ESTIMATION' AXD COXTKOL OF OPERATE TIME OF RELAYS 179 



Conductance Shunt 



In a companion article, a hyperbolic relationship between the flux 

 and current is used to represent the portion of the hysteresis looj) of 

 concern in release. This, with a conducting sleeve or shunted winding, 

 gives an excellent representation of the release waiting time. It also 

 provides an understanding of the controlling parameters, even with only 

 eddy cui'rents controlling the release. 'I'he form most useful for release 

 consideration is: 



^ ^ (.p" - ipo){Gs + Gc + G,) / \nz _ 1\ ^ ^3^^ 



NIo V^ — 1 2/ 



where 



<P — (Po 



z = 



(f — (fO 



In these experessions, <p is the flux corresponding to the ampere turns iV/o 

 at which the relay just releases, (po is the residual flux, and (p is the 

 asymptotic saturation flux. The conductance terms are the same as for 

 the operate case except Gc now is computed using the dc winding re- 

 sistance plus any winding shunting resistance. If the winding has no 

 shunt then Gc = 0. The operated contact spring load enters through ^ 

 and Nh ■ 



The function of z in brackets has a broad maximum in the region of 

 relay release. It therefore is appropriate to consider the releasing ampere 

 turns as the independent variable and plot the measured release time 

 with the total conductance as a parameter covering the range of interest. 

 As Ge is a characteristic of the structure and not subject to adjustment, 

 the curves are actually labelled in terms of just the sleeve, if any, plus 

 the shunted winding. For the wire spring relay, Fig. 14 shows data in 

 this form. If the function of z were truly constant, the curves would 

 all have a slope of — 1 in this plot. 



Strictly speaking, the above equation applies only to the time until 

 the magnetic pull has decayed to equality with the operated spring 

 load. Following this is the motion time, during which the pull decays 

 further. For convenience, however, the armature motion time through 

 the distance to the nearest contact is included in the chart for release 

 time as (a) the contact actuation is the means used to measure the time 

 and (b) this much motion always takes place before any contact is 

 actuated. 



The further displacement of the armature continues at almost con- 



