SECTIONAL TRANSACTIONS.— G. 359 



Dr. F. W. Carter, F.R.S. — Magnetic noise in dynamo-electric machines. 



The paper deals with vibrations in dynamo-electric machines caused by 

 the movement of the magnetic field with the rotor of the machine ; and 

 particularly with such as result in the objectionable howling noises which 

 certain machines emit. The vibrations are shown to be due to one or other 

 of the members having a toothed magnetic structure, which causes periodic 

 variation in the field, and consequently in the mechanical traction on the 

 structure. The conditions favouring vibration are elucidated, and the 

 relation between the frequency of vibrations and the numerical data of the 

 machine determined. It is shown, moreover, that certain harmonics in 

 the periodic magnetic forces which act on the teeth are essential to the 

 vibration ; and methods for computing the forces are discussed, applicable 

 particularly to synchronous machines. In an appendix a comparison is 

 made between two machines from the point of view of vibration, and it is 

 estimated that one has some eighty times the potency of the other in pro- 

 ducing sound. In a second appendix certain features of the sound from 

 travelling wave sources are discussed, and it is shown that the sound 

 produced is not effectively broadcast unless the pulse of distortion which 

 produces it travels around the machine with speed greater than that of 

 sound in air. 



Prof. E. G. CoKER, F.R.S. , and Dr. Ruth Levi. — French and British 

 cleavage tests of timber. 



The need of maintaining a sufficient supply of useful timbers is a matter 

 of world-wide importance, and the numerous scientific and other questions 

 which arise have led to the establishment of a number of State laboratories 

 for the study of these problems. Among these the mechanical properties 

 of timber take an important place, and numerous standard tests have been 

 devised which are in process of co-ordination, in order that the results from 

 all sources shall be strictly comparable. 



This co-ordination is especially difficult for a material like timber, vvhich 

 has unlike properties in different directions, and although the grain of 

 timber is one of its most prized characteristics, yet its presence renders a 

 scientific classification of its mechanical properties a task of the utmost 

 complexity. 



At the suggestion of Sir Alfred Ewing, many of these timber tests have 

 been investigated by photo-elastic means, using transparent and isotropic 

 materials. 



This procedure aflfords stress distributions of more simple types than 

 occur in actual test members of anisotropic bodies like timber ; nevertheless 

 they afford considerable aid in ascertaining the actual nature of the practical 

 tests employed, and also help to explain some of their anomalies. 



Cleavage tests are usually made on notched specimens of timber so designed 

 as to ensure fracture along the grain at a place of minimurn cross-section. 

 This does not always happen, and photo-elastic investigations of French 

 and British standards prove that the minimum cross-sections are not places 

 of maximum stress intensity. 



The paper describes these experimental observations in detail, and shows 

 that various cleavage tests now made are not properly co-ordinated. 



The suggestion is made that a better form of test might prove to be a 

 simple tension test with the grain of the timber arranged perpendicularly 

 to the line of the load. 



