14G THE BELL SYSTEM TECHNICAL JOURXAL, JANUARY 1954 



winding, in the definition of this article exists, as exhibited by the minima 

 on Fig. 1, was shown in Part I to be a consequence of equation (12). 

 This best winding and its determination is the basic subject of this 

 article. 



OPERATE TIME SIXGLE RELAYS 



As in Part I, the operating time of a relay is considered as made up of 

 three stages: (1) the waiting time while the armature remains at the back- 

 stop and the pull builds up to equality with the back tension, (2) the 

 motion time beginning at the end of the waiting time and continuing 

 while the armature moves from the backstop to the position of the earliest 

 contact, and (3) the stagger time during which the armature actuates 

 all of the remaining contacts. 



The mass controlled case, as a practical matter, simplifies to a determi- 

 nation of only the first two without regard for the spring load, with the 

 displacement of the armature taken to the latest contact in the array. 

 The armature pull builds up to values in excess of the load during its 

 early motion and the velocity is so high during the stagger time that 

 this small interval can be included as part of the motion time. This 

 treatment is essentialh^ that of the three stage approximation of Part I. 



The load controlled case simplifies to a determination of the time 

 required for the pull to build up to the maximum load, including the back 

 tension, with most of the relatively slow motion taking place during this 

 pull buildup. The remainder of the motion time is accounted for as an 

 empirically determined correction factor in the expression for time of 

 pull buildup. This treatment corresponds to the single stage approxima- 

 tion of Part I, with a correction term to account for the additional motion 

 time. 



Fig. 1 shows graphically the basic characteristics of the two tx'pes of 

 operation. This chart displays the operating time of the same relaj^ under 

 the conditions of 1.0 or 5.7 watts, and three contact load conditions: 12, 

 18, and 24 contact pairs. For each of the six conditions, a curve is drawn 

 showing the effect of the number of winding turns. It is clear that (1) 

 there is a best number of turns for each case, (2) for the 5.7 watts the 

 best number of turns is the same for all three loads, (3) for the 5.7 watts 

 case the difference in operating time is only 0.1 millisec in 5.6 millisec, 

 between 12 and 24 contact pairs, (4) for the 1.0 watt case the best number 

 of turns increases with the number of contact pairs, and (5) for the 1.0 

 watt case the operate time is double for 24 compared to 12 contact pairs. 

 These strikingly different behaviors form the basis for di\'ision of relays 



