﻿78 
  Prof. 
  H. 
  A. 
  Rowland 
  on 
  Electrical 
  

  

  Method 
  15. 
  

  

  ~WJW 
  or 
  ^ 
  2L 
  / 
  L 
  " 
  or 
  ~"c/ 
  = 
  

  

  R%XB^+ 
  gg 
  (R" 
  + 
  R 
  w 
  ) 
  -R^R 
  / 
  R 
  f// 
  R 
  i// 
  

   RR^-R'"^, 
  

  

  ®L 
  or 
  K 
  or 
  ,2 
  L0 
  ;r_ 
  R/X%- 
  R% 
  y 
  (R 
  7 
  + 
  RJ 
  

  

  C, 
  L" 
  ' 
  ~ 
  R% 
  , 
  (R" 
  + 
  R'") 
  - 
  R'R^R'"* 
  

  

  (A) 
  When 
  R^rrrco 
  we 
  have 
  

   L, 
  _R'R 
  „(R"+B'") 
  -K»B,R» 
  _ 
  Bq 
  , 
  K" 
  

  

  7.2T 
  rw 
  _ 
  xx 
  x 
  ^/ 
  •"'-"; 
  / 
  

   ' 
  " 
  R"R"' 
  ' 
  

  

  If 
  we 
  adjust 
  by 
  continuous 
  current, 
  we 
  shall 
  have 
  

  

  R'"R, 
  - 
  R'R 
  /y 
  = 
  0. 
  

  

  For 
  a 
  condenser 
  we 
  can 
  make 
  R" 
  = 
  provided 
  there 
  is 
  no 
  

   electric 
  absorption. 
  In 
  this 
  case 
  b^hfi" 
  is 
  indeterminate 
  and 
  

  

  • 
  we 
  can 
  adjust 
  to 
  find 
  y-^-. 
  However, 
  two 
  simultaneous 
  

  

  adjustments 
  are 
  required. 
  

  

  But 
  I 
  have 
  shown 
  that 
  the 
  presence 
  of 
  electric 
  absorption 
  

   in 
  a 
  condenser 
  causes 
  the 
  same 
  effect 
  as 
  a 
  resistance 
  in 
  its 
  

   circuit, 
  the 
  resistance, 
  however, 
  varying 
  with 
  the 
  period 
  of 
  

   the 
  current. 
  Hence 
  R" 
  must 
  include 
  this 
  resistance. 
  How- 
  

   ever, 
  the 
  value 
  of 
  R" 
  will 
  not 
  affect 
  the 
  first 
  adjustment 
  much 
  

   and 
  so 
  the 
  method 
  is 
  easy 
  to 
  work. 
  If 
  it 
  is 
  sensitive 
  enough 
  

   it 
  will 
  be 
  useful 
  in 
  measuring 
  the 
  electric 
  absorption 
  of 
  con- 
  

   densers 
  in 
  terms 
  of 
  resistance. 
  

  

  It 
  has 
  the 
  advantage 
  of 
  being 
  practically 
  independent 
  of 
  

  

  the 
  current 
  period 
  for 
  ~, 
  as 
  it 
  should 
  be. 
  

  

  For 
  comparison 
  of 
  capacities 
  the 
  same 
  simplification 
  does 
  

   not 
  occur. 
  

  

  Indeed, 
  the 
  method 
  is 
  of 
  very 
  little 
  value 
  in 
  this 
  case, 
  being 
  

   surpassed 
  by 
  16. 
  

  

  Method 
  16. 
  

   (A) 
  [R 
  / 
  R 
  // 
  -R. 
  // 
  R 
  / 
  ][W 
  + 
  r' 
  + 
  /']+W[R/'-r 
  / 
  RJ=0; 
  

   L 
  ' 
  or 
  9l-K 
  + 
  ( 
  W 
  ') 
  

  

  U 
  P./ 
  "" 
  -R 
  T 
  

  

  L, 
  a 
  R„ 
  R„(W 
  +r' 
  + 
  r")" 
  

  

  