NATURE 



217 



THURSDAY, JULY 4, 1907. 



TUE COXSTRl'CTKiX OF DYXAMOS. 

 The Construction of Dynamos {Altcrnaling and Direct 

 Current). By Tyson Sewell. Pp. xi + 316. (Lon- 

 don : Crosby Lockwood and Son, 1907.) Price 

 7*'. 6(/. net. 



S"" O many books have already been written on tlie 

 subject of the dynamo that any new publication 

 can only be justified on one of two grounds. Either 

 it must contain new matter of practical importance not 

 hitherto treated in such a way as to be useful to the 

 desii;ner, or it must be in comparison with existing 

 books an improvement in the wav of treating old and 

 well-known matter, so that the student and the prac- 

 tical engineer may have less difficulty in grasping the 

 subject than they often have with the existing books. 



A perusal of Mr. Sewell 's book will leave the 

 reader with the impression that the designer of 

 dynamos will learn nothing from it, and that the 

 student may with equal advantage read any of the 

 previous publications treating of the dynamo in a 

 popular style. The book is in this respect no better 

 and no worse than dozens of others. As regards its 

 use for the design of dynamos, the author himself 

 deprecates it, for lie says in the preface : — 



" The examples of design are introduced by way of 

 illustration only. The actual designing of dynamos 

 is the work of comparatively few men, most manu- 

 facturers having standardised particular lines which, 

 with slight modifications, meet most requirements ; 

 and in view of the labours of the Engineering 

 Standards Committee, further developments in the 

 direction of uniform practice may be expected." 



The idea underlying this sentence seems to be that 

 as only few men have to design dynamos, and as the 

 Engineering Standards Committee looks after these 

 things, it is not necessary to treat of design in books. 

 Yet in the first sentence of the preface the author says 

 that his work is an attempt to treat of theory, design, 

 and construction, and several chapters are devoted to 

 working out the details of design. 



The general arrangement of the book follows well- 

 known lines. We get first the fundamental principles 

 of direct currents. Ohm's law, its application to the 

 " Silvertown Set " and Evershed's " Megger," though 

 without description of the internal arrangement and 

 only an outside view of the box. Then come a few 

 pages on the magnetic field and on testing iron, but 

 the treatment is incomplete, and in one case also mis- 

 leading. This is the description of the magneto- 

 metric method, where the author fails to point out 

 that it is only applicable to thin and long wires, and 

 not to a turned iron rod, as he saj's ; and that the 

 effect of the solenoid must be eliminated by a special 

 coil. In the next chapter we. get all the well-known 

 illustration (but on a prodigiously large scale) to show 

 lines of force, and how by cutting them an E.M.F. is 

 produced, and the formula for the E.M.F. of a two-pole 

 dynamo. It would have been well if the author had 

 also given the formula for a multipole machine in 

 this place. The chapter following deals with the 



NO. iq66, vol. 76] 



fundamental principles of the alternating current, and 

 then we get to the alternating field. Here the author 

 introduces the hysteresis loop, and mentions that its 

 area is a measure of the work lost per cycle, but gives 

 no proof. The figures he gives for the hysteresis 

 coefficient of 0-002 to o-ooS are certainly much too 

 large, and he is also in error when he says that in 

 dynamo work the loss due to eddy currents is very 

 small. It is well known that the losses in iron 

 actually occurring in the dynamo are considerably in 

 e.xccss of the theoretically calculated losses, and that 

 the increase is in a large measure due to eddy currents 

 at the burred edges of the plates. When dealing with 

 capacity the author gives a neat hydraulic analogy to 

 show why the charging current is a quarter phase in 

 advance of the E.M.F. 



In the chapters on construction and theory of 

 bipolar dynamos we find the usual illustrations, but 

 all on a large scale, so that a good deal of space is 

 uselessly occupied and the text correspondingly re- 

 stricted. Although in the illustrations the author 

 follows well-known lines, this cannot be said as 

 regards his way of designing. Here we find some 

 novelty, but hardly improvement on the usual practice. 

 Thus, in giving the design of a 400 kw. 550 v. 

 machine, he starts with the rule that 100 kw. is a fair 

 average allowance per pair of poles, and finds thus 

 that eight poles is the right type. Then he calculates 

 the number of commutator bars on the basis of 5 v. 

 per bar, and by assuming a certain thickness he 

 arrives at the diameter, and again by assuming a 

 certain circumferential speed he obtains the number 

 of revolutions per minute. This freedom in the selec- 

 tion of speed may be convenient, but it is certainly 

 not the condition which prevails in practice. As a 

 rule the speed is given, and one has to design the 

 dynamo to suit it. Having thus found the speed, the 

 author proceeds to find length and diameter of arma- 

 ture by simply applying the so-called "output 

 formula," but as he does it without critical investi- 

 gation he gets rather unusual dimensions, namely, 

 93 inches diameter by isi inches long, whereas a little 

 common-sense reasoning untrammelled by adherence 

 to formula would have shown him that So inches by 

 20 inches would make a far better and cheaper 

 machine. The number of conductors is found from 

 a formula of circumferential current density, and so 

 the designing goes on simply by applying formulas 

 without criticism, not a method to be recommended 

 either to students or to " engineers who have occasion 

 to deal with technical matters," as the author says in 

 the preface. 



Another claim made in the preface is only partially 

 justified. The author says that " the available space 

 is almost exclusively devoted to machines representing 

 present standard practice." There are a few illus- 

 trations of the fly-wheel type of alternator, and also a 

 turbo-alternator, and in so far standard practice is 

 considered, but the " mono-coil-claw " type. Fig. 200, 

 the inductor machine. Fig. 201, and the ironless 

 m.ichines of Mordey, Ferranti, and Crompton, are cer- 

 tainly not present standard practice, whilst the details 

 of armature housing shown in Figs. 192 to 197 may 



L 



