A GOVERNOR FOR TELEPHONE DIALS 1287 



The additional terms, r, j and e in the limiting equation make it pos- 

 sible to design the governor for optimum regulation for any particular 

 value of n desired. Since the term r, the case inside radius, is controlled 

 by space reciuirements, the terms j and e assume added importance. They 

 are determined by the point at which the drive-bar arms act against the 

 weights and can be made anj^ value required to meet the design objec- 

 tive. 



I Figure 11 is a plot of doo/dG at various values of /x for the drive-bar and 

 fly-bar governors specified in Tables I and III. It indicates that for any 

 coefficient of friction the new drive-bar governor should exhibit less 

 change for a change in input torcjue than is possible with the fly-bar 

 governor. 



EXPERIMENTAL DATA 



To substantiate the theoretical conclusions drawn from the analysis 

 of the steady state speed eciuations, experimental data were compiled 

 from a nvimber of models of each type of governor. Drive-bar and fly-bar 

 governors, made to the specifications listed in Tables I and III were 

 investigated to determine their response to changes in input torque and 

 changes in the stud-to-case coefficient of friction value, m- The tests were 

 conducted on 7-type dials manufactured by the Western Electric Com- 

 pany as standard product. 



Dial Speed Versus Coefficient of Friction 

 The theoretical analysis of the equation 



doi Gd 



indicates that the two types of governors should exhibit identical fric- 

 tional characteristics. Fig. 12 represents the theoretical plot of dial speed 

 in pulses per second versus coefficient of friction. The single curve satisfies 

 both the fly-bar and drive-bar governors as specified in Tables I and III. 

 This cun-e shows the change in speed of a dial initially adjusted to 10.0 

 pulses per second, operating at normal torc^ue, as the coefficient of fric- 

 tion varies. It indicates that if there is a decrease in the value of n from 

 that which existed at the time of initial adjustment, there will be a cor- 

 responding increase in the dial speed. The curve is drawn with m = 0.25 

 as representing the normal stud-to-case condition and a normal governor 

 input torque of 7,500 dyne-cm. 



The experimental data were compiled for the following operating con- 



