iK.ixsMissiox cn.ik'.u I i:i<isiics or ii:ici:-iii.i i.ks h\i 



\\ II M >1K\I1\ I-. Ani II AIION (>!■ Till-: MlvTIIOl) 



III the lollowin^ illusiratioii a small iniinher of hand pass wavf- 

 lilUT sections ha\inK ilifferont rharaclcristirs is chosen ixirposely 

 so as to allow ,111 .ippreciahle interaction j'actor and llic u--f of all the 

 charts. 



The niid-part of the composite wa\e-lilter is made up of one mid- 

 si-ries section of lyiK- Vl\ and one mid-half section of .l/-type IV\. 

 the desijinations In-ini; those of a previous paper. The termination 

 .It one end is made Kx\ l>y aiiding (.v — .5)sit in series with ts'pe VI \ 

 anil at the other is Ki\{m), as is diagramniatically represented at the 

 top of Fig. 4. The values of all the parameters were chosen as follows : 



R = GOO ohms. .v = .80, 



/, = 4.(KM)~ . .U-type, /,v = S.OOO - , 



/, = 7.000-, 



and </ = .01 (assumed constant for computaiidii purposes). 



With these values the magnitudes and locations of inductances and 

 capacities are as shown in the center of Fig. 4, where the series imped- 

 ance parts have been merged together. 



The variables Uk and I'* for the "constant k" band pass wave- 

 hller as well as U„. T'„, and wof the .l/-t\pe arc gi\cii b\' formulae 

 i24l. (2tj). and (31). In the 3-element type 17, 



These variables have been computed in the present case for fre- 

 quencies on both sides of the transmitting band and are tabulated 

 below. The other tables including that of transmission losses are based 

 u(X)n this table and the charts. 



The next to the last, and the last columns give the total trans- 

 mission losses as obtained In- this chart method and by direct network 

 computation, respectively. C"omparison shows that there is a \cr\- 

 siitisfactory agreement In'tween them, the dilTerences al all Ire- 

 (luencies iH'ing negligible in practice. The greatest ditTerences of 

 approximately .0") attenuation units at fre(|iiencies 'M'A) and 7500 

 cycles [x-r secon<l, just outside the transmitting band, can reailih' 

 lie explained as due to the omission of dissii)ation in the two terminal 

 loss factors and the retUction coefficients. The lransmi>sion loss is 

 shown graphically at the Inittom of Fig. 4. 



It is believed that the use of this chart method will result in con- 

 siderable time economv with calculations of this kind. 



