TRANSACTIONS OF SECTION G. 709 



and, as a result, files are now required to reach a certain standard of cutting 

 power. A description of this machine with illustrations was given. 



Unfortunately very early in the history of the machine doubts were enter- 

 tained as to the accuracy of the results obtained by it, as files known to be 

 good were condemned by it. 



Eventually the writer was asked to report on the Herbert file-testing machine, 

 and a large number of tests were accordingly made. The results of these were 

 in many cases normal, but in others they showed extraordinary differences of 

 effectiveness of cutting power, not only among files said to be in all respects alike, 

 but between the two opposite sides of the same file. 



The writer reported that the machine appeared to be defective in one im- 

 portant point — namely, that it treated the file as a machine tool instead of as a 

 hand tool j thus in the machine the file moves across the face of the test bar 

 through an absolutely constant path, the respective teeth of the file each stroke 

 working in identically the same graves or furrows on the face of the test-bar 

 stroke after stroke. The result is that the face of the work occasionally becomes 

 glazed in appearance and the file ceases to cut, though the file itself may not be 

 worn out. In the case of hand-filing no two strokes are made in exactly the same 

 direction. The conditions, therefore, under which the tests are made in the 

 machine differ from those under which the file is worked in actual practice, and 

 this difference works, at least in some cases, to 'the disadvantage of the file. 



For the purpose of removing this objection the writer has devised an 

 addition to the Herbert machine, by means of which the path of the file in the 

 machine is no longer a constant one, but changes its direction stroke by stroke 

 as in the case of hand-filing. To secure this the file is no longer held rigidly at 

 its two ends, but is connected by ball-joints, the effect of which is equivalent to 

 that of a wrist movement at each end of the file. The variation of the path of 

 the file each stroke is obtained by slightly shifting the position of one end of 

 the file, relatively to the other end, by a simple mechanism during each return 

 stroke, so that on the following working stroke it moves in a different path from 

 that which it had in the preceding stroke. The means by which this movement 

 is obtained will be explained and illustrated by diagrams. 



The addition of this arrangement to the Herbert file-testing machine has 

 resulted in the removal to a large extent of the irregular results previously 

 obtained from files of similar quality. 



MONDAY, SEPTEMBER 5. 



The following Papers were read : — 



1. The Electrification of the London, Brighton, and South Coast Railway 

 between Victoria and London Bridge. By Philip Dawson. 



2. On the Use of an Accelerometer in the Measurement of Road Resistance 

 and Horse Rower} By H. E. Wimperis, M.A., Assoc.M.Inst.C.E. 



The author described the form of accelerometer recently invented by him and 

 constructed by Messrs. Elliott Bros. The instrument consists of a brass box 

 about four inches across containing a copper disc mounted on a vertical pivot 

 and ' damped ' in its motions by a permanent magnet. The e.g. of the disc is 

 purposely removed from the axis so that, when the box moves forward, one side 

 of the disc tends to lag behind, thus partially winding up a coiled spring and 

 actuating a pointer which moves ever a scale. To render the reading unaffected 

 by any accelerations at right angles to the direction of motion, a second parallel 

 axis is fitted, which is geared to the first one, and has attached to it masses having 

 the same mass moment as the disc itself. Couples about these two axes add up in 

 the direction of motion, but neutralise one another in any direction at right angles. 



1 Published in the Enc/iverr, September 16, 1910. 



