NATURE 



[November 25, 1909 



that the science is complete and that all the problems it 

 presents have been finally solved. Abstract as these in- 

 vestigations often are, there is ample room for the applica- 

 tion of those general principles of scientific research which 

 his earlier training will have helped to develop, and the 

 final test of his mathematical powers will be found in the 

 success with which he extends the scope and methods of 

 the science. 



Mathematics as we know it to-day is in living contact 

 with experimental science on the one side ; on the other 

 it borders on the domain of philosophy ; to each it has 

 some contribution to offer, and in the' words of Weier- 

 strass " a mathematician who is not something of a poet 

 will never be a complete mathematician." Is it not, then, 

 a subject worthy of a place in university studies? 



DEVELOPMENTS OF ELECTRICAL 

 ENGINEERING.' 

 "T^HIS address deals with a few only of the many recent 

 developments in electrical plant and its application 

 to industrial purposes. 



Generators. 



The modern tendency is to instal very large units. 

 This is partly due to the large demand made on the 

 power house and the desire to restrict the number of 

 units, and partly to the fact that the advantages of the 

 steam turbine over the reciprocating engine become more 

 pronounced with the increased size of the unit. The 

 General Electric Company of New York have built several 

 turbo-alternators of 14,000-kw., and the British Westing- 

 house Company inform me that it would be quite feasible 

 to build sets of 15,000 kw. up to 15,000 volts pressure. 

 In water-driven alternators, also, the tendency is towards 

 large units. Thus the power house of the' Norwegian 

 Nitrogen _ Company at Svalgfos, near Notodden, has been 

 fitted with four turbine-driven three-phasers, each for 

 10,500 kilovolt-ampere, and developing 7000 kw. at 10,000 

 volts. It is obvious that in these circumstances special 

 ventilatmg arrangements become necessary. Dr. Kloss, in 

 a paper read before our Institution about a year ago has 

 pomted out that the scientific wav of ventilating turbo- 

 dynamos IS to take the air from the outside and discharge 

 It to the outside of the engine-room. It is important that 

 only clean air be used, and for this reason air filters are 

 built into the inlet ducts. These are formed of pockets 

 of porous cloth extended over wooden frames, and so 

 placed that the dust which settles on the cloth mav be 

 removed by beating or with a vacuum cleaner. Washing 

 or chemical cleaning is only required after some years 

 of use. 



In most modern electricity works the circulating and air 

 pumps are driven by electric motors, but this method has 

 been replaced at the works of the AUgemeine Elektricitats 

 Oesellschaft by turbo-driven centrifugal pumps. No piston 

 purnps at all are used, and the feed may be regulated 

 without paying attention to the feed pu'nip. The feed 

 water obtained by this method is absolutelv free from air 

 and only 5 per cent, of make-up for the feed is required! 

 Since no piston engines of any kind are used, there is no 

 need for oil filters. 



An important development in turbo sets was initiated 

 about ten years ago by Prof. Rateau with his exhaust 

 steam turbine. _ The cost of adding exhaust steam turbo 

 sets to an existing installation of large size may be taken 

 at from 6/. to lol. per kilowatt exclusive of thermal 

 jtorage I he commercial advantage is considerable. Thus 

 in the Osterfeld Mine a Rateau plant installed at a cost of 

 53;00o! has resulted in an annual saving of about 20,000!. 



Ihe desire to reduce the cost and complication of switch- 

 gear and to make paralleling easy has led to the use of 

 non-synchronous machines as generators. The rotor may 

 be a squirrel-cage of very simple construction and requir- 

 ing hardly any insulation, no matter how high the pres- 

 sure produced by the stator mav be. The mechanical con- 

 struction IS easier than that of the revolving field of an 

 ordinary turbo-alternator, and since the air space can be 

 made small, the power factor is high. A sooo-kw. non- 



1 Abridged from an address delivered before the Institution of Electrical 

 Engineer, on November w by Prof. Gisbert Kapp, president of the ins.itutbn. 



NO. 2091, VOL. 82] 



synchronous generator was last year added to the 

 plant of the Inter-borough Rapid Transit Company, New 

 York. 



There is some difficulty in the design of turbo-alternators 

 for very low periodicity, since the speed becomes insuffi- 

 cient for the satisfactory working of the turbine. To meet 

 such cases Mr. E. Ziehl has devised a type of alternator 

 which he calls a " double-field generator." The principle 

 may be explained as follows : Imagine a non-synchronous 

 motor having precisely the same three-phase winding in 

 stator and rotor, and let the circuits be connected either 

 in series or parallel in such way that a three-phase current 

 sent through the machine will produce fields which in stator 

 and rotor revolve in opposite sense. If now the rotor be 

 driven by power in a sense opposite to that of its own 

 field and with a speed corresponding to twice the fre- 

 quency, the field produced by the rotor currents will in 

 magnitude and direction of motion be identical with that 

 produced by the stator currents. Thus each of the two 

 windings contributes one-half the field common to both. 

 At the same time the demagnetising action of each winding 

 is eliminated by that of the other. Since the E.M.F. is 

 generated in both windings, only half the flux as compared 

 to a synchronous generator is required ; hence less 

 hysteresis loss, smaller radial depth of stampings, and less 

 copper weight. The paralleling is easy ; the speed need only 

 be approximately right, and if coupled up in a wrong phase 

 position no damage is done, since the inductance is then 

 very great. 



Transformers. 



In transformers also there is to be noticed a general 

 tendency towards large units, which is not surprising if 

 one considers that for the calcium-carbide industry alone 

 about half a million horse-power in generating plant has 

 been installed throughout Europe, and that most of the 

 power has to flow through transformers to the carbide 

 furnaces. 



The General Electric Company of America have built 

 several io,ooo-kw. three-phase transformers working at 

 60 frequency, and giving a pressure of 100,000 volts. The 

 largest European transformers of which 1 could find a 

 record are some made by the Siemens-Schuckert Werke. 

 They are three-phase 6750-kilovolt-ampere capacity oil 

 cooled, for 66,000 volts on the high-pressure side. The use 

 of oil as a filling medium has made it possible to build 

 transformers for very high pressure. In one American 

 power-transmission plant now under construction the step- 

 up transformers are intended to raise the pressure to 

 1 10,000 volts, but even higher pressures can be obtained. 

 Transformers giving e.xtremely high pressure on the 

 secondary are used for testing insulators and insulating 

 material. A transformer of this kind has recently been 

 made by Messrs. Brown-Boveri. It is a 50-kilovolt-ampere 

 transformer wound for a primary pressure of 1000 volts 

 and giving on the secondary 250,000 volts, but even this 

 has been exceeded when the transformer was used in test- 

 ing the dielectric strength of insulators. From a curve 

 referring to such tests which the makers have sent me I 

 find that the highest pressure recorded was 310,000 volts. 



The reduction in weight of transformers due to the use 

 of alloyed iron, large units, and vigorous cooling is very 

 remarkable. As an example of good modern practice, I 

 take a Brown-Boveri transformer where the active material 

 weighs only 3-1 kg. per kilowatt, and the efficiency is 

 986 per cent, at full non-inductive load. In an Oerlikon 

 3500-kw. transformer the active iron only weighs 7 tons, 

 being at the rate of 2 kg. per kilowatt output. The largest 

 self-cooling oil transformers of which I know are some 

 i200-kw. three-phase 40-frequency 5000-volt transformers 

 made by the British Westinghouse Company, but for 

 larger unit artificial cooling becomes necessary. 



For furnace work it is well to allow a rather large 

 inductive drop so as to reduce the rush of current in 

 the event of a short circuit in the furnace. This means 

 wide spaces between primary and secondary poils, but it 

 also involves the necessity for good mechanical support. 

 The mechanical forces acting on the individual coils may 

 become considerable, and this is probably the reason why 

 some makers prefer the core type with concentric cylindrical 

 coils, the cylinder being the best shape for resisting radial 

 forces. 



