DYNAMIC MEASUREMENTS ON ELECTliOMAGNETIC DEVICES 1431 



break contact of Bl with the opening of tlie make contact of A. The 

 A relay winding has a somewhat similar But fixed set of Hhimts across 

 it, to slow it down with respect to B2 and to equalize operate and release 

 time. The adjustment of the two potentiometers prot-eeds as follows. 

 The time ^i is set at 0, flux decay is chosen, and potentiometer 1 is ad- 

 justed to the point where the meter just starts to drop from the full 

 reading. Then flux rise is chosen and the potentiometer 2 is adjusted to 

 the point where the meter just starts to rise from zero. A fine check on 

 these settings is to record the meter readings for a lew ((lual small time 

 steps and observe that the successive differences are consistent. 



For applying an on and off battery condition as shown in Fig. 9 for 

 the A contact, only one contact on one relay is used. The A contact 

 circuit actually is made of two pairs of relays operated in a push-pull 

 arrangement. The circuit may be changed to connect the four sets of 

 transfer contacts into a lattice configuration to supply battery reversals 

 to the apparatus under test. This is used in testing core materials to 

 eliminate residual effects and for polar relays and similar structures. 



The electronic control circuits which apply or stop the relay winding 

 currents will be described later. 



Timing Control System 



The timing control system consists of a frequency source, a pulse 

 forming circuit, three binary-quinary ring counters in tandem, a time 

 selection circuit, coincidence circuits, memory circuits, and three control 

 circuits, one for each of the A, BI-B2 and C relay circuits. A block 

 diagram of the system is shown in Fig. 1 1 . 



Oscillator. The frequency source is a bridge stabilized oscillator. This 

 is compared to the Bell System standard frequency for calibration. The 

 accuracy of measurement, both for magnitude and the time scale depends 

 directly upon this oscillator. The magnitude error enters through equa- 

 tion (5) where the cycle time T is exactly the interval for 1000 cycles of 

 the oscillator frequency by virtue of the 1000 discrete steps in the three 

 decades. The time scale is determined by these same steps. Hence the 

 accuracy and stability of the oscillator enters directly; 0.1 per cent can 

 be realized easily. This is as good as necessary as it exceeds the dc instru- 

 ment accuracy. 



Pulse Shaper— The pulse shaper is shown in Fig. 12. It consists of an 

 initial cathode coupled amplifier followed by three more direct coupled 

 stages. The square wave output of the third stage is differentiate by a 

 small condenser for voltage spire production. It is a Schmidt* type circuit 

 designed by I. E. Wood. 



