384 SECTIONAL TRANSACTIONS.— G. 



15. Prof. G. W. 0. Howe. — The B.E.S.A. Glossary of Terms used in 



Electrical Engineering : A Protest and a Suggestion. 



16. Mr. W. J. Kearton. — The Distribution of Pressure in Impulse Steam 



Turbines at Varying Loads. 



The nozzle areas in impulse steam turbines are determined in the process of design 

 so as to bring the ratio of blade velocity to steam velocity in each stage as near to- 

 that value which ensures maximum efficiency as economic conditions will permit. 

 At partial loads in steam turbines of normal design, and also under the very variable 

 conditions in turbines of the mixed-pressure and heat-extraction types, the nozzle- 

 areas are no longer correct and the distribution of pressure and hence of heat-drop in 

 the various stages becomes very different from that which exists at the economical* 

 load. 



The present paper outlines a method whereby the actual pressure distribution 

 may be calculated. The pressure distribution in a nozzle is largely a function of 

 the cross-sectional area at different points along the nozzle axis, and may be- 

 determined readily. In a turbine, however, the velocity of the steam is a dis- 

 continuous function. The method makes use of (a) a curve showing the actual total 

 energy available by expansion from the initial to any given pressure, (6) a curve- 

 showing the approximate specific volume of the steam at any pressure, and (c) the 

 equation of continuitj'. The calculation is by a step-by-step process beginning at 

 the exhaust pressure. 



The paper illustrates the application of the method to high-pressure steam, 

 turbines with nozzle-control governing in the first stage, to mixed-pressure turbines,, 

 and to heat-extraction turbines. The ratio of blade speed to steam speed is affected 

 in certain stages to an appreciable extent, with a reduction in efficiency as a con- 

 sequence. Relative steam velocities are altered and shock losses introduced. In. 

 certain types of plant, considerable free-expansion losses are involved. 



17, Prof. C. H. BuLLEiD. The Fatigue of Cast Iron. 



18, Dr. A. P. Thurston. — Classification of Patent Specifications, 



Patent classification has to achieve various somewhat conflicting objects, i.e. - 

 (a) to enable a manufacturer or individual to ascertain whether he may manufacture- 

 or use an article or process with impunity ; (b) to enable a manufacturer or individual, 

 in the event of receiving a threat of an infringement, to ascertain what his chances 

 are if he fights the action ; (c) to enable an incjuirer to ascertain the whole existing 

 public knowledge of any art or manufacture ; (d) to enable an inventor or his agent to 

 claim as much as possible under his patent ; (e) to enable the staff of H.M. Patent 

 Office to protect the public by seeing that no more is claimed than is permissible, 

 t.e. that the claims are novel ; (/) to enable a manufacturer or individual who finds 

 that his competitors have obtained patents to demonstrate the invalidity of such 

 hostile patents or to restrict their scope. 



Patent classification should obey the basic laws of classification, e.g. : {a) it 

 should proceed from wide headings to narrow headings, from terms of great extension 

 and small intention to terms of small extension and great intention ; (b) it should be 

 adapted to admit interpolations : (c) the characteristics chosen for subdivision of 

 any one class or sub-class must be consistent throughout that class or sub-class ; 

 (rf) the headings must be mutually exclusive ; (e) the terminology used must be 

 consistent, the meaning of terms being invariable throughout. 



Patent classification is a classification of subject-matter rather than of literary 

 material, and the subject-matter is available generally in penny numbers. Therefore 

 it is not necessary to follow another law of book classification, namely, to allot a 

 book to the nearest heading of the classification that will wholly contain it. The 

 American schedule obeys this rule, but the British schedule, in which every sub- 

 division is self-contained and in which there are no ' superior ' and ' inferior ' 

 headings, does not. 



The system adopted by the British Patent Office is.in theauthor'sopinion,a8uperb 

 achievement. It has now been in daily operation for over twenty years, namely, 

 from the 1st of January, 1905, and has enabled searches to be made with precision 



