478 PROFESSOR W. E. AYRTON, MR. T. MATHER AND MR. F. E. SMITH: 



The castings, drawn tubes, screws, &c., intended for making the beams, pillars, and 

 other parts of the physical balance, were forwarded by Mr. OERTLING to the National 

 Physical Laboratory for magnetic test to determine their suitability, or otherwise, for 

 the purpose. With the exception of the castings, all the first samples showed distinct 

 paramagnetic properties, and were rejected, as also were several specimens submitted 

 subsequently. It was found necessary to obtain the tubes and rods from special 

 sources before satisfactory ones were secured. Many samples of brass screws were 

 purchased and tested, but none were sufficiently free from magnetism, so Mr. OERTLING 

 was obliged to cut all used in the instrument in his own shops. The locks and keys 

 for the balance case had also to be specially made, phosphor bronze being the chief 

 material used. No trouble was experienced with the paraffin wax used to coat the 

 windings. 



In the magnetic tests on the metallic parts of the physical balance the eddy- 

 current effects were small. When a brass rod was quickly inserted as a core to the 

 primary, a swing of 4 millims. resulted, the direction being the reverse of that 

 corresponding to increased permeability. To overcome this effect a thick brass rod 

 was cut in two and the metallic substance placed between and in contact with the 

 two portions, the whole being connected together by means of a metal tube so as to 

 realise as nearly as possible a continuous metallic conductor. The brass rod was so 

 long that when introduced into the primary it projected about 40 centims. from the 

 far end when the metal under test was about to enter the coil. An axial motion of 

 the metallic rod did not give rise to eddy currents capable of producing a swing of 

 0'2 millim. 



Further tests on the suspended system were made after the erection of the balance 

 by sending a current through one of the fixed coils when lowered so that the 

 corresponding suspended cylinder was without it. The rest-point of the balance was 

 unaffected thereby, and remained unchanged when the current was reversed. Similar 

 observations were made when the current was sent through the other fixed coil, but 

 no change was detected. 



SECTION 6. CONSTRUCTION, MEASUREMENT, AND INSULATION OF COILS. 



Preliminary tests at the Central Technical College and subsequent ones at the 

 National Physical Laboratory led to the choice of " First Statuary " Carrara marble for 

 the material of the cylinders. The tests showed this to be an excellent electrical 

 insulator and of negligible magnetic susceptibility. The preliminary insulation tests 

 were made on a small cylinder 4 inches in diameter and 2 inches in axial length. A 

 double screw thread (36 turns to the inch) was cut on this, and helices of No. 24 

 bare copper wire wound thereon. The insulation resistance between adjacent strands 

 was low at first, but rose to 4000 megohms when the cylinder was immersed in hot 

 liquid paraffin wax, removed, and allowed to cool. The magnetic tests have already 

 been described. 



