the one which receives the external signal is usually in the primary side and the other, which activates the relay, on the secondary side. When there is only sensitive device it usually shunts the transformer secondary (which has a high impedance so as not to reduce the performance of the device), thus creating an efficient feedback loop.
A general feature of all Tesla's receivers is their delicacy. Very careful adjustment was necessary to get the sensitive device at the threshold of avalanching. Most of the sensitive devices were rotated (see June 23rd) so that they were only good conductors during the action of a signal. In some cases, however, this did not achieve satisfactory deactivation of the coherer. Then he used an electromagnetic buzzer to periodically interrupt the excitation of the sensitive device (see Fig. 2 of August 8th). Probably the circuit in Fig. 2 gave him the idea for that in Fig. 3, where the rotary interrupter is replaced by a buzzer as an electromagnetic interrupter. He then used a buzzer in various other configurations (Figs. 5 and 6 of August 8th), with the aim of reliably biasing the condenser, and hence also the sensitive device, to threshold.
Tesla did not measure the sensitivity of his receivers by any definite method, but there is no doubt that he did compare them in some way. From his notes very little can be deduced about their sensitivity, i.e. the power required to activate them. A rough idea is given by data from July 4th, when he used similar receivers to register electrical discharges. He estimated that he registered waves produced by lightning at least 200 miles away, and continued to receive signals (at periodic intervals) later when the weather had already cleared. He records that with the receiver shown in the figure of July 28th he was in one instance able to register lightning discharges at a distance of 500 miles. He estimated the distance from the periodicity of the signals as the storm moved away.
