1 64 



NA TURE 



[June i8, 1903 



economic effect of interheaters, through which the steam 

 is passed on its way from one cylinder to another, of a 

 compound or triple-expansion engine. During the past 

 half-century, numerous types of interheaters have been de- 

 signed and applied more or less spasmodically ; while, in 

 recent years, the use of such appliances has become a 

 prominent feature in certain branches of American practice. 

 The data on which the use of such heaters is founded, how- 

 ever, are far from being of a satisfying character, and they 

 present discrepancies which certainly require clearing up. 



What is really required is accurate information as to the 

 e.xtent to which our most advanced steam-engine practice 

 can — especially in the case of large power units — be im- 

 proved by the use of superheated steam, and as to the 

 manner in which such improvement can best be realised. 

 In connection with this matter, I may point out that we 

 are much in want of a thorough determination of the 

 physical properties of superheated steam, extending over 

 the range of temperatures and pressures likely to be em- 

 ployed in practice. Such a determination may, I hope, 

 soon be undertaken. Equally desirable also is the thorough 

 investigation of the action of steam — both saturated and 

 superheated — in the various types of turbine motors, a 

 matter which has, as yet, been by no means dealt with so 

 exhaustively as its great, and rapidly growing, practical 

 importance deserves, and respecting which many lessons 

 undoubtedly remain to be learnt. 



In addition to the various points already mentioned, the 

 question of the economy to be secured by the use of still 

 higher pressures of steam than are now used requires in- 

 vestigation. We are without any direct determination of 

 the latent heat, volume, and temperature corresponding to 

 pressure in the case of steam of pressures exceeding 350 lbs. 

 per square inch. The published data relating to steam of 

 higher pressures have been obtained by extrapolation, and 

 are by no means strictly to be relied upon. 



The thorough investigation of the theory and practical 

 working of internal-combustion engines presents for solu- 

 tion problems at once so numerous and so varied as to tax 

 to the utmost the skill and ingenuity of the experimenter. 

 There appears to be good ground for believing that with 

 an increase of temperature there is a very substantial in- 

 crease in the specific heats of such gases. While, how- 

 ever, the general fact may be regarded as proved, the 

 numerical data necessary to enable that conclusion to be 

 turned to practical account are far from having been fixed 

 with certainty, and further determinations are greatly 

 wanted. 



The value of experiments on internal-combustion engines 

 depends in a most important degree upon the accuracy with 

 which variations of temperature can be observed, both in 

 the cylinder before and during explosion, and in the walls 

 of the chamber in which the explosion occurs. As Prof. 

 Callendar has pointed out, the temperature assumed by the 

 platinum wire of an electric resistance thermometer ex- 

 posed to such gases must necessarily be less than that of 

 the gases themselves. Moreover, the rate at which heat is 

 communicated from the gases to the wire is dependent not 

 only upon the difference of temperature, but also on the 

 pressure, in a way not yet accurately known ; and thus the 

 accurate determination of the results of explosions in in- 

 i^ernal-combustion engines means not merely the skilful use 

 of known appliances, but the determination of certain 

 physical constants involving much expenditure of time and 

 labour. Then, again, the effect of the injection of water 

 or water-vapour into the cylinder in itself offers much 

 scope for investigation, as does also the influence of the 

 quality and quantity of the -lubricating oils on the gaseous 

 mixture. The governing of internal-combustion engines 

 and the regulation of the powers developed by them at 

 various speeds and under varying conditions are also 

 matters which present many unsolved problems. 



In the case of large bridges, roofs, and structural work 

 of that class, there is ample scope for aid to be given by 

 the better determin&tion of the amount and effect of wind- 

 pressure — a branch of experimental inquiry which is at 

 present far from being in a satisfactory state. What is 

 greatly required is a thorough investigation of the action 

 of the wind on surfaces of different areas and shapes, and 

 particularly its effect on partially shielded areas. Amongst 

 other points requiring settlement is the action of wind on 

 the lee-side of roofs — a matter on which the experiments of 

 NO. 1755, VOL. 68] 



Irminger have thrown much light, but which still requires 

 further investigation. 



In the determination of the stresses induced in the 

 elements of a structure by the forces applied to that struc- 

 ture, there still remain many problems of importance im- 

 perfectly solved. The theory of the plate-web girder, for 

 instance, is in a far from satisfactory state, particularly as- 

 regards the action of web stiffeners, the stresses on the 

 web itself, and those on the connections between the weib 

 and the flanges. The whole subject of resistance to com- 

 pound stresses — such, for instance, as those existing in the 

 web of a plate girder or a flat stayed plate, forming part 

 of a steam boiler — is one urgently requiring further experi- 

 mental investigation. 



Then, again, we are now largely using hollow shafts for 

 marine and other purposes, and the relation of these to 

 solid shafts of the same nominal strength, as regards the 

 power of resisting repetitions of varying or alternating 

 stress, has not yet been systematically investigated. 

 Another point is the effect of oil-tempering and different 

 modes of annealing on the endurance of fatigue, a matter 

 which, in view of the effect of similar treatments on the 

 ultimate strength and limit of elasticity of steel, is one of 

 much importance. 



The great problem we have still to face — and it is a 

 problem which will tax to the utmost our powers of re- 

 search — is the determination of what the change which we 

 call elastic fatigue really is. The indications of ordinary 

 testing machines do not reveal any change in the behaviour 

 of a material which has certainly exhausted a large pro- 

 portion of its " life " under repeated applications of stress,, 

 and we must evidently, to solve the problem, have recourse 

 to other modes of inquiry. What is the change of struc- 

 ture produced by fatigue, and in the case of any but pure 

 metals is this change accompanied by any rearrangement 

 of the constituents? How is this change of structure 

 affected by variations of treatments, by annealing, or, ira 

 the case of steel, by tempering? 



It is sometimes of considerable importance to ascertairr 

 whether a certain object, as, for instance, a propeller shaft, 

 or a portion of a bridge structure, or a steel rail, has or 

 has not been injured by the repeated applications of stress- 

 to which it has been subjected ; and at present the only 

 method of determining this is the testing to destruction of 

 the object respecting which the information is desired. 

 But if we knew accurately in what part of the object the- 

 stresses to which it had been subjected would first cause 

 injury, and if we further knew in what way the existence 

 of such injury would be indicated by change of structure, 

 it would follow that the microscopic examination of a small' 

 portion, cut from the most sensitive part of the object, 

 would afford a valuable indication of what was going on. 



There are other questions which appeal directly to the 

 users of steel. Amongst such questions are the oil-temper- 

 ing of mild steel forgings and of steel castings ; the in- 

 vestigation of the treatment during manufacture and 

 hardening of spring steel ; the examination of the qualities- 

 of special steel alloys, suitable for the construction of engine 

 or machine details, in which exceptional strength and light- 

 ness are essential ; and the production of alloys capable of 

 resisting corrosion and withstanding great changes of 

 temperature, and thus specially suitable for the construction 

 of superheaters and other apparatus in which such changes- 

 occur. 



We have in new steels a series of materials which promise 

 to revolutionise a very important percentage of our machine- 

 work, and to necessitate very material alterations in the 

 proportions of our machine tools, involving very heavy 

 outlay, if we wish to advance with the times. Now these 

 arc facts pointing to the necessity for extensive research- 

 conducted in a thoroughly systematic way. 



I have endeavoured to show how desirable it is that the 

 engineer and the physicist should work together in dealing 

 with certain investigations which I have enumerated, and I 

 have done so because, although engineers generally now fully 

 appreciate the aid which physical science can afford, there 

 ha? not hitherto been such an intimate association of the two 

 classes of workers as is really desirable. But with elec- 

 trical engineering the case is quite different. We are 

 accustomed to speak of the extraordinarily rapid development 

 of electrical engineering, and the marvellous way in which 

 it is assuming such a paramount position in civilised life. 



