bars held in place by several thin vertical strips. A part section of the grid, made by cutting the plate in two STORAGE OR SECONDARY BATTERIES 107 between two of the vertical strips, is shown in Fig. 93, in which A is a side view of part of one of the vertical strips. The small horizontal bars are shown in cross-section at A, B, C, D, E, and F. There are of course a great many more horizontal bars in the entire grid than shown in this section. The exposed sur- face of each horizontal bar appears as a line, as shown in the preceding figure. The object of making the grid in this form is FIG. 93. Section of Grid for Battery shown in Fig. 92. FIG. 94. Phantom View of Exide Storage Battery. to expose as great a surface of the paste to the electrolyte as possible, and at the same time provide a light-weight grid which holds the paste securely in place. A phantom view of an Exide battery intended for ignition usage is shown in Fig. 94. This battery is of a slightly earlier form than that shown in Fig. 92, but is the same in a general way, lacking only some of the improvements in detail that appear in the latter form. The battery, Fig. 94, is made up of three cells connected in series. Each of the three cells has three positive plates and four negative ones. Each cell is provided with its own vent plug. 91. Charging the Storage Battery. A storage battery of the ignition type is generally charged and ready for use when sent 108 ELECTRIC IGNITION out from the factory. After it has been discharged it can be charged again by connecting to some exterior direct-current source of electric supply and sending current through it in the reverse direction from that in which it discharges. The positive side of the source of supply must be connected to the positive terminal of the storage battery, and the negative of the supply to the negative of the battery. An alternating current cannot be used directly for charging a storage battery, but it can be rectified by suitable apparatus for transforming it into a direct current which can be sent through the battery to charge it. In charging the battery, as in discharging it, the amount of current must be kept within the maximum safe amperage of the battery. This is ordinarily accomplished by the use of suitable regulating apparatus inserted in the supply circuit. A rheostat is generally used for regulating the amount of current. Gas is formed in the battery while charging it, slowly at first, and then more rapidly. The formation of gas is especially rapid when the battery has become almost completely charged. (See also Chapter XXVI.) 92. Chemical Action in a Lead Storage Battery. When a storage cell is in a fully charged condition and ready for use, the active material in the plates connected to the positive ter- minal is in the chemical form of dioxide of lead (PbO 2 ) and has a brown color. In the negative plates the active material is in the form of porous, or spongy, metallic lead and has a gray color. Although this has been stated before, it is repeated here to bring it fresh in mind. During the discharge of the cell, the dioxide of lead in the posi- tive plate is partly changed to monoxide of lead (PbO) by the loss of part of its oxygen. The metallic lead in the negative plate is partly changed into monoxide of lead (PbO) also, by the addition of oxygen to it. The amount of sulphuric acid in the electrolyte is reduced by decomposition into sulphur and water, so that the electrolyte becomes weaker and has a lower specific gravity. During the charging of the cell the above reactions are reversed and the elements of the cell are restored, more or less completely, to their first-mentioned condition of the charged cell. STORAGE OR SECONDARY BATTERIES 109 93. The capacity of a storage cell is measured in ampere-hours. An ampere of current flowing for one hour is an ampere-hour. So is half an ampere flowing for two hours, or four amperes flow- ing for a quarter of an hour, etc. Four amperes flowing for one hour are four ampere-hours, and the same amount, four amperes, flowing for two hours are eight ampere-hours. In general, the current in amperes, multiplied by the number of hours during which it flows, equals the number of ampere-hours. Amperes of current X hours of time = ampere-hours. In order to fully specify a storage battery, its voltage, or the number of cells in it, must be stated in connection with its capac- ity in ampere-hours. The following table refers to lead storage batteries for ignition use, as made by one manufacturer. SIZE AND CAPACITY OF IGNITION STORAGE BATTERIES. All of these batteries are 9 inches high and 6f inches wide over all. Number of Cells in Battery. Volts Pressure. (Approximate. ) Ampere-hours Capacity at Ser- vice Rate. Length over All. Weight. Inches. Pounds. I 2 "I f 3 i 8| 2 4 i J 5ii I 7 3 6 7^ 25! 4 8 I 9& 34 i 2 i r 4tt 12 2 4 60 J 7& 24 3 6 f 1 9rl 351 4 8 J I W* 47^ i 2 1 r 5if 151 2 3 4 6 80 lift 3t 46 4 8 J I ?sA 61 i 2 1 r 6^| 19 2 3 4 6 100 i4fl 37^ 55^ 4 8 J I I9T* 74 CHAPTER XI. FLOATING THE STORAGE BATTERY ON THE LINE OF A DIRECT-CURRENT GENERATOR. 94. A storage battery can be kept in continuous service by the method of operating known as " floating the battery on the line." A direct-current generator which will give a voltage somewhat higher than that of the battery is ordinarily used, but modified forms of the system use generators giving a pressure very much higher than that of the battery, sometimes several times that of the battery. It is assumed in the following discussion that the generator maintains a constant, or nearly constant, voltage slightly higher than that of the storage battery on open circuit when fully charged. One arrangement of the apparatus for the above method of operation is shown diagrammatically in Fig. 95, in which A FIG. 95. Direct-current Generator and a Battery Floated on the Line, with a Light Load. represents the commutator and brushes of a direct-current gen- erator which maintains a constant voltage, or nearly so; B is the storage battery, and C is any piece of electrical apparatus, such as an incandescent lamp, through which current is sent. The current flows from the positive (+) brush of the generator to the junction D, where it divides, part flowing through the no FLOATING THE STORAGE BATTERY III lamp C and the remainder through the storage battery B in the direction to charge it. These two currents come together again at the junction E and flow through the same wire to the negative ( ) brush of the generator A. The direction of the current is indicated by the arrowheads on the lines representing the cir- cuits. This action continues as long as the conditions remain unchanged. Now suppose that several additional lamps are added to the circuit, as shown in Fig. 96, which are the maximum number that the system is intended to operate. The battery now dis- ~l FIG. 96. Battery Floated on the Line, with Heavy Load. charges so as to furnish current to the lamps, thus aiding the generator which still supplies current, all of which flows through the lamps. The battery and generator now operate in conjunc- tion, both sending current through the lamps. The generator delivers more current when all of the lamps are in the circuit than when only one is in the circuit. If all of the six lamps are alike, they will take about six times as much current as any one of them alone. The greater number of lamps requiring more current than one, lowers the voltage at the lamps and also the difference of pressure at the junction points D and E. When the difference of pressure between D and E drops to a lower value than the voltage of the storage battery on open circuit, the battery begins to deliver current instead of receiving it, as in the case where only one lamp was in circuit. The lowering of the pressure between D and , due to increasing the number of lamps, causes the genera- tor to deliver more current. 112 ELECTRIC IGNITION Another method of arranging the apparatus is shown in Fig. 97 for one lamp, and in Fig. 98 for several lamps. The operation of this system is in a general way the same as that of the preced- ing two figures. The direction of current is indicated by the arrowheads on the circuits. If the circuit is broken between the generator and other apparatus, the generator will of course become inoperative so FIG. 97. Modified Form of Fig. 95. FIG. 98. Modification of Fig. 96. far as the other parts of the system are concerned. The battery will then furnish all of the current required for the lights. This within the limits of the battery. Still another method of arrangement is shown in Fig. 99. The dynamo A is placed between the storage battery B and the load C. When the load is small, as represented by one lamp C, and the generator is at its proper voltage, it sends current through both the storage battery to charge it and through the lamp. The current from the generator divides at D, part going to the lamp and part to the positive side of the storage battery, as indicated FLOATING THE STORAGE BATTERY 113 by the arrowheads. These divided currents unite again at E and flow together to the negative brush of the generator. If the full load is put on, as represented by several lamps in Fig. 100, then the storage battery discharges into the circuit, thus aiding the generator. Both send current through the lamps. The direction of the current is indicated by the arrowheads. The two currents unite at D and flow together through the lamps. FIG. 99. Electric Generator between the Load and the Battery which is Floated on the Line. Light Load. m I r4 .7*N -PF* c FIG. 100. Electric Generator between Heavy Load and a Battery, which is Floated on the Line. They then separate at E, part going to the negative brush of the generator and the remainder to the negative side of the battery. 95. An automatic cut-out is used on some storage-battery and generator systems in which the storage battery is floated on the line. This has been mentioned in connections with Figs. 63 and 64. The purpose of this cut-out is to prevent the flow of a large reverse current from the battery through the generator in case the latter slows down so as not to give a sufficiently high voltage, or in case of its complete stoppage. It is often desirable, ELECTRIC IGNITION and especially so for ignition service where the motor runs inter- mittently, to have the cut-out also operate automatically to put the generator into circuit when its speed is again sufficient to give the necessary voltage. An electric system with an automatic cut-out of the last- mentioned type is shown diagrammatically in Fig. 101. The cut-out consists of an electromagnet with a double winding and an armature with a contact-point at one end. The cut-out armature as shown consists of a blade spring to which is fastened a contact point and a soft steel disk, the latter opposite the end of the magnet core. One end of the blade spring is fastened to a DTG7Generator> la C [ ' D At C Core--: Shunt ^ Coil Seriesx^ Coil ^ itomatic ut-Out 11 Storage Battery 9~ ~* ^T^ k ' F _^ f b E i ^kcot ^-kfJ tArmatuie Po FIG. 101. Complete Dynamo and Storage-battery System with an Automatic Cut-Out. stationary block G in such a way that the elastic action of the spring tends to draw the armature away from the core and sepa- rate the contact-points. When the core is magnetized it attracts the cut-out armature and holds it in the position shown with its contact-point pressed against the mating contact-point to which is connected one end of one of the magnetizing coils of the cut-out. If the magnetism of the cut-out core becomes weak, the armature then springs away from it so as to separate the contact-points. One coil of the cut-out is permanently connected in series with the field-magnet coils of the shunt-wound generator. This coil, marked " shunt-coil " in the figure, has a comparatively great number of turns of insulated wire large enough to continuously carry the current that flows through the field-coils of the genera- tor. The other coil of the cut-out, marked " series-coil," is in series with the main circuit of the generator. This coil has com- FLOATING THE STORAGE BATTERY 115 paratively few turns of insulated wire large enough to carry all of the current from the generator. When the system is operating in the regular manner, the cur- rent flows through the circuits in the direction indicated by the arrowheads on the lines representing the circuits. Since the direction of flow in the two lines leading from the storage battery to the points D and E may be first in one direction and then in the other, it is indicated by a pair of opposed arrowheads on each line. The currents in both coils of the cut-out flow in the same direction around the core, and both magnetize the core in the same direction so as to keep the contact-point of the cut-out armature drawn up against its mate. If the generator slows down so that its voltage drops below that of the battery, then the battery sends current back to the generator and through it and the cut-out coils. This back cur- rent flows from the positive (+) terminal of the battery to D and then to the positive (+) brush of the generator, where it divides, part flowing through the field-coils of the generator and the shunt-coil of the cut-out in the same direction as before to the junction F. The remainder of the back current flows through the armature of the generator to the negative ( ) brush and thence to F. From F all of the back current flows through the series-coil of the cut-out in the opposite direction from that in which it flowed before. The direction of the back current through the main circuit is indicated by the arrows alongside the circuit. The back current through the series-coil of the cut-out opposes the magnetizing action of the current in the shunt-coil and demag- netizes the core of the cut-out enough to allow the cut-out arma- ture to spring back so as to separate the contact-points. The current through the series-coil stops as soon as the circuit is opened by the separation of the contact-points. A weak current continues in the shunt-coil as long as the generator keeps running at slow speed. This cur rent xeases as soon as the generator stops running. There is then no current in any part of the system to the right of the points D and E. The battery keeps sending current continuously through the part of the system to the left of it. n6 ELECTRIC IGNITION If the generator is started again, the contact-points of the cut-out still remaining apart, it will at first send current through only its field-coils and the shunt-coil of the cut-out, including of course the generator armature and the connections of this circuit. When the voltage at the brushes of the generator be- comes somewhat higher than that of the battery as the speed increases, the current sent through the shunt-coil is great enough to magnetize the core sufficiently to draw the cut-out armature toward it and thus close the circuit at the contact-points. This establishes the normal condition of operation. The size of the cut-out as shown in the figure is much larger in proportion to the other apparatus than it is in the constructed apparatus. It is shown large in order to make its construction appear plainly. A compound-wound direct-current generator can also be used with a cut-out of this nature. In Fig. 102 a storage battery is floated on the line of a variable- speed direct-current dynamo. A volt-ammeter is included in VOLT AMMETER U=-11CONNECT(-1)AND(+4) WIRES THROUGH PROPER SWITCHES FOR IGNITION AND LIGHTING THE SAME. ^4JAS FROM A BATTERY APLCO 10* PATENTS DYNAMO WITH LOAD APLCO LIGHTING STORAGE BATTERY PENDING REGULATOR & CUT-OUT FIG. 102. APLCO Electric System with Battery Floated on the Line. Apple Electric Company, Dayton, Ohio. the system, for measuring the voltage of the battery and the amount of current flowing through the battery. The hand, or pointer, of the volt-ammeter is shown pointing to the zero of the scales on the dial. When the dynamo is send- FLOATING THE STORAGE BATTERY 117 ing current through the storage battery to charge it, the indicator hand moves to the left and points to the lower scale, which gives the reading of the amount of charging current. When the battery is discharging, the indicator hand moves to the right of the zero and indicates the current rate of discharge. To obtain the voltage of the battery, the push-button (" press-button ") must be pressed in. The indicator hand then points to the pressure on the upper scale. The current through the battery is indicated continuously except during the time the button is pressed in to obtain the reading of the voltage. The dynamo is provided with an automatic cut-out and a load regulator. The latter regulates the current delivered by the dynamo within a limited range. It does this by automatically varying the resistance in the field-coil circuit. The load regu- lator makes it possible to drive the armature of the dynamo at a rotative speed proportional to that of the crank-shaft of the motor, even when the speed of the crank-shaft is extremely variable, as in automobile and boat motors, and still keep the current from the generator approximately constant as long as the armature rotates fast enough to generate sufficient voltage. The automatic cut-out opens the dynamo circuit when the speed of the armature falls below the requisite amount. FIG. 103. Two-voltage System with Two Storage Batteries Floated on the Line. 96. A two-voltage system with two storage batteries floated on the line is shown in Fig. 103. Two storage batteries of the same voltage are connected in series between the positive and negative sides of the circuit in the same manner as the one Il8 ELECTRIC IGNITION storage battery in Figs. 99 and 100. This system gives two voltages, one double the other when the two storage batteries are of the same voltage, as stated. A two-point switch in the ignition circuit can be closed on either of the two contact-points by placing the pivoted arm in the corresponding position. The flow of current through the system depends on the amount of electrical resistance of the ignition apparatus relative to the voltage of the storage batteries. The resistance of the ignition apparatus is ordinarily low enough for the method of operation to be as follows: When the voltage of the dynamo brushes is higher than that of the storage batteries in series, as measured between the points T and U, and the switch is closed as shown, then while the cir- cuit is closed in the ignition apparatus, the current from the positive brush of the dynamo flows to O, then through the switch and ignition apparatus to S, thence through battery 2 and to the negative brush of the dynamo. At the same time current flows from the positive terminal T of battery i to 0, then through the switch and ignition apparatus to S, and thence to the negative terminal of battery i. While the circuit is broken in the igni- tion apparatus in the usual manner of operation, the dynamo sends all of its current to T and through both storage batteries in series so as to charge them. If the dynamo stops and is cut out of circuit, then battery i supplies the current to the ignition apparatus, and battery 2 is idle. When the switch is closed as shown, and the voltage at the dynamo brushes is higher than that of the two batteries in series, then battery 2 continuously receives charging current, and battery i is alternately discharged and charged in accordance with the closed and open positions of the ignition apparatus. When the switch is closed on its other contact-point, the bat- tery 2 is alternately discharged and charged, and battery i is continuously charged. 97. A two-voltage system with lamps and ignition apparatus is shown in Fig. 104. The voltage at the lamps is approximately twice that at the ignition-apparatus terminals. At the lamps FLOATING THE STORAGE BATTERY 119 the voltage is approximately equal to that of both batteries in series; at the ignition apparatus the voltage is approximately equal to that of one battery. If there were no loss of voltage Battery 1 1 +o Ignition Lar & ips & C ND.C. 'Generator T3 1 Apparatus F -1 / It ^ p2 1 1' -J Two-PointC Switch Battery 2 - FIG. 104. Two-voltage System with Lamps and Ignition Appliances. in the connecting wires, the voltage at the lamps would be the same as that of the two batteries in series; and that at the ignition apparatus would be equal to that of one storage battery. The manner of operation of this system is essentially the same as that of Fig. 103. 98. A switchboard for a two-voltage system, a direct-current dynamo, and two storage batteries are connected together in Fig. 105. The switchboard has an ammeter for indicating the amount of current, and a voltmeter for measuring the pressure of the system. There is also a pilot lamp which glows while its circuit is closed by pressing the push-button at the left-hand side of the board. The voltmeter is made to register by pressing the push-button at the right-hand side of the board. The opera- tion of the system is essentially the same as that of Figs. 103 and 104. The board has five switches, all of the blade, or knife, type, whose handles are shown at i, 2, 3, 4, and 5. When the double-pole double-throw switch i is closed on the dynamo side of the switchboard, as shown, the dynamo sends current through both storage batteries in series to charge them, and at the same time supplies current to the 12 -volt lamp circuits. Opening the switch breaks both the armature circuit of the 120 ELECTRIC IGNITION FIG. 105. (See also Figs. 106 and 107.) Switchboard Apparatus and Connections. dynamo and its field-circuit, and also breaks the battery circuit to the i2-volt lamps. When the switch is closed in its left-hand position, the battery discharges through the i2-volt circuits if the lamps are turned on. The connections to the ignition circuit are not affected by opening or reversing the main switch i. FLOATING THE STORAGE BATTERY 121 Q< Volt Meter ->Q_ ^ ^ 1 "? 6 Volts for Ignition Coils a>| r f ^_ 12 Volts-> Lights Indicates a Closed Switch FIG. 106. (See also Figs. 105 and 107.) Wiring Diagram showing Connections at the Back of the Switchboard. The single-pole double-throw switch 2 is for cutting the am- meter into or out of circuit. The ammeter is in circuit when this switch is in the position shown. Opening this switch breaks either the main circuit of the dynamo or the battery-discharge circuit, according to the position of the main switch i. The single-pole double-throw switch 3 is for reversing the 122 ELECTRIC IGNITION direction of the 6- volt current through the ignition apparatus. When the switch is in the right-hand position, as shown, the ignition apparatus takes current from battery 2 if the main switch is closed in the battery-discharge position. If the ignition switch 3 is closed in its left-hand position, then battery i supplies current to the ignition apparatus. The single-throw switches 4 and 5 are for opening and closing the lamp circuits. The wiring diagram of the switchboard is shown in Fig. 106, which also shows the automatic cut-out for protecting the dy- FIG. 107. (See also Figs. 105 and 106.) Dynamo, Storage Battery and Switchboard. Dayton Electrical Manufacturing Company, Dayton, Ohio. namo. The switch-blades are represented by broken lines. The double-throw switches are represented as closed in the reverse positions of the preceding figure. The diagram shows the igni- tion apparatus connected to the positive side of battery i. The ammeter is out of circuit, and both the main circuit and the field circuit of the dynamo are open. Fig. 107 is a photographic illustration of the dynamo, switch- board, and two 6-volt storage batteries such as are used in a system of the nature just described. It is suitable for ignition and lights on a small boat. CHAPTER XII. MECHANICALLY OPERATED MAKE-AND-BREAK IGNITERS AND KICK-COILS FOR LOW-TENSION IGNITION. 99. A mechanically operated igniter in the form of a low- tension ignition plug is shown in Fig. 108 together with the means of operating it. The illustration is elementary in its FIG. 108. Make-and-break Igniter or Spark-Plug. Elementary Form. nature. Part of the metal plug A is cut away to show the con- struction. The front end of the igniter remains outside of the cylinder of the motor when the igniter is in place, and the back end either projects into the combustion chamber or forms part of its wall. 123 124 ELECTRIC IGNITION A metal rod B extends through the plug and is enlarged at the inner end as shown at C. The hole through which this bar passes is considerably larger than the bar and is coned at the ends to fit correspondingly shaped insulators C and .D, which hold the rod in place and insulate it from the plug A. When the nut E is screwed down it draws the insulating cones C and D into the taper ends of the hole in the plug so as to make a gas- tight joint, and suitable packing around the rod next to the enlarged inner end and under the nut makes tight joints at these points. A knurled nut F at the outer end of the insulated electrode B affords a means to attach the wire or other form of electric con- ductor which brings the current to the plug. The screw G can be used for attaching the other wire when it is desired to bring both sides of the electric circuit to the igniter in this manner. The more general practice is to connect one side of the circuit to the metal of the motor at the most convenient point. The metallic contact between the plug and the motor metal gives electric connection also between them. A movable spindle H fits in another hole through the plug so as to have metallic contact with the plug, and at the same time be free to rotate, or rock, in the body of the plug. The inner end of the spindle has a metal arm / rigidly attached to it. This arm is sometimes called the movable electrode. Another rocker- arm J is fastened rigidly to the outer end of the spindle H. A blade-spring K is fastened to the rocker-arm J at the end next the spindle H and is of such a form that its free end stands away from the arm when the spring is not stressed. This completes the igniter proper. When the outer arm J is raised it rocks the inner arm / down so that it makes contact with the inner end C of the insulated electrode B and thus closes the electric circuit between the in- sulated electrode and body A of the plug. Then when the arm J is moved downward so as to separate the movable electrode from contact with the insulated electrode, an electric arc is drawn between the electrodes C and / at the point where the contact is broken. MAKE-AND-BREAK IGNITERS AND KICK-COILS 125 The means for operating the igniter consist, as shown, of a cam L and a cam-follower M. The latter is in the form of a push-rod with a collar N and an enlarged spherical upper end O. The push-rod passes freely through suitable openings in the ends of the arm / and spring K. The spring presses lightly down- ward against the collar when the parts are in the position shown. The guide P is for keeping the push -rod in position. The cam is driven by the shaft on which it is mounted. As the cam L rotates in the direction indicated by the arrow on it, the projecting lobe of the cam lifts the push-rod M and then allows it to drop when the edge of the lobe passes from under the push-rod. This occurs when the cam has passed through about three-quarters of a revolution from the position in which it is shown. The upper movement of the push-rod first lifts the rocker-arm / so as to bring the movable electrode I down against the station- ary electrode. The movement of the rocker-arm is stopped as soon as the movable electrode makes contact with the stationary one. The continued upward movement of the push-rod bends the spring K and the rod slips up through the arm so that the enlarged end rises above the arm. When the edge of the cam passes from under the push-rod, the reaction of the spring snaps the push-rod down quickly so that the knob on the upper end strikes the rocker-arm a sharp blow and drives its free end downward so as to cause a rapid separation of the electrodes. This is known as the hammer-break method of interrupting the electric current. The rapid separation of the contact-points of the igniter is very essential to the successful operation of the igniter. Com- pared with slow separation of the contact-points, the rapid separation produces a better arc for ignition and causes less fusing, or burning, of the contact-points. The insulation used in the igniter is generally either mica or steatite (soapstone). The mica should be pure and especially free from any metallic substance or metallic compounds. When steatite is used, it is generally first machined to form and then baked at a high temperature to bring it to the condition in which it is commonly used for the tips of gas burners. 126 ELECTRIC IGNITION The contact-pieces (points) of the igniter are generally made of either platinum, an alloy of platinum and iridium (platino- iridium), of steel alloy, especially a steel alloy containing a large proportion of nickel together with less amounts of other elements. While platinum and its alloys are excellent for the purpose, they are extremely expensive. Only small pieces are set into the electrodes, and are generally removable. 100. The duration of contact between the electrodes of a mechanically operated igniter should be as short as possible to establish current flow to the necessary amount when the source of electricity supply is a battery. Long duration of contact is wasteful of electricity and soon exhausts the battery. On the other extreme, when an alternating-current generator supplies the electricity in the usual manner, the electrodes may be kept in contact continuously except during the time necessary to separate them to form the arc and to close them again immedi- ately, so far as current supply is concerned. When electricity is supplied by a direct-current generator, it is generally advisable to have a short period of contact, since imperfect contact main- tained for some time may cause fusing of the electrodes. Other conditions, such as the use of several igniters in the different combustion chambers of a motor with several cylinders, may make a short period of contact necessary. It is generally undesirable to have the electric circuit closed through two or more mechanically operated igniters in different combustion chambers at the same instant. In some ignition systems it is impossible to operate when two igniters in different combustion chambers have their electrodes in contact at the same instant. 101. Bosch Mechanically Operated Igniter. Fig. 109 shows two views of an igniter and operating mechanism as constructed by the Bosch Magneto Company. The side of the igniter is shown in (a), and the external end in (b). A coiled tension-spring A acts on one end of the external rocker-arm B so as to press the movable electrode C against the stationary electrode D when the operating rod E is lifted by the cam F. The operating rod E is pressed downward by a coiled compression spring G, whose lower end bears against a collar on MAKE-AND-BREAK IGNITERS AND KICK-COILS 127 the rod and whose upper end bears against the stationary sup- port H. As the cam F rotates it lifts the operating rod E against the resistance of the compression spring so as to first allow the coiled tension spring A to pull the contact-point of the movable electrode against the stationary one, and then to push the rod up still farther so that it slips through the hole in the external rocker-arm. The enlarged upper end of the rod is thus lifted FIG. 109. Bosch Make-and-break Igniter. Bosch Magneto Company, New York City. free from the rocker-arm. As the cam lobe passes from under the cam follower, the operating rod E is forced downward by the expansive action of the compression spring G, and the ball at the top of the rod strikes the external rocker-arm and moves it down so as to quickly separate the contact-points of the elec- trodes. The compression spring G is made strong enough to overcome the resistance of the tension spring A and keep the 128 ELECTRIC IGNITION cam follower in continuous contact with the cam during the highest speed at which the igniter is to be used in any particular application. The lower end of the operating rod E is pin-connected to one end of each of a pair of links 7 whose opposite ends are similarly connected to a short rocker-arm on the shaft J. These links carry the roller K which bears against the cam and follows its outline. The short rocker-arm just mentioned can be rocked by a control lever L fastened to the same shaft. The shaft J of the controller and the cam shaft are supported by bearings which are maintained in fixed positions relative to the igniter plug. The control lever L is used to vary the instant of separation of the contact-points of the electrodes relative to the rotative position of the cam and to the position of the piston of the motor in its movement; in other words, to advance the ignition by causing it to occur earlier in the revolution of the cam, or to retard it by causing it to occur later. The ignition is advanced by moving the control lever to the right, and retarded by moving it to the left. The dotted outline of the control lever to the right is its position for early ignition, and the dotted outline at the left is its position for late ignition. 102. Truscott Boat Manufacturing Company's Igniter. This igniter is used especially in motor boats. It is shown in Fig. no. The outer end of the stationary insulated electrode appears at A . This electrode extends through the plug in the usual manner and projects inward near the end of the movable electrode B, which is fastened to a rotative spindle whose outer end is shown at C. A hammer-break arm D fits freely rotatable on the spindle C and is normally pressed against a stop on the spindle by a coiled torsion spring E. One end of E is fastened to a taper pin which passes through the rocker-shaft. When ignition is to occur, the free end of the rocker-arm D is raised by the lifter G, which is bored to fit freely rotatable on the end of the push-rod H. As the arm D is lifted it rocks the spindle C and contact arm B with it until the contact-point in B strikes against the stationary electrode. This prevents MECHANICAL MAKE-AND-BREAK IGNITERS 129 further movement of both the rocker-arm B and the spindle C, but the hammer-break arm D is lifted still higher and turns on the spindle C so as to separate itself from the stop on the spindle. The lifting of D winds up the spring E to a slight extent more than it is normally. The lifter-arm H continues rising till it disengages from D. The spring E then snaps the hammer-break H FIG. no. Make-and-break Igniter. Truscott Boat Manufacturing Company, St. Joseph, Michigan. arm D down quickly so that it strikes a sharp blow against the stop on the spindle C and causes rapid separation of the ignition points. The downward movement of D continues till it strikes the arm F and is stopped in a horizontal position. The lifter-rod H then descends, carrying with it the lifter G, and the end of G strikes against the bevel / on D. The end of G is also beveled where it strikes the bevel /. The action of the bevel twists G around on H as they descend, so that G slips down past D and is then snapped back under D by the coiled tension spring / so that the lifter G is again brought into position to lift D. K and K are nuts on the stud-bolts L and M for fastening the igniter to the motor. The lower stud-bolt M is extended outward 130 ELECTRIC IGNITION and serves as a binding-post for holding one of the wires of the external circuit. Fig. in shows the mechanism for operating the push-rod. This is accomplished by an eccentric N on the crank-shaft of o FIG. in. Operating Mechanism of Fig. no. the motor. The eccentric strap P is connected to the member Q by the pin R. The member Q is fastened to the lower end of the push-rod H and is limited to vertical movement by the plunger S which fits in a suitable guide. The eccentric has a free rotative fit on the crank-shaft and is connected to the arms of a fly-ball governor (not shown) also mounted on the crank- MECHANICAL MAKE-AND-BREAK IGNITERS 131 shaft. The eccentric is thus caused to rotate with the crank- shaft so as to move the push-rod, but its angular position on the crank-shaft is changed by the action of the governor as the speed of rotation varies. This shifting of the eccentric varies the time of ignition so that it occurs earlier at high speeds of rotation than at slow speed. When the motor stops, the governor brings the eccentric to a position such that ignition cannot occur before the crank of the motor has passed its dead-center position just after compression of the combustible charge. This prevents ignition at such an instant as to drive the crank-shaft of the motor backward when starting it. 103. The Fay & Bowen low-tension igniter, Fig. 112, has an adjustable hammer-break device as part of the igniter. Both FIG. 112. Fay & Bowen Make-and-break Igniter. the stationary electrode A and the movable electrode B are provided with inserted contact-pieces, or points. The hammer for breaking the circuit comprises a head C and a rod, or plunger, D. The latter slides through holes in the outer end of the body of the igniter, and is pressed against the cam E by a coiled com- pression spring F which is wound around the plunger and bears against a collar on it. The head of the hammer has an adjust- able striking piece G which is pressed against the external arm H of the movable electrode so that the contact-points of the electrode are kept apart except while the plunger is pushed back. When the plunger is pushed back by the rotating cam E, the coiled tension spring 7 draws the contact-point of the movable 132 ELECTRIC IGNITION electrode against the contact-point of the stationary electrode. The tension spring / is connected to one end of the external arm H of the movable electrode. The cam forces the hammer back far enough to remove its striking piece to some distance from the arm //. When the edge of the rotating cam passes out of engagement with the plunger, the hammer, including the plunger, is snapped down by the spring F and the striking piece hits the arm H so as to drive it around and separate the contact-points of the electrodes quickly. 104. Westinghouse make-and-break igniters are shown in Fig. 113. One is right-hand and the other left-hand. The mica FIG. 113. Low-tension Igniter of the Westinghouse Machine Company, Pittsburg, Pa. washers for insulating the stationary electrode at the outer end are visible at A . The inner end of the plug is recessed to receive similar insulation. The contact-points of the electrodes are pressed together by a coiled torsion spring wound around the rocker-spindle of the movable electrode and pressing against the outer rocker-arm. These plugs have removable contact-points in the electrodes. MECHANICAL MAKE-AND-BREAK IGNITERS 133 105. The Snow Steam Pump Works mechanically operated make-and-break igniter for low-tension current is shown in Fig. 114. The cast-iron plug i has the metal cut away in the middle portion to secure lightness and ease of construction. A flange at the outer end serves as a means of supporting some of the external parts and for fastening the plug to the engine. The movable electrode 2 is of high-grade nickel steel and is one piece with the inner rocker-arm 8 which carries the removable contact-point 9. The inner end of the spindle rocks in a station- ary bronze bushing 15, and the outer end is provided with a tight- fitting bronze sleeve 17 which rocks in a steel bushing 16. The external rocker-arm is fastened to the spindle by a bolt-and-nut lock. The stationary electrode 3 is insulated from the plug by mica washers 4 and 5, and lava bushings 6 and 7. It is held in place by nuts 12 and 13, which also hold the terminal n into whose shank is soldered the wire for bringing electricity to the electrode. A removable contact-point is set into its inner end. Two oil pipes 20 connect the outer end of the plug with oil passages leading to the inner bronze bushing 15. These oil passages are shown most clearly in the " section on B-B." A threaded hole 22 through the flange of the plug is for the- insertion of a cap-screw, or a set-screw, which can be screwed down to press its point against the metal of the engine and thus forcibly withdraw the plug from the engine to a short distance after the fastenings have been removed. The outer ends of two of these igniters with all attached parts are shown in Fig. 115. The reference numbers in this figure are not the same as in the preceding one. 106. The mechanical make-and-break operating mechanism of the Snow Steam Pump Works for a pair of igniters in the same combustion chamber is shown in Fig. 115. The igniters are like the one shown in the preceding figure. The head of the upper igniter is shown at i, and that of the lower one at 2. The ends of the insulated stationary electrodes of the two igniters are at 3 and 4, and the wire terminals for the electric conductors are shown in place fastened to these elec- 134 ELECTRIC IGNITION FIG. 114. (See also Fig. 115.) Low-tension Igniter. Mechanical Make-and-Break. For Large Engine. The Snow Steam Pump Works, Buffalo, N. Y. 1. Body of plug; cast-iron. 2. Movable electrode; steel, not insulated. 3. Insulated stationary electrode, steel. 4. 5. Mica washers for insulating stationary electrode. 6, 7. Lava insulating bushings. 8. Inner rocker-arm; integral part of rocker-shaft 2. 9. Contact-point; movable and removable. 10. Contact-point; stationary, removable. 11. Terminal to which electric conductor (wire) is soldered. 12. 13. Nuts for fastening stationary electrode and terminal in place. 14. External rocker-arm end. Shown better in Fig. 115 as parts 5 and 6. 15. Bushing for rocker-arm bearing; stationary, bronze. 16. Bushing for rocker-arm bearings; stationary, steel. 17. Sleeve, tight on 2, loose in 14; bronze. 1 8. Oil hole. 19. Groove for oil pipe. 20. Oil pipe leading to inside bearing of rocker shaft. 21. Stud-bolt and nut for fastening igniter to engine. Two used. 22. Threaded hole in flange of plug. For bolt to start (loosen) the plug to remove it from the engine. MECHANICAL MAKE-AND-BREAK IGNITERS 22. OUTER END OF PLUG 22 1 j ,1 ll II j l! | i] i -i i i i i ? -i !' '' l! !i ' i ten ), and (). This igniter has two pairs of contact-points, of which both movable points, i and 2, are attached to the same double- ended rocker-arm 3. The stationary contact-points, 4 and 5, are inserted in lugs 6, 6, which project from the inside end of the body of the igniter, and are part of the body casting. The body of the igniter fits into a cylindrical hole which pierces the cylinder wall of the engine; the body makes a tight joint near its inside end. The material of the engine cylinder is represented by the short herring-bone lines. The body of the igniter is in metallic (electric) contact with the metal of the engine cylinder. The rocker-arm 3 is carried on the live spindle 7 which passes through the insulated metallic tube 8 from end to end of the igniter body; the tube forms a bearing for the spindle. The tube 8 is insulated from the body of the igniter at the inner end by means of the mica washers 9, and at the outer end by a wood-fiber bushing 10. A nut n fits on the outer end of the tube 8 to hold the tube in place. A coiled compression spring 12 and a pair of metallic washers 13 are placed between the nut ii and the insulating bushing 10. The expansive force of the coiled spring acts to keep the flange of the inner end of the tube 8 tightly pressed against the mica insulating washers 9 so as to maintain a tight joint unaffected by different amounts of expansion in contiguous parts on account of the heat of combustion. Two arms, 14 and 15, are rigidly attached to the outside end of the spindle 7 for rocking the contact-points. These arms are insulated from the spindle by the flanged bushing 16. The two cap-screws, one on each side of the spindle, are tightened to grip the arms on the spindle 7. i 7 6 ELECTRIC IGNITION ELECTROMAGNETIC IGNITERS AND IGNITION SYSTEMS 177 Two solenoid coils, 17 and 18, each with a soft-iron or mild- steel plunger-core, are used for separating the contact-points at which the arc is formed for ignition. The plunger-core of coil 17 is shown by dotted lines in view (5); the end inside of the coil is tapered and bored to receive one end of a non-magnetic rod 19, whose opposite end is fastened to a bar 20 that forms part of a yoke. The end of the steel plunger-core that extends beyond the end of the solenoid 17 is fastened to the yoke-bar 21 by means of the nut 22. The end bars, 20 and 21, of the yoke are also connected together outside of the solenoid by the rod 23, which engages with the forked end of the arm 14 that is fastened to the insulated spindle 7 of the rocker-arm, as already described. A coiled compression spring 26 is placed between the yoke end- bar 21 and the end of the solenoid coil 17. The expansive force of this spring keeps the collar 24 pressed against the arm 14 when no current is flowing through the solenoid. The parts 19 to 26 are duplicated in connection with the solenoid 18. The action of the two springs 26 is to keep the ignition contact-points pressed together. When the ignition timer (not shown) closes the electric circuit so that current flows through the two magnet coils in parallel, the steel plunger-core of each solenoid is drawn farther into the coil than when no current is flowing. This drawing in of the plunger brings the collar 25 against the side of the arm 14, and likewise the duplicate of collar 25 against the side of arm 15. This moves the arms 14 and 15 so as to rock the rocker-arm 3 and thus separate both pairs of contact-points, 1-4 and 2-5. The plunger-cores move rapidly and gain considerable speed before the collars 25 strike the arms 14 and 15. Consequently the collars strike a hammer-blow against the arms so as to cause rapid separation of the contact-points at which the ignition arc is to be formed. The terminal (binding post) to which the wire from the source of electricity is connected is shown at 27. The current divides at this terminal and flows through the two magnet-coils in parallel and thence to the metallic washers 13 on the insulated electrode rod 7. 178 ELECTRIC IGNITION The ignition contact-points, i, 2, 4, and 5, can be driven out for renewals by using a punch or piece of small rod inserted in the slightly reduced extension of the hole into which each point is fitted. The makers of this igniter find that the use of two pairs of contact-points doubles the life of the points as compared with an igniter using only one pair of contact-points. The igniter is constructed for use on no- volt circuits. The thickness and number of turns of wire in each magnet-coil of course determine the voltage that is suitable. 131. One-ring Timer for Large Engine with Four Combustion Chambers. The complete constructional form of a timer for use with electromagnetic igniters is shown in Fig. 144. The end view is shown at (yl), the side view at (J3), and one of the brush-holders at (C). The mechanism for varying the time of ignition is shown in Fig. 145 in connection with the ring, or spider, which carries the brush-holders. This timer embodies the slip-ring and segmental ring shown in elementary form in Fig. 142. In Fig. 144 the current is brought to the timer by the wire i connected to the brush 2 which bears on the slip-ring 3. Part of the slip-ring 3 is broken away at the top in view (A) in order to expose the upper part of the long segment 4 of insulating material and the side of the short metallic segment 5, which together form the ring on which the four brushes 6, 7, 8, and 9 bear. The slip-ring 3 and the composite ring 4-5 are carried by the heavy metal ring 10 which fits on the shaft n. The slip- ring 3 is broken away under the brush 2 in view (A) so that the brush apparently bears on the heavy ring 10, but this is not actually so, since the brush is farther forward than the ring. The segment 5 is metallically connected to the slip-ring 3, but is electrically insulated from all of the other parts of the rotor. The brush 6, of the four similar brushes, fits into the brush- holder 12, which is hinged to the bracket 13. This bracket is fastened to a piece of insulating fiber 14. A coiled compression spring between the fiber and the brush end of the holder presses the brush against the rotor of the timer. The fiber 14 is fastened to a metal segment 15, which is flanged to fit into a circular ELECTROMAGNETIC IGNITERS AND IGNITION SYSTEMS 179