THE EFFECTS OF HEAT TREATMENT ON FINE 
METALLIC SUSPENSIONS 
By N. N. ZirBewu 
DEPARTMENT OF Puysics, THE Rice InstTituTE 
(Received February 15, 1932) 
ABSTRACT 
When a suspended system is supported by a fine wire the equilibrium position 
usually changes slowly for a long time after the load is applied. The equilibrium posi- 
tion also changes with temperature. It is found that both of these disturbing factors 
can be eliminated by a suitable heat treatment of the wire. Observations have been 
made on tungsten and platinum-iridium wires of sizes suitable for use in the Eétvos 
torsion balance. 
INTRODUCTION 
ANY types of physical apparatus involve the use of a rotating system 
suspended by a thin filament of metal or other material. Numerous ir- 
regularities in the behavior of an arrangement of this sort are often observed, 
such as changes in the equilibrium position with time and temperature, 
changes in the torsion constant of the suspending filament and failure of the 
suspended system to return to its initial position after-a displacement. These 
irregularities are especially serious in the Eétvos torsion balance, in which 
the suspension is very thin and is heavily loaded. Investigations of various 
sorts on the behavior of suspension wires have been published by Shaw and 
Lancaster-Jones,! Boys,? Threlfall,? Kénigsberger* and others. The present 
work is devoted to an attempt to remove by suitable heat treatrnent the 
changes in equilibrium position with time and temperature. On this particu- 
lar phase of the subject very little has been published. 
APPARATUS AND PROCEDURE 
Observations have been made on several tungsten and platinum-iridium 
wires of diameters and lengths suitable for torsion balance use. The furnace 
used to anneal the wires and for the time drift and temperature coefficient 
determinations was a brass tube wound with an electrical heating coil and 
well insulated with asbestos. A glass window inserted in the side allowed the 
angular position of the suspended system to be observed with a lamp and 
scale, and a thermometer inserted at the top allowed the temperature to be 
read. In all the work the scale distance was 120 cm and the scale divisions 
millimeters. 
1H. Shaw and E. Lancaster-Jones, Proc. Phys. Soc. London 35, 151 (1923). 
2 V.C. Boys, Phil. Mag. 23, 489 (1887). 
* Richard Threlfall, Phil. Mag. 30, 99 (1880). 
4 J. Kénigsberger, Zeits. f. Physik 40, 729 (1927). 
32 
176 
