SAGE rfl UttiMMtg % ;tfc»g ME FUND ///£>//#£ BOUGHT WITH THE INCO FROM THE ENDOWMENT ' THE GIFT OF Bcnrg W. Sage 1891 Cornell University Library arV18S72 The storage battery; 3 1924 031 236 999 olin.anx The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924031236999 THE STORAGE BATTERY THE STORAGE BATTERY & practical treatise ON THE CONSTRUCTION, THEORY, AND USE OF SECONDARY BATTERIES. BY AUGUSTUS TREADWELL, JR., E.E. ASSOCIATE MEMBER A. I. E. E. Neto Ijorft THE MACMILLAN COMPANY LONDON : MACMILLAN & CO., Ltd. I8 9 8 ft All rights reserved Copyright, 1898, By THE MACMILLAN COMPANY. Norhjooto $reB2 J- S. Cushing & Co. - Berwick & Smith Norwood Mn«B. U.S.A. PREFACE In pursuing his work with storage batteries, the author found himself greatly hampered by the lack of any compact data concerning the construction of the many cells which have been, and are on the market, and by the paucity of reliable discharge curves. Be- lieving that a book containing such data and curves, together with rules for the handling and maintenance of cells, would be of great value, not only to the stu- dent and manufacturer, but also to all interested in storage batteries, the author began the compilation of the latest and most accurate data concerning the sub- ject. He believes that the list of American and foreign patents, which is given as foot-notes for the various types, will prove of inestimable value to the inventor. It must not be supposed, however, that the list given is at all complete, or that the full list of patents, covering each cell described, is given. So many patents relating to storage batteries have been granted, that a complete list would require a volume of its own ; only the prin- cipal patents, therefore, have been given for each cell. The chapter on the chemistry of secondary batteries will be found to give the latest and most generally accepted theory concerning the chemical reactions taking place in an accumulator. This chapter was vi PREFACE submitted to, and approved by, a prominent chemist, Dr. Sewal Matheson, whose courtesy the author desires to acknowledge. The table of data, which is to be found in the Appendix, may be relied upon as giving the latest and most accurate figures which could be obtained, of all the batteries. In the Appendix will also be found methods for the measurement of the E.M.F. and internal resistance of a storage battery, also data from which the theoretical and practical capacity of an accumulator may be determined. The majority of the cuts illustrating Chapter VII., were obtained through the courtesy of Mr. Charles Blizard, of the Electric Storage Battery Co. The author also desires to acknowledge favors extended by the various storage battery companies who have so kindly aided 1 him in the preparation of this work, by freely giving information of every sort, and by furnish- ing electrotypes. Thanks are also due to Dr. Samuel Sheldon and Mr. John J. Rooney for valuable aid. AUG. TREADWELL, Jr. New York, March, 1898. INDEX PAGE Accumulators for annex stations 145 Accumulators for cold climates 237 Accumulator installations : Belfast 151 Berlin 180 Birkenhead 153 Boston 158 Brooklyn . . . . . . . . . .164 Burnley 179 Chester 148 Chicago Board of Trade 180 Commercial Cable Building, New York .... 177 ' Edinburgh 149 Hartford 155 Isle of Man — Douglas-Laxey R.R 148 Isle of Man— Mt. Snaefel R.R .148 Manchester ■ . . 150 Merrill 154 New York — Bowling Green . . . ._ . .161 New York — 12th Street . 161 New York — 59th Street 160 Philadelphia — Philadelphia Edison .... 168 Philadelphia — Union Traction Co. ... . 173 Rome 178 Zurich 146 Zurichberg 180 Accumulators in place of resistances 146 Accumulators in telegraphy 184 Albuquerque 189 Atlanta . .187 Baltimore 190 vii Vlll INDEX Accumulators in telegraphy — Continued Paris Washington Wilmington Accumulator stations in Europe Accumulators to act as reservoirs Accumulators to carry the peak of the load Accumulators to carry the entire load at minimum hours Advantages of accumulators during an emergency Advantages of overcharging Alkaline-zincate genus .... Alternating currents with storage batteries Baillache Berthelot's discovery of persulphuric acid Boiling Buckling Cadmium plate test Capacity, data for the practical calculation of Capacities of modern batteries, table of . Capacity, theoretical calculation of . Celluloid in accumulator construction Central stations, with and without accumulators Characteristic curves Charging at constant voltage, advantages of Charging, hints concerning Charge, length of Charging, best rate of ... . Chemical changes during charge Classification of batteries, ordinary . Classification of batteries, Reynier's . Color of plates when charged . Commelin Comparison of thick and thin plates . Conclusions regarding storage battery traction Connecting cells Cosgrove on measurement of E.M.F. Cost of construction and operation of electric roads 189 188 190 140 143 141 142 144 136 9, 102 192, 224 103 134 227 222 ■ 239 ■ 253 ■ 254 252 • 239 • 194 . 118 • 223 • 237 . 226 221, 222 • 131 8 8 . 228 ■ 103 . 129 . 218 2 35> 245 • 249 • 215 INDEX ix PAGE Cost of horse and storage battery traction compared . . 200 Cost of overhead and underground trolley systems . . . 203 Cost of storage battery installations 191 Darrieus on persulphuric acid 134 Darrieus' theory 134 Data of operation of railways : Compressed air line in Paris 205 Steam in Belgium 206 Steam in Denmark . 206 Steam in Saxony 206 Steam in Paris 206 Storage battery in Birmingham 208 Storage battery in Paris 205 Trolley in Havre 205 Trolley in Marseilles 206 Defects of lead-sulphuric-acid storage battery . . . . 115 Definition of primary battery 1 Definition of storage battery 1 DeVirloy 103 Description of batteries : Acme 83 American 33 Barber-Starkey 80 Barker 99 Basset 107 Beaumont and Biggs 43 Blot 92 Boese 46 Boettcher 25, 99, 105 Brush 59, 88 Buckland 64 Chloride 39 Correns 76 Crompton 43 Currie 37 Darrieus 108 D'AFsonval 28 De Kabath ......... 26 X INDEX PAGE Description of batteries — Continued: De Meritens 25 Desmazure 103 Desruelles 47 Drake and Gorham 29, 70 Dujardin 25 Duncan 25, 37 Eickemeier 68 Electro-Chemical 22 El well-Parker 16 Engel 95 E.P.S 66 Epstein 17, 100 Erving 98 Faure 48, 49 Faure-King, E.P.S. traction cell 67 Fitzgerald 43, 45 Ford-Washburn 77 Garassino 25 Gaudini 29 Gelnhausen 95 Gibson 74 Grout, Jones, and Sennet 89 Gruenwald 96 Gulcher 84 Hagen 50 Haid 107 Haschke 86 Hatch c 2 Hauss 74 Hering 9 ! Hess 8 2 Hollingshead I0 8 Hough g4 Jacquet Sh 7° James S o ; g 9 Johnson and Holdregge 78 Julien c, Kalischer JO y INDEX xi PAGE Description of batteries — Continued: Khotinsky 64 Knowles 51 Kowalski 85 Krecke 94 Lane-Fox 35 Lehman 108 Lelande-Chaperon 102 Lloyd 65 Lugo 100 Maloney 108 Marx 105 Mason 97 McLaughlin 45 Metzger 48 Monnier 43 Montaud 35 Monterd 93 Nevins 5 1 New York Accumulator 31 Oerlikon 9° Ohio Storage Battery 3° Paget 19) 88 Payen 3 8 Percival 44 Peyrusson 29 Plante" 14, 16 Platner 106 Pollak 44. 63, 89 Pumpelly 81 Pumpelly-Sorley 35 Reckenzaun 9° Remington 3^ Reynier 28, 97, 99 Reynier Elastic 79 Ribbe 8 5 River and Rail i°° Rooney 34, 55, 66 Schaeffer-Heineman ......•■ 95 xii INDEX PAGE Description of batteries — Continued: Schenek-Farbaky °9 Schoop 18, ios Shultz 3 6 Silvey 3 6 > 94 Simmen ......•■•• 2 ° Sola-Headland 79 Sorley 94 Standard 3° Starkey 2 5 Sutton 97 Tamine ioo Tauleigne 107 Theryc-Oblasser 82 Thompson-Houston , 102 Tommassi 75 Tribe 43 Tudor 61, 88 Union 58 Van Emon 55, 66 Van Gestel 78 Verdier 46 Waddell-Entz 104 Willard 20 Winkler 55 Woodward 37, 64 Worms 51 Difference of potential between lead-antimony and active material 120 Different chemical combinations on positive plate . . . 128 Difficulty in making chemical analysis loo Discharge, duration of 228 Discharge, duration of, reasons for ...... 228 Discharge, effects of too prolonged a 225, 228 Distribution of current 24; Eating away of lead salts . 10, 37 Effect of too high a discharge rate 129, 230 Efficiency of a storage battery no INDEX xiii PAGE Efficiency of accumulators, average 203 Efficiency of accumulator installations 194 Elbs and Schb'nherr on persulphuric acid 138 Electric carriages 219 Electro-chemical methods of formation .... 10, 1.6 Electrolysis of a lead salt 10, 35 Electrolyte, alteration in the 114 Electrolyte, density of 238, 240 Electrolyte, purity of 235, 241 E.M.F., calculation of, Streintz 251 E.M.F., calculation of, Wade 250 E.M.F., measurement of, Cosgrove 249 E.M.F., measurement of, Negrenau 249 End plates 243 Examples of storage battery installations .... 146 Faure 48, 66, 67 Formation of lead-peroxide on negative electrode . . 127, 130 Forming charge, duration of 227 Future improvements 245 General theory of storage battery no Generation of heat due to 124 Grassi, measurement of internal resistance .... 248 Griscom and Fitzgerald on active material . . . -137 Grooves 10, 59 Hanover, mixed system at 213 Higher capacity of positives over negatives .' . . .129 History of the storage battery 4 Brush 6 Davy 5 Erman 5 Faraday 5 Faure 6 Gautherot 5 Grove 5 Jablochkoff 7 Jamins . 7 xiv INDEX PAGE History of the storage battery — Continued: Marianini 5 Maxwell 6 Metzger 6 Niaudet 6 Nicholson and Carlisle 5 Plante" 6 Ritter 5 Rue, de la 6 Schoenbein 5 Sinstedin 6 Volta 4, 5 Wheatstone 5, 6, 98 Improvements in Plante" type 10, 16 Improvements in Faure type 10, 50 Influence of acid on open circuit voltage 117 Installations, storage battery . 140 Installing batteries 237 .King, E.P.S. patents 66, 67 Lead-copper genus 9> 97 Lead-sulphuric-acid genus 8, 14 Lead-zinc genus 9, 98 Life of a storage battery, average • . 204 Losses in a storage battery 119 Loss on open circuit . . 120 Mance, measurement of internal resistance .... 248 Manhattan Elevated R. R 184 Manufacture of traction cells 217 Mechanical improvements 10, 25 Method of making copper-oxide electrode .... 102 Miscellaneous cells g, I0 j Negrenau on measurement of E.M.F 249 Occlusion of hydrogen theory j™ Operman's copper-oxide electrode . . . . . .102 Organic matter in active material 239 INDEX xv Parker, E.P.S. patents .... Passage of current from one plate to another Per cent load factor . Per cent maximum demand factor Perforations .... Perforations, kinds of Peroxide plate called positive . Persulphuric acid theory . Berthelot .... Elbs and Schonherr . Robertson and Darrieus . Plante" and Faure types compared Plates, construction of Potential of a cell due to . Position of cells Preparation of active material . Prevention of buckling Primary and storage batteries compared Private plants Reason that storage cells are not near perfection Recuperation Reduction of weight by alloys by "non-conducting grids . Relation between capacity and discharge rate Relation between capacity and specific gravity Relation between E.M.F. and acid concentration Resistance measurements, Grassi Resistance measurements, Mance Resistance measurements, Sheldon Resistance of a cell, Streintz Retaining case made of conducting material Retaining case made of non-conducting material Retention of paste Reynier's classification Reversed polarity Robertson on chemical theory Room, choice of battery . PAGE . 66 121 . I96 . I96 11, 66 11, 66, 241 1 134, 138 ■ 134 • 138 ■ 134 . • 48 241, 243 . 118 • 234 • 94 11, 90 . 185 . 192 184, . 109 206, 215 10, 50 10, 50 10, 52 230 238 117 248 248 247 114 n, 75 11, 79 10, 59 244 127 233 xvi INDEX PAGE Salomons, on best rate of charge 222 Salomons, on process of charging I3 1 Sellon, E.P.S. patents 66 Separators 2 35 Setting up a battery 2 34 Sheldon, measurement of internal' resistance .... 247 Solid active material IO > 44 Specific resistance of acid "4 Spraying 2 3^ Storage battery, economy of 3 Storage battery traction — Berlin .210 Birmingham 208 Brussels — Tervueren 207 Chicago — Englewood 212 Dubuque 214 Hagen — Vienna 208 Hague — Sheveningen 211 Hanover 213 Madison Avenue, New York City 212 Paris 206 Streintz, calculation of capacity 25 1 Streintz, theory 112 Sulphating 241 Superiority of lead-lead over other types 11 Swan, James W., originator of perforated plates ... 66 Telegraphy, storage batteries in 3, 184 Temperature variations during charge and discharge . -123 Testing a battery 231, 239 Tests of Theryc-Oblasser battery for traction .... 207 Tests of traction systems 205 Theoretical energy of lead 1 1 j Theory 109, 126 Theory of Darrieus 112 Theory of primary and storage batteries identical . . .110 Theory of Streintz . . . ... 112 Traction, storage battery 4, 198 Traction, storage battery, in Europe and America . . -199 INDEX XVll Transportation of battery after being used Trolley and accumulator systems compared Trolley roads, disadvantages of Uses of accumulators Volckmar, E.P.S. patents Variation in capacity with temperature Wade, calculation of capacity . Weight of storage battery cars, increased 244 214 199 2, 141 66 125 250 201 ILLUSTRATIONS Batteries : American 33 Blot 93 Brush 60 Chloride 40, 41, 42 Correns 76 Currie 38 De Kabath 26 Drake and Gorham 70 Eickemeier . 69 Electro-Chemical 23 E.P.S 67 Epstein 18 Gibson 74 Hatch 53 Hauss 7 1 Hering 91 Jacquet 74 Khotinsky 65 Montaud 36 New York accumulator 31 Ohio storage battery 30 Percival 44 Pumpelly-Sorley 35 Reckenzaun 9° xviii INDEX PAGE Batteries — Continued: Rooney 34; S 6 Schoop 1 9 Tommassi 75 Tudor °i Union 5^ Willard 20, 22 Battery curves : Bradbury-Stone 7$ Chloride 43 De Kabath 27 Electro-Chemical 24 E.P.S. ' 68 Ford-Washburn 77 Gadot 224, 225 Gulcher 84 Haschke 87 Hatch 54 Hauss 72) 73 Pollak 63 River and 'Kail io'i Rooney 57 Tudor 62 Willard 21 Curves : Berlin load curve . 181 Capacity and specific gravity at close of discharge . -130 Charging at constant current .... . 224 Charging at constant voltage ...... 225 Chicago Board of Trade, load curve . . . .182 Effect of accumulators on voltage curve, Merrill . 154,155 Effect of accumulators on voltage curve, Union Traction Co., Philadelphia. . . . Frontispiece, 174 ^ Elevator-load curve . . . . . . . .183 E.M.F. and per cent of acid 116 E.M.F. of charge and discharge compared . . • H3 E.M.F. and specific gravity at close of discharge . . 131 Hartford load curve . . . . . . -157 INDEX Curves — Continued : , Influence of acid on open-circuit voltage . Load curve for Union Traction Co Per cent load factor for English stations . Per cent maximum demand factor for English stations Philadelphia load curve ...... Relation between capacity and discharge rate . Relation between capacity and specific gravity Specific resistance of acid ..... Temperature variation during charge and discharge . Theoretical load curve ...... Installation illustrations : Accumulator plant, Bowling Green, New York Accumulator plant, Brooklyn .... Accumulator plant, Philadelphia Edison . Accumulator plant, Union Traction Co., Philadelphia Cell regulators, Philadelphia Edison Switchboard connections — Atlanta Switchboard connections — Bowling Green Switchboard connections — Burnley Switchboard connections — Chicago Board of Trade Switchboard connections — Zurichberg . Switchboard connections — 12th Street, New York Switchboard, Philadelphia Edison . 165 167 170 175 172 188 163 179 182 180 162 171 Measurements — diagram of connections : Internal resistance — Grassi Internal resistance — Mance Internal resistance — Sheldon E.M.F. — Negrenau E.M.F. — Cosgrove 248 248 247. 249 249 ABBREVIATIONS E. W Electrical World, New York. N. Y. E. E Electrical Engineer, New York. L. E. R Electrical Review, London. L, E The Electrician, London. El. Anz Elektrotechnischer Anzeiger. Wied. Ann. ... . Wiedemann's Annalen. Trans. A. I. E. E. . . . Transactions of the American Institute of Electrical Engineers. A. P American Patent. B. P British Patent. F. P French Patent. G. P German Patent. L Lighting. T. . . Traction. G Glass. R Rubber. W Lead-lined wood tank. Cd Cadmium. Alka— Zn Alkaline — Zincate type. Pb — Zn Lead — Zinc type. Pb — Cu Lead — Copper type. Chlor Chloride battery. THE STORAGE BATTERY oJ e 1 1 ) 1 1 1 1 2 1 3 1 1 IS Hours Discharge Fig. 10. from an N plate (spun), 6 x 7 x 0.4375 inches, weight 4.5 pounds. In each case the discharge was at a con- stant current, 4 amperes for A, 7 amperes for B, 12 amperes for C, and 10 amperes for D. LEAD-SULPHURIC-ACID GENUS 2$ VARIOUS CHEMICAL METHODS OF FORMATION Dujardin 1 uses as a forming bath a solution contain- ing 10 kilos of water, 2 kilos of sulphuric acid, and 1 kilo, of alkaline nitrate, soda, potassium, or other suitable alkali. Boettcher uses a bath of sulphuric acid, acetic acid, and water. Garassino 2 places spiral perforated lead plates in a nitric acid bath for 12 hours, then in caustic potash,, and finally in pure water, after which electrolytic spongy lead is formed in a hot caustic potash solution, and the plates are pressed. By mis- take Starkey 3 placed some plates in dilute sulphuric acid, which contained a small quantity of a solution of chromic acid. He found that they set harder and quicker, and became more deeply peroxidized than in ordinary acid ; they have since behaved in a most satis- factory manner. By cooking his plates in a solution of litharge in caustic potash or soda, Duncan produced a thick dense deposit of spongy lead. It should be remembered that in order to have much success with electro-chemical methods of formation, only pure, soft, rolled lead should be used. 2. Mechanical DE MERITENS' CELL Many of the early investigators with storage batteries worked along the lines of mechanical improvement in order to obtain the maximum amount of active surface 1 F. P., 174,761; 1886. B. P., 16,408; 1886. 2 B. P., 12,665; l8 9 2 - 3 E - w -> Vo1 - 2 7> P- 4 66 - 26 THE STORAGE BATTERY for a given weight. Prominent among these stood 'M. de Meritens, 1 who constructed plates formed of thin lead laminae 2 mm. thick. These lead laminae were folded one upon the other somewhat in the form of a book, and the whole was soldered to a stout framework of lead. Having a large surface, a large amount of oxygen was required ; the capacity, consequently, being increased, with the attendant disadvantage, however, of the local action being also increased. The film of per- oxide was extremely thin. Because of the rapid local action this accumulator was a failure, especially where required to maintain its charge for a lengthened period ; but where a rapid rate of discharge was required, this type gave excellent satisfaction. DE KABATH This cell 2 is made in the form of a number of shallow perforated lead boxes, as shown in Figs. 11 and 12. Each box contains between 180 and 190 lead strips, alternately straight and corru- gated. Lead strips, So cms. long, 1 cm. wide, and 1 mm. thick, are so corru- gated that their Fig. 11. Fig. 12. 'B.F, 1 173; 1882. 2 A. P., 263, 124; 1882. B. P., 287 ; 1883. G. P., 21,689, 22,690; 1882. LEAD-SULPHURIC-ACID GENUS 27 length is reduced to 36 cms., the straight strips being of the same dimensions as the corrugated strips. The complete element measures 38 cms. in length, 9 cms. in width, and nearly 1 cm. in thickness, and weighs 2.2 pounds. The box may also be made of cardboard, caoutchouc, parchment, or other acid-proof material. 70,000 60.000 □ 2 60.000 _l u. (3 10,000 DC UJ Q. » 30,000 CO 2 HOURS FORMATION FIG. 13. Fig. 13 shows the rate of increase of 1 kilo of lead when made up into this form of element at any period of time. This curve was obtained from a cell weighing 30 kilos, the weight of the elements being 2 1 kilos, and that of the electrolyte 6 kilos. 28 THE STORAGE BATTERY REYNIER Reynier sought to increase the exposed surface by making lead "plaits." His electrical connections being bad, he took a steel frame which had been dipped into an alloy of lead and antimony and placed the " plaits " in that. He found that when the plates were charged that they were less dense; they therefore buckled. To obviate this, longitudinal slits were cut in the plates, thus allowing for expansion. d'arsonval Hoping to obtain the largest surface for a given weight, d'Arsonval substituted lead shot for the solid lead plates. While this expedient would seem to give good results, it must be remembered that in order to be effective the shot must be in good electrical contact, but since they would soon become oxidized, it would be impossible to obtain good results. SIMMEN Simmen substituted lead wire for the shot, with the result that his accumulator became much more efficient than that of d'Arsonval. The lead wire was obtained by pouring molten lead through a cullender into a pan of cold water. This sudden cooling makes the sur- face of the wire very rough, and the wire itself very light and porous, so as to be easily acted upon chemi- cally. Masses of this wire were taken and pressed to the desired shape and placed in a perforated lead chamber, thus forming, one electrode. Owing to the LEAD-SULPHURIC-ACID GENUS 29 chemical action on these leaden chambers the com- pressed wire was placed in a metal frame, which was but slightly acted upon by oxidation. This improved form has been termed the Simmen-Reynier cell. DRAKE AND GORHAM These elements, usually known as the D. P. plate, 1 consist of a large number of narrow strips of lead, hav- ing points or projections on their faces ; they are built up one above the other. A large working area is thus produced, and it is claimed that a cell of this type can be charged or discharged at a high rate without buckling or disintegration. GAUDINI In this accumulator the plates are formed of a mix- ture of lead and coke and retort carbon. By the use of carbon the manufacturers claim to obtain greatly in- creased porosity, besides accelerating the formation. The electrodes are separated from each other by porous partitions. These partitions are made of any solid or gelatinous acid-proof material, and may be either straight or curved. PEYRUSSON This accumulator 2 contains but two electrodes, placed one within the other. The positive element is in the form of a central rod, from which radiate thin sheets of lead, half a millimetre thick, and placed within a hollow cylinder, which forms the negative electrode. This cylinder is composed of sheets of lead, bent so as 1 B. P., 10,608; 1892: 12,650; 1894. 2 B. P., 8226; 1886. A. P., 523,371; 1894. 30 THE STORAGE BATTERY to expose both surfaces, and united by bands into the form of a cylinder. THE STANDARD BATTERY PLATE This plate, which has been brought out by J. H. Robertson, 2 is made by mixing pumicestone with lead while the latter is in a semi-molten condition. When thoroughly mixed, the mass is compressed to the de- sired shape and left to cool. This gives a very porous plate, having a large active surface. Owing to the nature of the pumicestone it is unnecessary to " eat it away." The makers claim that the heat of the molten lead renders the pumicestone more porous than it is ordinarily. OHIO STORAGE BATTERY In this battery a plate made from chemically pure rolled lead is passed through a machine which raises circular grooves, i inch in diameter, fg inch deep, and -jk inch thick over the sur- face, each circle being inde- pendent, and separated from each other by ^ of an inch, as shown in Fig. 14. After the grooves are formed, the plate is put in a special solution, and the grooves are filled by electro-chemical action. 1 A. p., 546,739 ; 1895. LEAD-SULPHURIC-ACID GENUS 31 For ordinary purposes the plates are separated by a hard rubber comb, but in cases where the cells will re- ceive extra hard usage each positive plate is covered by a porous flexible envelope, thus absolutely preventing short circuits. It is evident that as one groove expands after usage it cannot crowd any others, except the ones in its own circle. All buckling is thus prevented. NEW YORK ACCUMULATOR AND ELECTRIC CO. S PLATE This plate, which was brought out by Harris and Holland, 1 is shown in Figs. 15, 16, and 17. As will be 1 Fig. 15. Fig. 16. Fig. 17. seen, it is an open-work, ribbed and- grooved plate, with the ribs of one side crossing those of the other, and bodily, and therefore electrically joined to them. The 'A.P, 574.417; '896. 32 THE STORAGE BATTERY plate is made either by casting in a mould, or else by pressing, rolling, or sawing it out of a piece of pure rolled sheet lead. The grooves are made at an angle of 20° to the horizontal. Where strength and dura- bility are first considerations the plates are made heavier, every tenth rib being made thicker and the adjoining grooves wider. For some uses, the positive plates are protected by perforated sheets of insulating material secured to the plate by rubber bands. These plates are formed by immersing them in a bath consisting of a mixture of dilute sulphuric acid (practi- cally a non-solvent of lead) and nitric or acetic acids (both solvents of lead), which produces a coating of lead sulphate. They are then removed and subjected to the action of an electric current in an electrolyte consisting, preferably, of a moderately strong solution of magnesium sulphate, or its equivalent (as aluminum sulphate), and proportionately small quantities of sul- phuric and acetic acids, and magnesium acetate, or equivalents therefor. Thus two actions are constantly taking place : the formation of lead sulphate by chemi- cal action and the peroxidation of this sulphate by the current, the magnesium sulphate and acetate being used to facilitate the peroxidation of the difficultly peroxidizable lead sulphate. When the formation has penetrated to a sufficient depth, the plates are thoroughly washed, and subjected to the action of the current in an electrolyte of dilute sulphuric acid, and a proportionately small quantity of an acid sulphate, as that of sodium or potassium. This treatment com- pletes the formation by converting any remaining por- LEAD-SULPHURIC-ACID GENUS 33 tions of the lead sulphate to peroxide, but does not increase the depth of formation. When fully formed, such of the plates as are intended for negatives, are reduced by connecting them to the negative terminal of the charging current. The electrodes are then ready to be assembled to form cells, where they are charged in the usual way. By this process well-formed plates are obtained from the raw material in from 30 to 50 hours, depending on the strength of current. It is not advisable, however, to use too strong a current in forming. THE AMERICAN BATTERY CO. This plate is made from a solid sheet of pure rolled lead, ^ of an inch thick, and grooved on both sides, as shown in Figs. 18 to 20. These grooves are -j^ of an inch wide and -^ of an inch FIG. 18. FIG. 19. Fig. 20. thick. The active material, formed electrically in a strongly oxidizing solution, completely fills the grooves. D 34 THE STORAGE BATTERY For long life, in constant and severe service, this con- struction will be found to give excellent satisfaction. In a former plate, brought out by Morrison, 1 lead ribbon was folded loosely upon itself, forming a square plate ; the lead ribbon, \ of an inch wide, was grooved on both sides. The strips were solidly united at the ends by a process which left the edges of the plate a solid bar of metal, practically acid proof ; foot and terminal pieces of the same metal were also cast on. The grooves served, not only to key in the active material when formed, but also as canals, permitting the electro- lyte to act freely on every part. The lead strips were so folded that they were about ^V °f an mcn a P art ; this allowed expansion to take place freely in a vertical direction. Rubber combs were used as separators. ROONEY For his central station, or high discharge type, Mr. Rooney builds his plates as in Fig. 21, which, as will be seen, is very similar to the De Kabath plate. Corru- gated strips of lead ribbon, alternated with straight strips, are burned at one end. A number of these are then burned to a lead conductor, and the whole is electrically and mechani- fig. 21. cally connected to a lead » A. P., 512,514-522479 ; 1894. LEAD-SULPHURIC-ACID GENUS 35 frame. In this way all expansion and contraction are provided for, and buckling is unknown. As a further protection, the positive plates are surrounded by a per- forated hard rubber retaining case. The plates are formed in a few hours by a nitrate solution. VARIOUS MECHANICAL METHODS Lane-Fox 1 makes his plate by alternating lead lami- nae with sand. Pumpelly-Sorley 2 constructed theirs by clamping a sheet of pure rolled lead between two Fig. 22. plate forms, and then bringing it against a gang of saws, thus producing the plate shown in Fig. 22. The manufacturers claim to have produced the first integral slotted rolled lead plate. 3. Electrolytic (1) Electrolysis of a Lead Salt montaud's elements This investigator coated a lead sheet with electrolytic lead, from a solution of lead in potassium and water. 1 A. P., 285,807; 1883. 2 A. P., 521,897-467,522; 1894. 36 THE STORAGE BATTERY Fig. 23. The bath was heated to ioo° C, so that the current density might be high. With a current of 600 am- peres, only 30 minutes was required for making the plates. After washing, the plates were ready to be formed. The shape of the plates is shown in Fig. 23, a rod of white metal connecting the plates of like polarity. For this cell, the best charging rate is about 10 amperes per square metre of active surface, and the discharging rate about 20 amperes per square metre. SILVEY Silvey 1 places lead plates in a solution containing a mixture of acetic acid and potassium, and passes a cur- rent through the cell. This decomposes the anodes, depositing them in a metallic state on the cathodes. The cathodes are then removed, and the deposit is com- pacted by pressure ; after- which the plates are placed as positives in a storage cell and formed. VARIOUS METHODS Shultz 2 covers the lead with sulphur, and then heats it to form lead sulphide, after which the plates are placed in a bath and made spongy by electrolysis. Remington 8 immerses lead plates in a saturated solu- 1 A. P., 512,757-523,689; 1894. 2 G. P., 21,454. 8 A. P., 342,855; 1886. LEAD-SULPHURIC-ACID GENUS 11 tion of lead in caustic alkali, and then deposits a coating thereon hy electrolysis. Duncan 1 produces a coating by making the plates the anodes in a bath con- taining a solution of oxide of lead in potassium. (2) Eating away of Lead Salts WOODWARD Woodward 2 pours molten lead on common salt, and while still pasty, the salt and the lead are thoroughly in- termixed, and the plastic material is compressed to the requisite shape. The salt is then dissolved out and the plates are formed. In another type of battery, designed especially for traction purposes, the plates are placed horizontally, and sheets of porous earthenware, or other suitable porous insulating material, are placed between - them. CURRIE In the manufacture of this grid, 3 a brass rod is placed in an asbestos tube. Fused lead chloride is poured into the mould, around the rod, and into the meshes of the tube, thus filling up the interstices of the asbestos, and forming a thin-walled tube.- The brass rod is then withdrawn, and an alloy of lead and antimony is poured into its place, a connecting rod of the metal being cast at the same time. The chloride is then reduced by electrolysis, after which the plate is ready to be fitted 1 B. P., 15,433; 1888. 2 A. P., 392,373-392,374-393,954-393,955; 1888: 406,969; 1889. 8 A. P., 447.279-4S .834-4S3.99S-459.49i; i»9'- 38 THE STORAGE BATTERY up and formed. Fig. 24 shows the arrangement of the plate. The elements can be used with positives of simi- lar shape, or with flat positives of the ordinary style. FIG. 24. THE PA YEN ACCUMULATOR There has been much controversy, of late, as to the originator of chloride plates. It has been settled by giving to Marchenay 1 the honor of first mentioning lead chloride for accumulator plates ; to Maxwell- Lyte, 2 the honor of constructing the first chloride of lead plate; and to Andreoli, 3 that of introducing the first chloride of lead grid plate. It is to the Maxwell-Lyte type that the Payen 4 battery belongs. 1 L. E. R., May 25, 1894. 2 A. P., 422,308; 1890. B. P., 3452; 1883. 3 B. P., 8842-12,595; 1886: 18,807; 1892. 4 A. P., 440,267-440,277-440,575; 1890. LEAD-SULPHURIC-ACID GENUS 39 In the manufacture of these plates, an intimate mix- ture of asbestos fibre and fused lead chloride is formed, and the molten mass poured into a mould, crystallizing as it cools. The result is a chloride of lead plate, bound together with asbestos fibre. By making this the -cathode to an ordinary lead anode, the lead chloride is transformed into spongy lead, after which the plates are ready to be formed in the ordinary manner. A mixture consisting of 90-95% lead chloride, and 10-5% zinc or cadmium chloride is employed in the- manu- facture of the plates. THE CHLORIDE STORAGE BATTERY This battery, which is manufactured by the Electric Storage Battery Co. 1 of Philadelphia, has come into prominent notice of late years, and stands among the best of the batteries which are upon the market. In the manufacture of this cell, commercial lead is reduced to a fine powder, dissolved in nitric acid, and then precipitated by hydrochloric acid. The lead chloride thus obtained is washed and fused with zinc chloride. The molten metal is poured into a mould, and allowed to cool, forming pastelles, about -^ of an inch thick. For the positive plates, the pastelles are circular in form,^f of an inch in diameter, with a bevelled V-shaped periphery; for the negatives, they are square, f of an inch on an edge. These pastelles are placed in a mould, and a lead-antimony grid is cast around them under pressure. By packing these grids 'A. P., 415.329-41S.330-415-331-415.333; "889: 477,182; 1892. B. P., 1229; 1893: 10,836-12,953-12,954; 1895. G. P., 57.053; 189°- 4° THE STORAGE BATTERY between zinc plates in a tank containing a dilute solu- tion of zinc chloride, and short-circuiting them, most of the chlorine is extracted. The last traces of chlorine Fig. 25. can now be removed by thoroughly washing the plates in running water; a pure spongy lead plate results. The positive plates are then formed by packing them tightly between perforated ebonite boards, in dilute sul- LEAD-SULPHURIC-ACID GENUS 41 phuric acid, and passing a current through them in one direction for two weeks. Lately, however, the Electric Storage Battery Co. has abandoned this method of making positive plates, and constructs them of a lead-antimony grid -f^ of an inch thick, and having circular holes | of an inch in Fig. 26. diameter; the grid being cast under pressure to make it dense. A corrugated, soft lead ribbon, also ^ of an inch wide, is bent into the form of a spiral, and is pushed into these holes; the active material being formed from this ribbon by electro-chemical process. The expansion of the active material, during use, tends to wedge the spiral more tightly within the hole, thus 42 THE STORAGE BATTERY improving the contact. This plate, known as the " Manchester " plate, was brought out by Rhodin. 1 It W Miiiiiiiiiiiiiiiiiiniiiiii iiiiii Fig. 27. is constructed essentially like the original Brush plate, though the "active material, or material to become 1 A. P., 567,044-567,045-; is LEAD-SULPHURIC-ACID GENUS 43 active," is a spiral of ribbon lead, instead of the paste or cement used by Brush. In Figs. 25, 26, and 27 are shown the E-n Portable cell, the ordinary Portable cell, and the Central Station 2.1 2.0 "Chhr de" 5 6 7 Hours Discharge Fig. 28. cell. Fig. 28 shows the charge and discharge curves of this battery. They were taken from a 75 A. H. cell, which was charged with a constant , current, and dis- charged through a constant resistance. VARIOUS METHODS OF FORMATION Monnier 1 makes an alloy of lead with about 4% of zinc; the alloy is then cast into the desired shape, and the zinc eaten away. Tribe 2 used various salts of lead, mostly, however, the arsenides, phosphides, or sulphides; the salt was then reduced to spongy lead by electrolysis. Messrs. Beaumont and Biggs, 3 Fitz- gerald, 4 and Crompton 5 used alloys of lead containing 1 B. P., 1556 ; 1883. 2 B. P., 2073 ; iS * B. P., 29 ; 1882. A. P., 524,710 ; 1894 3 B. P., 12,818; 1886. 6 G. P., 22,816; 1882. 44 THE STORAGE BATTERY tin, iron, or antimony from which the foreign material was eaten away by electrolysis. Pollak, 1 with some of his plates, uses lead carbonate and caustic potash mixed to a thick paste. They are then moulded to the re- quired form and dried, after which a current is passed to reduce them to spongy lead. 4. Plates of Solid Active Material percival's secondary pile In April, 1866, George G. Percival took out American Patent 53,668, for a secondary electric pile; this was the first United States patent granted for storage batteries. It is constructed of solid active material, and is shown in Fig. 29. In this figure, B is a wooden containing box, and C is a porous partition. A, A, are layers of powdered gas car- bon, powdered lead, or other suitable conducting powder. These two layers constitute the two electrodes, and when in use, are wet with dilute sul- phuric acid. By placing these layers in a horizontal posi- tion, a layer of non-conducting powder may be substituted for the partition. --C FIG. 29. IB. P., 813; 1893. LEAD-SULPHURIC-ACID GENUS 45 FITZGERALD S LITHANODE Professor D. G. Fitzgerald : makes his " compressed active material " plate, by thoroughly mixing litharge and ammonium sulphate, and then allowing the mass to dry slowly under pressure, in the requisite shape. The plate is converted into peroxide, by placing it in a bath of magnesium hypochlorite, or other suitable chlorine compound, for a preliminary coating ; this insures uni- form peroxidation in the forming. The forming is effected electrolytically in a bath of sulphate of mag- nesium, for a lengthened period. The preliminary coat- ing, may, however, if desired, be applied mechanically. The contacts are of some unoxidizable metal, such as gold or platinum. The plate is used as a positive to a negative of ordinary lead. By the use of lithanode, a positive plate is obtained which is unusually free from local action, and the manufacturers claim that fully 90% of the positive plate is lead peroxide. In discharging a battery containing lithanode plates, care must be taken not to run the E.M.F. down below 1.8 volts per cell. Professor Fitzgerald claims that when used according to directions, the electrical capacity of lithanode is almost exactly one ampere-hour per ounce ; the best rate of discharge being ^ f a n ampere per square inch of lithanode plate. MCLAUGHLIN The electrode for this battery 2 consists of a conduct- ing core having a ledge or seat at its lower extremity, 1 A. P., 524,710 ; 1894. B. P., 3731 ; 1890. 2 A. P., 424,809 ; 427,785; 432.202; 1890: 475.335; l8 9l- 4 6 THE STORAGE BATTERY on which is placed a block of active material, so that the entire surface of the active material, except the lowest side, is exposed to the electrolyte. The makers claim for this type of battery that there will be no twist- ing, warping, or buckling, and that the battery will stand a high rate of discharge. BOESE The Boese 1 plates consist of a solid slab of active material, enclosed in a lead frame, similar to the frame of a slate. The slabs are formed of minium for the positive, and minium and litharge for the negative, these being formed into a paste with alcohol, containing cer- tain hydrocarbons, such as anthracene, obtained from a distillation of coal tar. After being moulded and pressed into the frame, the plates are pierced with a large number of small holes, for the escape of the gas ; they are then baked, and placed in a dilute sulphuric acid solution to harden, after which they are formed. These plates are said to be extremely porous, and to have a conductivity nearly equal to that of lead. More than 20,000 of these cells are in use on the German and Hungarian postal railway cars for lighting, where they give excellent satisfaction. VERDIER Verdier 2 mixes a lead oxide with a vegetable oil, or a mixture of glycerine and water, to a thin paste. The material is dried in air, and perforated. The plates are 1 G. P., 78,865 ; 1892. 2 B. P., 8973 ; 1889. LEAD -SULPHURIC-ACID GENUS 47 then transformed to spongy plates by treatment in a solution of sodium sulphate, and a mixture of glycerine and water ; after which they are formed in dilute sul- phuric acid. DESRUELLES Desruelles 1 mixes 60 parts of lead peroxide, 40 parts of graphite, 25 parts of pulverized porcelain, and 10 parts of white of egg, presses into shape, dries, and heats until the albumen coagulates. A very porous and extremely hard plate results. 1 G. P., 61,620 ; 1891. B. P., 4877; 1891. CHAPTER III LEAD-SULPHURIC-ACID GENUS I. — B. Faure Type The formation of accumulators by means of reversals being an expensive as well as a troublesome process, Metzger and Faure maintained that, if the active ma- terial were mechanically applied, the previous tedious formation would be saved, and that better results would be obtained than with the old process. Experience has proved the wisdom of these views, and to-day the use of these two types is evenly divided, European practice favoring the Faure type, and American practice the Plante type. The advocates of the pasted type claim that a larger percentage of the total weight of the plate consists of active material, and that this, while not con- • ducive to high rates of discharge, permits the Faure cell to take in a greater charge, and to obtain a greater staying power than is possible with the Plante" cell. On the other hand, the Plante cell will do what the Faure, with equal weight or surface, will not do, — produce rapid discharges in currents of great volume. , metzger In 1878, R. L. Metzger, 1 of Alt-Breisach, Germany, took a sheet of perforated lead, 1.5 mm. thick, formed 1 El. Anz., 1892; 651. 48 LEAD-SULPHURIC-ACID GENUS 49 it into the shape of a box, about 7 mm. deep, and filled it with a paste composed of a lead oxide, dilute sulphuric acid, and potassium silicate. After the paste was thor- oughly dry, a perforated lead cover was soldered to the box. Two such elements, placed in dilute sulphuric acid, composed the battery. From 48 to 72 hours was required for the formation. Somewhat previously to this Metzger had used the two paste-filled boxes in the form of a hollow cylinder, placed the one within the other. The lower part of each was filled with some insulating material. This, not proving satisfactory, was abandoned for the flat electrode type. FAURE Although used in 1878 for the first time, the pasted type was not universally known until 1880, when Camille A. Faure, 1 working independently of Metzger, tQok out patents on that type. He took a lead oxide, in the form of a paste, spread it upon a spiral Plants plate, and allowed it to dry. It was insulated from the nega- tive element by means of felt, or parchment- paper, and the whole was replaced in dilute sulphuric acid. Ac- cording to some tests by Lord Kelvin, this type gave about 12,000 foot-pounds per pound of cell complete. In a cell composed of 16 plates, each 17 x 12.5 inches, the amount of lead oxides present was 50 pounds, and the total weight of the cell, 135 pounds. The capacity 1 A. P., 252,002; 1882: 309,939; 1884. B. P., 129: 1676; 1881: 1769; 1882. G. P., 19,026; 1881. So THE STORAGE BATTERY was 176 ampere-hours, at an 8-hour rate, and 299 am- pere-hours at a 14-hour rate. The great disadvantage with this cell is that " lead-trees " are soon formed. Improvements in the Faure Type 1. Reduction of Weight (1) Use of Alloys HAGEN In this battery, brought out by Gottfried Hagen, 1 of Germany, the lead frame consists of two halves, each composed of ribs crossing each other at right angles. Each rib is in the form of a triangular prism, with its base outwards. The two halves are not cast solid along the inner edge of the ribs, but are some distance apart, and are held together by a series of short cross-bars, the entire frame being cast in one piece. In this way a very light and yet strong frame is secured, and one which is capable of holding firmly a large amount of active material. The plates are kept apart in the cell by means of corrugated, perforated celluloid separators, which offer considerable resistance to lateral pressure. For stationary batteries the ratio of the weight of the active material to the total weight of the plate is 50%, while for transportable batteries it is 60%. The dis- charge should never be continued under 1.88 volts. JAMES In this accumulator the positive plate consists of an alloy of lead with 1% of cadmium, the negative 1 G. P., 52,880 ; 1889. LEAD-SULPHURIC-ACID GENUS 51 containing 2% of antimony. Both plates are pierced with circular holes, in which the active material is placed. For the positive plate the active material consists of the following mixture: minium, 85%; litharge, 10%; carded asbestos, 4% ; and powdered carbon, 1%. That for the negative contains : litharge, 94% ; asbestos, 4% ', sulphur, 1% ; and powdered carbon, 1%. KNOWLES The grid for this battery 1 is composed of an alloy of 82% of lead, 16% of tin, 1.9% of antimony, and 0.1% of arsenic. Solid blocks of active material composed of 75% of red oxide of lead (minium), and 25% of yellow oxide of lead .(massicot), first thoroughly mixed, and then treated with sulphuric acid, and hardened, are placed between these perforated grids, which are held together by rivets. The makers claim that by this means a very light, as well as an unoxidizable, grid is obtained. VARIOUS METHODS In the Jacquet 2 battery the grid is composed of a white metal alloy. Worms uses 96.5% of lead, 2.2% of antimony, and 1.3% of mercury for his grid. Julien employs 92% of lead, 4.5% of mercury, and 3.5% of antimony. Nevins uses 30 parts of lead, and 100 parts of tin. The usual alloy used is 96% of lead to 4% of antimony. 1 A. P., 480,266-482,979-483,562-483,563 ; 1892: 538,919; 1895. 2 B. P., 18,028; 1889. j2 THE STORAGE BATTERY (2) Use of a Non-conducting Support HATCH In the old form of Hatch 1 battery the lead salts were contained within the corrugations, or grooves, of a highly porous zigzag plate of earthenware. When these grooves were so filled as to present an even sur- face, the requisite number of packed plates for a com ; plete cell were bound together between stout wooden boards by means of flexible rubber bands, each plate being separated from its neighbor by a sheet of two- pound lead, which served as a conductor. Its internal resistance varied from 0.01 to 0.042 ohm. Each packed plate was capable of absorbing 50% of its weight of water. 54% of the total weight of the plates was active material. In the latest type of plate 2 porous unglazed earthen- ware is used, having square receptacles on its face side, and grooves on its reverse side, as shown in Figs. 30 to 33. The face of each plate is packed with red lead, filling the small squares, and rising ^ of an inch above the surface of the plate, so as to secure an agglomera- tion with the electrode during the forming process. The packed plates are then placed back to back, with the grooves crossed, as shown in the figure. As in the older type, two-pound sheet lead is used for the elec- trodes, and is placed between the packed surfaces of two adjacent plates. Fig. 33 shows a completed accu- mulator, bound by means of stout rubber bands between 1 a. p., 441,413 ; 1890. 2 a. p , 585.472-585,473 ; 1897. LEAD-SULPHURIC-ACID GENUS 53 FIG. 30. Fig. 32. Fig. 33. 54 THE STORAGE BATTERY rigid backs. By this means a circulation of the electro- lyte is obtained, and an escape for the gases formed is provided. The elasticity of the element allows for the expansion and contraction of the active material, with- out closing the pores of its own mass. The manufac- turers claim that as no amount of current in either direction can possibly injure the element, it is admirably adapted for central station purposes, and that the abso- 2.1 2.0 "1.9 j o > 1.8 1.7 1.6 I 1 i i i 4 t 6 ' i ! S 1 Fig. 34. lute confinement of the active material avoids deteriora- tion by loss of that important part of the battery. The internal resistance of the cell, at a io-hour rate, is 0.006 ohm. Taking account of the weight of the central conductor, it has been found that 65%, or, omitting this factor, 77%, of the weight of the plates is active material. The earthenware plates used will absorb 40% of their weight of water. In Fig. 34 is shown the discharge LEAD-SULPHURIC-ACID GENUS 55 curve of a 100 ampere-hour cell, discharged at a con- stant current of 10 amperes. WINKLER An element of the Winkler 1 battery consists of a series of troughs, of an acid-proof non-conducting mate- rial, such as celluloid or vulcanite, in a frame of the same material. The element is dipped in the active material, which is in a semi-liquid condition ; any excess of the active material being brushed off after drying. A lead conducting wire is placed in the bottom of the trough, and is embedded in the active material. The electrolyte is a gelatinous mixture of sodium silicate, sulphuric acid, and ammonium sulphate. The forming charge lasts 26 hours. Where lightness is not a prime desideratum, the element is made of lead, and the elec- trolyte dilute sulphuric acid. VAN EMON In this electrode, 2 a non-porous, non-conducting grid is perforated, usually with square holes. Perforated plates of lead are placed on one side of the electrode, the other side being open to the electrolyte. Ribbed separating frames are placed between the positive and negative plates. ROONEY In this plate, 3 a cut of which is shown in Figs. 35 and 36, the grid is built up of layers, or strips, of wool- felt, which are placed in parallel rows at right angles 1 A. P., 471,590-471,591-471,592 ; 1892. 2 A. P., 524,656 ; 1894. 8 A. P., 549,023-549,077 ; 1895: 574,826; 1896. 56 THE STORAGE BATTERY to each other, thus leaving square holes, or pockets, for the reception of the active material. In order to make metallic contact with the paste, perforated lead conduct- ing strips are placed midway between the faces of the plate. In the construction of the plate, sheets of wool-felt, painted on one side with a cement, which has been softened by heat, are cut into narrow strips, about ^ of an inch wide, and of the required length, these strips Fig. 3s. Fig. 36. being laid in parallel rows in a form. When the grid has reached half its thickness, the lead strips are laid in place, and when the form has been filled, heat is applied to soften the cement. The plates are cooled slowly under heavy pressure, which is maintained until the cement has hardened, after which the grids are lifted from the form. The grids are then pasted as though metallic, and when set, the lead strips are firmly embedded in the LEAD-SULPHURIC-ACID GENUS 57 active material, leaving no metallic surface exposed to the action of the electrolyte. Lead connectors are burned to the centre conducting strips, and the whole plate is surrounded by a perforated non-conducting retaining case. During expansion the paste presses more firmly into the felt, thus absolutely preventing all buckling. 2.3 2.0 1.9 1.8 3 4 5 6 HOURS DISCHARGE FIG. 37. This grid will absorb 230% of its own weight of water. Mr. Rooney has thus succeeded in obtaining a very light yet strong plate, and one which his been found to give excellent satisfaction under the most severe conditions. Fig. 37 shows a curve taken from a 5-plate cell, charged at a constant current of 7.8 amperes, and discharged through a constant resistance, the average current being 8.92 amperes. 58 THE STORAGE BATTERY THE UNION STORAGE BATTERY In this battery, 1 which was put upon the market in 1898, the plates consist of horizontal porous earthen- ware concave dishes, or saucers, filled with the active material, the active material being placed in the plates in the condition of a dry powder (see Fig. 38). The conductor, of thin sheet lead, stamped into a shutter- like form with lips and slots, is laid upon and pushed down into the active material. Extension strips, lead- FlG. 38. Fig. 39. ing from the conducting plates on alternate sides, reach to the top of the cell ; these strips are brought together, one-half on each side, and constitute the two terminals. The dishes, thus filled and arranged, are stacked on top of one another, and make up the complete element, as shown in Fig. 39. A porcelain or earthenware dish is laid on top, and the whole is bound together by vulcanite rods. 1 A. P., 534,603 ; 1895. LEAD-SULPHURIC-ACID GENUS 59 It will be seen that this arrangement effectually pre- vents the plates from short-circuiting, and that any tendency to buckle on the part of the conducting plates is prevented by the porous dishes. No strain of any amount can fall upon the dishes, as the conducting plates are thin, and the associated active material is in a more or less plastic condition. Professor Langley, 1 of the Case School of Applied Science, tested this cell, and found that after less than 250 ampere-hours' formation, at a 12.5-ampere rate, the cell gave 3.36 ampere-hours per pound of total weight, at a 5-hour rate ; the internal resistance, when charged, being 0.007 ohm. 2. Devices for the Retention of Paste (1) Grooves or Recesses THE BRUSH GRID To Charles F. Brush have been granted American patents 2 covering the broad principle of mechanical 1 Professor Langley gives the following test of one of the cells : Weight of element 11.5 pounds. Weight of cell, complete, with electrolyte . 18.75 pounds. Height of cell, over all 1 1.0 inches. Width of cell, over all 4.5 inches. Length of cell 5.0 inches. Discharge current 12.5 amperes. Average E.M.F. 1.86 volts. Time 5.0 hours. Capacity per pound total weight .... 6.2 ampere-hours. 2 A. P., 266,089-266,090-261,512 ; 1882 : 337,298-337,299 ; 1886. B. P., 3579; 1884. 6o THE STORAGE BATTERY application to the support plate, no matter in what form ; they also cover receptacles or perforations in the support plate. After long litigation, he was also awarded the priority of invention, in this country, of the pasted type of plate. The form of his support plate is shown in Figs. 40, 41, and 42. The plates are cleaned chemically, and immersed either in lead acetate or lead nitrate, and Figs. 40, 41, and 42. spongy lead is deposited electrolytically. They are then washed and placed in dilute sulphuric acid until the spongy lead has been converted into peroxide, after which they are ready for use. Another method of making the plates active is to fill the grooves with lead sulphate, place them horizontally in a solution of common salt and ammonia, with a zinc plate hung above, and short-circuit them. LEAD-SULPHURIC-ACID GENUS 61 TUDOR STORAGE BATTERY This accumulator, brought out by the Tudor Bros., 1 and manufactured by The Akkumulatoren-Fabrik Actien Gesellschaft, Hagen, is much used on the Continent, especially in Belgium, and to a small extent in this country, the American patents for this battery being controlled by the Electric Storage Battery Co. Rolled lead plates are grooved, as in Figs. 43 and 44. The thickness of the plate be- tween opposite grooves is 3 mm. for the posi- : fig. 43. Fig. 44. tive, and 1.5 mm. for the negative plate. The width of the grooves at the edge is 3 mm. for the positive, and 2 mm. for the negative plate; the distance be- tween the grooves at the edge being 1.5 mm. for 1 A. P., 478,661 ; 1892. B. P., 11,543; 1887. 62 THE STORAGE BATTERY both plates. The grooves are first coated with a thin layer of lead peroxide by electrolysis, and then packed with the usual oxides ; after which the plates are rolled, to key in the active material. By this treatment the formation of lead sulphate at the junction of the grid and the active material is reduced to a minimum, o > Char £6 f D/sche rsre 1.7 5 6 HOURS FIG. 4S . and the active material itself is less liable to scale and disintegrate. The internal resistance Varies from 0.015 t0 °-° 2 ohms. Fig. 45 gives the curve for this cell as obtained by Kohlrausch; the cell was charged at a constant current of 5 amperes, and discharged with a constant current of 6.5 amperes; the specific gravity varying from 1.115 to I-147. LEAD-SULPHURIC-ACID GENUS 63 POLLAK In this accumulator 1 the lead plates are covered with an electrolytic deposit of spongy lead, after which they are worked in such a manner as to produce the appear- ance of a short-bristle hair-brush. These hair-like pro- trusions are about 2 mm. long, with a space of about 1 mm. between the points. They are then coated with electrolytic lead, washed, and covered with a mixture 2.1 o > 2.0 1.9 3 HOURS Fig. 46. of lead sulphate and salt water. After being pasted, they are packed between zinc plates in an acid bath, and the sulphate is reduced to spongy lead. After the sulphate is thoroughly reduced, the plates are rolled so as to bend the hair-like rods, and key the active mate- rial in place. The forming charge lasts about 45 hours. 1 B. P., 7428; is G. P., 67,290-73,548; 1892. 6 4 THE STORAGE BATTERY Fig. 46 shows the curve of a 9-plate cell, total weight 24.75 pounds, which was discharged at a constant cur- rent of 18 amperes. WOODWARD In the manufacture of these elements 1 a lead plate is moulded with some substance, usually rock salt, which is afterwards removed. This leaves a plate having a highly porous surface. The active material is then pressed into the cells and pores. BUCKLAND An element for this battery 2 consists of a plain cast lead plate, containing holes and slots, in the latter of which are secured projecting pieces of some acid-proof, non-conducting material, such as ebonite. These pro- jecting pieces form, with the plate, horizontal troughs, in which the active material is placed in the form of a paste. By this means no part of the lead grid is exposed to the action of the electrolyte, it being pro- tected by the active material and by the supporting troughs. KHOTINSKY Figs. 47, 48, 49, and 50 show the plate brought out by Captain de Khotinsky, 3 of Holland; Fig. 48 showing the cross-section of the inner plates, and Fig. 49 'A. P., 392,373-392,374-393,954-393,955; 1888: 406,969; 1889. * A. P., 550,480; 1895: 556,660; 1896. 8 A. P., 345,511-347,231; 1886. B. P., 4490-4756; 1882: 3261-8416; 1885: 17,160; 1891. LEAD-SULPHURIC-ACID GENUS 65 that of the end plates. These plates are cast under a pressure of about 300 atmospheres. For a rapid dis- charge, the ribs are made much shorter than where a slow rate is to be employed, and the pos- itive plates contain more ribs than the negatives. For a slow discharge, the 8-hour rate, the positive and negative plates con- tain an equal number of ribs ; but in both cases the ribs on the posi- tive are thicker than those on the negative plates. The plates are covered on both sides with parchment paper, asbestos, or thin perforated celluloid sheets. Fig. 47. Fig. 48. Fig. 49. Fig. 50. LLOYD Robert McA. Lloyd 1 treats lead plates with a hot nitric acid bath. This produces a " honeycomb " struc- ture in the surface of the plate, which is then filled with active material. 1 A. P., 491,684; 1893. B. P., 8534; 1890. 66 THE STORAGE BATTERY (2) Perforations As stated before, the holes or perforations in lead plates may be of three general types : a, where the diameter is the same throughout ; b, where it is larger at the surface than at the centre ; and c, where it is larger at the centre than at the surface. Of these three general types, the second is undoubtedly the worst, since the plugs of active material, having nothing to hold them in place, tend to become loose and fall out, especially during a heavy discharge; while the oppo- site shape, or the last, is the best. Of these three gen- eral types, the Van Emon 1 may be used to illustrate the first, the Rooney, 2 the second ; and the majority of perforated plates the third ; although the Rooney may as correctly be classed with the third type, because of the methods employed to hold the active material in place. According to a United States Court decision, to James W. Swan 3 belongs the honor of originating the perforations which extend entirely through the plate. THE E.P.S. GRID The patents* taken out by Messers. Sellon, Volck- mar, King, Parker, Swan, and others, combined with Faure's 6 patents, cover what is known as the E.P.S. battery, manufactured by the Foreign and Colonial Electric Power Storage Co. of England. The grid in 1 Vide p. 55. 2 Vide p. 55. 8 A. P., 312,599; 1885. B. P., 2272; 1881. *A. P., 259,657; 1882: 321.759-324.597; 1885: 454.187; 1891. B. P., 4781; 1887: 24,127; 1892. 5 Vide p. 49. LEAD-SULPHURIC-ACID GENUS 6 7 its latest form is made of pure lead with perforations, the shape of which is shown in Fig. 51. A thin per- forated strip of metal runs across each aperture midway between the edges. Fig. 51. The paste, composed of minium and dilute sulphuric acid for the positive ; and minium, litharge, and dilute sulphuric acid, or a solution of magnesium sulphate for the negative, is pressed into the grid and dried. The plates are then hardened in dilute sulphuric acid, after which they are ready for forming. A strong current for 48 hours is required for the positive, and 24 hours for the negative plates. To prevent short circuits, a hard rubber ring is placed around the negatives, or glass rods are placed between the plates. The internal resist- ance of this battery at 9 amperes is 0.0045 ohm, and at 10 amperes, 0.0038 ohm ; 9 amperes being the normal discharge rate of the cell tested. Fig. 52 gives the results of a complete test of this battery. In a cell put upon the market by this company in 1897, for autocar purposes, and designed by Camille Faure and Frank King, 1 the weight, as compared with the E or EK type, has been reduced fully one-half. A grid very similar in form to the ordinary E.P.S. grid is used. This pasted grid is first wrapped in a cloth of 1 A. P., 501,728; 1893: 544,673-552,425; 1895: 568,447; 1896. 68 THE STORAGE BATTERY silicated asbestos, and then placed in a close-fitting envelope of perforated celluloid. The envelope is held to the plate by means of celluloid pins, which are 2.4 2.3 2.2 2.1 2.0 1.9 1.8 1.7 1.6 1.5 1.1 co 1.3 > 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 r ^- -Oh: rp;e r Dis char ge V -S, \ \ \ °'\ iS 'p li W CM rg,« v A -\ ^— — -~ \ ■s s k \ 3 i 6 7 8 9 10 HOURS Fig. 52. 11 12 13 14 15 16 17 18 100 90 80 70 O 60 S. 50fc ^ 30^ 20 10 cemented on either side to celluloid diamond-shaped washers, the pins taking different positions on the posi- tive and negative plates. EICKEMEIER In this accumulator, 1 a flat lead foundation is pierced with polygonal holes. (See Figs. 53 and 54.) The active material, packed around plugs, is inserted in these holes, 'A. P., 413.339; 1889. LEAD-SULPHURIC-ACID GENUS 6 9 and the plugs are afterwards removed. An insulating plate, having its holes considerably smaller than those in the leaden plates, is placed between the elements; this prevents the active material from overlapping and short-circuiting by contact with an adjacent plate. Each vertical line of holes constitutes a chamber containing A ,< fig. S3- the electrolyte ; the various chambers being connected by channels in the base plate (see ff, Fig. 54), so that the electrolyte has a constant circulation through the battery. The chambers are filled by means of a funnel which fits tightly in the feed aperture at the top. The lead grid is thus protected from all action by the elec- trolyte, except through the active material. SCHENEK-FARBAKY The positive plate for this type of cell 1 consists of a lead frame with a trellis formed of circular intersect- 1 A. P., 344,959-348,625; 1886: 359,248; 1887. 7o THE STORAGE BATTERY ing bars ; thus giving polygonal holes which are filled with active material. The smaller holes between two of the intersecting arcs are left open in order to regu- larly interrupt the continuity of the. packing mass. The active material for the positive plate contains 47. S % each of minium and litharge, and 5 % of coke, treated with dilute sulphuric acid (10-15%); that for the negative plate, 95% litharge, and 5% coarse powdered pumicestone, with dilute sulphuric acid (10-15%). The grids are filled with this paste and are partially dried. The surplus paste is then scraped off, the plates are entirely dried in air, are moistened with dilute sulphuric acid, and again air-dried; are placed in sulphuric acid for from 10 to 12 hours, and finally air-dried. The elements are bound between paraffined wooden rods. The internal resistance of this cell averages 0.00 1 ohm. VARIOUS TYPES In the Drake and Gorham 1 grid a double dove-tail, shown in Fig. 55, is obtained by passing the grid be- Fig. 55. tween rollers. The Jacquet 2 plates are shown in Figs. 56 and 57. In making these grids, Jacquet casts them entire from a white metal alloy. To facilitate removal, the plates are not burned together, but are bolted to !B. P., 3986; 1888: 17,655; 1895. 2 B. P., 18,028; it LEAD-SULPHURIC-ACID GENUS 71 72 THE STORAGE BATTERY \ \ 1 \ ] \ J 1 1 1 oil Ml °1 1 / h « = 1 * 3 C I j\J 5 C s c ? t 4 i g snoA LEAD-SULPHURIC-ACID GENUS 73 sanoH-3a3dwv M AiiovdVO 74 THE STORAGE BATTERY a cast white metal pole piece. The Gibson 1 battery, shown in Fig. 58, consists of a lead plate, whose per- forations contain buttons with enlarged heads. The active material is packed around these buttons. Figs. 59 and 60 show the D. J. Hauss plate. As will be seen, Fig. 56. Fig. 58. r Fig. 57. it consists of an ordinary perforated pasted lead grid. The active material is made by mixing sulphate of calcium, or the sulphates of other light metals with the litharge and alkaline solution, so as to form a plastic mass. This is tempered in a slightly acid solution, and then packed in the grids, and the pasted plates formed in a saline solution. Fig. 61 shows the curve for this plate, and Fig. 62, the relation between the capacity and discharge rate. 1 A. P., 388,668; 1888: 397,796; 1889: 439,240; 1890. LEAD-SULPHURIC-ACID GENUS 75 (3) Active Material surrounded by Conducting Material TOMMASSI The latest form of this cell 1 consists of a perforated conducting tube, filled with the active material, and 'containing a conductor B. (See Fig. 63.) An insulating plate A is placed at the bot- tom of this tube, which is usually made rectangular in form. The active material for the positive electrode is composed of a lead oxide, mixed with dilute sulphuric and phosphoric acids to form a paste. Precipi- tated or spongy lead is used for the nega- tive. Short-circuiting between two adjacent tubes is prevented by means of insulating perforated retaining walls. In the older forms the tubes were made of insulating material, and the conductor was given the Fig. 63. Ch irge D/sc'f arge 12 3 I 5 C 7 8 9 10 11 12 HOURS Discharge Curve of Bradbury-Stone Battery. Class B 2, (3). Fig. 64. 1 A. P., 454,091; 1 89 1. 7 6 THE STORAGE BATTERY shape of a screw, or rod with branching arms. The formation of the Tommassi elements is said to require 220 hours. CORRENS In this accumulator, 1 as will be seen from Fig. 65, the two frames are composed of lattice work, and are Fig. 6 S . so placed that no two interstices fall opposite one an- other. The two frames are connected by means of small rods. The active material, minium for the positive, and litharge for the negative, contain a lead silicate, neutral- ized with ammonium chloride, and magnesia or white clay. 1 G. P., 51,031; 1888: 52,853; 1889: 54,371; '890= 63,433; 1891. LEAD-SULPHURIC-ACID GENUS 77 FORD-WASHBURN In this accumulator, which is manufactured by the Ohio Storage Battery Co., 1 of Cleveland, Ohio, the ele- ments consist of a flat bar of lead for the positive pole, placed within a perforated conducting cell of sheet - lead ; the cell being filled with lead dioxide. The con- ducting cell is placed within a non-conducting porous chamber, usually of earthenware. The space between ZL=.~( ,-B u\ V C \b \A D 1 ! ! 4 1 H i Durs 1 ! >ischa rge ) l 1 l i 2 1 3 li Fig. 66. the lead conducting cell and the porous non-conducting cell is also filled with lead dioxide. Outside of the earthenware cell is placed another perforated lead con- ductor, having the space between it and the earthenware cell filled with litharge. These two lead conductors form a single element. The electrolyte consists of dilute sulphuric acid, containing, by weight, 1.49% of sodium sulphate. » A. P., 451,541; 1891: 1,233; "892- 7 8 THE STORAGE BATTERY Professors Langley and Mayberry gave this battery a thorough test. They found that during a two months' test, neither heavy charges or discharges, nor a pro- longed course of vibration, imitating that of a street car, produced any measurable deterioration of the bat- tery; the internal resistance during the entire test averaging 0.0048 ohm. Fig. 66 shows the curve obtained by them from a 5-element cell, the total weight of the battery being 70 pounds, and that of the ele- ments 42 pounds. The battery was discharged with a constant current, the rates being 10 amperes for curve A, 15 amperes for B, 20 amperes for C, and 30 amperes fori?. VAN GESTEL Van Gestel 1 uses a lead tube, perforated, and filled with active material. A lead-covered copper conduct- ing wire passes through the centre of the tube, which is bent upon itself, until it forms, practically, a square plate. JOHNSON AND HOLDREGGE In this battery, 2 perforated lead plates, which are ribbed only on one side, are covered on that side with active material. Two plates are then taken, and so bolted together that their ribbed sides are adjacent. Conical steel pins are placed in the perforations of the plates, before the introduction of the active material, and are removed when the plates are thoroughly dry, thus giving the electrolyte free access to the active material. 1 A. P., 358,092; 1887. B. P., 12,376 ; 1888. * B. P., 13,274; 1893. LEAD-SULPHURIC-ACID GENUS SOLA-HEADLAND 79 This plate 1 is composed of perforated rectangular lead tubes, into which the active material is pressed. The makers claim that since equal surfaces are exposed on the four sides to the current, buckling is impossible, even on short circuits. The ratio of the weight of the active material to the total weight of the plate is less than 50%. These batteries have been introduced in London for the propulsion of autocars. (4) Active Material surrounded, by a Non-conducting Material REYNIER ELASTIC Reynier, 2 in his latest sulphuric acid cell, has endeav- ored to so construct the elements as to permit them to expand or contract during charge or discharge, without damage to themselves. To accomplish this, the plates are , separated by porous sheets of silica, which, he found, only slightly increased the internal resistance. Each element contains one positive and two negative plates, and four sheets of silica, the end sheets being fluted to increase the available space for the electro- lyte. The containing vessel is made of pure lead, sur- rounded by an expansible corrugated case. The plates themselves are made from a very fine lead wire net, slightly compressed into the desired shape. In the latest type, the cells are composed of a num- ber of elements in series, placed between two rigid end 1 B. P., 15,120; 1892. 2 A. P., 438,827; 1890. F. P., 181,698; 1887. 80 THE STORAGE BATTERY pieces, which are drawn together by means of strong india rubber springs. Owing to the action of these springs, expansion and contraction of the elements can take place freely without causing disintegration. BARBOUR-STARKKY Mr. Barbour-Starkey 1 simply makes the cells of the ordinary form solid by filling in the space between the plates and the cell with a dry mixture of sawdust and plaster of Paris, in the proportion of 25 : 10, and then saturating the mixture with dilute sulphuric acid. A non-resinous sawdust is found to be the best. Mr. Barbour-Starkey claims that this method of treating cells will effectually prevent all warping and buckling, and will preserve the plugs of active material from being detached. Recently a battery of E.P.S. traction cells were treated in this way, and then put to regular work. A loss of current capacity and general ineffi- ciency was experienced, and the battery had to be ultimately abandoned. OERLIKON In this cell, brought out by Dr. Schoop, 2 the amount of active material on the grids is only about two-thirds as great as in the ordinary type. In the manufacture of the electrolyte, dilute sulphuric acid, specific gravity 1.250, is mixed with dilute sodium silicate, specific gravity 1.180, in the proportion of 3:1. When freshly made, this mixture is quite fluid, but it gradually solids 1 B. P., 15,754; 1887 : 7619; 1889. 2 A. P., 529,199; 1893. B. p., 7719; 1889. LEAD-SULPHURIC-ACID GENUS 8 1 fies, and in about 24 hours it becomes a hard, jelly-like mass, having a slightly bluish tinge. When bubbles of gas are formed, as they sometimes are, they simply push the gelatine aside and escape. When fully charged, a small amount of acid is forced out, and floats on top, the acid being again absorbed during discharge. This cell has received some very elaborate tests by Dr. Kohlrausch, at the Hanover University, showing very good storage capacity. Dr. Kohlrausch believes that, by the use of this gelatinous electrolyte, the cells may be in constant use for two years, giving their full current capacity, and that they may be used for another year with excellent results. PUMPELLY In this battery 1 the grids are placed horizontally. They are cast with stout legs on one side, so con-, structed as to bear the entire weight of the plate. The legs also serve as conductors between plates of like polarity. In building up a cell, the bottom plate is laid upon a foundation of cellulose, or wood-pulp fibre. The interstices of the grid are then filled with red-lead or litharge, according to the polarity of the plate. The requisite hardness of the active material is obtained by hand pressure. Upon this plate is placed another layer of cellulose or fibre, then another grid, packed, and so on. The cell is then filled with the electrolyte until the packing is thoroughly saturated, considerable free liquid being left in the cell. 1 A. P., 416,299; 1889: 442,390-442,391; 1890. G g 2 THE STORAGE BATTERY THERYC-OBLASSER In the manufacture of this battery 1 a perforated envelope of celluloid, or similar material, is filled, while in a plastic condition, with the oxides to be used, a core of lead-antimony being placed in the centre. The open- ing is then closed, and the whole is subjected to heavy pressure. The core is thus protected from all electro- lytic and chemical action. THE HESS STORAGE BATTERY This battery 2 differs from other existing types, through the employment of a double electrode. In the construction of the battery, lead plates contain- ing square perforations are used. These perforations and one side of the plate are covered with an extremely porous non-conducting material, composed of quartz sand, held together by asphalt. Two plates are placed side by side, about ^ of an inch apart, with their exposed lead surfaces facing each other, and so arranged that the exposed lines, both vertical and horizontal, are half a space removed from the corresponding lines on the other plate. The plates are provided with projecting ribs, and are cemented around the edges, thus forming a pocket for the introduction of the active material. The elements are then assembled, the double electrode representing one element. Hard rubber strips, with buttons at 1 A. P., 500,978-502,643; 1893. B. P., 5059-24,834; 1895. 2 A. P., 525,017-525,018; 1894. LEAD-SULPHURIC-ACID GENUS 83 intervals of 2 inches, are used as separators, leaving a clearance space of ^ of an inch between the electrodes. The assembled elements are now placed in the con- taining cell, and the electrolyte is introduced, after which they are ready for the introduction of the active material. This is accomplished by means of an appli- ance called a conveyer. The conveyer forces the active material into tubes or conductors, which register with the pocket of each electrode. The internal resistance of this battery, as obtained by Houston and Kennelly, varies from 0.0038 to 0.008 ohm. The manufacturers claim for this battery, that it is the only one in which there are no exposed metallic surfaces ; that the conducting plates are protected from consumption ; that there is no possibility of the active material becoming disintegrated and falling out; that the active material obtains a degree of porosity impos- sible with other batteries ; and that buckling or warping is absolutely prevented. In a cell containing 15 plates, each 9 inches square, the total surface for each elec- trode is 7.875 square feet. THE ACME BATTERY In this battery, which was brought out by Kennedy and Diss, 1 each plate consists of a thin, slotted sheet of rolled lead, which is covered on both sides with active material. The active matter is held in place by per- forated plates of insulating material, placed on each side of the plate. Bolts at the corners hold all the elements together. 1 A. P., 482,043-482,044 ; 1892. 8 4 THE STORAGE BATTERY GULCHER Giilcher 1 uses a frame of parallel lead wires, around which he weaves elastic glass wool, the method of weaving resembling that of a wicker basket. The active material is held in a finely divided state on the lead wires by means of the glass wool. The plate is saturated with a concentrated solution of lead acetate and dilute sulphuric acid, and is placed between zinc plates, covered with filter paper, and placed obliquely 2.5 / Ch irg . — Dis cha ■%e HOURS Fig. 67. in salt water * or very dilute muriatic acid for forming. His patent reads: electrodes for electric accumulators, comprising a fabric made of lead threads as warp, and glass or quartz threads as woof, a frame of lead, and a covering of spun glass. These plates were tested for traction purposes by two large German firms, and it was found after three months of hard usage that the plates were uninjured and their IB. P., 6947; 1894. A. P., 562,396 ; 1896. LEAD-SULPHURIC-ACID GENUS 85 capacity unchanged. Fig. 67 shows a curve taken from a 40 A. H. cell charged with a constant current of 7.4 amperes and discharged through a constant resist- ance. The average discharge current was 7.58 amperes. KOWALSKI The Kowalski, or I. E. S. plate, which is manufac- tured by the International Electric Storage Co. (Ltd.), of London, consists of a perforated envelope of celluloid, in which is placed an electrode consisting of a number of antimonous lead wires. The space between the elec- trode and the envelope contains paper pulp and pulver- ized oxides of lead. These are soaked in an electrolyte, whose composition is kept a secret. The liquid is absorbed by the powder, and after the electrode has been immersed for 24 hours it is removed and dried, and is then a compact mass. This battery is used on 50 cars on a French railway for lighting. RIBBE This grid 1 consists of a sheet of rolled lead T a g of an inch thick, with elongated perforations. It is coated on both sides with active material, which fill the perforations and make a uniform layer. The grid is then covered on both sides with ribs of celluloid of a conical cross-section, about 4 mm. in breadth, so as to divide the surface of the plate into long narrow rec- tangles. Each pair of adjacent ribs on opposite sides of the grid are cemented by acetone solution at inter- 1 A. P., 553,596 ; 1896. 86 THE STORAGE BATTERY vals through holes in the grid. In order to further secure the active material in close contact with the grid, finely perforated plates of celluloid are cemented to the free surface of the ribs. Each plate then consists of a continuous lead core, a layer of active material on each side, and a close fitting cover of perforated celluloid, which is strengthened at intervals by vertical ribs, the ribs being cemented together through the grid. HASCHKE This plate is made from chemically pure sheet lead, perforated with holes \ of an inch in diameter and £ of an inch apart. The plates are first slightly disintegrated by chemical action, and are then pasted with the active material. The positive plates are about three times as thick as the negative plates. Each positive plate is enveloped by specially prepared insulating material. The plates are then assembled and enveloped by stout rubber bands, the plates being separated from each other only by the thickness of the insulation (^- inch). The positive plates rest on the bottom of the containing cell, the negatives being held 1 inch above the bottom by the compactness of the elements. The insulating medium of this battery is specially prepared cardboard, subjected to an electro-chemical process. Certain chemicals, the composition of which is a secret, are decomposed by electrolysis in a contain- ing vat, and the gases therefrom rise and saturate the insulation. A current of 35 amperes at 20 volts is used to vaporize or treat 45 sheets of this board, each sheet LEAD-SULPHURIC-ACID GENUS 87 being 26 inches by 16 inches in size. In Fig. 68 are given some interesting curves, A being taken from a 250 A. H. cell, and B and C from a 100 A. H. cell. The curve C was obtained during the burning of a f-inch hole through a 2-inch sheet of steel. In 1897 Mr. Haschke overhauled the old New York Accumulator Co.'s cells in the family residence and 2.7 2.6 2.5 2.1 2.3 [2 2.2 O 2.1 2.0 1.9 1.8 1.7 1.6 1.5 1.4 c a\ SP- \v _u \ \ \ \ \ s^ A B C 1 2 3 Tl * ME 5 6 7 8 9 19 Curve A Time in Hours Curve B Time in Minutes Curve C Time in Seconds Fig. 68. hotel of Mr. Potter Palmer, of Chicago. The small amount of active material remaining in the cells was removed; and after the plates had been subjected to heavy pressure they were repasted, and the positives were wrapped in the insulating material just described. The plates are now in excellent condition, and are be- lieved by many to be better than when new. 88 THE STORAGE BATTERY 3. Devices for Securing Better Contact BRUSH In his patents, Brush 1 specifies the use of pressure sufficient to " weld the packing mass " to the sup- port plate. In the manufacture of his battery, spongy lead is deposited electrolytically before the applica- tion of the active material. A current is afterwards passed, thus forming a coherent mass of the active material and the support plate, TUDOR In the Tudor 2 element, the plate is first oxidized, either wholly or in part, before the application of the active material. This method has the advantage that the plate itself supplies active material to replace that lost in the cell. PAGET In the method brought out by Dr. Leonard Paget, 3 and used by the MacReon Storage Battery Co., fused lead oxide is subjected to the action of a reducing agent, such as carbon mixed with nitre. This reducing agent is placed in the mould in such a manner that when the fused substance is poured in, metallic lead is produced by the reducing agent, thus giving a perfect union between the active material and the support plate. 1 A. P., 262,523; 1882: 264,211; 1882. 2 A. P., 413,112; 1889. 3 A. P., 397,607; 1889. LEAD-SULPHURIC-ACID GENUS VARIOUS METHODS 8 9 In addition to the three methods mentioned above might be given those due to James, 1 Pollak, 2 and Grout, Jones, and Sennet. 3 1 Vide page 50. 2 Vide page 63. 8 G. P., 21,376. CHAPTER IV i. The Prevention of Buckling RECKENZAUN These plates 1 are formed of tubes made from porous compressed active material (see Fig. 69). The tubes are placed in a mould, and pure molten lead is poured around them to form a plate \ of an inch thick. The space between the tube is -j^g of an inch, the diam- eter of the tubes being ^ of an inch, and their length 1 ^ inches. In another form of plate 2 an arc is passed between the plates and an arc car- bon, thus transforming the surface of the plate into peroxide ■ to any desired depth. By this method an amount of chemical combi- nation is produced in a few minutes that it would take days to produce by the slow process of electrolytic action. It has been found that Fig. 69. 1 A. P., 385,200; 1888. 2 A. P., 475,797; 1890. 90 THE PREVENTION OF BUCKLING 91 when the oxides are formed in this manner they do not show any disposition to scale or fall off when placed in an acid bath, or when subjected to electrolytic action. The object in view in both of these methods is to pro- duce a plate whose expansion will be in the direction of the axis rather than at right angles to it. HERING This battery a contains but four elements, two positives and two negatives, as in Fig. 70. The two outer plates are composed of solid blocks of lead peroxide, and the two inner ones of spongy lead, The peroxide plates are made by mixing dry powdered lead peroxide, minium, and lead carbonate or sulphate, with a solution of acetate of lead to a stiff paste. The paste is then pressed into a mould and allowed to dry. Plates of con- ducting material are placed against the flat sides of these porous blocks. Perforated strips of non-conducting ma- terial pass over both sides of FlG the electrodes, thus securing good contact between the conducting material and the porous blocks, and at the same time securing the ele- ments against short circuits. 1 A. P., 429,272-429,273-429,274; 1890. g 2 THE STORAGE BATTERY Instead of using the usual sulphuric acid electrolyte, Mr. Hering prefers sodium or potassium sulphate in the acid. By this means the local action is greatly reduced. This battery, although not on the market, is an excel- lent one, the buckling or warping being almost entirely stopped. THE BLOT ACCUMULATOR This plate, the invention of M. G. R. Blot, 1 is of the Plante type, and contains no pasted oxides whatever. Each plate is made up of several longitudinal coils of lead ribbon (see Figs. 71 to 75), the planes of which are perpendicular to that of the plate. The windings of these coils are alternately of embossed and corrugated lead ribbon wound round a shuttle, and so fixed as to be free to expand. The thickness of the ribbons and the shuttle varies according to the capacity of the cell and the rate of charge or discharge required. Only the lower ends are soldered to the frame. The plates are suspended in a somewhat complicated frame, in which they rest on glass, and nowhere do they touch the bot- tom of the cell. The manufactures claim: (1) the maximum electri- cal surface obtainable, the surface being one-third of a square meter for each kilogramme of plate ; (2) rapid charge and discharge, the cells being able to stand a rate of 18 amperes per kilogramme of plate, and hav- ing been charged at constant potential in one hour; (3) absolute immunity from buckling, even on short circuit ; (4) absolute and efficient conductivity between ia. P., 535,885; 1895. THE PREVENTION OF BUCKLING 93 the frame and the active material; (5) durability and low first cost. Figs. 71 to 75. MONTERD In this battery, the plates consist of concentric cylin- ders, one side of which is grooved, and contains the active material. g4 THE STORAGE BATTERY 2. The Preparation of Active Material SORLEY Sorley * prepares his active material by taking a lead oxide, treating it with sulphuric acid to produce expan- sion, and afterwards drying with heat. This, it may be noted, is the usual method of prepa- ration. HOUGH Hough 2 prepares his active material as follows : a mixture of dry monoxide of lead and sulphate of mag- nesia is first formed ; this is reduced to a paste by mix- ing with ammonium sulphate and water. The plates are coated with the paste, after which the magnesium sulphate is eaten away. SILVEY In one of his batteries Silvey 3 uses for the negative plate a paste composed of a low oxide of lead, finely divided metallic lead, and superficially oxidized parti- cles of metallic lead, and water; for the positive, sul- phuric acid and water, a high oxide of lead, finely divided metallic lead, and superficially oxidized- parti- cles of metallic lead. KRECKE In a note in the Elektrotechnischer Anzeiger? it was stated that Krecke has succeeded in producing a very hard mass by mixing lead oxide with tannic !A. P., 419,728-423,091; 1890. 2 A. P., 512,283; 1894: 535,541; 1895. » A. P., 538,628 ; 1895. " E. W., Vol. 28, p., 733. PREPARATION OF ACTIVE MATERIAL 95 acid and albumen of glue, or by mixing lead oxide with uric acid. THE ENGEL SYSTEM For the positive plate, 1 litharge is mixed with sul- phate of magnesia, washed grease, and hydrochloric acid. For the negative plate, litharge is mixed with calomel, or bisulphate of mercury and ammonia. These materials are applied to the plate and hardened by im- mersion in water for several days. The plates are then formed in a solution of common salt, after which they are used in the ordinary sulphuric-acid electrolyte. SCHAEFER-HEINEMAN These manufacturers 2 make a paste of glycerate of lead, mixed with certain fatty acids ; this is applied to a plate of the grid form. The positive plates are formed in an acid bath containing potassium perman- ganate, with a current density of 19 to 25 amperes per square metre. The negative plates are formed more slowly in an ordinary acid bath, without the addition of the potassium permanganate. GELNHAUSEN In the Gelnhausen or "lead-dust" accumulator, the active material consists of- lead-dust and powdered pumicestone, thoroughly mixed and moistened with water. As this material shows a decided tendency to set, the pasting must be done quickly. !B. P., 16,162; 1894. 2 G. P., 80,420-82,787-82,792 ; 1894. 96 THE STORAGE BATTERY GRUENWALD Gruenwald makes his active material by mixing pul- verized lead with linseed oil and borate of magnesium. Formation changes the oil into a resinous substance, which constitutes the binding material. CHAPTER V BATTERIES IN WHICH ONE OR BOTH ELECTRODES ARE OF SOME OTHER METAL THAN LEAD II. — Lead-Copper Genus The advantages peculiar to accumulators of this type are that they are easy and economical of construction, and that they keep their charge fairly satisfactorily. The voltage, however, is low, averaging about 1.25, and the capacity per unit of weight small. These batteries are not used in commercial practice, and are of little interest save for laboratory purposes. REYNIER Reynier used an ordinary lead cell in which the electrodes were placed horizontally, the electrolyte be- ing composed of copper sulphate, MASON Edward J. Mason 1 used metallic plates containing lead peroxide for the positive and iron plates coated with copper for the negative electrode. The electro- lyte was copper sulphate in a solution containing free sulphuric acid. SUTTON . In this battery a copper containing-vessel served as the negative electrode ; the positive electrode was amal- 1 A. P., 439.324 ; 189°- h 97 9 8 THE STORAGE BATTERY gamated lead, and the electrolyte was copper sulphate. As might have been foreseen, the copper vessel was soon eaten away. The electrodes were separated by wood. ERVING The cathode of this form of cell is of copper, con- nected to a sheet of zinc, and placed outside of a porous cup. The anodes are of lead with 2% of silver, placed inside of the porous cup. A paste of lead peroxide and aluminum is packed between the porous cup and the lead. The electrolyte consists of ammonia and acid bisulphate of mercury. III. Lead-Zinc Genus Wheatstone, in 1843, appears to have been the first to have advocated the use of zinc instead of spongy lead for the cathode. Since then many investigators have taken up the problem, and have introduced sec- ondary batteries of this kind. These have attained considerable prominence of late years, and are in com- mercial use to-day, although it is extremely doubtful if they will ever supersede the lead-sulphuric-acid genus in engineering work. By the use of zinc and lead peroxide, it was found that the E.M.F. was increased, but that new difficulties were introduced, principal among which was the eating away of the zinc by the electrolyte. Reynier has found, however, that if the zinc be either chemically pure, or thoroughly amalga- mated, local action is reduced to a minimum. The construction of the lead-zinc cell is economical, and it is the lightest, theoretically, 'of all types. The capacity LEAD-ZINC GENUS 99 in watt-hours per pound of working substance, is, ac- cording to calculation, 57% higher than in the case of the lead-sulphuric-acid batteries. BOETTCHER In 1882 Emile Boettcher 1 constructed a cell with thin corrugated lead sheets for the positive and ordi- nary zinc plates for the negative electrode. The lead plates were covered with a paste of lead oxide moistened with zinc sulphate, the electrolyte being zinc sulphate, 1 : 3. The E.M.F. was 2.2 volts. REYNIER M. Emile Reynier, in 1883, constructed a battery with four peroxide of lead plates of the Plants type for the positive, and three smooth sheet-lead elements, covered with chemically pure electrolyzed zinc, for the negative electrode, in an electrolyte of zinc sulphate. The total area of the positive active surface was 200 sq. dcm., and that of the negative 1 50 sq. dcm. The total weight of the cell was 17.16 kg., that of the elements being 9.6 kg. His cell gave 152 ampere-hours at a 6-hour rate, the average internal resistance being .02 ohm. The formation required 200 hours. BARKER In this accumulator amalgamated copper and zinc plates riveted together are used for the negative elec- trode, and leaves of lead foil, coated with graphite and clamped together, for the positive. An acid solution of !G. P., 21,174; 1882:23,916; 1883. IOO THE STORAGE BATTERY zinc sulphate forms the electrolyte. A minimum of local action is claimed for this battery. LUGO 1 In the Lugo cell a negative plate of zinc, coated with a lead oxide, and a positive plate of lead, also coated with a lead oxide, are used in a solution of borate of ammonium. EPSTEIN Ludwig Epstein 2 sets the cathode in rotation during charge and discharge. His cathode consists of zinc amalgam, on a copper wire net, which is fastened to a conducting shaft. The anode is lead peroxide. TAM1NE Tamine 3 places lead and zinc electrodes, similar in form to Plante plates, in a solution of iooo parts of concentrated zinc sulphate, 500 parts of sulphuric acid (10%), 40 parts ammonium sulphate, and 50 parts of mercuric sulphate. In constructing the anodes, 20 parts of electrolytic lead peroxide, 75 parts of lead filings, and 5 parts of resin are cemented together under a pressure of 300 atmospheres. THE RIVER AND RAIL SECONDARY BATTERY In this accumulator, which was brought out by Main i and Meserole, 5 the positive electrode consists of a num- ber of thin lead plates, fastened by lead rivets to thicker »A. P., 458,424-458,425; 1891. 2 A. P., 543,680; 1895. 8 B. P., 12,824; 1884. * A. P., 359,934; 1886: 401,289-401,290-401,291; 1889. 6 A. P., 359,877-361,660; 1886: 381,941; 1887. LEAD-ZINC GENUS IOI outside lead plates, all the plates containing a number of fine holes. These plates' are formed in the usual Plante manner; that is, by electro-chemical action. The negative plate consists of zinc amalgam, deposited electrolytically on a tray-shaped copper plate. These plates are made in a U form, the positive plate being hung within, and separated from the negative by a hard i, 6 ^ a 15 b^ 1) ~ 14 6 a \\ a- a = Volts \ 12 b- -b = Amp ires \ \\ \\ \ b a> 1 i 1 i ! HOI < RS C ISCH i &RGE ( l a i i i ! FIG. 76. rubber ring. The electrolyte contains zinc sulphate and mercuric sulphate. In an older form, the plates were placed horizontally. Fig. 76 shows a curve taken from one of these batteries, which contained 14 plates, and was built especially for traction purposes. The weight of the cell was 45 pounds, and that of the plates 27 pounds. 102 THE STORAGE BATTERY IV. Alkaline-Zincate Genus LALANDE AND CHAPERON This cell, which is known both as the Edison-Lalande and Lalande-Chaperon 1 accumulator, has copper oxide as one electrode, and almost anything that can be plated with zinc for the other; the electrolyte being either caustic potash or caustic hydrate. Although it is claimed that almost no local action takes place, it has been found that the zinc is actually dissolved in the solution. Opperman gives the following method for making a very cheap copper oxide electrode. He immerses car- bon plates in a saturated solution of copper nitrate for a short time, and then dries them. This operation is repeated until the plates are thoroughly saturated. They are then carefully dried and heated, slowly at first, until the copper nitrate is changed into copper oxide. Care must be taken that the copper oxide is not reduced to copper. These plates, he claims, are very porous, and consequently very active. They may be regenerated by washing with water, and air-drying for several days. With these plates, the voltage, according to Mr. Opperman, is 1.2 as against .0.8 for the ordinary electrodes. THOMSON-HOUSTON The Thomson-Houston 2 accumulator consists of a glass vessel which is divided into two sections by a po- rous diaphragm placed horizontally. In each section is placed a copper electrode parallel to the diaphragm, 1 B. P., 1464; 1882. 2 A. P., 220,507; 1879. ALKALINE-ZINCATE GENUS 103 and the cell is filled with a saturated solution of zinc sulphate. The capacity of this cell is claimed to be independent of the extent of surface of the electrodes, and only dependent upon the mass of the material to be acted upon. The process of charging might be described as the working of a gravity battery backwards. Zinc is deposited and copper goes into solution as cop- per sulphate under the action of the charging current. The duration of the charging action is only -limited by the amount of zinc sulphate present, and by the thick- ness of the copper element. When ready for use, after charging, this cell constitutes a copper-zinc gravity battery. DESMAZURE This type of cell was first experimented upon by Lalande before he became identified with the Lalande- Chaperon battery. He did not, however, obtain any practical results. Desmazure 1 then took up the prob- lem, and with the aid of De Virloy, Commelin, and Baillache succeeded in producing a secondary battery which gave fair results, but which, for various reasons, was not a commercial success. The problem of pro- ducing a commercial copper-zinc storage battery was finally taken up by Philip and Entz, who produced the cell now known as the Waddell-Entz accumulator. In the Desmazure battery, copper mud, compressed into blocks, covered with brass gauze, and placed in a parchment envelope, is used for the positive, and iron gauze for the negative element. This copper mud is 1 A. P., 345,124; 1886: 402,006; 1889. io4 THE STORAGE BATTERY formed into blocks under a pressure of about iooo atmospheres ; a very porous block of copper, of about two-thirds the specific gravity of sheet copper, being the result. The containing chamber is thin tinned sheet steel. The resistance of I sq. dcm. of electrode surface is about 0.35 ohm. The solution usually employed is composed as follows : Water 1000. o parts. Zinc 14467 parts. Combined potash 200.82 parts. Free potash 3!3-7 2 parts. In the Waddell-Entz 1 battery the positive element is composed of copper wire gauze, surrounding finely divided copper, and enclosed in cotton bags. The nega- tive plates are a network of fine iron wire, on which the zinc is deposited. By heating the cells to a temperature of 86° F., while charging, it was found that the copper oxide was not dissolved. While charging, zinc is abstracted from the solution, and deposited upon the negative plates. At the same time oxygen is developed, uniting with the positive element, and forming a copper oxide. During the dis- charge, the reverse action occurs. Zinc from the nega- tive element reenters into solution, and oxygen is abstracted from the copper oxide, till at length the couple becomes quite inert. It is claimed that the plates do not lose mechanical strength by the repeated chargings and dischargings. 1 A. P., 421,916; 1890 : 440,023-440,024; 1890: 461,823-461,858- 467.573; 1891:425,260; 1892. ALKALINE-ZINCATE GENUS BOETTCHER ' 105 On the bottom of an iron containing-vessel, but insu- lated therefrom, is placed a zinc plate. About this is a heavy layer of potassium solution in zinc oxide. A block of porous copper oxide, serving as the anode, is soldered to the iron vessel. The electrolyte is a 50% solution of potassium hydroxide saturated with zinc. As the zinc is taken from the above layer, the copper is raised. The E.M.F. is 1.1 volts, and the internal resist- ance 0.5 ohm. schoop 2 The anode consists of 64 vertical copper rods, 8 mm. in diameter. The two ends are provided with a cover- ing of magnesia, the bottom covering serving to insulate the electrode from the containing-vessel, and as a sup- port for the surrounding diaphragm. A parchment paper, or cotton-wool envelope, is fastened to the mag- nesia rods by cotton-wool threads. The steel contain- ing-vessel is divided into 64 square cells, by steel plates. In these stand the copper rods, their tops being con- nected to a copper plate. Zinc is deposited on the steel sheets. The containing-vessel, together with the zinc deposit, serves as the cathode. V. Miscellaneous Types THE MARX LIQUID BATTERY In this battery 3 the energy is stored, not in the elec- trodes, as is customary, but in the electrolyte itself, 1 G. P., 57,188; 1890. 2 B. P., 7711; 1893. 8 A. P., 440,175; 1890. I0 6 THE STORAGE BATTERY which is, therefore, termed electroline. The electrodes are ordinarily of carbon, 2 negatives to 1 positive. The electrolyte, or electroline, is made up as follows : Perchloride of iron 450 grammes. Water 900 grammes. Hydrochloric acid 500 grammes. The passage of a current between the plates causes the liquid to assume a greenish tint. It then turns yellow, and, finally, a yellowish brown. When the cell is fully charged, which is indicated by the color of the liquid, the electrodes are removed. In order to dis- charge the cell, it has been found best to use electrodes of varying conductivity, such as zinc or iron, in conjunc- tion with carbon, iron being usually employed in prefer- ence to the charging electrodes. When a highly porous carbon block is placed in the electroline, between the two metallic electrodes, and the outer circuit is com- pleted, the liquid decomposes, passing through the same series of colors, but in the reverse order, that it does during the charge. PLATNER 1 Zinc and carbon plates are placed horizontally in a concentrated solution of pure ferricyanide of sodium. When the cell is discharged, a pulverized coating of ferrocyanide of zinc is formed on the zinc plates, and the solution is reduced to ferrocyanide of sodium. The reverse action takes place during charge. 1 G. P., 81,494-82,100; 1886. MISCELLANEOUS TYPES 107 HMD 1 Iron electrodes, covered with tin or lead, and having apertures for the reception of the active material, are used. The active material is formed of Prussian blue and oxide of lead, a suitable covering being placed over the electrodes in order to retain the active material. BASSET In this cell 2 each electrode is formed of carbon, cov- ered with peroxide of iron and wrapped in blotting- paper, and the electrolyte is a solution of protochloride of iron. The containing-vessel is lined with a mixture of wax, paraffin, and pulverized colcothar. TAULEIGNE Tauleigne uses carbon in a porous cup as the negative electrode, lead chloride being packed firmly around the carbon. Carbon also surrounds the porous cup, thus serving as the positive element. The electrolyte con- sists of a 60% solution of protochloride of iron. KALISCHER In 1885 Dr. Kalischer 3 brought out a cell which was intended to overcome the usual disadvantages of lead accumulators. As an anode, he used iron, and as a cathode amalgamated lead. The electrolyte was a con- centrated solution of nitrate of lead. The iron resisted the corroding action of the solution. The E.M.F. of the cell was 2 volts. 1 A. P., 271,628; 1883: 294,464-296,164; 1884. 2 A. P., 306,05.1 ; 1884. 8 A. P., 311,007-311,008; 1885. I0 8 THE STORAGE BATTERY MALONEY The electrodes 6f a cell described in a patent issued to J. F. Maloney, 1 in 1883, are composed of black oxide of manganese and carbon, the electrolyte containing ammoniacal salts. HOLLINGSHEAD 2 The positive plate of this cell is composed of dioxide of manganese, and the negative plate of iron or steel. The electrolyte is composed of water containing an iron salt, which, on decomposition, deposits an insoluble com- pound on the negative and a soluble compound on the positive plate. LEHMAN Lehman 3 places commercial barium superoxide, in a soft paste, on his plates, the electrolyte being a solution of barium chloride, barium bromide, barium iodide, or such an acid as will produce an insoluble, or nearly insoluble, barium salt, (as nitric acid (?), sulphuric acid, or phosphoric acid). DARRIEUS According to some German patents granted to Dar- rieus, 4 spongy antimony plates are used for the negative and lead peroxide, or oxidized antimony, for the positive electrodes, in a dilute, sulphuric-acid electrolyte. The advantages claimed, are : that sulphate is not formed on the negative plates from local action, that the mechani- cal strength is greater, and that the weight is less than is the case with ordinary lead elements. 1 A. P., 271,880; 1883. » G. P., 70,708-72,199; 1893. 2 A. P., 422,126; 1890. 4 G. P., 81,080. CHAPTER VI THEORY OF THE STORAGE BATTERY Although much has been accomplished in the direc- tion of the practical development of the storage battery, during late years, and much valuable information has "been obtained, yet, at the present day, very little as regards the precise chemical reactions which occur have been definitely settled. E. J. Wade 1 says: "It is probably for this very reason that storage cells are still so far from practical perfection, as compared with dynamo electric machinery, and other apparatus in whose development theory and practice have gone hand in hand." It should be remembered, however, that the chemical problems to be solved are exceedingly difficult. Dr. Frankland 2 called attention to this fact during the discussion of Professor Ayrton's paper on the " Chem- istry of Secondary Batteries." "The physical qualities of the cells are capable of very accurate estimation and investigation. But when you" come to attempt to ascertain the chemical changes that occur in the charging and discharging of a storage cell, you encounter formidable difficulties. The outsider has no idea of these difficulties. Nothing seems more simple than to determine the chemical changes that take 1 London Electrician, Vol. 33, p. 625. s London Electrician, Vol. 26, p. 177. 109 IIO THE STORAGE BATTERY place in either the positive or the negative plate of a storage battery. It is not so in reality. The substances used as active materials are in the first place mixtures, and the materials obtained at the end of the reactions are also mixtures, and these mixtures are insoluble in any reagent which does not decompose them. They cannot be volatilized; they cannot be subjected to any process of solution and crystallization in order to separate and purify their elements." The general theory of the storage battery is almost identical with that of the primary battery. It is subject to the same general laws, and is coupled up in the same way as the ordinary voltaic cell, and when charged, it becomes simply a primary battery. It, however, pos- sesses this immense advantage, in that when used up, its component materials can be brought back to nearly their original condition by passing a current through the cell. The general theory of the storage battery may be briefly stated as follows : During the discharge, both electrodes are converted into lead sulphate, with the extraction of sulphion from the electrolyte, thus reducing the density of the solution. The action on the positive plate is supposed to take place in two stages : first, the reduction of the peroxide to monoxide, and then the conversion of the monoxide into sulphate. On charging, the action is reversed, the sulphate being converted into peroxide on the positive and metallic lead on the negative plate. Many investigators believe that it is hydrated peroxide of lead (H 2 Pb 2 5 ), rather than lead peroxide (Pb0 2 ), which is formed on the positive plate. THEORY OF THE STORAGE BATTERY riI Many manufacturers now use lead sulphate as the active material in pasting both their positive and negative plates, instead of following the older method of apply- ing minium to the positive and litharge to the negative plates. The amount of electrical energy which can be thus stored by the conversion of the lead sulphate into peroxide, or hydrated peroxide of lead, as the case may be, is proportional to the amount of active material formed and capable of being acted upon. "Theoretically, the amount of energy which one pound of lead would generate, if wholly converted into lead sulphate, could be produced by a quarter of a pound of zinc or iron, or about half a pound of copper. Practically, the very property of lead which at present constitutes its superiority to other metals, that is, the insolubility of its sulphate, at the same time limits its efficiency, by reducing the energy obtainable per pound of metal to a small fraction of the amount theoretically possible. In the first place, the active material requires a grid or support of inactive material, which, even in the best form, will weigh nearly as much as itself, and in the Plantd type may be many times its own weight. Secondly, under the best conditions, not more than one- half of the active material is really acted upon, because the sulphate formed on its surface effectually screens the inner portions. The combined effect of these two causes is that not more than 5% to 15% of the weight of the electrodes is usefully employed." a The theoretical value of lead peroxide has been esti- mated by Plante" to be 4.48 grammes for one ampere- 1 London Electrician, Vol. 33, p. 605. 112 THE STORAGE BATTERY hour. A later investigator gives the value as 4.44 grammes per ampere-hour. The former value is prob- ably the correct one, as the majority of investigators have obtained that result. This gives approximately 100 ampere-hours per pound of active material. As- suming that the positive and negative plates are identi- cal, the result would be 50 ampere-hours per pound of peroxide and spongy lead. Plates of the highest capa- city do not, however, yield more than 16 ampere-hours per pound of peroxide and spongy lead. This difference is due to the fact that all the active material cannot be used, and that support plates of nearly equal weight must be employed. Faure has calculated that with a total thickness of 5 mm., the action penetrates to a depth of J mm., thus making 80% of the weight dead weight. With a greater porosity, of course, the per- centage of dead weight would be much smaller. Dr. Streintz * believes that the chemical energy in an accumulator is due to the sulphating, neglecting the secondary reactions, such as the absorption of hydrogen at the negative plate, the generation of free gases, and the formation of hydrated lead oxide'. This theory assumes as its foundation that metallic oxides cannot exist in the presence of free acid. Dr. Darrieus 2 has, however, come to the conclusion that the oxides can so exist. It has been found that when an acid acts on an insoluble oxide, the product itself being insoluble, the action is never complete. He believes that the sul- phate which is to be found on the positive plate after discharge is always variable in quantity, and is due 1 Wied. Ann., Vol. 53, p. 698. 2 L'Electticien, May 18, 1895. THEORY OF THE STORAGE BATTERY 113 only to the local action of the acid on the oxide, and that it is never included in the principal reactions of the discharge. If the chemical reactions occurring during the charge and discharge of a cell were exactly the reverse of each other, then the E.M.F. of charge and discharge would be the same. Professor Ayrton 1 has found that the E.M.F. for about two-thirds of the charge is very nearly 0.14 volts higher than that during the corresponding 30 40 so ampere-hours Fig. 77. periods of discharge, and that from this point onwards the difference continually increases. Even after full allowance has been made for the internal resistance of a battery, the E.M.F. of charge is always higher than that of discharge. Fig. 77, which is taken from Pro- fessor Ayrton's paper, illustrates this ; curve a shows the variation of E.M.F. during the charge, and curve b that during the discharge. Curve b is expressed in rela- tion to the ampere-hours contained in the cell, the point at which the discharge is stopped being assumed to cor- 1 Jour. Inst. Elec. Eng., England, Vol. 19, p. 699. II4 THE STORAGE BATTERY respond to emptiness. It is, in reality, therefore, the ordinary curve of discharge plotted backwards. These curves were taken from an E.P.S. cell. The persul- phuric acid which is the primary product at the positive electrode during the charge is, according to Wade, the cause of the high E.M.F. It is now generally believed that there is no possibility of doing away with the waste reaction in the formation of persulphuric acid. It is interesting to note here that Dr. Streintz 1 has found that the internal resistance of a battery is a func- tion of the current. For small currents it is higher than when on open circuit ; while for strong currents it falls, and becomes less than when on open circuit. The internal resistance of the accumulator when on open cir- cuit increases with the number of discharges. Accord- ing to Schoop, the internal resistance is the smallest when the cell is about half discharged, after which it increases, sometimes reaching as high as 15 times its minimum value. In Fig. 78 are given curves showing the specific resistance of the acid of a storage cell at different strengths and temperatures. It will be seen from these curves, that while an acid, the strength of which corresponds to the specific gravity 1.250, has the least resistance, it does not give the highest voltage. The use of such acid possesses the disadvantage that during the discharge the specific gravity diminishes, and consequently the resistance of the cell increases and the voltage falls. The largest change that takes place in the electrolyte is, naturally, an alteration in the degree of concentra- 1 Zeit. fur Elektrotechnik, Nov. 1, 1893. THEORY OF THE STORAGE BATTERY "5 tion. "The proper proportion between the active hydrogen and that which appears electrolytically seems to bear some relation to the capacity of the plate. It does not, as Gladstone and Tribe suppose, vary inversely with the current strength, but it is highly probable that, for every plate, there exists a current density for which this proportion becomes a maximum." 1 9 E x o z 5 < I- - i UJ a. I* o 111 ■//, / y '/4X a. O z UJ 0. 2.1 O z O CO > 1.9 • ~^JJ« Zona m Electrician, XXXV, G87. i 8 12 16 20 INTERVAL IN HOURS AFTER CHARGE Fig. 80. 24 mixture of yellow and puce-colored lead oxides, while on the other parts a white substance which is easily detachable is deposited. Messrs. Gladstone and Hib- bert have further proved conclusively that the E.M.F. of a cell does depend, in some way, upon the strength of the acid employed. Fig. 79 shows the result of their observations. In Fig. 80 a set of curves is given, showing the fall of the open-circuit voltage for 2 1 hours after the completion of the charge, and the influence of different strengths of acid. H8 THE STORAGE BATTERY In a communication to the American Institute of Electrical Engineers, in May, 1894, Mr. Griscom states that the potential of a cell is partly due to the degree of charge of the positive, partly to that of the nega- tive plate, and partly to that of the electrolyte. If a negative plate be taken from a fully charged cell, which indicates, for example, 2.65 volts with the normal charg- ing current, and be coupled with the positive from a partly discharged cell, indicating 1.9 volts at the normal rate of discharge, it will be found that the resulting E.M.F. lies between the two. If the couple be removed to other electrolytes, it will be found that the E.M.F. will rise or fall according to the greater or less density of the new solution as compared with the old. From this it will be seen that the measurement of the poten- tial difference at the terminals of a cell is not a sure indication of the charge, unless both plates are equally charged, and this is a condition which rarely obtains in practice. It will be found, however, that the variations of the specific gravity of the liquid are approximately proportional to the useful capacity of the cell, if we take account of the local action, the short-circuiting, the changes in temperature, and the general sulphating of the positive plates while idle. The examinations of the characteristic curves of a cell show very plainly the variations of the E.M.F. It is found, for example, that, in a storage battery whose plates are of nearly equal capacity, the changes in the positive plate determine the characteristic curves of potential on discharge, and the changes in the negative those at the end of the charge. THEORY OF THE STORAGE BATTERY ng Although, according to the majority of laboratory tests, an accumulator possesses an energy efficiency of from 80% to 85%, yet, in commercial practice, an efficiency of more than 70% is seldom realized. Besides the regular transmission loss, about 15%, there is a loss due to leakage, to local action, to the cells not being in the best condition, and to several other causes which do not occur in the laboratory tests. Crosby and Bell divide the losses incurred into four groups : 1. The direct losses due to heating. 2. The losses due to local action between the sup- porting grid and the active material. 3. The losses due to local action in the active mate- rial itself. 4. The losses due to the unreversed chemical action. It is the last two sources of loss which are generally the most formidable. Of the losses due to irreversible chemical actions, a portion is ascribable to the produc- tion of irreversible chemical compounds, and a portion to the electrolytic action producing free hydrogen, oxy- gen, ozone, and hydrogen peroxide. It has been found that thick grids, with heavy plugs of active material of corresponding thickness, are the most likely to suffer from the various losses, except the first, because the chemical action in a large and dense mass of material is by no means uniform throughout, and consequently differences of potential probably occur between differ- ent portions of the same plug. The amount and character of the by-products formed is very largely determined by the rate of discharge and I2 o THE STORAGE BATTERY the working temperature of the cell. During the dis- charge of a cell, as has been stated, free oxygen, hy- drogen, ozone, and hydrogen peroxide are formed in the solution, and attack the plate, without materially assisting in the discharge. Other and more compli- cated substances are also produced, — basic sulphates and the like. Darrieus 1 has found that an antimony-lead grid and lead peroxide will give 1.4 volts, and that the potential difference between an antimony-lead grid and spongy lead in dilute sulphuric acid is 0.52 volt. It is a matter of common experience that if an accu- mulator be discharged slightly, before being allowed to rest for any considerable length of time, the local action will be increased. In order to avoid local action, contact between the conducting grid and the electrolyte must be avoided. This is accomplished by having an unbroken layer of peroxide on the surface. A slight discharge breaks or destroys this layer ; hence the increase in local action. If the materials used in the plates be pure, the electrolyte also pure and of the proper strength, and if the conditions favorable to the forma- tion of persulphuric acid be avoided, local action will be found to be greatly reduced. According to a test by Epstein, 2 the average loss of charge of several accumu- lators tested was only about 20% during a period of three months ; and Sir David Salomons s has found that of some discharged plates, which had been left idle for four years, the only fault was a bad color. In all other 1 L'Electricien, Nov. 17, 1894. 2 London Elec. Rev., Feb. 1, 1S95. 8 London Electrician, Dec. 15, 1893. THEORY OF THE STORAGE BATTERY I2 i respects they were as good as ever. " He feels assured that, if the discharge has not been too rapid, no harm would be done by allowing them to stand for years. The plates are, however, certain to buckle if they are not charged in the usual way. A sure method to prevent this is to brush the positives with a stiff brush, in order to remove the surface scale, after which they may be charged, but slowly at first." Gaston Roux 1 has also found that if an accumulator of the pasted type be charged to saturation, and then left on open circuit, that the local action will be very slight, and the cell will lose, at a maximum, not more than 6% of its capacity in three months. Many investigators, including Messrs. Gladstone and Tribe, were of the opinion that the local action between the active material and the support plate in the positive electrode led to the disintegration of the latter. If this theory were correct, it would be advantageous to leave the film of .sulphate which covers the grid intact; and a battery should never be overcharged, as this would tend to decrease the film in question. Later investi- gators have, however, found that the film of peroxide which is formed on charging from the lead sulphate is the real protective coating. According to this view, overcharging is, to a certain extent, beneficial, as it tends to increase this coating, besides bringing ail the active material into the condition of lead peroxide. One of the most interesting phenomena in the dis- charge of a cell is the passing of current from one plate to another. It has been found that, with plates 1 London Electrician, Vol. 25, p. 754. 122 THE STORAGE BATTERY manufactured rigorously alike, kept in parallel, and subjected to the same treatment, the current variations often amount to more than 30%. It has also been found that the E.M.F. of different plates of a cell, connected in parallel and discharged through equal resistances, will vary from 1.6 to 1.85 volts. This gives rise to a rather peculiar phenomenon, that of the exchange of charge. It had been previously held that if one plate had less capacity than another, that at a certain point it would cease to discharge ; but that its E.M.F. would be the same as the rest, and conse- quently there would be no flow of current. Mr. Griscom has, however, found that, on breaking the circuit at the end of the discharge, it was hours before the batteries reached equilibrium, owing to the considerable flow of current which passed. He explains this by saying that "the deficient plate keeps on discharging at a lower rate than the perfect plate, and finally reaches a much lower point of discharge. On interrupting the current, the plate which has not been discharged so completely rapidly recovers a higher voltage than the other, and therefore discharges into it. This effect will also take place in different parts of the same plate, and may be a cause for the formation of peroxide on the surface of a negative plate after discharge." 1 Mr. Griscom also found that, when left at rest, the positive plates in a section will discharge themselves. Sir David Salomons, in a communication to the Ameri- can Institute of Electrical Engineers, in May, 1894, stated that " there are probably two causes for this : J Trans. A. I. E. E., Vol. 11, p. 302. THEORY OF THE STORAGE BATTERY 123 first, the slight leakage which exists in every installa- tion ; and secondly, a leakage in the cell itself, apart from any local action which may take place in conse- quence of the material employed in building up a sec- tion." Many users of the storage battery have found that the addition of caustic soda, or of sulphate of soda, greatly reduces the slow automatic discharge. Indirect evidence as to the nature of the chemical changes taking place in an accumulator may be derived from an examination of the curves representing the variations in the temperature of a cell during charge and discharge. In both cases, if all the reactions were absolutely electrolytic, no heat would be generated, except that due to the internal resistance of the cell. Any rise in the temperature, therefore, except that which may be accounted for in this manner, must be ascribed to wasteful, " unelectrolytic, heat-producing action," including local action. Curve a, Fig. 81, shows the rise in temperature during the charge, and curve b the fall during discharge. Upon calculating the amount of heat liberated, to which the fall in tem- perature corresponded, it was found to far exceed the internal resistance of the cell, and in fact it was equiv- alent to 17% of the total amount of energy put into the cell while charging. Professor Ayrton has found that the working temperature of a cell is always above that of the air, even when its temperature is falling in discharge. In Fig. 82, taken from the same article, will be found curves representing the rise in temperature, in degrees per ampere-hour. Curve a represents the rise during charge, and curve b that during discharge. 124 THE STORAGE BATTERY As before stated, Duncan and Weigand 1 found that the loss of energy exhibits itself in two ways, one of which is a generation of heat. This rise in tempera- ture they found to be due to : i. The Joule effect. 2. The current set up by local action between the active material and the support plate. 13 1.1 5- = 1-0 0.9 l-O 0.8 ;s 0.7 0.6 0.5 1^- 12-3 i 5 6 7 8 Time in Hours from- Commencement of Charge and Discharge FIG. 8i. 3. The current set up by local action in the plugs. 4. The heat losses corresponding to the electrolysis of the solution. It has been generally held that the temperature of a cell varies with the internal resistance, and therefore 1 Vide page 116. THEORY OF THE STORAGE BATTERY 125 that the chemical or cooling effect will at times predom- inate, and at others the I 2 R, or heating effects; and that for every cell there exists a point where the two factors will balance each other, and beyond which heat- ing or cooling will ensue. The experiments of Ayrton, Griscom, and Reckenzaun prove that this so-called law very seldom holds. In one case, according to Recken- zaun, the temperature of the cell was at least 4° below 0.3 ;o.2 ;o.i ^ 10 30 30 40 50 60 70 Ampere- Hours Fig. 82. 80 90 100 110 120 that of the normal, while the internal resistance was at least double that of the average. Besides this, the heating effects cannot vary as the square of the current since the internal resistance diminishes as the current increases, and what is still more remarkable, the E.M.F. itself seems to rise. The average variation in capacity in a cell whose tem- perature ranges from 0° to 22 C. is about one-half of 1% for each degree change in temperature. 126 THE STORAGE BATTERY Since the tendency of the acid in the electrolyte is to form sulphate of lead, from both the spongy lead on the negative and the peroxide on the positive plate, the generally accepted theory at present is that of the direct formation of lead sulphate at both electrodes. Each molecule of the peroxide is supposed to lose an atom of oxygen, and each atom of spongy lead to gain an atom of oxygen. Two atoms, or molecules, of hydrogen sulphate are thus abstracted from the elec- trolyte to react with the peroxide or spongy lead, and their place is taken by two atoms of water. The reac- tion for the positive plate, according to this theory, is, therefore, Pb0 2 + H 2 S0 4 = PbS0 4 + H 2 + O ; and that for the negative plate is Pb + O + H 2 S0 4 = PbS0 4 + H 2 ; or, including both reactions in one equation, Pb0 2 + 2 H 2 S0 4 + Pb = PbS0 4 + 2 H 2 + PbS0 4 . Thus the final result of the complete discharge of a cell is to form lead sulphate and water by removing sul- phuric acid from the electrolyte and depositing sulphate of lead upon each plate. The above is, fortunately, a self-limiting process, since the sulphate is a poor conductor. All the peroxide is therefore not acted upon, and at the end of the dis- charge we have peroxide of lead crystals covered with a coating of sulphate. It has been estimated that not more than 50% of the peroxide is converted into sul- phate. THEORY OF THE STORAGE BATTERY 127 In Professor Ayrton's paper on the " Chemistry of Secondary Cells," to which reference has already been made, the following important conclusions were drawn by his assistant, Professor Robertson: 1. The particles of the peroxide soon get coated in the discharge with a layer of lead sulphate, which pro- tects the peroxide from further action. 2. The analysis also shows that a proportion of active material still remains at the end of the discharge. 3. The loose, powdery surface of the peroxide plate seems to be thoroughly converted into lead sulphate. 4. When the peroxide on the surface of the plate falls to about 31%, the cell very rapidly loses its E.M.F., owing to the inactive layer of sulphate, which impedes the action of the sulphuric acid on the active material beneath ; and also to the formation of peroxide on the negatives. The diffusivity of the acid is decreas- ing, and it has to penetrate further and further into the plate to find the active material. When the whole of the paste approaches the composition of 31%, the cell loses its E.M.F. entirely. 5. The action seems to take place most rapidly where the current density is the greatest; the plate becoming hard there from sulphate soonest during dis- charge, and oxidizing there the quickest during charge. Plante, and Gladstone and Tribe, have noticed the formation of lead peroxide on the negative electrode during the discharge of a battery, and have pointed out that when it commences to form more rapidly than it is reduced, the two electrodes will rapidly approach equilibrium. Since when the circuit is broken local I2 8 THE STORAGE BATTERY action alone can take place, the. peroxide on the nega tive plates will be reduced, and on making the circuit again the cell will once more give a current. In this way Messrs. Gladstone and Tribe account for the re- suscitating power of the storage battery, as well as for the rise of E.M.F. on breaking the circuit. If the positive plate of a lead secondary battery be examined microscopically, there will be found soft, porous crystals of a very dark color, which are proba- bly electrolytic peroxide. There will also be found some brilliant red crystals, probably Frankland's red sulphate ; also the yellow sulphate crystals, and finally the well-known white sulphate. The negative plate will show metallic lead, with one and sometimes two sulphates. The production of the diverse chemical products is probably attended by the production of different potentials, and the final E.M.F. of the bat- tery must therefore be a resultant with one or more chemical reactions predominating at various parts of the discharge. In Mr. W. W. Griscom's paper on " Some Storage Battery Phenomena," read before the American Insti- tute of Electrical Engineers in May, 1894, several very interesting curves of charge and discharge for both positive and negative plates were shown. An exami- nation of the curves for the negative plate of a Faure cell showed at the end less capacity than that for the positive plate. Within the working limits of charge or discharge the negative plate did not vary over 2% of potential difference. The positive plate showed a fluctuation of about 6%. The total fluctuation of the THEORY OF THE STORAGE BATTERY 129 cell after the first few minutes was 6% in discharge down to 1.9 volts. When a cell is supposed to be fully discharged, it is often noticed that at least 30% of peroxide is still to be found in the positive plate. From this it has been inferred that the negative plate has 30% less capacity than the positive. In one sense this is true ; but, more strictly speaking, it is probable that the negatives have the same capacity as the positives, but discharge 30% sooner. It is a well-known fact that a high rate of discharge is injurious to a battery. Messrs. Duncan and Weigand have found that when the rate of discharge is too rapid, acid is taken from the solution inside the plug, thus weakening the solution, which only gains acid by diffu- sion from the outside. This diffusion increases greatly as the capacity is lowered, the acid in the plug becom- ing very weak, and a phenomenon which was noticed by Gladstone and Tribe occurring. They found that at a certain dilution the chemical action changes, and a new compound is formed in place of the peroxide of lead ; also that the plate becomes greatly corroded. In Figs. 83 and 84 will be found curves showing the capacities per pound of active material for different specific gravities at the end of the discharge, and cor- responding curves for the voltage. These were taken from two Chloride cells, one having thin and the other thick plates, the rate of discharge for each being 0.45 ampere per pound of plate. The rate per pound of active material was approximately the same for both, but the rate per unit of area was nearly 40% greater K 130 THE STORAGE BATTERY for the thick plates. It will be found from an examina- tion of the curves, that the capacity practically varies inversely as the thickness. Curve a was taken from plates 0.24 inch thick, and curve b from plates 0.4 inch thick. After a certain portion of the active material in a plate has been converted into lead sulphate, further sulphation is attended with a production of normal ■5 "S 0. 4. S-S. >•■» •5 -a a 5 Q_ la < A B/ 'A ^B 1100 1200 1300 MOO 1600 1600 Specific Gravity of Electrolyte at End of Discharge FIG. 83. white sulphate. At this stage in the discharge, the diffusion of the persulphuric acid which remains over from the last charge, and its decomposition with the formation of hydrogen peroxide, leads to the produc- tion of peroxide of lead on the negative plate. The formation of such a compound on discharge explains the rapid fall of E.M.F. to be noticed at the close of the discharging period. When the discharge is very rapid, sulphation will THEORY OF THE STORAGE BATTERY 131 take place rapidly, and the sulphuric acid which is dis- tributed through the mass of the active material will be, to a large extent, withdrawn before it can be replaced by diffusion. It is evident that in such a case the E.M.F. will fall, and that as soon as diffusion can take' place, a higher E.M.F. will follow. We are able to explain thus the higher voltage of a cell after a period of rest. Before charging accumulators, the positive and nega- tive plates which have been formed, either by the Plante" «t a ^A ^B E \rf *tT 1100 1200 1300 1100 1500 1600 Specific Gravity of Electrolyte at End of Discharge Fig. 84. or Faure process, contain a sulphated salt of lead, usually termed lead sulphate (PbS0 4 ). During the charge the lead sulphate is changed into peroxide of lead (Pb0 2 ) on the positive plate, and spongy lead (Pb) is formed on the negative plate. The liberated sul- phion (S0 4 ), combining with the H 2 of the water, in- creases the density of the electrolyte, and the liberated oxygen combines to form Pb0 2 . Sir David Salomons 1 has divided the process into three stages, and considers it as follows : " The first stage indicates the discharged cell. The second indicates what might be termed the 1 Elec. Light Installations, 7th edit., V61. I, p. 100. I3 2 THE STORAGE BATTERY rational change which takes place, though in all proba- bility a number of equations would be necessary to represent what really happens between the first and last stages. In this stage molecules of water have been taken from the electrolyte, and an equal number of molecules of sulphuric acid added, thus increasing the strength of the acid solution. In the third stage — the charged cell — the specific gravity of the electro- lyte may decrease slightly." Positive. Electrolyte. Negative. 1st stage PbSo 4 H 2 S0 4 + H 2 PbS0 4 2d stage PbO H 2 S0 4 + H 2 PbO 3d stage Pb0 2 H 2 S0 4 + H 2 Pb Besides this, an additional chemical action takes place during the charging, gas being given off at the negative, and when charging is nearly finished, at both electrodes. Gladstone and Tribe, 1 who have investigated the sub- ject, say : "The principal, if not the only, function of the hydrogen of the water is that of reducing the lead com- pounds." Messrs. Streintz and Neuman, arguing from this fact, claimed that the occlusion of the hydrogen was the chief factor in the charge of an accumulator. Strecker has proven, however, that the charge is based chiefly on the reduction of the sulphate of lead, rather than on the occlusion of hydrogen. According to the occlusion theory, the oxygen and hydrogen, dissociated during the charging, are occluded at the electrodes, the hydrogen at the negative and the oxygen at the positive pole ; and these gases on recom- bining give the phenomena of discharge. This theory 1 Chemistry of Secondary Batteries, p. 48. THEORY OF THE STORAGE BATTERY 133 was strengthened by the fact that the activity of the Grove gas battery was known to be due to the recom- bination of the gases which covered the electrodes. According to this theory, the presence of sulphuric acid in the electrolyte is merely to give conductivity to the water. When dilute sulphuric acid is electrolyzed with plati- num electrodes, the dissociation of hydrogen and oxy- gen is accompanied by the absorption of energy ; which energy, less that lost in overcoming the resistance, is yielded upon the recombination of these gases. The same E.M.F. which is needed to overcome the affinity of the oxygen for the hydrogen will be developed when these gases recombine. An E.M.F. of approximately 1.5- volts is needed for electrolyzing dilute sulphuric acid. This result could be anticipated from the fact that 34,180 e.g. calories are liberated when oxygen and hydrogen combine to form water. According to the method given by Lord Kelvin for calculating the " volta- motive-force " from chemical union, this energy corre- sponds to 1.49.2 volts. Since the normal E.M.F. of a lead-sulphuric-acid accumulator is very nearly 2 volts, something more than the tension produced by the occlu- sion of hydrogen and oxygen is necessary to explain the E.M.F. of secondary cells. Of the hydrogen liber- ated during charging, only traces are found at the negative electrode. Dr. Frankland has also proven that neither oxygen nor hydrogen is occluded during charging. Although Plants, Dr. Oliver Lodge, and other firm believers in the occlusion theory had noticed the for- I34 THE STORAGE BATTERY mation of lead sulphate throughout the active material, and on the grid itself, they ascribed it to local action. Dr. Lodge believed that, to obtain the best results, the amount of acid in the electrolyte should be small. Many investigators, even at the present day, hold that water plays the most important part in the primary reactions of electrolysis. It is now generally assumed that the acid breaks up under the action of the current into hydrogen (H 2 ) at the negative and sulphion (S0 4 ) at the positive elec- trode. According to the modern views of electrolysis, some portion of the hydrogen sulphate in the solution already exists in a dissociated form as free molecules of H 2 and S0 4 . As soon as a difference of potential is set up between them, these molecules are attracted to either electrode, and their place is immediately taken by others. If this assumption be correct, it is easily seen that the water merely serves as a solvent, and as a medium in which dissociation can take place, rather than playing a direct part in the process. In 1878 Berthelot discovered persulphuric acid [H 2 (S0 4 ) 2 ], and showed that it was the primary product at the positive electrode when dilute sulphuric acid was subjected to electrolysis. Later Messrs. Robertson and Darrieus, working independently of each other, obtained the same result. Their experiments were conducted with ordinary lead cells in actual use, and under normal con- ditions. According to the theory advanced after these discoveries, the freed sulphion, which cannot exist in a free state, nor, in the case under consideration, enter into combination with the substances on the positive THEORY OF THE STORAGE BATTERY • 135 electrode, combines with the sulphuric acid, rather than with the water ; thus : H 2 S0 4 + S0 4 =H 2 (S0 4 ) 2 . It is this reaction which is the cause of the high E.M.F. which is required to produce decomposition. "For persulphuric acid is one of those comparatively rare compounds, termed exothermic, whose formation is accompanied by an absorption of energy, and which liberate energy in decomposition. This acid is very unstable, and almost immediately decomposes at the electrode, reacting with the water, and passing back to normal sulphuric acid, with the formation of hydrogen peroxide, and then of water and liberated oxygen " ; thus : H 2 (S0 4 ) 2 + 2 H 2 = 2 H 2 S0 4 + H 2 2 , H 2 2 = H 2 + 0. At the end of the discharge the positive plate con- sists of small particles of lead peroxide, covered with lead sulphate, the free sulphion combining with the lead sulphate, forming persulphate of lead [Pb(S0 4 ) 2 ] : PbS0 4 + S0 4 = Pb(S0 4 ) 2 . This reacts with the water, forming lead peroxide, and normal hydrogen sulphate : Pb(S0 4 ) 2 + 2 H 2 = 2 H 2 S0 4 + Pb0 2 . At the commencement of the charge, when there is a large quantity of lead sulphate present, it is probable that a very large percentage of the sulphion is absorbed in this manner, and that only a small part reacts with the hydrogen sulphate, forming hydrogen pe'rsulphate. I3 6 THE STORAGE BATTERY After a certain point in the charging has been passed, the latter reaction will increase, because less and less of the sulphate remains to be acted upon by the sul- phion, most of the sulphate having been converted into lead peroxide. When the platd has been thoroughly peroxidized, and cannot be further attacked, it is prob- able that the whole of the sulphion goes to form per- sulphuric acid, with immediate redecomposition and the liberation of oxygen. From this it would appear that the difference between the curves of charge and dis- charge, in Fig. 62, is in reality a measure of the amount of persulphuric acid formed. Unfortunately, no data are available that would enable it to be stated in abso- lute figures. On the negative plate, nascent hydrogen is liberated, and exerts a powerful influence on the salts of lead that are present, reducing them to the metallic state with the liberation of hydrogen sulphate : PbS0 4 + H 2 = Pb + H 2 S0 4 . When the reduction of lead sulphate is nearing com- pletion, the surplus hydrogen commences to pass off in a gaseous state, thus indicating the end of the charge. From these facts it would appear that there is no advantage to be gained by continuing the charging after the hydrogen or oxygen has ceased to be ab- sorbed freely, because the presence of some unoxidized sulphate, although it increases the internal resistance, rather impedes than promotes local action. On the other hand, it is absolutely necessary that the minium on the opposing plate should be thoroughly reduced, THEORY OF THE STORAGE BATTERY 137 because a mixture of peroxide and metallic lead is very conducive to the production of lead sulphate, thus increasing the resistance and diminishing the E.M.F. Griscom and Fitzgerald do not, however, believe that lead peroxide is present on the positive plate, but rather that the active material consists of hydrated peroxide of lead. In support of this theory, Griscom J says : " The material on the charged positive plate is com- monly called peroxide of lead, but it certainly differs from it in its ability to generate E.M.F., and in its appearance; and Fitzgerald has pointed out that its composition corresponds to the hydrated peroxide of lead (H 2 Pb 2 O s ). He further intimates that a higher oxide, such as perplumbic acid (H 2 Pb 2 7 ), may be present. The fact observed by Gladstone and Tribe that 34% more of oxygen was absorbed by the positive plate than could be accounted for by the production of peroxide of lead, becomes, by this means, explicable. This large percentage of oxygen has probably been used in converting the hydrated peroxide of lead into perplumbic acid. Their suggestion that it has probably been absorbed by local action between the grid and peroxide is utterly untenable. There is no such action, and if there were, the grid would not last through a dozen charges. The conversion of H 2 Pb 2 5 into H 2 Pb 2 7 would account for the abnormal rise of E.M.F. at the end of the charge, and if it be assumed that the H 2 Pb 2 7 is unstable, yielding ozone gradually, thus accounting for the odor of a freshly charged posi- 1 Trans. A. I. E. E., Vol. 11, p. 302. I3 8 THE STORAGE BATTERY tive plate, it would account for the steady fall of E.M.F. on interrupting the charging current." As shown above, persulphuric acid diffuses through the electrolyte, and undergoes decomposition, with the production of heat, into hydrogen peroxide and normal hydrogen sulphate ; the hydrogen peroxide being after- wards decomposed into oxygen and water. That per- sulphuric acid is formed during the charge, is indicated both by analysis and by the temperature changes occur- ring during that period. For the first two-thirds of the period of charging, the temperature is constant, but after that it steadily increases. This is exactly what would be expected; for, during the first two-thirds of the charging period, the amount of persulphuric acid in the electrolyte is nearly constant, but during the last third of the charge, the amount continually increases, until a point is finally reached where the persulphuric acid is decomposed as fast as it is formed. Although the persulphuric acid theory, which is practically the same as that advanced by Darrieus, is the generally accepted one at the present day, Messrs. Elbs and Schonherr, 1 who have investigated the subject, oppose it. With a specific gravity of 1.300, and a cur- rent density of 2800 amperes per square metre, they found that the yield of persulphuric acid was only 24% of the theoretical yield, while with a current density of 1 300 amperes per square metre, the amount could only just be detected. They claim that since the current densities in accumulators are much smaller than this, the amount of persulphuric acid formed would 1 Zeitschrift fur Elektrotechnik und Elektrocheraie, 1895, PP' 4 1 7 an d 468. THEORY OF THE STORAGE BATTERY 139 be so small as to have no effect in the production of the peroxide. According to their experiments, lead and lead sulphate are not converted into peroxide by solu- tions of sulphuric and persulphuric acids, no matter what the concentration. They find that a clean lead plate in such a solution is rapidly sulphated, without the formation of peroxide, and that a peroxide plate, in the same solution, is converted into sulphate with the evolution of oxygen. They admit that persulphuric acid is sometimes formed in accumulators, but they claim that its formation is accidental, and that it is only a secondary product. Although these results are important, they seem to have but little bearing upon the theory of Darrieus. It must be remembered that the experiments of Darrieus and Robertson 1 were made with lead accumulators in actual use, and under normal conditions, while those of Elbs and Schonherr were with platinum electrodes. That there are only traces of persulphuric acid in accu- mulators is no proof that it is not a primary product, since but little is actually known of the products formed in lead accumulators. There is no method known by which the potential difference of the electrodes in an accumulator, as compared with an auxiliary electrode, may be measured while the current is passing, unless we except the cadmium plate test of Appleton. Further, in order to disprove Darrieus' theory, it would be neces- sary to show, by repeated experiments, that his results were wrong. Since no proof has yet been given, it is probable that Darrieus' theory will remain as the gen- erally accepted one for some time to come. 1 Vide page 134. CHAPTER VII APPLICATIONS, — STORAGE BATTERY INSTAL- LATIONS Storage batteries are beyond question, to-day, a commercial part of the central station lighting business, and an important factor in the regulation at the power station. As remarked by Mr. A. E. Childs at a meeting of the American Institute of Electrical Engineers in 1895, "The great variations and fluctuations of the load on power circuits, especially those power circuits sup- plying trolley lines, are among the greatest difficulties which engineers have to contend^ against, and any appli- ances that will aid them to arrive at a satisfactory running of their station is looked upon with favor by them." Some engineers advocate the use of gas- engines, and others the use of two classes of ma- chinery, — one, the most expensive and of a type giving the highest efficiency obtainable, and the other, consisting of much cheaper and comparatively inefficient, although perfectly reliable, machines. A constantly in- creasing number of engineers, however, believe that in the use of storage batteries lies the true solution of this problem. As an example of how important a part accumula- tors are playing in central station development, it may be stated that of 189 new lighting installations in 140 STORAGE BATTERY INSTALLATIONS 141 Switzerland, 87 contain storage batteries. On March 1, 1897, of 265 central stations in actual operation in Ger- many, 77% employ continuous currents, and 80% of these use accumulators, whose total output is 3 1 % of the total power of the direct generators of these stations. The total power represented by the continuous stations is 59,160 kw., while that represented by the ordinary alternating current and three-phase stations is 19,087 kw. Of the 36 central stations belonging to the Association of Representatives of German Electric Supply Under- takings, representing Norway and Sweden, Denmark, Germany, and Austria, 25 use accumulators, ranging in power from 65 to 1746 kw. hr. The ratio of the gen- erator output to the output of the accumulators is as 124: 119. As shown in a previous chapter, the uses of storage batteries may be classed under four great heads : 1. To carry the peak of the load at maximum hours. 2. To carry the entire load at minimum hours. 3. To act as an equalizer or reservoir. 4. For the equipment of annex stations. These four principal heads include all or nearly all the uses for which a storage battery may be employed. 1. To carry the peak of the load. — In all systems of lighting, whether it be gas or electricity, there is a large percentage of the connected load, which is used for only a short period of the 24 hours. This period of maximum demand, it has been found, varies from 1.5 to 4 hours per day. It is to take care of this load that gas-engines and inexpensive machinery have been pro- posed. The Boston Edison station have given both 142 THE STORAGE BATTERY inexpensive machinery and storage batteries a thorough test, and have found that the batteries give by far the best results; so much so that in 1897 they placed an order for a fourth plant. In the Boston station, 90% of the total output is produced by means of multipolar generators driven by vertical triple expansion engines ; and yet the total capacity of this apparatus is not 50% of the maximum load of the station during the winter. If a steam plant be installed to take care of this 50% of the maximum load, which is but 10% of the entire output, measured in kilowatt-hours, it is evident that the station will find itself running with an extremely small load factor, and consequently with very low efficiency. 2. To carry the entire load at miniimim hours. — If a station have a very small "motor load factor," as is the case in many of the European stations, then it would pay, perhaps, to shut down during this period, thus saving one shift of men, and making a reduction in the boiler room expenses, from drawing the fires. In the majority of American stations, however, the period of minimum load is so short, — in the Boston station being only 6 hours, — that it is found to be more economical to charge the battery during this period of light load, than to draw the fires and to throw the work of the station upon the batteries. In a communication to the American Street Railway Association, Mr. McCullough gave a theoretical curve, (shown in Fig. 85), which illustrates these two preced- ing conditions. In this, the period of light load is found to be 5 hours, from midnight to 5 a.m., while STORAGE BATTERY INSTALLATIONS 143 during the remainder of the day, 5 a.m. to midnight, the steam plant is operating at full capacity. 3. To act as an equalizer or reservoir. — It is this use of the storage battery which will especially appeal to engineers, whether they be connected with lighting or power stations. In modern stations electricity is delivered both to the "bus-bars" in the central station, 4000 12 Midni 13 3 4 5 6 7 ght A.M. 8 9 10 1] 12 1 2 3 4 5 6 7 8 9 10 11 12 Noon P.M. Midnight FIG. 85. and to " bus-bars " in the annex stations, thus being delivered at varying potentials. During the period of light load, the drop will be scarcely noticeable, and the pressure at the lamps practically constant. As the load increases, extra dynamos are thrown in, until the maxi- mum load is reached. Where a station is not equipped with batteries, it is almost an hourly question as to what it will be necessary to do next; whether to start dyna- mos or to stop them, or to change "boosters." This 144 THE STORAGE BATTERY trouble is intensified by the uncertainty as to what the load is going to be at any given time. When, because of a sudden storm, a heavy load is suddenly thrown on the machinery, it is impossible, unless the station be equipped with batteries, to prop- erly care for this load for a considerable time. This point may be illustrated by means of Fig. 86, which shows an actual load curve — from the Philadelphia 12 i a Midnight A.M. 15(100 1 .11! Ps rts of J "4 ^ re ire en' erierg ' sbs dec Hi! etl is §s$ m gi\ en fro: nB att ;rk s \ iU h P Lrts of Dia sr. y/a % re pre sen Ler erg y ^i sb dec lil e t lis m %. gn 'en to 3at ero \ ? \ V/f, % n w 1 it" 1 f f m w 4 / V t> % Y •w 'fa 5000 A ■// \y fy \ (/ */ > ,f <> f ? §5 ^ ^ ^ S t S S ^ ^ t p m N 12 1 Moon P.M. TIME FIG. 86. 12 Midnight Edison station — for a day in May, 1897, when, because of a sudden storm, the load was thrown on very quickly, the peak of the load being higher than usual. It should also be noted that in case of a breakdown, the battery is ready to carry the entire load for the short time necessary to start up another unit. This is an advan- tage which will not appeal so much to an outsider, perhaps, as to a man intimately acquainted with station management. STORAGE BATTERY INSTALLATIONS 145 Where accumulators are used as storage reservoirs, it has been found that a floor space, approximately 10 yards square, will suffice for storing iooo kw. hr. Such a reservoir will cost less than those used for either air, steam, or water ; the depreciation will be less, and the life will be longer. The pressure, moreover, will be constant, thus enabling lamps of 2.5 watts per candle to be used, which represents a saving of 20%. More- over, steam storage is not a true reserve, only relieving the boilers, while batteries are a true reserve, delivering the electricity directly and under conditions of high economy. Many engineers claim that the great trouble with storage batteries is their low efficiency. In the Boston station, however, it has been found that the watts lost by the inefficiency of the battery is made up more than fourfold by the better efficiency of the steam plant. In Hamburg, the loss in the accumulators is only 1.2% of the average amount of electric energy generated during the year. Manufacturers will guarantee the efficiency of a battery, when operated in parallel with a dynamo, to be greater than 75%. Moreover, foreign experience has proven that the use of accumulators in- volves a saving of at least 15% in fuel. According to the New York Electrical Review for Oct. 16, 1895, a battery plant, costing $150,000, when worked in con- nection with the trolley system, will show a net increase of $100,000 in earnings. 4. For the equipment of annex stations. — In cases where a heavy load is required, at some distance from the central station, it has been the practice in the past 146 THE STORAGE BATTERY either to install a separate station at that point, or to run heavy feeders to the centre of distribution. The consequence of the latter alternative being taken has been heavy losses in the feeders, owing to the large cur- rents transmitted, and heavy interest charges for the con- ductors. Now, however, by the use of batteries, much smaller conductors are required, on account of its not being necessary to carry the current equal to the maxi- mum load, thus reducing the interest charges. Moreover, a high tension current would probably be employed, by which the line losses would be still further lessened. The battery in this case would act as a transformer, being charged in series and discharged in parallel. Manchester, New York, Boston, Brooklyn, and Phila- delphia afford notable examples of accumulator sub- stations, New York having two such stations. Another method which has been developed by M. Nodon is to use accumulators in place of resistances, particularly with arc lamps. In such cases a consider- able portion of the current is irrecoverably wasted in heating the rheostat, while if a battery of accumulators be substituted for this dead resistance, all the waste current may afterward be utilized. The plan possesses the further advantage of affording a much more con- stant light. Examples of Storage Battery Installations ZURICH The first example of accumulators being used in a railway power station was at the power plant of the Zurich-Hirslanden Railway, in Switzerland. The in- STORAGE BATTERY INSTALLATIONS 147 stallation consists of 270 Tudor cells, connected in parallel with the generator. When the line was first installed, complicated automatic switches were used, to regulate the number of cells in circuit, but in February, 1 895, these switches were removed, and the battery was connected directly in parallel with the line. The bat- teries have been in use for 2\ years, with as yet no signs 600- _ O 510 ■ 500 210 1 | 1 1 1 190 180 A i \V1 f» Vf \ ^ V / UX 120 2: no £ 100 \VS \ \\V \ ■ a N~ \\v . •1 P sW \ \" ■sW V .A \ \ s S \ > '>7 Tt 2 >N $ " -//, r, - + J \ '//. V; A, • 7L \v '//, A / , / \v VA A/ A ^ ■ti '// 1 I v w A ■//, 7 A// 1 u A, '/ M , A /. '/A 'A \ \ // / A f/'i 7 10 30 20 u< \ z ■( r V '/ 7/1 y K/ V V / •A V , 15 6 7 8 U 10 11 12 13 11 15 16 17 18 19 20 21 22 23 21 MINUTES Fig. 87. of deterioration. The coal consumption is 3.9 pounds per car-mile, which, when the smallness of the entire plant — a steam engine of only 90 H.P. being used — and the exceptionally heavy grades are considered, is extremely small. Fig. 87 shows a load curve taken from this plant. The line ab represents the battery load curve ; cd the generator load curve, and ef the voltage curve. It will I4 8 THE STORAGE BATTERY be noticed that the E.M.F. varies only between 535 and 560 volts, and that ^/ averages 85 to 90 amperes. DOUGLAS-LAXEY RAILWAY, ISLE OF MAN On this road the battery station is at Groudle, nearly midway between the two generating stations. A bat- tery of 240 Chloride cells is employed, capable of yielding, at 500 volts, 70 amperes for 9 hours, 140 am- peres for 3 hours, or 300 amperes for 45 minutes. The battery is placed in parallel with the line, charging or discharging according to the demands of traffic. By means of a " booster " in the station, it can be brought up to full charge at any time. During the winter, when only two cars daily are run, the accumulators furnish the entire power, being charged once a week. THE MOUNT SN^EFEL LINE, ISLE OF MAN Here 246 Chloride cells are used, capable of furnish- ing, at 550 volts, 176 amperes for 3 hours, 112 amperes for 6 hours, or 72 amperes for 12 hours. The battery is connected, and used in a manner similar to that of the Douglas-Laxey road, including the minimum winter load. CHESTER, ENGLAND In this installation the three-wire system is employed, and three direct-driven, shunt-wound, Parker generators are used, giving 184 amperes at 440 volts. Two batter- ies, each of 115 K-type E.P.S. cells, are employed, their normal capacity being 300 ampere-hours at a 5-hour rate. They will also give 240 ampere-hours at a 3-hour STORAGE BATTERY INSTALLATIONS 149 rate, and 60 ampere-hours at a 0.5-hour rate. These results are obtained with an electrolyte temperature of 55° F. Each cell contains 25 plates in glass jars. The cells are supported by oil-insulators on dry-wood bear- ers, and are placed in parallel rows of two tiers each. In each battery circuit, on the switchboard, is a regu- lating switch interlocked with a "booster" reversing switch. This reversing switch enables the "booster" E.M.F. to be added to that at the "bus-bars" on charge, or, on discharge, to that of the battery. All the switchboard connections are of copper, but all bat- tery connections are of brass, painted with an acid-proof enamel, blue for the positive and red for the negative connections. EDINBURGH Six direct-connected, bipolar, shunt-wound machines, with drum armatures, deliver current to the two sides of the three-wire system at 270 volts. Two similar gener- ators, of smaller size, used as balancing machines, de- liver current at 135 volts. The battery consists of 132 Crompton-Howell cells, 31 plate type, with a capacity of 1000 ampere-hours at a 5-hour rate. Lead-lined containing-cells are used, resting on glass oil-insulators on wood bearers. The battery is divided into two half- batteries, positive and negative, arranged in four rows of two tiers each, on cast-iron stands, the 8 " hospital " cells being on a separate stand. The 26 cells in each half nearest the middle wire are used as regulating cells, and are placed in parallel, with each other as occasion de- mands, the connections to the regulating switchboard ISO THE STORAGE BATTERY being by solid copper rods. The battery room is a well-lighted, well-ventilated room, with a fire-proof floor which is covered with acid-resisting asphalt. Provision is also made for a second battery, of similar size, in case of necessity. This method of connecting the central cells in paral- lel, while necessitating a more complicated system of connections than with the usual method of cutting out cells on the outer end, renders the manipulation exceed- ingly simple, and does away, to a great extent, with the troublesome charging of the back cells. By this arrangement of battery regulation, all of the cells in the battery are always used, which is not the case with back E.M.F. cells. In order that charging may go on at light load, when there is only a small drop on the feed- ers, the "hospital" cells are connected, four on each side of the system. Under normal circumstances, the battery is charged during light load, the charging current varying with the external load, so as to keep the engine load constant. During heavy load, the batteries are kept, as far as possible, idle, it being only in the case of a "peak," and when the generators are shut down, that the batteries are subjected to a high rate of discharge. MANCHESTER The electric-light works of Manchester, England, have recently added to their plant a storage-battery sub- station, situated about one mile from the generating station, the five-wire system being used. A motor STORAGE BATTERY INSTALLATIONS 151 generator, which carries the normal charging current of 300 amperes, is used to raise the E.M.F. from 410 to 590 volts. There are 224 cells divided into 4 series of 56 cells each, each battery being again subdivided into a main battery of 44 cells, and a regulating bat- tery of 12 cells, with a total capacity of 1250 kw. hr. The normal rate of discharge is 300 amperes for each battery, and the maximum, 600 amperes. The battery carries the entire load after midnight, the third shift of hands, from 10 p.m. to 6 a.m., being almost dispensed with. On Sundays, the battery carries the entire load up to 4 p.m., in addition to its regular night load, thus relieving the second shift of men. The plates are supported in the cells by means of glass hangers, and are separated from each other by solid glass rods. The containing cells are supported on iron and timber stands, the battery being insulated throughout with duplex oil-insulators. Connections are made to the switchboard by means of solid copper rods carried around the walls on porcelain insulators which are fixed to iron supports. BELFAST The plant in the engine room consists of four 120 I.H.P., tandem, double-acting, horizontal gas-engines, rope-connected to four 60 kw. Siemens' generators. There are, likewise, two 60 I.H.P., single-cylinder, double-acting, horizontal gas-engines, rope-connected to two 26.4 kw. Siemens' generators ; also, two 150 I.H.P., 152 THE STORAGE BATTERY four-cylinder, single-acting, high-speed, vertical gas- engines, direct-connected to two 72 kw. Siemens' gen- erators. Ignition is made by means of hot tubes, which are of ordinary wrought-iron, nickel-steel, and porcelain. For starting, the generators are connected to the bat- teries, and run as motors. The dynamos give 240 amperes at 240 volts, 220 amperes at 120 volts, and 300 amperes at 240 volts respectively. All the generators are shunt-wound, bipolar machines, with drum armatures. The two small machines have double-wound armatures, the windings being connected to separate commutators. By means of a plug switch- board, these windings can be connected either in series or parallel, so that the machines can be run at either 120 volts for balancing, or at 240 volts across the system. The three-wire system is used. The battery consists of 126 E.P.S. cells, of the 34-K, or heavy-discharge type, divided into two sets of 63 cells each; the normal capacity of the cells being 500 ampere-hours at a 5-hour rate. There are also 8 " hos- pital " cells, which can be used either to assist weak cells, or can be put in series with the main battery. The regulation in this station is accomplished by put- ting the cells at the middle-wire end of the positive bat- tery in parallel with cells at the middle-wire end of the negative battery, the other end of the batteries being connected through ammeters and switches to their re- spective "bus-bars." In addition, the neutral wire is shifted either toward the positive or negative end of the batteries. The plates are contained in lead boxes, which are supported, by means of glass oil-insulators, STORAGE BATTERY INSTALLATIONS IS3 on four rows of wooden stands. The connections from the battery consist of bare copper rods, supported by oil-insulators, suspended from the roof. BIRKENHEAD This station is arranged for the three-wire system, but uses at present only two wires with a pressure of 230 volts. Three 80 H.P. engines are direct connected to three 50 kw., shunt-wound, Crompton dynamos, with an output of 200 amperes at 250 volts. The battery consists of 136 E.P.S. cells, of the K-50 type, with a capacity of 750 ampere-hours at a 5-hour rate, or 400 ampere-hours at a i-hour rate. Lead containing-boxes are used in pitch-pine trays on glass insulators. The battery is subdivided into the main battery of 126 cells, and the " hospital " battery of 10 cells. Regulation is accomplished by cutting in or out the end cells, of which 19 at each end, or 38 in all, are connected to the regu- lating switches. The cells are arranged in three long rows of two tiers each, on pitch-pine bearers, supported by cast-iron standards. These stands rest on thick stone slabs, the slabs resting on a concrete foundation. Over this con- crete foundation, and surrounding the slabs, is placed a layer of roofing felt, then a layer of bitumen, and the whole floor is then covered with slate. This acid-proof floor is ridged to give drainage. The end cells are connected from row to row by cop- per rods, and regulating leads, also of copper, run over the cells in wood cleats on ebonite insulators, and are 154 THE STORAGE BATTERY supported by bearers resting on iron girders. Space is provided for a similar battery, when the change to the three-wire system is made. MERRILL In this station, which furnishes power for both light- ing and railway circuits, the generators are all run from the same shaft, the prime mover being water power. It will be readily seen that under such circumstances "BEFORE" 400 i>*. — n i /* M 1 / A V A n [ ^ 1 *~ — ■ u IJ 1^ \(\f )[ f \ i\ { y vy \J \/ / V ■ V — \ . _-^ . 12 3 16 7 S 9 10 11 12 13 14 15 MINUTES A=Railway Voltage B=Lighting Voltage FIG. 88. the voltage would be far from steady. The advantages derived from the introduction of storage cells are illus- trated in Figs. 88 and 89, the curves being taken before and after the installation of the batteries. Before the batteries were installed, only one car could be run dur- ing the heavy lighting hours, and then far from satis- factorily. The battery was installed for the double purpose of increasing the lighting capacity, and for regulating the railway voltage; water-wheel governors had previously been tried for this latter purpose, but STORAGE BATTERY INSTALLATIONS 155 were found to be unsatisfactory. The battery is situated in a sub-station, about f of a mile from the power station, and about {of a mile from the centre of distri- bution for the lighting service. The battery contains 240 Chloride cells, of 11-F type, with 500 ampere-hours capacity at a 10-hour rate. It is divided into 4 series of 60 cells each, which can ■AFTER" h/i /I s\h r~^ fv \A /)/ '1/N k\ 1 J- k IS r" ' L " / % \ is- 1 I V^ V V -I 12s ^ * 122 116 1 ! i i . MIN i JTES i ) 1 1 1 li Fig. 89. be connected 4 in series for the railway circuit, or 2 in parallel for the lighting circuit which is arranged upon the three-wire system. A variable resistance is put in circuit between the battery and railway, for tut- ting down the voltage when the battery is fully charged. Two large, double-throw, three-point switches connect the battery to the railway, or to the three-wire lighting system. HARTFORD The Hartford Electric Light Co. generate current at 500 volts, — three-phase system, — at the power station on the Farmington River, 10.8 miles distant from the city. This is raised by means of " step-up " transform- I5 6 THE STORAGE BATTERY ers to io.ooo volts, and is transmitted over six #o wires, — two in multiple for each phase, — 1200 kws. being transmitted in this way. At the Pearl St. station the pressure is reduced to 2400 volts, and the current is changed from three-phase to two-phase. Most of the energy received at the Pearl St. station is used for three purposes. It supplies the current to all the alternating current lighting and power system. It supplies the current to the rotary transformers at the State St. station for charging the storage battery, and it supplies the current to the three-wire, direct-current system. At the State St. station is placed a battery of 1 30 Chloride cells, 65 on each side of the system. Each cell contains 30 positives and 31 negatives, the plates being made in two parts, each 15.5x15.5 inches. Manchester positives and Chloride negatives are used in a lead-lined oak containing-cell. The normal ca- pacity is 8500 ampere-hours at a 5 -hour rate, a 1.25- hour rate being allowable if necessary. Twenty end cells on each side are connected to the regulating switches, the switches being mounted vertically, and similar to those installed by the Electric Storage Bat- tery Co. The diagram (Fig. 90) shows the station operation for Dec. 12, 1896. The line enclosing B and C spaces show the current taken from either the rotary con- verter or battery for line use. As shown, at midnight the consumption had fallen to 350 amperes, and ran along at this rate till 6 a.m., when it began to rise, and at 8.30 had an output of 1550 amperes. During STORAGE BATTERY INSTALLATIONS 157 this time the batteries had been charging from the converters (see A space). At 8.30 the battery took the peak of the load, and carried it for 1.5 hours. At noon the demand dropped from about 1400 amperes to 750 amperes for one hour, during which time Noon Fig. 90. 8 9 12 P.M. Midnight If. r. Ma. Eng., XXIII. -88. the battery was charging. From 1 to 3 p.m. the river current, as it is termed, was all required by the lines, and the battery remained quiet. At 3.30 p.m. the rotary transformers were shut down and the bat- tery took the entire load till 11 p.m., when it was shut down and charging began, the rotary transformers also furnishing all line current. I5 8 THE STORAGE BATTERY BOSTON In 1897 the Boston Edison Co. already had three battery plants in service, and were installing a fourth one. The First, Second, and Third generating sta- tions are situated at the points of an approximately equilateral triangle ; the distance from the First to the Third stations being 3800 feet, and from the Second to the Third 4200 feet. The First and Second are each connected to the Third by tie lines of 3,000,000 circular mills capacity, on each side of the three-wire system. The batteries are run in parallel with the generators. The first battery plant was installed at the Third, or main, generating station, 90% of the total output being generated at this station. The battery consists of 140 Tudor cells, the capacity being 4125 ampere- hours at 1.25 hours, 5148 ampere-hours at 3 hours, and .6940 ampere-hours at a 10-hour rate. In this station are five sets of " bus-bars," to give the different pressures as required ; therefore five sets of regulating switches are employed on each side of the system. The switches consist of 31 contacts to which 30 of the cells are connected in each battery, and are entirely automatic. The second battery is installed at the First station, and is the largest of the four batteries. It consists of 144 Tudor cells, the capacity being 8250 ampere-hours at 1.25 hours, 10,296 ampere-hours at 3 hours, and 13,880 ampere-hours at a 10-hour rate. The contain- ing-cells are lead-lined wooden boxes, each containing 37 frames. Each positive frame contains 16 plates, STORAGE BATTERY INSTALLATIONS 1 59 7 inches square ; and each negative 4 plates, 14 inches square; the plates being secured in their frames by soldered lead strips. The elements stand on glass plates set on edge, leaving a space of 6 inches below the elements for the sledge. Each cell contains 640 litres of acid, and measures 3 feet 10 inches x 3 feet 4 inches x 3 feet; the cells are arranged in six rows on three floors of the building, and stand on porcelain insulators. The floor of the room is made of slate. Connections from the cells to the switchboard are by copper bars, 0.5 inch thick, and from 3 to 6 inches broad. This station has three sets of "bus-bars," and consequently three sets of regulating switches on each side of the system. As in the other stations, the regu- lation is entirely automatic, being accomplished by 0.25 H.P. electric motors. The third battery is installed at the Fourth, or Scotia St. station, and is distinguished by being situated in a fashionable residence district, and is entirely a sub- station, containing no generating machinery whatever. It is 7000 feet from the First station, and nearly 11,000 feet from the Third or main station. The current is sent from the Third to the First, and from there to the Fourth station, at a high potential; the same generators charging the First station batteries as well as the Fourth. The battery consists of 140 Chloride cells, resting on four "petticoat" insulators, of which the capacity is 3000 ampere-hours at 1 hour, 4500 am- pere-hours at 3 hours, and 6000 ampere-hours at a 7.5- hour rate. The containing-cells are hard pine tanks, 18x22x38.75 inches, inside measurement, lined with 160 THE STORAGE BATTERY 5-pound lead. Each cell contains 21 plates, Manchester positives and Chloride negatives being used. Twenty- four cells on each side of the system are connected to the regulating switches by hard-drawn copper bars, 3x0.5 and 1.5x0.5 inches cross-section. These bars are supported on porcelain insulators, resting on wooden hangers, the hangers being supported from the floor by iron pipes. The battery room floor is 12 inches con- crete, with a regular cement finish. The fourth battery is at the Second station, and will have a capacity of 4000 ampere-hours at 1 hour, 6000 ampere-hours at 3 hours, and 8000 ampere-hours at a 10-hour rate. NEW YORK The Edison Electric Illuminating Co. of New York have at present three storage battery plants, — one at the 59th St. station, another at the 12th St. Annex, and a third at the new Bowling Green station. The generating plants of the entire system, with the ex- ception of that at the 26th St. station, have been shut down several times over night, or over Sunday, in order to test the practicability of running the system from one generating plant, with the aid of the 12th St. bat- teries, during the hours of minimum load. The results show that this is quite practicable, and with the aid of the Bowling Green station it is expected that consid- erable economy will be developed in this way. The 59th St. station contains the first successful storage battery installation in American Electric Light stations. The battery is of the English Crompton- STORAGE BATTERY INSTALLATIONS 161 Howell type, with a capacity of 3000 ampere-hours at 3 hours, or 2000 ampere-hours at a i-hour rate. In the 1 2th St. station, 150 cells, having a capacity of 8000 ampere-hours at 10 hours, 6000 ampere-hours at 3 hours, and 4000 ampere-hours at a i-hour rate, have been installed. Each cell contains 39 elements, the positive plates being of the Tudor and the nega- tive of the Chloride type. The battery is arranged in four rows of two tiers each, resting on wooden bear- ers supported by iron standards. By a system of tie feeders, this battery can be charged either from the 26th St. or the Duane St. stations. The 12th St. station contains, besides the battery plant, a complete steam equipment, which is run only during one watch of maximum demand. The battery supplements this, besides taking the entire load during the remainder of the day. Fig. 91 shows the diagram of connections for this station, and Fig. 92 that for the Bowling Green Annex. The Bowling Green Annex likewise contains 150 cells, 75 on each side of the three-wire system. In this case, each cell contains 33 elements of the Man- chester positive and Chloride negative type, in a lead- lined, poplar, containing-tank, which measures 40.75 x 20.5 x 30.5 inches. The tanks rest on four "petti- coat " insulators, which in turn rest on 6-inch vitrified tile. The capacity of the battery is 2000 ampere-hours at 1 hour, 3000 ampere-hours at 3 hours, and 4000 ampere-hours at a 10-hour rate. There are 20 regu- lating cells at each end of the battery, each of which is separately connected to a block on the ordinary regu- 162 THE STORAGE BATTERY STORAGE BATTERY INSTALLATIONS 163 1 64 THE STORAGE BATTERY lating switch. Two regulating switches are connected in multiple on both sides, thus enabling the battery to be discharged at two potentials, or to be charged and discharged at the same time. The battery room floor contains, first, a layer of con- crete 1 8 inches thick, then a layer of pitch and felt, then a series of vitrified hollow tiles, for conducting the feeding cables, then more pitch and felt, then a 12-inch layer of concrete, and finally a layer of vitrified white tile. This elaborate flooring is necessary because the battery is situated in the sub-basement of the Bowling Green Building, the floor of which is below the water line. The waterproofing on the side walls is retained in place by a 'wainscoting of slate, securely fastened to the brick walls at the top, and embedded in the concrete at the floor level. All connection between the cells, and between the regulating cells and the switchboard, is by means of copper bars 0.5 inch thick by 3 inches wide. The area of all joints is 18 square inches, and each connec- tion is made by means of two ^-inch bolts. These copper strips are supported on porcelain insulators, which rest on hangers, or horizontal iron beams. Fig. 93 illustrates the battery room for this station. BROOKLYN The Brooklyn Edison Co. have, at the time of writ- ing, one battery plant in complete operation, and are installing a second and larger battery. The former is located in what is known as their Second District STORAGE BATTERY INSTALLATIONS 165 1 66 THE STORAGE BATTERY station, located in Lexington Ave. The generating machinery is only run during the period of maximum load ; at other times, the station is used as a battery sub- station, in connection with the First and Third District stations, to which it is connected by tie lines. The plant, illustrated in Fig. 94, consists of 160 Chloride cells, containing Manchester positive and Chloride negative elements. The capacity is 8000 ampere-hours at 10 hours, 6000 ampere-hours at 3 hours, and 4000 ampere-hours at a i-hour rate. Thirty end cells, on each side of the three-wire system, are connected to the; cell-regulating switches, which are located in the battery room, and the main distributing board is placed on the main floor of the building, immediately under the "battery room. The regulating switches are oper- ated, as usual, by electric motors, connected thereto by worm gearing. Two 16-C.P. lamps are connected in each motor-control circuit, one lamp being located on the main distributing board and the other in the bat- tery room. When the regulating brush is travelling from one contact to another, these two lamps are in series, and burn dimly ; but when the brush is entirely on the contact, one cell having been cut in or out, one of the lamps is extinguished, and the other burns brightly, thus giving an indication of when the motor should be stopped. In addition to this, there is an indicator telling the number of cells in circuit, which works on the principle of the Wheatstone bridge. Charging is accomplished either by means of the machines at the station, or by means of tie lines from STORAGE BATTERY INSTALLATIONS 167 1 68 THE STORAGE BATTERY the other stations, a "booster" being used in either case, as necessary. : ,, rrr ari i The battery being installed is to be located at; what is known as the Citizens' station, and is to consist of 156 Chloride cells, with a capacity of i2,ooo } ampere- hours at 10 hours, 9000 ampere-hours at 3 hours>jand 6000 ampere-hours at a i-hour rate. . ) PHILADELPHIA The Philadelphia Edison Co. have lately installed the largest individual storage battery plant in the world, At Hartford the cells are larger, but they are fewer in number, and their total output is, therefore, less. The Philadelphia plant consists of 160 Chloride cells, containing 57 plates each, 31 x 15.5 inches. The capa- city of the battery is 15,200 ampere-hours at 10 hours, 11,250 ampere-hours at 3 hours, and 7500 ampere-hours at a 1 -hour rate. The plates are supported on glass sheets, in ash containing-cells, which are lined with 4-pound lead. Thirty cells on each side of the three- wire system are connected to the regulating switches by rolled copper bars, with a sectional area of 2 square inches. These are supported on porcelain insulators, which are mounted on channel irons hung from the ceil- ing. All joints are bolted together by four |-inch bolts, and are treated with Edison-Brown plastic alloy. All other connections are made by means of lead "bus-bars," reinforced with copper strip. All copper connections are painted with an acid-proof enamel paint. The regu- lating switches, which are operated by electric motors, STORAGE BATTERY INSTALLATIONS 169 are placed on the outside of the battery room wall. The main switchboard is located in the dynamo room, on : the second floor,, immediately under the battery room : . Figs. 95, 96, and 97 illustrate, respectively, the battery room, the switchboard, and the cell-regulators, for this station. On the battery room floor was first laid concrete, then three thicknesses of paper coated with pitch to render it waterproof, then chemical brick laid in cement for one-half its depth, the other half being filled with hot pitch. The aisles are graded so that all liquid will drain off in channels provided for that purpose. The cells are supported on eight double "petticoat," porce- lain insulators, set in vitrified tile, to raise them slightly above the floor. That portion of the main switchboard which is de- voted to the battery connections is supplied with the following instruments : 2 motor-control switches. 2 cell-regulating switch indicators. 2 ammeters ; for the cell-regulating switches. 1 ammeter ; for the battery control. 1 voltmeter with five-point switch ; to indicate pressures on the two "bus-bars." 2 cell-regulating switches (on the " charging-bus "). 1 low-reading voltmeter with 30-point switch (to indicate . the voltage of each regulating cell). 7 knife switches. These instruments are all in duplicate, one set for each side of the system. As a general rule, the battery 170 THE STORAGE BATTERY STORAGE BATTERY INSTALLATIONS 171 Fig, 96, 172 THE STORAGE BATTERY /008 juodi; ! 01 M A FIQ. 97. STORAGE BATTERY INSTALLATIONS 173 carries the entire load from 12.30 a.m. to 6.30 a.m., about 8000 ampere-hours being taken out; from 6.30 a.m. to 4.30 p.m. the battery is charged; at 4.30 the peak commences, and lasts till about 6 p.m. ; from then on till 12.30 a.m., the battery is charged, to be ready for the night load. Fig. 86, which was taken from this station, shows a load curve. This is not the normal load curve, however, because of a sudden thun- derstorm. In 1890, when the Union Traction Co. of Philadel- phia decided to extend their lines, it was found that either a new power-house, or an accumulator annex, would have to be built, as the needed addition to the existing feeder system would require such an expendi- ture for copper as to render it commercially impracti- cable. The cost of copper alone would have been about four times as great as the cost of a battery sufficient to meet all requirements. The erection of a new power- house was out of the question, because of the heavy operating expenses. The capacity of the power-house would have to be about 750 kw., and this, at an esti- mated cost of #85 per H.P., would £ost about $85,000. The total cost of the annex, including the changes in cables, was about $50,000. Before the extension was made, the pressure at the end of the system was barely sufficient to operate cars on schedule time, the pressure varying as much as 50%. This condition is clearly illustrated by the frontispiece, and by Fig. 98. After the installation of the batteries the only variation in pressure was between 1 and 5 a.m., when the battery 174 THE STORAGE BATTERY was taken out of circuit. The installation consists of 248 G-13 Chloride cells, in lead-lined boxes, mounted on two tiers of oil-insulators. All connections are made by continuous weld, no mechanical contacts being em- ployed. The maximum rate is 400 H.P. for one hour. Fig. 98. Figs. 99 and 100 illustrate, respectively, the battery room, and a load curve for this station. The follow- ing data regarding the operation of the installation may be of interest ; they are taken from a communica- tion to the Engineers' Club of Philadelphia, made by Mr. Hewitt in December, 1896. STORAGE BATTERY INSTALLATIONS 175 176 THE STORAGE BATTERY Fig. 100. STORAGE BATTERY INSTALLATIONS 177 The saving in favor of the battery for the year is *i 5.633.96- Highest charge for 24 hours 456,000 watt-hours. Lowest charge for 24 hours 312,000 watt-hours.' Average charge for 24 hours 385,161 watt-hours. Highest discharge for 24 hours 456,000 watt-hours. Lowest discharge for 24 hours 216,000 watt-hours. Average discharge for 24 hours 329,419 watt-hours. Average efficiency for the month 85.5% Maximum specific gravity at 6 P. M 1.201 Minimum specific gravity at 6 p.m. 1.184 Average specific gravity at 6 P. M '. . . 1.192 Maximum specific gravity at 12 p.m 1.194 Minimum specific gravity at 12 p. m 1.182 Average specific gravity at 12 p.m 1. 188 This last specific gravity is approximately that pre- ceding the night charge, and is an indication of the amount taken out during the day ; it should be borne in mind that at 1.160 the battery is practically empty. THE COMMERCIAL CABLE BUILDING, NEW YORK. The battery plant in this building consists of 120-2500 ampere-hour Chloride cells, whose internal dimensions are 20.5 x 26.5 x 20.75 inches. The containing-cells are made of ash. They are lined with lead and painted with asphaltum, and are mounted on glass insulators, resting on hardwood blocks on 4-inch I beams. The cells are arranged three tiers high on one side of the room and two tiers high on the other in order to economize space. Strong running boards are bracketed out from the frames for the use of the attendants. The plates in the cells rest on glass strips, which in turn rest 1 7 8 THE STORAGE BATTERY on leaden legs. Connections between adjacent cells are made by burning the plates to lead " bus-bars " ; all copper leads are dipped in lead, as is all steel work and all bolts and riveted joints, to protect them from the battery fumes. The battery currents are controlled by a 25-H.P., 75-volt "booster," driven by a 240-volt motor. The field of the " booster " is wound with a series and shunt coil. During the day, when the battery is being charged, the series field works against the shunt field, the battery terminal being outside the series coil, that is, between the series coil and the motor load. In this case, if the elevators take a heavy current, the " booster " weakens,, which prevents the " peak " being taken from the dy- namo, and throws it all on the battery. At night the generators are thrown off, and the battery terminal is changed to a position between the "booster" brush and the series coil. In this case the field connections work cumulatively; and if the elevators draw a heavy current, the series coil builds up the "booster" voltage, so as to compensate for the drop in the batteries. Thus the lamp pressure is maintained at a constant value. OTHER INSTALLATIONS For the electric railway in Rome, Italy, alternating current from the Tivoli-Rome transmission plant is sup- plied. The pressure is first reduced by means of a stationary transformer, and is then converted to a direct current by means of a motor-generator having a multi- polar field and a single armature. The armature has STORAGE BATTERY INSTALLATIONS 179 collecting rings on one side and a commutator on the other. The battery plant consists of 300 Tudor cells, with 720 lew. hr. capacity. Pressure is kept constant by means of an automatic switching-in apparatus. I~4£ S6S A K> %r ihHHI'I'HI'h O B.F.S. -104-V » N Etcc HWU, XXIII, »7. FIG. IOI. In Burnley, England, the three-wire compensating de- vice, shown in Fig. 101, is used. The dynamo generates 230 volts. The compensator consists of two armature coils wound on a single drum, each for an output of 200 amperes at no volts. The battery is of 120 cells, of about 500 amperes each. The bars represent regu- lating switches; thus balancing with only one engine and one dynamo. i8o THE STORAGE BATTERY S5- On the Zurichberg railway, in Switzerland, gas- engines are used in conjunction with a battery of ^~ 300 Tudor accumulators. The diagram, Fig. 102, shows the connections. "There is an automatic regulator ab, and further a hand regulator b', for the 90 control cells di- vided into 30 groups of 3 each. The cells are in parallel with the dynamo Z^; and D 2 is an auxiliary dynamo, whose field coils are placed between b and b', and which charges all cells beyond a. The excitation of Z> 2 changes with the number of cells in circuit. The inter- rupters are solenoids with mercury cups." Fig. 103 shows the load curve for Berlin, LH3)-0oLQ^. Elec. World, XXJ3ir=337 Fig. io2. in which a represents the current curve, b the watt- age curve, and c the charging curve. The maximum charging current is 650 amperes, and the maximum discharge current 1420 amperes. Fig. 104 shows the load curve for the electric plant in the Chicago Board of Trade, and Fig. 105, the switch- board connections. STORAGE BATTERY INSTALLATIONS 181 Secondary Batteries in Elevator Work A growing use for secondary batteries exists in con- nection with elevator work. Many office buildings have been compelled to install their own generating plant, because of the unwillingness of central station managers to accept the load. When the character of the load due to elevator service is considered, this unwillingness will be clear. Because of the large amount of energy con- 6 7 8 9 10 11 12 1 A.M. Noon 10 11 12 1 2 3 i 5 Midnight A.M. 3 i 5 C 7 P.M. Hours Fig. 103. sumed, the customers require that the current be given them at the lowest rates ; but because of the fluctuating character of the load, which is well illustrated in Fig. 106, 1 central-station men cannot afford this practice. Of late, however, storage-battery plants have been installed, thus changing the load from an undesirable to a most desirable one. The battery suffices for the entire ele- vator and lighting work, besides putting a steady load on the engines. In some cases the storage battery has 1 Electrical Engineer, New York, Vol. 23, p. 497. 182 THE STORAGE BATTERY ~F>a.rh 0/ S h,a,oLe.-3- Fio. 104. = "Paris of jp1a.4ra.rrt Sb.Orole.J- ikc -f hit 'Refrcsetft e-tteryy gi ten, h lift.lsm.ntl motors cLirect front tZgrrtnitjo ~ 7°MrT S o f JJlCLOfO. ij shuitil li Ae i T?ef resent energy ftvert-^io 6 o.iie:-ies fori}- cJTfteut»Lo .~Pm,r \ s ef -* 2l*2Z 27rt - Sha-tteeZ Zy&eiu'lis 7?e ires »■? e "■ergj g ivtvi io Izghi s, mo ('irs <*««? t\Zev* ?or-s titre : t w /y , > ra he. tteri es 55 «v;ih tutat 104. STORAGE BATTERY INSTALLATIONS 183 been installed and maintained at the sole expense of the owners of the central station, who have placed the h- 240 % l h~ ^ 608 576 1 1 I 1 2 3 |i Is MINUTES 1 1 1 8 9 1 1 1 1 ill 1 1 116 1ln~ I liiri ., j T Itirrfl LU S20 CC LU 1 ILL Hill 11 rl n |f]j 1 11 ili < 856 224 In nil |- 1 Fill 1 1 f n' ' ^ II H Fl Mil 1 1 160 128 96 64 11™ rl 1 |! -1 ' 11 III III iJJ. 1 i 5 MINUTES Fig. 106. meter between the battery and wiring mains in the building and have charged only for the current actually used by the customers. r 84 THE STORAGE BATTERY In some cases " recuperation," 1 as practised on the Paris Accumulator Lines, is being employed to advan- tage. In such cases the car, descending by gravity, charges the battery, thus restoring much of the energy used during the ascent. An interesting experiment is being conducted on the Manhattan Elevated Railway, of New York City ; that of carrying the battery plant on the train itself, rather than installing it at sub-stations along the line of the road. The battery consists of 248 Chloride cells of 400 ampere-hours' capacity, the total weight being 10 tons. The batteries are all connected in parallel by a third rail, so that all work in unison for any variation of load in their immediate vicinity. In this way the total amount of battery is evidently less than if it were installed in a sub-station. Moreover, the load on the feeders will be the average load and steady, the battery taking charge of the heavy load that comes with start- ing. In this way the cost of copper for the feed wires is greatly reduced and the pressure at the motor is kept constant. On all curves, crossings, and in and out of the car-barns, the third rail can be dispensed with, since the battery is capable of doing the entire work at these places. The Storage Battery in Telegraphic Work The field which, from a practical, scientific, and economic point of view, holds the largest possibilities for the employment of accumulators is' the field but lately exclusively occupied by the primary battery, as 1 Vide pages 206 and 215. STORAGE BATTERY INSTALLATIONS 185 applied to telegraph and telephone stations, police and fire alarm, and all signalling systems. Those stations whose business is not large enough to warrant the in- stallation of a generating plant, are fast equipping with a battery of accumulators. The low first cost, the small amount of space occupied, — about one-fourth of that required of a primary battery to do the same work, — the constancy of the E.M.F. and internal resistance, the obviation of the troubles due to creeping salts and cor- roding connections, are a few of the advantages which recommend a storage battery for any of the above pur- poses. It is a fact, as McRae has said, that "the poorest type of lead secondary battery that is on the market to-day is capable, if properly installed, of giving better service than the best type of primary battery." In July, 1895, 3 1 16 storage cells were performing the work previously required of 20,407 primary battery cells. The estimated cost of the latter, not including freight, is $10,203.50, while that for the former is be- tween $8400 and $8500. When the element of freight- age is considered, the difference is still greater. A small amount will have to be added to the accumulator cost, because of auxiliary apparatus required by a stor- age battery installation, and not by a primary battery. This addition will, however, be very slight, being seldom over 1 5 % of the cost of the battery. As regards the cost of maintenance of the primary battery, it may be stated that this will average about $1.50 per year. William Finn states that "the mate- rials consumed in a single cell of the gravity battery furnishing current for the quadruplex circuit would, in !86 the storage BATTERY the course of a year, amount, at the lowest estimate, to #1.10." J. B. Stewart puts the cost of the gravity bat- teries used in the "West Shore" railway telegraph system at $1.65 per annum. The maintenance cost of the storage battery is made up of three factors : the cost of charging current, the interest on first cost, and the depreciation of cells. The first factor is somewhat diffi- cult of estimation. Taking as a basis, however, the cost at which this current could be bought from regular electric lighting stations, it has been estimated that the cost of current equivalent to that furnished by a gravity cell during one year would be about 9 cents. The addi- tion of 2.5 cents for each of the other two factors brings the total maintenance cost up to 14 cents, as against $1.50 for a primary battery. According to Mr. .Finn, "the elaborate series of en- gines and dynamos now furnishing the electromotive power in many of our telegraph offices could, in the opinion of expert authorities, be replaced with accumu- lators with considerable economy in capital expenditure, and an ultimate saving in the cost of maintenance." The great flexibility of an accumulator is one of the main points of superiority of a storage-battery plant over a dynamo. Storage batteries are being introduced in telephone work, the battery being installed at the subscriber's station, and charged, when the line is not in use, from a generator at the exchange. When a call is made, and the line is switched for conversation, the central genera- tor is automatically cut out, and the battery switched in. Perhaps the earliest storage battery to be used for STORAGE BATTERY INSTALLATIONS 187 telegraphic work was introduced, in 1892, in the Burry printing telegraph, — the system used by the Stock Quotation Telegraph Co. At that time electrical ex- perts were almost unanimous in their opinion against the use of storage batteries for this purpose. Mr. Burry, however, believed in it, and installed one on six months' trial. It proved to be a complete success, in regard to quality of current, labor-saving, and general economy. Storage batteries are also being largely introduced in fire-alarm work, as evidenced by the fact that 10,422 battery cells have been introduced during the last two years, and 47 cities are now equipped with storage systems. The first large storage battery installed by the West- ern Union Telegraph Co. was in its offices at Atlanta, Ga. In this installation 700 Chloride cells perform the work previously requiring 8000 gravity cells. This plant includes 344 cells of 75-ampere-hours capacity ; 172 50-ampere-hour cells; and 172 25-ampere-hour cells, — all used on the main line; also 12 cells of 250- ampere-hours capacity, exclusively for the local circuits. These 700 cells are divided into two equal sets, which are charged and discharged on alternate days ; the sets in turn being divided into 8 groups of 43 cells each, which are charged in multiple series, and discharged in series. The cells are charged from the 500-volt mains of the Georgia Electric and Power Co. ; two motor dynamos being used to reduce the line E.M.F. to the correct charging pressure. One of these, a 7.5-H.P. motor dynamo, used for the main line battery, transforms to 1 88 THE STORAGE BATTERY no volts; and the other, a i-H.P. for the local cells, transforms to 16 volts. Fig. 107 shows the connections. At Washington, the second large Western Union in- stallation, 724 Chloride cells are used, divided as follows : 398 cells of 50 ampere-hours, 320 12.5-ampere-hour -+■ 6 VOLTS 6 VOLTS — Wiftof X LOCAL APPARATUS ^ Fig. 107. cells, and 6 125-ampere-hour cells. The main battery of 640 cells is divided into 16 groups of 40 cells each. This is divided into two sets, which are charged and dis- charged alternately, as in the Atlanta installation. The 78 cells comprising the " short wire " battery are a part of the 50-ampere-hour set. They are likewise divided into two sets, differing, however, from the preceding by being charged in series and discharged in multiple. STORAGE BATTERY INSTALLATIONS 189 The main battery is both charged and discharged in multiple. No transforming machinery is needed in this station, as the charging E.M.F. is no volts. These 724 accumulators replace 7300 gravity cells. At Albuquerque, N.M., is probably the first large installation dependent entirely upon storage batteries. This plant consists of 360 Chloride cells, of 20 ampere- hours' capacity at service discharge, divided into 9 sec- tions for convenience in charging. There are also 12 cells of 100 ampere-hours capacity at service rates for supplying local circuits; the smaller cells being for main line service. The cells are charged to 2.5 volts per cell, and discharged to 1.8 volts. A resistance is placed in each discharge circuit to prevent cells from short-circuiting. The cells are placed in two racks, each having three rows of double shelving, the highest being 4.5 feet above the floor. The total floor space occupied meas- ures 13 feet by 5 feet. These cells supersede 900 Callaud cells. Formerly, to have increased the service would have required an additional equipment of Callaud cells, while now an additional service only requires more frequent charging. The charging current is taken from the local lighting company's incandescent circuit, and is controlled by dynamo regulators. At the central station in Paris, 1 1,000 Callaud cells of the larger type were required to do the work now per- formed by 340 accumulators. This battery is divided into six distinct sets; two sets each of 50-60 ampere- hour Laurent-C61y cells, and four sets each of 60-72 ampere-hour Tudor cells. Two of the Tudor sets — one 190 THE STORAGE BATTERY each for the positive and negative side — supply day cur- rent for main line and local circuits, and to five Baudot distributors. The other Tudor sets act as a reserve, while the two Laurent-C61y sets furnish current for the night service, which is comparatively light. The cells are placed on two strong trestles, arranged in parallel rows of 15 cells each, two tiers high. All connections are thoroughly soldered. As the charging generator gives only 70 volts, the batteries are charged in multiple series; three groups of 20 elements being arranged. In tbe fire and police telegraph system at Wilming- ton, Del., 470 B-3 Chloride cells, 6 ampere-hours ca- pacity are used. They are divided into 22 separate units, of from 7 to 30 cells each ; each of the 1 1 units having a battery in duplicate. Current at 119 volts pressure passes through a knife switch, a fuse, a polar and neutral relay, and finally through lamps and resistance coils to the six charging sets ; the lamps and resistance coils varying with the cells in circuit. In case of re- versed polarity, the armature of the polar relay falls on the back stop, thus completing a circuit through a bell, which rings until the polarity is corrected. If the dynamo should stop, the neutral relay opens, and the circuit is broken, thus preventing - batteries from dis- charging through the generator circuit. In the Baltimore office of the Chesapeake and Poto- mac Telephone Co., the outfit consists of two portable 4-volt batteries. The sets are used alternately, and are carried by two men a distance of three squares for charging. Notwithstanding the provision of a duplicate set, and the trouble of carrying batteries, far more satis- STORAGE BATTERY INSTALLATIONS 191 factory results have been obtained than with the pri- mary battery of 80 cells, which was formerly employed. Many have contended that the first cost and the maintenance charge for storage batteries are so great as to prohibit them from all use in private plants. That this idea is a fallacious one may be readily seen by any one who will take the trouble to look up the cost and details of a few of the many private plants now in existence. In a plant in New South Wales, the total cost of working is $35, not including the attendant's time, who in this case is the gardener. The total cost of installa- tion of everything except the building was $1000. The plant consists of an oil engine, dynamo, and battery of sufficient size to light 25 lamps two days in winter, or three in summer. In a French plant consisting of a 0.75-H.P. gas-engine and dynamo, and a battery of 85 available ampere-hours, with an installation of 25 lamps, the total cost was only $440. For a small gas-engine plant, using accumulators, the cost, as estimated, is : 1 Building, $2420 ; engine and 20 kw. dynamo (no volts), $7500; 62 cells, 572-ampere- hours capacity, $4000; total, including switchboard ac- cessories, cranes, tools, etc., $15,000. For the running cost, 750 lamps are assumed for 900 hours per year; also that 40% of the total energy is delivered by the accumulators ; 5 % of the first cost is allowed for their maintenance. With this as a basis, it is calculated that from 12.9% to 17.1% of the capital invested will be earned per year according to the price for gas. With 1 Elektrotechnischer Anzeiger, Nov. 29, 1894. I g 2 THE STORAGE BATTERY the direct system, a 500-H.P. plant will cost about $47,000 ; with a storage-battery system, about $46,000, the battery representing $22,500, and the steam plant being just half the former size. The average cost, according to the quotations from leading battery- makers, for a battery to give 1 kw. for 3 hours, will be $65, or including switchboard, instruments, build- ing, etc., $95. A i-kw. steam plant will also cost about $95, but if everything is considered, foundations, flues, chimney stocks, etc., the cost will be in favor of the accumulator plant. Among the more pretentious private plants may be mentioned that at Ellerslie, the home of ex- Vice-Presi- dent Levi P. Morton, which comprises a 35-H.P. steam engine, two C. and C. generators, of 12.5 and 25 kw. respectively, and 67 G-i 1 Chloride cells. That of Mrs. Hearst, at Sunol, Cal., comprises a twin-cylinder gaso- line engine of 22 brake H.P., a generator, and 60 F-11 Chloride cells. That in Mr. Charles T. Yerkes' New York residence is probably the largest private plant in existence. In this installation is a 35 actual H.P. Otto gas-engine, belted to a 30-kw. Siemens-Halske generator — shunt-wound — and a battery of 60 G-25 Chloride cells, 2500-ampere-hours capacity at a 10-hour rate. The maximum discharge is 2000 ampere-hours at a 4-hour rate. A 7.5-kw. 4-pole "booster'' is also provided. During the erection of a large music hall in Zurich, Switzerland, which required an installation equivalent to 2000 16-C.P. incandescent lamps, it was found that the night load at the central station, an alternating- STORAGE BATTERY INSTALLATIONS 193 current station, had nearly reached its full capacity, that the mains carried their full load, and that the day load at the station was comparatively small. It was decided, therefore, to install accumulators, which were to be charged during the day, and carry the full music- hall load at night. As only that portion of an alter- nating-current wave can be utilized in charging in which the voltage is higher than that of the battery, two improved Pollak rectifiers were installed, by which the regulation of the precise point of the wave could be obtained by simply moving the brushes. Two bat- teries of accumulators, in all 113 cells, of the Pollak type were connected to the two sides of the three-wire system ; the capacity of the battery being 1528 ampere- hours at a 4-hour rate. Current is supplied through two transformers of. 30 kw., and two auxiliary trans- formers of 15 kw., connected in series with the others, thus adding their 55 volts to the 105 volts of the larger transformers. The auxiliary transformers have ten con- tacts on different parts of the coil, by means of which the voltage is controlled. Excellent regulation is ob- tained, the brushes on the rectifier being seldom moved, and the efficiency of rectification has been found to be about 94%. Mr. Pollak claims that such a rectified current has the peculiar property of accelerating elec- trolytic processes; which property has not yet been satisfactorily explained. A novel scheme has been proposed in France, which, if successful, will open up a comparatively new field for accumulators. This is to furnish all towns along or near the river front with a storage-battery plant, and o 194 THE STORAGE BATTERY then to charge these batteries successively by means of a floating central station. By this means many towns, too small to have a complete installation of their own, may be furnished with electric power with but few of the attendant expenses. In a paper read by Mr. B. J. Arnold before the Northwestern Electrical Association, in St. Paul, Minn., the following interesting table of data was given. Available power in kw. . Expenditure in dollars per kw. Duration of working in years - Profit, % capital .... Total energy in kw. hr. distributed Energy per pound of coal in kw. hr. distributed . Cost per kw. hr. in cents distributed Power utilized divided by power available in % . Use of Accumulators Energy expended for charge in kw. hr. . . . Energy furnished by dis- charge in kw. hr. . . . Industrial efficiency of ac- cumulators in % . . . Loss in accumulators in % of total energy distrib- uted Direct Current Eber- feld 5°° 545-53 5 14.09 305.794 5-7° 80 Ham- burg 580 816.93 4 18.05 513,183 140 5.18 79 D. C. with Accumulators Barmen 22$ 908.50 5 7-65 122,026 90 6.85 61 59.573 42,584 71-5 14 Han- 600 785-70 2 "•34 181 5.02 48 194,733 154,836 79-4 11 Dussel- dorf 600 920-35 I 7 337,285 129 4-54 51 279,506 216,561 77-5 13 Alt. C Cologne 680 693-55 1 7.2 307,074 71 6.65 5o At the station in Kijew, Germany, employing 72 E.P.S. accumulators, a current efficiency of 83% and an energy efficiency of 68% is reported, for the three years ending November, 1894. At Cassel, England, for STORAGE BATTERY INSTALLATIONS 195 1895, the energy efficiency was 72.5%, and the current efficiency 85.3%. At the Clichy Sector, in Paris, two sets of batteries of 250 cells each have been installed, with a capacity of 2000 ampere-hours at an 8-hour rate ; one set of batteries are of the Chloride type, and the other set, the Tudor. For the five months ending July, 1896, the mean efficiency was 69%, the mean daily output was 75% of the total capacity at the nor- mal rate, and the minimum daily output, 70% of the total capacity at the normal rate. At Dusseldorf, for 1894 and 1895, the average output for one year was 56.5% of the total output, the total output being 3>6So,730 ampere-hours. The following table gives the monthly efficiencies for the same station. April .... . . . 64.7% October . . . • ■ 52-7% May .... . . . 78.0 November . . • • 48.3 June .... . . . 82.2 December . . ■ • 44-9 July .... . ■ ■ 73-8 January . . . . . 48.4 August . . . . . . 80.7 February . . • ■ 54-6 September . . . . . 72.2 March . . . . . 56.8 On the Zurich-Hirslanden railway, the introduction of accumulators has effected a saving of 2.2 pounds of coal per hour, or about $2500 per year. The coal con- sumption on this road is from 30% to 40% less than on a similar road where storage batteries are not used. In Belfast the total efficiency of the station is 91%, and the battery loss, 6.7% of the total energy generated ; the total generating expenses are 6.48 cents per kilo- watt-hour sold. In Edinburgh the total expenses are 3.12 cents per kilowatt-hour sold. At present, in installing a storage-battery plant, the ig6 THE STORAGE BATTERY battery should be about one-third the maximum full-load capacity of the station. Figs. 108 and 109 show the per cent maximum demand factor, and the per cent load- factor, for some English central stations, both with and without storage-battery equipments. It may not come amiss to state that the load factor indicates to i"E liW V) u0 a. O 1- O < 100 b. □ a> c O O >< 7 3 m O 3 Z < S 80 1U Oswesti _ 1 Oxfo "5 T ■a I ! g 00 c E 13 13 a> CO 3 1- 1 0) > a so k Batle y s fster ns -» -Non -Ba1 tery Sys ems Fig. 108. what extent the hours have been filled up by work, and the maximum demand factor assists in showing what there is superfluous in the generating plant. In a battery station, the maximum demand factor may, of course, exceed 100%, as is the case in Bradford; but in a non-battery station, never, unless its rated ca- pacity is lower than its real capacity. STORAGE BATTERY INSTALLATIONS 197 In conclusion, it may be stated as a fact that in all stations that have to meet a fluctuating demand, a com- bined accumulator and steam plant will be found cheaper in first cost, and cheaper and far more satisfactory in o h- o < D < O en i to 0) 3 ■a m- X O > 03 Lor d i HOURS Fig. no. such cases, a machine called a rectifier must be used for changing the alternating to a uni-directional current. By means of this machine, only that portion of the current wave is utilized in which the voltage is higher than that in the battery. It is claimed that such a rectified alternating current has the peculiar property of accelerating electrolytic processes, and is, therefore, peculiarly adapted for charging purposes. It has not yet, however, been made clear that such rectified cur- CONCLUSIONS 225 rents really have this property. A central station in Zurich, Switzerland, has had occasion to make use of such a machine, with gratifying results. 1 1 1% 2 m 3 i TIME IN HOURS Fig. in. When a cell has been discharged below a certain point, the excessive loss of conductivity, owing to too large a proportion of the active material having been converted into sulphate, will be apt to injure it, and the 1 Vide page 192. 226 THE STORAGE BATTERY contact between the peroxide and the support plate may be broken. The recharging of such an exhausted cell, when, indeed, it is possible to recharge it at all, must be carried on very slowly. It should be recharged as soon as practicable, since the cell is in a weak state, and is the seat of slow reactions, in the nature of those occurring during discharge. When it is possible to re- charge an exhausted cell, the current used should be at least 30% below the maximum charging rate. Care should be taken that the scale, or powder, which falls off does not stick between the plates. If it were not for the fact that the adherence of the white sulphate to the peroxide beneath, in an exhausted cell, is very weak, it is probable that the recharging of a battery, when in such condition, would be impossible. Herr Briiggeman has concluded from some investi- gations he has made that the charging should be stopped at the beginning of the sharp bend of the curve; since before that point is reached the energy lost by the evolution of gas is very small, while after that it becomes very great. This corresponds to Salo- mon's rule, that every cell should boil in an equal de- gree when charging is stopped. It should be borne in mind that this rule refers only to the regular chargings, and not to the first or forming charge. In the latter case the charge should be continued for at least 30 consecutive hours, without interruption. If, however, such a run be impossible, it should be continued for at least 10 hours a day, for 3 consecutive days; the former method being preferable. When the evolu- tion of gas becomes excessive, which will occur toward CONCLUSIONS 227 the end of the charge, it is well to decrease the charg- ing current. With the pasted type of plate, the question of the duration of the first charge is far more vital than with the Plant6 type ; since, with the former, there is great danger of a coating of sulphate being formed between the active material and the grid ; especially if the plates are allowed to stand in the acid before charging, or if the charging current be stopped before the active ma- terial be thoroughly formed. If this coating of sulphate be once formed, it is almost impossible to get the plate in first-class condition, as the sulphates insulate the active material from the grid, and thus cause the action to take place on the grid itself, instead of in the active material. This applies to the positive plate, rather than to the negative, since it is easier to reduce the sulphate to spongy lead than it is to oxidize it. The term "boiling" does not indicate the rise in tem- perature of a battery, but rather the great evolution of gas which occurs when a cell is nearly charged. It is evident that as charging proceeds the amount of sul- phate to be converted into peroxide becomes less and less, and the plates therefore become virtually smaller, so that the current becomes too large for the work de- manded of it. The result is that that part of the cur- rent not actually used in the formation of peroxide decomposes the electrolyte into its constituent elements. It will be noticed that when an accumulator has been in use for a considerable time, the gases evolved do not produce such a milky appearance of the liquid as be- fore. The reason for this is that the plates are better 228 THE STORAGE BATTERY formed ; consequently a larger charging current can be used without producing " boiling." The color of the positive plates should be, when formed, of a dark red or chocolate, but when fully charged their color will be much darker, resembling more that of a wet slate ; the negatives, although also of a slatish color, are always considerably lighter than the positives. To one who has become accustomed to studying the plates, it is a comparatively simple matter to tell the relative amount of charge that a cell may contain. In using an accumulator care should be taken that, at the close of the discharge, at least 25% of the total capacity of the cell remains unused. The best modern practice is to leave 30%. In other words, the voltage of the cell at the close of the discharge should be never lower than about 1.8 volts, under load. All manufact- urers indicate the point at which the discharge should be stopped, and the battery should be run in accordance with their directions. Mr. Griscom 1 gives the following reasons why it is undesirable to run a battery lower than about 1.8 volts. " It should be understood that a full discharge, i.e. a discharge to a point situated at the beginning of the steep part of the voltage curve, is working a battery to the danger limit, and is undesirable for the following reasons : • " 1. Regulation is troublesome. " 2. The efficiency is low. " 3. Dangerous molecular changes take place, as indi- 1 Trans. A. I. E. E., Vol. 11, p. 302. CONCLUSIONS 229 cated by the changes in the internal resistance and in the E.M.F., as well as by "buckling." "4. Uneven plates discharge into one another after the circuit is interrupted. "5. The life of the battery is shortened. " By classifying the failures and successes of a num- ber of observations made on various batteries, the truth dawned upon us that whenever a battery was exhausted to its full capacity daily, its life did not exceed 500 charges ; but whenever it was worked within two-thirds of its capacity, complaints were unknown. It is only necessary for the engineer to remember to add 50% of the capacity, as a factor of safety, to the maximum load, just as he allows several hundred per cent in calculating the strength of a bridge or axle. " This additional amount is not a dead loss in invest- ment. It produces many countervailing advantages. It provides a very effective and safe reserve for cases where the charging breaks down, and it increases the actual efficiency of the battery, which rises from about 80% to nearly 90%, when used with sufficient reserve. And for cases where it is necessary to maintain a con- stant potential difference, it raises the efficiency much more, because in these cases the commercial efficiency must be rated, not from the average point of E.M.F., but from the lowest point to which the battery falls on discharge, and when used in this way the potential difference drops only 2.5%." When the plates of a cell are discharged beyond the maximum discharge permitted, nearly all the material of the positives becomes lead sulphate, which is soon 230 THE STORAGE BATTERY decomposed into the higher sulphates, which ruin the plates and cause them to " buckle " while charging. A series of experiments has been lately conducted with various batteries to determine in what manner they are affected by heavy discharges. The conclusion which has been drawn from the work is that there is no objection, as a rule, to short high discharges, as long as the maximum rate is not affected. It is the prolonged discharge which is especially injurious to a battery, since the sulphating, which then takes place, is the cause of " buckling," especially when the discharge is not immediately followed by a charge. If, however, the discharge is too large, it is likely to drive the paste out of the plates. It is the quantity of gas which is driven from the plates at the moment of sudden heavy discharges that causes the injury. Although all makers give the best rates of discharge for their plates, it may be well to state, as an approximate rule, that a good rate of discharge is about 8 amperes per square foot of positive plate. It is on this point — that of being able to withstand" heavy discharges — that the superi- ority of the Plante type of plates depend. A plate having a large surface, covered with a thin layer of peroxide, freely exposed to the action of the electrolyte, will be found to have a far greater capacity per pound of plate, for rapid discharging, than is the case with a plate having a thicker layer of active material. The capacity of such a cell will evidently be but little affected by the rate, and the watt-efficiency will not be nearly so much affected by the dilution of the electrolyte in the im- mediate neighborhood of the active material. Fig. 112, CONCLUSIONS 231 which was taken from a Chloride plate, shows the vari- ation in capacity at different rates of discharge. In testing a cell, care should be taken that the source of current be steady. A battery of cells in multiple series will be found best, or, if that be not convenient, a source of current of considerably higher voltage and I 70 o =--===== Httt |_J__L = --^ ■■^ ^. ^ ^ r ^ r =t ^ K ^-_ £ -V — = IA V o 10 10 98 7 6643210 Rate in Hours Fig. 112. • a large resistance in series with each cell will be found to give equally good results. The time interval should be so arranged that about twenty readings may be taken. The record should contain a column each for the time, volts, amperes, ampere-hours, watt-hours, and remarks, in which the specific gravity, temperature, " gassing," etc., should be noted. The specific gravity should b'e measured at regular intervals, say about ten times during the test. Three efficiencies should be cal- culated, — the volt-, the ampere-, and the watt-efficiency. f r if 232 THE STORAGE BATTERY In plotting the curves, it would be well to make the horizontal or time distances small, and the vertical scale units large. In comparing cells of different types, sizes, or volt- ages, — especially where the weight is an important factor, — the vertical scale should be made watts per pound rather than volts. A very useful curve for gen- eral practice is the " straight-line curve " proposed by Carl Hering in the Electrical World. In this, the ordi- nates are the capacities in watt-hours, and the abscissae, the rates in watts. In obtaining the data for this curve, the cell should be fully charged for each test, and the discharge should be continued until the E.M.F. has fallen — in each case — a certain fixed percentage of what it stood a few minutes after the beginning of the discharge; the discharge being at a constant current. As the result is different for different makes of accu- mulators, each manufacturer will have to determine whether this relation approximates sufficiently close to a straight-line function for the calculations occurring in practice. When this curve does approximate suffi- ciently to a straight line, it will be of the general form watt-hours = a — b watts, a and b being constants to be determined for each particular type. By making the abscissas the time in hours, the curve will resemble in general outline the shape of a magnetization curve. The more nearly this curve approaches to being a horizontal straight line, the more perfect (so far as the dependence of the capacity on the rate is concerned) is the accumulator. CONCLUSIONS 233 Although in practice accumulators may be discharged discontinuously, for testing, the charging and discharg- ing periods should be continuous, without any intervals of rest. The most rational way to charge is to charge at a constant voltage. This means using a diminishing current, and as this will be found to be a difficult mat- ter, it is better to use a constant current until the volt- age shows signs of rising appreciably, then to reduce the current suddenly to a lower value — ■ taking voltage reading for both values — and so on to the end of the charge. M. Simon advises using a constant power, that is, a diminishing current and an increasing voltage, the rates being such that the product is always constant. He argues that charging at a constant current has the objection of taking too long, the current being weak at the beginning and too strong at the end. With a constant potential, on the other hand, the charge is almost too great at first, and so small at the end that it prolongs the time of charging considerably. A bat- tery, in the latter case, receives, in two or three hours, more than three-quarters of its capacity. The discharge should be made under conditions of constant current, using a rheostat, stepping down by small increments for the purpose. If this is impracticable, the next best method is to discharge through a constant resistance. Where the tests are wanted for power purposes, it would be best to discharge at a constant wattage. All tests should be repeated until at least two like discharges, under the same conditions, are obtained. In the installation of a battery, the first point to be considered is the selection of a room. This should 234 THE ST0RAGE BATTERY be dry, well-ventilated, and of a moderate temperature ; otherwise the evaporation will be found to be very large. The floor must be of some acid-proof material, and so made as to drain rapidly ; an outlet being pro- vided for the liquid. If the floor is already put in, and of wood, it should be covered, especially where the battery stands, with a lead tray. The room should be located as near the generating room as possible, so as to reduce the wiring cost to a minimum. The battery should be placed in as few tiers as possi- ble, and in such a manner that the direct rays of the sun are not allowed to fall upon the cells. The rays of the sun are likely to crack the glass. This is probably due to the unequal expansion of the glass, for it has been found that jars which are carefully annealed never crack in this manner. Of course, the latter precaution does not apply to large batteries, where lead-lined wooden tanks or solid lead boxes are used. All exposed metal work should be protected with an acid-proof paint. When a battery is received, the elements and con- taining-cells should be carefully unpacked, and all dust and any foreign particles removed. One should be sure that all insulators and distance pieces are in posi- tion, and that the plates are in their proper alignment. In connecting the cells, sufficient sectional area should be provided; otherwise the various plates will be worked unevenly, and the full capacity of the battery will not be obtained. The most satisfactory method of connect- ing up is to " lead-burn " or weld the positive plates of one cell and the negative plates of the next to the same lead "bus-bar," thus ensuring good connection between CONCLUSIONS 235 the various plates of a cell, and also between any two consecutive cells. If it is'desired to solder rather than "lead-burn" the electrodes, the following method will be found to give the best results : Strips of lead for making the joints are placed for some time in a strong potash solution, after which they are thoroughly washed and scraped. The electrodes themselves should also be scraped. The two are then held tightly in a mould, in the form of tongs, and molten lead poured around them. If the elements are to be bolted together, one should see that all bolt connectors are thoroughly screwed up; otherwise resistance and consequent heating will result. In setting up a battery, it should be remembered that plates deteriorate on standing exposed to the air. They should, therefore, be unpacked and set up immediately on arrival. When they are entirely connected up, they are ready for the addition of the electrolyte, and for the forming charge, which they should receive immediately. In mixing the electrolyte, one should use only chemi- cally pure acid, and always pour the acid into the water. Before placing the electrolyte in the cells, Lucas treats it with basic sulphide. It is then allowed to rest for 24 hours, after which it is filtered and ready for use. When filling the cells, the top of the plates should be covered with the liquid by at least half an inch, and the electrolyte should never be allowed to fall below this point. If it is desired to separate the plates from each other, and no regular separators are at hand, perforated porous paper, saturated with paraffin wax, will be found to give 236 THE STORAGE BATTERY good service, and to be practically unacted upon by the acid. When glass jars are used, it is well to paint them at the top, for about an inch, with paraffin wax to prevent the creeping of the solution. A new battery will never give its full capacity till after about twenty discharges. During this time it should be given about 25% overcharge. After that, 10% over- charge, that is, 10% more charge than was taken out, will be sufficient for ordinary work. In mounting cells in cars, or in any place where there is a liability of breakage, it is best to use the Boese system. In this system, a set of glass cells are inserted in a box, between which and the cells, and between the cells themselves, is poured a melted mass of some insulating material which, when cold, will be rigid, but elastic, and will retain the liquid even if the glass cells should break. Drake and Gorham use a method for stopping the spray from an accumulator, which it would be well to adopt in all cases. This is to float particles of a light substance on the acid to the depth of about \ of an inch. In all cases where it is necessary to get at the acid, this substance can be easily brushed aside. In soaking up spilled acid, many attendants use either ammonia, saw-dust, or soda; it has been found that whiting is better than any of the other remedies. Before beginning to charge a storage battery, it should be gone over carefully, and any cell that is not up to the standard should be disconnected and put in working order before being replaced. CONCLUSIONS 237 If the accumulators are to be used in a cold climate, it would be well to adopt the device of M. Varennes. 1 He places a small incandescent lamp, covered with a black varnish, in each accumulator. These lamps are connected to an automatic device which puts them in or out of circuit, according as the temperature is above or below a certain predetermined point. In installing plants where expert attendance is not to be had, it is well to place in the circuit two mag- netic cut-outs, one set for maximum current, and the other for minimum voltage, so that the battery cannot be discharged too low. It might be well, in some cases, to use a resistance instead of the regulating cells. Owing to the different lengths of time that the regulating cells have to be in circuit, it is ex- ceedingly troublesome to keep track of them, and when neglected, their spraying becomes a disagreeable feature, all of which would be obviated by the use of resistances. To obtain the best results in charging a battery, the following points should be watched. The rate of charge should be normal, except in cases of emergency. At such a rate, unless the constant potential method be employed, the cell may be considered full when the voltmeter reads 2.5 volts during charge. The electro- lyte should be kept at uniform density throughout the cell; when water is added, because of evaporation, it should be added by means of a funnel reaching to the bottom of the cell. Care should be taken never to add acid after evaporation; otherwise the electrolyte will l La Lumiere Electrique, Jan. 6, 1894. 238 THE STORAGE BATTERY be too heavy. Hydrometer readings should be taken regularly ; this is an excellent indication of the amount of the charge in the battery. These readings are use- less, however, unless the precaution be taken to keep the electrolyte of uniform density. Fig. 113 shows the relation between the specific gravity and the capacity, and indicates clearly how close a guide hydrometer readings, when intelligently observed, are. 100 90 80 ??70 8 S S.36O O 1- ■S£50 sf» 20 10 1.: \ ^ s% "6 V 1.210 1.200 1.190 1.180 Specific Gravity Fig. 113. 1.170 1.160 About once a week, each cell should be tested with a low reading voltmeter and hydrometer. If any cell should read low, it should be at once cut out and care- fully examined to see if any material has been intro- duced which could short-circuit the cell. If no such trouble can be found, the cell should be disconnected from the discharge circuit and given an extra charge. If a couple of extra chargings do not bring the cell up to condition, and nothing is known concerning the cause of the trouble, the manufacturer should be consulted. CONCLUSIONS 239 Mr. Joseph Appleton 1 has devised a very interesting test for determining the condition of cells. He takes a plate of cadmium, mounted in a hard rubber frame, immerses it in the electrolyte, and reads the E.M.F. between it and the positive or negative plates of the cell. The cadmium should be shaken occasionally to free it from any bubbles of gas which may be formed on its surface. The cadmium plate should be washed with water every time it is taken from the cell. " By this method," says Mr. Appleton, "it is possible to ascertain at any time during the charge or discharge whether the positive or negative plates are in proper condition or otherwise, thus locating at an early stage any sign of irregularity or trouble. " During charge, the cadmium plate reads negative to the negative plate, until tbe cell is about full, when the reading should be zero ; the charge should be continued until the cadmium reads 0.2 volt positive to the nega- tive while charging at the normal rate." It is best never to allow any organic matter or oxi- dizable substance to come into contact with the peroxide element of the battery. When cellulose is used, the effect is to convert it into grape sugar, which is decom- posed. The latter is converted into plumbic formiate and carbonate, sulphate being the ultimate product. It has been the author's experience that celluloid should be used very sparingly in cells, and never in connection with the positive plate, since its general tendency is towards decomposition. Others have found that when celluloid is used for forks and bearers, it soon becomes 1 Storage Battery Engineering Practice, N. Y. E. E., Vol. 23, p. 454. 240 THE STORAGE BATTERY coated with a shiny black coating, which is formed from the material itself. It is always an easy matter to increase the capacity of a battery by mixing organic materials with the lead oxide; but, as any such mixture is always accompanied by a rapid deterioration of the plates, the increase in the capacity is concomitant with a decrease in the life. Besides this, those binding materials which are used to harden the plates are good as long as they are not decomposed by the current ; but this decomposition will take place with every successive charge. It is evident that, unless an undecomposable body is formed with the lead oxides, the life of the plates will be greatly decreased. Many investigators have used manganese, either in the electrolyte or as a part of the active material, as peroxide of manganese. All trace of such compounds should be avoided, as the tendency is to reduce the capacity of the battery by carrying oxygen from the positive to the negative plate. When filling the cells with acid, one cannot be too careful to have the acid of the proper strength ; for, if too strong, the plates will be found to sulphate more rapidly, and the sulphate will be harder to reduce. It has been found that with plates 0.4 inch thick, the maximum capacity will be obtained when the acid is about 1.270 specific gravity, and with plates 0.25 inch thick, acid of 1.240 specific gravity. Should the plates sulphate from any cause, it may be stopped and further prevented by using Urquhart's 1 remedy, which is made up as follows : 1 Electric Light Fitting, J. W. Urquhart, p. 47. CONCLUSIONS 241 To a quart of strong solution of common washing soda, add slowly, during agitation, 12 ounces of con- centrated sulphuric acid. This should be added to the electrolyte in the proportion of 1 : 25. Although the best modern practice is to use as the electrolyte nothing but the pure dilute acid, some investigators believe that if the electrolyte be either neutral or slightly alkaline, that the forming process will be much more rapid. To obtain this result, such salts of the light metals are used as will produce no decomposition on the positive electrode, and will there- fore not interfere in any way with the formation of the peroxide. Luckow x is the originator of the above process. In the manufacture of storage-battery plates, nearly every conceivable shape has been tried, and it has been found that an approximately square plate gives the best results. The plate should not be made too deep, else it will be subjected to different degrees of chemical action. Where the large central station plates are used, — approximately 15x30 inches in size, — some means have to be employed to keep the electrolyte of a uniform density. This is usually accomplished by means of a blast of air. As to the shape of the perfora- tions, it has been already pointed out, 2 that that hole which is larger at the centre than it is at the surface, is the best. Many manufacturers believe that it is best to use some alloy of lead, which is unaffected by the chemical reactions which take place in the cell, thus allowing the plate to be more rigid and lighter in 1 G. P., 84,423 ; 1894. 2 Vide page 66. R 242 THE STORAGE BATTERY weight than if constructed from pure lead. Mr. J. K. Pumpelly, however, believes that all alloys of lead should be avoided, and only chemically pure materials used. In this view he is supported by a constantly increasing number of manufacturers ; in fact, it is now the exception to use alloys, except where the weight is an important factor. The plate should be so constructed as to be able to expand with the active material, without destroying the contact between the two. The electrolyte should have free access to all parts of the active material; great porosity is therefore necessary. Zacharias 1 obtains this by pricking the active material with needles at the rate of about ioo holes per square decimetre. Many believe, notably among them Fitzgerald, that during the con- struction of the plates, it is best to make only the sur- face porous, so as not to sacrifice mechanical strength. Probably one of the potent causes for the destruction of the plates is the gas which is formed in the pores of the active material ; the harder and denser the active material, the quicker will it be destroyed. It is for this reason that the negative plates are the most difficult to construct. The contact between the active material and the con- ducting plate must be good ; for if poor, a white sul- phate will be formed at the surface, which practically forms an insulating layer. As pointed out by Mr. Her- ing, it is well to make the positive plates light, cheap, and easily replaceable. By this method, although the life will be shortened, the great difficulty caused by the 1 G. P., 84,810; 1894. CONCLUSIONS 243 gradual washing away of the peroxide will have been settled. When the perishable parts have been renewed, the battery is practically as good as new. This is the plan followed with the plates used in the Paris accumu- lator lines. The life will be soon known to the user, and he can readily determine for himself how much is to be allowed for amortization. It has heretofore been the general custom among manufacturers to construct the end plates — of the perforated pasted type — like the other negatives, thus giving them twice the necessary surface and capacity. Although this decreases the internal resistance, it will be found that the end positives discharge more rapidly than they should, thus producing buckling. An easy way of overcoming this difficulty would be to punch the active material out of alternate meshes on the end plates, leaving the plates half empty and half full. Batteries which are constructed in this manner have been found to give excellent service. Plates of the Plante type, and of the grooved pasted type, are so constructed as to have half the capacity of the other negatives. F. Zacharias x has come to the following conclusions concerning the manufacture and construction of plates : 1. No portion of the metallic frame should pass through the paste. 2. The paste should not be retained at the upper edge by the frame, but should be free to expand and to form gas. 3. Whenever the frame covers the active material, it should be perforated to allow the gas to escape. 1 London Electrician, April 17, 1896. 244 THE STORAGE BATTERY 4. The frame should distribute the current evenly, and have a minimum weight consistent with strength and securing the paste. 5. The frame should be so constructed that the mate- rial on the negative plates should not lose contact, even when disintegrated. When it is desired to transport the battery to a dis- tance, after having been in use, it should be taken apart, washed thoroughly, and the plates pressed together, so that only one face of each of the end ones is exposed to the air. Each batch is then wrapped in oiled paper. If it is not possible to do this, wrap each plate as much as possible with the oiled paper, and stuff the intervals between the plates with hay wrapped in oiled paper. The battery should be set up again as soon as possible, and then treated as though new. If the battery is to remain idle for any considerable le.ngth of time, it should be first given a full charge at normal rates, and then given a recharge — till it com- mences to boil — at least once a week. If for any reason this recharge is impossible, the directions for the battery used should be followed. If by any means the connections have become re- versed, so that the negatives assume a chocolate color and the positives a slate color, the only, remedy is to discharge the battery completely, so that the cell gives no E.M.F., or a very slight one. The connections are then changed, and the battery recharged; but slowly at first, as there is no counter E.M.F. to overcome. When a battery has become run down, Trowbridge 1 1 A. P., 551,565 ; 1895. CONCLUSIONS 245 charges it, removes the negative plates, and replaces them with zinc plates. The battery is then discharged, the zinc plates are removed, and the lead negatives replaced. In connecting up cells, or on soldering connections to the plates, thin connections, and all connections made by soldering on the plate, should be avoided, where the electrolyte, falling below the joint, may expose it to the air, or to the action of electrolysis at the point where nascent oxygen or ozone is set free. The relatively low conductivity of the peroxide must be considered ; if the expansion is weak, cracks will be produced in which a white sulphate is formed. The distribution of current should therefore be uniform throughout the plate. The distance between each plate should be the same at all points, and the electro- lyte should not be allowed to consist of layers of differ- ent densities. A frequent mistake is made in adding fresh acid when the specific gravity of the electrolyte has fallen, due to heavy sulphating. The cells should be given a prolonged charge instead. The addition of concentrated acid to the cell is liable to rot the grids. Sir David Salomons, 1 in speaking of future improve- ments, says : "There can be no doubt that the improvements in the future will take the form of a modified electrolyte, which, according to Mr. Robertson, must be of such a nature as to prevent the formation of or at once break up any deleterious substances which may be formed during the charge or discharge." It must especially 1 Electric Light Installations, Vol. I, p. 105. 246 THE STORAGE BATTERY break up and prevent the formation of all those lead- trees, or fine lead-needles, which are the greatest trouble with the batteries at present. It is also probable that batteries will be so constructed that the diffusivity of the acid will be greatly increased. Messrs. Gladstone and Hibbert, in a paper " On the Cause of the Changes of E.M.F. in Secondary Batteries," in Vol. 29 of the London Electrician, say : " The fall of E.M.F. at the end of the discharge leaves a large percentage of active material unacted upon. This is mainly due to the weakness of the acid against the plates, on account of the interstices being much clogged, and it would be counteracted to considerable extent if the diffusion could be increased. When a cell has been discharged below its minimum useful voltage, there occurs the destructive action called scaling. This is probably due to the abnormal chemical action arising from very weak acid. Diffusion would prevent this. APPENDIX Measurement of the Internal Resistance of a Storage Cell SHELDON In measuring the internal resistance of a storage bat- tery by this method, see Fig. 1 14. Care should be taken to have the standard resistance nearly equal to that of Fig. 114. the accumulator to be tested ; the Wheatstone bridge should have a resistance of about 10 ohms, or capable of carrying \ of an ampere ; and the alternator should give 10 or more amperes. The contact D' should be placed successively at the points 1, 2, 3, and 4, and D shifted till the minimum sound is produced in the telephone. Calling a, b, c, and d the readings of the bridge wire for the points 247 248 THE STORAGE BATTERY i, 2, 3, and 4, respectively, and r the resistance of B, then x= r a — b c-d mance's method, improved This improvement, suggested by Dr. Perrin, is shown in Fig. 115. The resistance should be made equal to the normal load of the battery, so that the meas- urements are made under normal working conditions. A small re- sistance is also inserted in the gal- FlG. 115. vanometer circuit. GRASSl'S METHOD The following, which is given by Professor Grassi, 1 is a combination of Mance's improved method and the methods of Hopkinson and Mathieson. In Fig. 116, x is the accumulator having an internal resistance x ; a, b, c, d, are four resistances, so determined that bd=ac, d being highly inductive; G is a galva- nometer; r, a standard re- sistance ; BC, a calibrated stretched wire ; and D, a sliding contact. In measuring the resist- ance, the wire F is con- nected successively to the terminals 1, 2, 3, and 4. For each connection, the position of D is adjusted until the 1 L'Elettricista, May 1, 1895. Fig. 116. APPENDIX 249 galvanometer remains at zero, when the switch 5 is opened and closed. Calling e, /, g, and h the readings on the calibrated wire BC, for the positions 1, 2, 3, and 4, we have for the resistance required h — e x = r 2- f-e Measurement of the E.M.F. of a Storage Cell Negrenau J gives the following method for measuring the E.M.F. of cells : "The current from a standard cell passes through a varia- s r a ble resistance r as far f as the point a (see Fig. 117), where the circuit branches. The first branch contains a resistance r", the other a galvanometer, a resistance r', and a cell connected up in series to the standard cell. The resistances r and r' are adjusted till the galvanometer shows a con- stant deflection on opening and closing the first branch. The ratio of the E.M.F.'s is then given by the ratio of the two resistances." Cosgrove improves this method by placing the gal- vanometer in the first branch. This gives a zero de- flection on balance instead of a constant deflection, and readings can therefore be made with greater accuracy. This second method also has the advantage that a tele- phone receiver may be substituted for the galvanometer if convenient. 1 E. W., Vol. 29, p. 739. T A 250 THE STORAGE BATTERY FORMULjE for the calculation of the e.m.f. of secondary cells In the following formula, from E. J. Wade's " Chemi- cal Theory of Accumulators," 1 W— the work in joules. Q = the coulombs of electricity that are passed through the electrolyte. H = the number of calories liberated by the recom- bination of a unit weight of one of the decomposed ions. e = its electro-chemical equivalent. c = its chemical equivalent. h = the electro-chemical equivalent of hydrogen = .00001038. J = Joule's coefficient = 4.2. . E = the E.M.F. required. W=QE, W=QJeH; therefore E =JeH and e = he; therefore E=JkcH= 4.2 x .00001038 c/f = .0000436 cH. Now cH — h eat ot formation . valency therefore E = • 000 °436 x heat of formation valency 1 L. E., Vol. 33, p. 657. APPENDIX 251 Since nearly all the battery equations are expressed in terms of the transfer of two atoms of hydrogen, or their equivalent (that is, they are bivalent), and since .0000436 x 46,000 . — :z — — = 1 volt, 2 we have E = neat °f formation, in calories 46,000 In the application of the above law it should be re- membered that the effects due to variations of the density of the electrolyte, allotropic modifications, and alterations in the state of the substances taking part in the reactions must be taken into account. Von Helmholtz claims that a temperature correction will also have to be applied, although from the result of some investigations that have been conducted by Preece, it would appear that the corrections are so small that it will not be necessary to take them into account. Dr. Streintz 1 gives the following formula for the cal- culation of the E.M.F. of an accumulator: E= 1.850 + 0.917(5-^-); where 5 = the specific gravity of the electrolyte, s = the specific gravity of water at the tempera- ture of observation, and E = the E.M.F. required. 1 Zeit. fur Electrotech., May 1$, 1895. 252 THE STORAGE BATTERY FORMULA FOR THE CALCULATION OF THE CAPACITY OF A STORAGE BATTERY IN AMPERE-HOURS It is well known that the current in ampere-hours maintained by the consumption of any given chemically active substance varies with the change of valence where oxidation and reduction occur, and inversely with the molecular weights of the transforming substance. The combustion or liberation of I pound of hydrogen cor- responds to 12,160 ampere-hours. The theoretical current capacity in ampere-hours may, therefore, be obtained as follows : Let V= the change of valence of the ions, W = the sum of the molecular weights affected, and 12,160 = the capacity per pound of hydrogen. 12,160 x V Then capacity per pound = W Calling lead sulphate, which is the ultimate product at both electrodes, the real active material, we obtain by the use of the above formula 40.24 ampere-hours, or 80.48 watt-hours, per pound of lead sulphate, with the lead-lead-sulphuric-acid battery. With the lead-zinc cell, the active working substance is both lead sulphate and zinc sulphate, and the theo- retical capacity then obtained is 52.54 ampere-hours, or 1 26. 1 watt-hours, per pound of active working substance. Considering the positive and negative plates as equal, which they practically are, the capacity per pound of working substance- on either plate alone would be 80. 5 APPENDIX 253 ampere-hours for the lead-lead-sulphuric-acid type, and 105.1 ampere-hours per pound for the lead-zinc type. According to Monnier's and Guiton's estimate, it re- quires 565,600 coulombs to peroxidize 1 kilogramme of minium. As stated previously, Plante and Paget agree on 4.48 grammes of lead peroxide as the equiva- lent of 1 ampere-hour, and this corresponds to 0.158 ounce per ampere-hour. Fitzgerald has found 0.135 ounce per ampere-hour. This, it must be remembered, is the theoretical equivalent, on the supposition that all the active material on either plate is transformed into lead sulphate ; that is, that the battery is completely dis- charged. As this is not accomplished, the best prac- tice is to allow anywhere from 0.53 to 0.86 ounce of lead peroxide, and from 0.5 to 0.8 ounce of spongy lead to the ampere-hour, according to the discharge rate, thickness, and density. Fitzgerald gives as the safest rule, and the best practice bears him out, that a weight of 0.53 ounce per ampere-hour on each plate for a 10-hour rate, 0.6 ounce for a 5-hour rate, 0.7 ounce for a 3-hour rate, and 1.0 ounce for a i-hour rate of discharge, for the ordinary thickness, and an electro- lyte density of 1.200 will be found to afford the best results. While, of course, these rules are only approxi- mate and will have to be modified for different plates and different conditions, it will be found that under ordinary circumstances it is perfectly safe to use the above values. 254 THE STORAGE BATTERY C fl U 3 3 EJ SUSP'S « 3 « u "J .£ .5 G a,3)p-a, & 4) So g«22 g I— i u X >-i M4 5-1 l-i 1-1 E K P^ fc g PL, PL, Ph 3 3 .5 a, aJ3 J3J3J^ S3 p< cu a g EP^Ph^^^^P^P-.P-iPhU 33ei|0A aSe-WAy aSjBipsiQ jo aje^i u->inO OOOOC)OOOOC>000 m*r»o O O tnt^o 6 6 6 «t-3 -4 d od d rorod d « od i-J dod SJBU 3A1J aAiiisOj jo iqSpAV ■ ■ u-j ■ VO ■ t ! ro ! vd '. ', '. '. '. ' '. '. '. '. ' '. '. '. ' . . "-> . vO • ■ ■ ■ u-j • « O uno • • • . . - r-. • • CO 00 vO • • Noti o >o ; !oo\; ; * • *. ; : * ^o oo m o M M ; ; m w ; ; ; ; ; ; ; ; « <-,^ ; m m : m 3 ^3 ^3 ja "U M- Q .3 ^fe m ciu 2 a is a > £ c c c c > M W flj flj « SB..-S 5§qS< 2 3 0.3 S B ■3 3 MPS BfflBUUUUUUU S S 2 ^ S QQ APPENDIX 255 j? >< s u B B V ro ■a .5 i 1 B « ■jsa 1 1 ►> ^ (A 3 3 en 3 3 3 08 Sj OJ ■"&, E" bJ3b£ tsts Vj lH in u 8 O Cham . . Treac Treac .. E. W vt 1* sr M m j- *? P. en O .5 S WW in in in "a 'tS O O O J) 3 w k, 1, vO 00 r-«00 (N <*> : 00 °n : tJ-O * ; vi t^oo ! W VO N «^o O ^W ^-vO mvO on : so w ; *-»\o « « : . >HK1 »H N CO m m rn c* •H fO ■3- O vo cv "M-" r«i« NM fO« Ov POVO 00 n- ^ «*■ CO M _ « m loo 1- MH OOflflHH « « 1-1 « N« W ^-\o M -*■ : M n« co ; CN 1-* ; m ro O <« *» ■ 0000 • O • • \0 u-» ■ O • ■ v} CVOO min 5 - t^ h WIN • m vo ONVO • 00 • • W « ■ w • w tN « « d ' xn*n ■* w . ci . . rn« . POvO . co "^ . VO fOii wimifl Of>!««(HM • . • .B.dHlOOl.3B ■9=2 :3 .a j! U CJ JO B B j3 in" in"~ Si MM" 93d "J 256 THE STORAGE BATTERY o < Robertson Robertson Lea Forbes Reynier D'Arsonval Prospectus Prospectus Robertson Prospectus Prospectus Geraldy E. W., 5-26-'94 sSbjjoa. aSe»Ay • O "in&QOOO^O fovo *r> q o p Q Q ■ ; ^ 31BI(J 3AI? -isoj ji[3«jtt. l B *°X -i-[EII3JEJ^ 9A113y sauisoj jo ihSiOjYv a . ci d Aouapujjj AjJiaug a inojr -J3AVO J-3BJO} J 13(1 spunoj; ui ?i{3f3.M. . .0000 Iro! !t^t^. ,fO. . .O . ■* ON ai3ldui03 'tP3J°P uno ■ : : : : : : J : j : • : : : j *j 73 O » Jacquet .... Kabath, de . . . Khotinsky, de . . Legay Lehman & Main . Lehman & Main . Lithanode, Fitzgerald Lithanode, Frankland Monnier .... N. Y. Ace. & Elec. Co Ohio Stor. Bat. Co. Peyrusson . . . Phillipart .... Pollak APPENDIX 257 P4« 3 g c a o u in 2; « « ■£ JE a a STSS in S S in 8 °v£tzi °o u -S -0 ■5 °- ■9 p< -o OS _ o o ■» "5 S J2 o ? N HI 3 sH in 3 u u rvi o Cua . a. £ 55 Jx in WW uWWcei hlJ fcfcfcfl, q -<*■ m q q • in\d 1 •qq ■ q n^o p no o I d ^- ; rood 6 d en ** »■« i>- ! ; >-Qoo q q q q q ! !«5 d ci h m lown q q *n ■ O rod i-i q q t>« '. \ '. '. - * ! I cc in !/■ o • • *• • o o ■ t^. . . . . i-»oo * • t^. t^. q q * ■ * * ■ q '. *>■ l>- C*» t>- '. '. I '. ! N , \o \Q \q \£> , t t , ,ro ; ; o o • • * * - • • • -oooo i w inoo O | O <4* 1 ■**■ « : \o o> *-»m win vo o 3 hi W « , n . h. M «n . q\ rooo vqooinqinin-Noo . J con ir> irioq ^-nn n nq\ • *° "1" O 4 " »oNio«nrnfi w^d in ■ n «' * £ so ci m m co\d in 2 ci h« ■ ir> ro ■ OiO HN^ ft 0\0 J iJ iJ ■-]<-} H UJhlHH S3 BJ