men l-RF.QUEXCy AMl'l.niV.RS 



189 



as a cliokc inil <>r auti>-tiaii>l(>rmi'r (willumi aii\- i-\tr.i rniidriisiT). 

 This, howevir, will tl(.|H-iul upon the kind of wire used in making the 

 coil. The loil usvtl in the nu-asiirenients al)o\e was made of No. 28 



Fig. 6 — Effective Resistance of Choke Coil 



solid wire, hut earlier results oljtained by other in\estigalors ha\e 

 shown that solid wire is superior to stranded wire at high frequencies, 



and thus it may be expected that the maximum of the ratio— ^^^ for 



a given inductance will occur at a lower frequency when the coil is 

 made of stranded wire. 



For constant frequency the niaxinuini aniplitication is pr()pf)rtione(l 

 L ' , . , 



to the ratio /^ as already menlidiird. It is tluis (k'sirai>le to adopt 



a construction for the coil, which will increase L without increasing 

 V/?' proportionally. The highest amplification will in general be 

 obtained when L is as large as possible for the frequency in question; 

 in other words, it will be possible to obtain a higher amplification 

 when the tuning condenser in the tuned circuit ampliticr is reduced to 

 zero, giving a simple choke coil amplifier. 



For a tuned circuit amplitier with an ordinary good inductance 

 coil made of stranded wire and of an inductance of, for instance, 

 200 microhenries and a high-frecjuency resistance of about 5 ohms, the 

 amplification at 800 kilocycles will not be higher than about 9 times, 

 according to formula (0) (using the .same kind of tubes as in the ex- 

 periments above), while with a choke coil an amplification as much as 

 18 times was obtained. This means that in order to gel high am- 

 plification, small coils made of fine, solid wire and with large induct- 

 ance and small distributed capacity should be used, rather than large 



