DEVELOP MENT OF A II.WDSET FOR TELEPHONE STATIONS 255 



output and the degradation of the quality are limited to a rather nar- 

 row range of positions in which the handset is seldom held. A loss in 

 output and a serious impairment of quality occurs when the usual type 

 of deskstand transmitter is held in the positions which would fre- 

 quently occur if it were used for handset service. This effect is shown 

 by curve B, Fig. 8. 



Response 



The diaphragm of the transmitter is made from thin duralumin 

 formed into a truncated cone with radial stiffening ribs. This reduces 

 the effective mass to about one-tenth that of the deskstand transmitter 

 and provides sufficient rigidity to insure vibration as a unit throughout 

 the frequency range of interest. A number of impregnated paper 

 rings each approximately four ten-thousandths of an inch in thickness 

 support the edge of the diaphragm. The dimensions of the recess into 

 which these rings assemble are so chosen that they separate slightly 

 from one another. This construction provides a resilient support for 

 the diaphragm and adds a certain amount of mechanical resistance due 

 to the viscosity of the air films between adjacent layers of paper and 

 the friction between the layers. An appreciable improvement in re- 

 sponse results from the lower mass and stiffness and the higher damp- 

 ing. The range in response of the handset transmitter in the fre- 

 quency range from 300 to 3000 cycles per second, is about 20 db as 

 compared with approximately 40 db for the deskstand transmitter 

 over the same frequency range. The more uniform response of the 

 handset transmitter causes a marked improvement in articulation over 

 that obtained with the deskstand instrument. 



A ging 



As previously mentioned, after a granular carbon transmitter has 

 been in use for a time, changes take place which often cause an increase 

 in resistance and carbon noise and a loss in sensitivity. These changes 

 in the contact conditions have been traced to two principal causes, the 

 abrasive action of one granule on another, and the deterioration of the 

 surface of the granules due to high temperatures resulting from exces- 

 sive contact voltage. The former is usually referred to as "mechanical 

 aging," the latter as "electrical aging." 



The aging of the deskstand transmitter is primarily electrical, for 

 although it is picked up occasionally and moved from one location to 

 another on the desk or table, it is not in general subjected to shocks or 

 jars which cause an appreciable motion of the carbon granules. Be- 



