486 THE BELL SYSTEM TECHNICAL JOURNAL, MAY 1952 



element which is carried at its base by a diaphragm and at its apex by 

 cantilever springs. These furnish the restoring force and restrict the mo- 

 tion of the moving system to a single degree of freedom, motion parallel 

 to the cone and coil axis. The second or feedback coil is secured to the 

 cone near its apex, at which point the stylus or drive pin is attached. 

 The coils move axially in annular air gaps polarized by a single magnet. 

 In the space between the two coils, copper ring shielding (a shorted turn) 

 is provided to minimize inductive coupling between them. The output 

 of the driving amplifier is supplied to the drive coil, while the feedback 

 coil is connected in proper (negative feedback) phase to the amplifier 

 input. 



The voltage generated in the feedback coil is proportional to the 

 instantaneous velocity of the moving system, and by virtue of the nega- 

 tive feedback, the amplifier-recorder system becomes a high force, high 

 mechanical impedance generator of mechanical motion, with the veloc- 

 ity very nearly proportional to the input voltage over a large range of 

 frequency and mechanical load. Measurements of the voltage generated 

 in the feedback coil provides a means of monitoring the velocity. Enough 

 power capacity is present in the amplifier so that large changes in the 

 load will not cause changes in the motion. 



The samples of the materials to be tested for wear resistance are 

 carried by a grooved aluminum beam, one end of which is hinged, the 

 other being driven b}^ the record stylus. The rubbing member, in this 

 case 25-mil nickel silver wires, are tensioned against the test samples as 

 they might be in switching apparatus. The wires can be removed for ob- 

 servation and measurements of the wear, and accurately replaced as the 

 parts are do welled together. 



Fig. 13 shows a measurement of a number of materials for a normal 

 force of 30 grams (0.0665 pounds) and a slide of 2 mil inches. The A 

 phenolic, which is the same as that tested and recorded in Fig. 10 by 

 the 18,000-cycle barium titanate transducer, produced essentially the 

 same wear showing that the wear is approximately independent of the 

 rapidity of motion for these materials. Nylon showed a rather erratic 

 wear curve due to the fact that it has a low melting point and tends to 

 ball up on the wires. This effect was considerably more pronounced at 

 18,000 cycles, where a very large indentation was found. 



Only three materials show low wear at reasonably uniform rates out 

 to a large number of cycles. These are C phenolic, a fabric filled phenolic, 

 the B phenolic, a wood flour filled molding phenolic and the D phenolic, 

 a cotton flock phenolic with graphite added. At lower forces and shorter 

 slides the wear at 10^ cycles is approximately proportional to the force 



