622 



Popular Science Monthly 



signals. If a single-pole switch is placed 

 in the lead to Ci of Fig. 2 (as shown in 

 Fig. I of p. 306 in the February issue) it 

 becomes possible to 

 use either the broad 

 or the sharp-tuned 

 secondary system, 

 as may be desired. 

 For any given wave- 

 length more induc- 

 tance on the sec- 

 ondary will be re- 

 quired to get loud 

 signals with the ar- 

 rangement of Fig. 



5, than for Fig. 2; "^ 

 this is because the 



secondary circuit of Fig. 5 actually is 

 broadly tuned by the capacity of the 

 detector, blocking-condenser, telephones, 

 etc., acting with the total inductance of 

 the secondary. Since the natural capac- 

 ity of these other elements is small, a 

 larger secondary inductance is made 

 necessary to reach the desired wave- 

 length. 



Where still less closeness of tuning is 

 necessary, the arrangement of Fig. 4 

 may be modified by omitting the loading- 

 coil Li which permits variation of 

 coupling, and by doing away with the 

 tuning-condenser Ci, as shown in Fig. 6. 

 This results in the ordinary close- 

 coupled direct tuner, which is useful for 

 picking up signals when interference is 

 not severe. By connecting in the tuning- 

 condenser Ci, as shown by the dotted 

 lines, it is possible to improve the 

 selectivity of the system in some meas- 

 ure, especially if the blocking-condenser 

 C2 is made of very small capacity or 

 even left out altogether. 



It will, of course, be seen at once that 

 in tuning the secondary circuit of any 

 of the receivers described above, one 

 may choose a great many combinations of 

 inductance and capacity in order to have 

 resonance to a certain frequency. For 

 instance, the wavelength of 5000 meters 

 is reached when the secondary inductance 

 is 4 millihenrys and the condenser 

 0.00173 mfds. If the inductance were 

 2 millihenrys, twice the former capacity, 

 or 0.00346 mfd. capacity would be re- 

 quired. The best ratio of inductance to 

 capacity depends largely upon the type 

 of detector used; for most crystals, the 



Fig. 5. By opening the 



secondary tuning - condenser 



the primary may be tuned 



independently 



condenser may be about 0.003 mfd., 



maximum for wavelengths from loco 



to 5000 meters, and correspondingly 



smaller or larger 



for shorter or longer 



waves. For the au- 



dion, where the 



highest possible 



voltage should be 



applied to the grid, 



it is best to use 



comparatively large 



values of secondary 



inductance, with 



the corresponding 



small secondary 



condensers; Ci, for 



had best never be 



mfd., even for the 



1 



c' 



audion working, 

 larger than o.ooi 

 longest waves. 



The size of the stopping-condenser C2 

 is also a matter of interest. For crystal 

 detectors, it is customary to use capaci- 

 ties of from o.oi to 0.04 mfd. at this 

 point in the circuit. By making the 

 stopping-condenser variable in steps of 

 about 0.005 mfd., it is possible to select 

 a best value for each particular operating 

 condition; in general, the higher the 

 telephone resistance and the higher the 

 incoming spark-frequency, the smaller 

 the stopping-condenser may be. The 

 smaller this condenser is made, after it 

 passes below about o.oi mfd., the less 

 is the damping of the secondary circuit, 

 and the sharper is the tuning. Too 

 great reduction of the capacity, however, 

 in the attempt to gain selectivity, re- 

 sults in weakening the response to the 

 signals. The size of the blocking or 

 grid-circuit condenser for the audion is 

 much less than for the crystal detectors; 

 Co is then best made variable, with a 

 range including values as small as 0.000 1 

 mfd. or less. 



In operating any of the sharply tuned 

 circuits shown in the foregoing, it must 

 be remembered that the best settings of 

 primary inductance, coupling, and sec- 

 ondary inductance and capacity are 

 largely dependent upon each other. In 

 tuning-out interference and "bringing 

 in" a particular station, the best plan 

 is first to open the secondary tuning- 

 condenser circuit to give the arrange- 

 ment of Fig, 5 ; this makes it possible to 

 tune the primary independently and ac- 



