< Previous - Table of Contents - Next >

SPECIAL NOISE CONTOURS

  The  curves   of   the  6AM4  tube  include   a   contour  of   minimum   noise
figure.   This   has   been   obtained  from  the   manufacturer's   data   which
indicate  the  correct  bias  for  use  with  the  tube ( with   a   low-impedance
grid  circuit )  in  a   grounded-grid   connection. This  contour,  which   indi-
cates  the  bias  required  for  obtaining  the  highest   signal-to-noise  ratio,
is found to lie near the negative edge of  the  contact  potential  area  ( where
the grid  will  bias  itself  with  an  infinite  grid  impedance ) for  the  tube.
Operation  of  the  tube   on   the   minimum   noise   contour   should   be  at-
tempted only in a grounded-grid connection.

LOGARITHMIC DATA

  The triode characteristic curves as normally presented do not give enough
data in  the  low  current  range  for  the  design  of  such  circuits  as  multi-
vibrators,  flip-flops,  and  relaxation  oscillators.   Therefore   characteristic
curves of  tubes  for  switching  applications  are  plotted  on  the  basis of a
linear  plate  voltage  scale  and  a  logarithmic   plate   current   scale.  The
small-signal data  are  also  plotted  on a logarithmic scale.  As an example,
the  curves  for  the  5965  tube are plotted on  the  ordinary  basis  and  the
logarithmic basis in this set.

PENTODE DATA

  The  G-Curves  provided on pentode tubes are curves of  constant bias
as  a  function  of  screen-grid  voltage  E_c2, and  nominal plate current I_p.
The  small-signal  data included are  in the  form  of contours of constant
value of nominal transconductance  G_m1.  Data on plate conductance are
not included since  they  rarely require  consideration with pentode tubes.
One  of  the few examples where the plate conductance data  are  useful
is in connection with series-pass tubes in regulated power supplies.

  An  examination  of  the  relations  of  the  various  voltages   and   currents
for  pentode  tubes  shows  that  the  voltages  having  the  greatest  effect on
plate  current  are  grid-to-cathode  voltage  e_c1,  and   screen-to-cathode vol-
tage E_c2  with  plate  voltage  e_b,  having  a  rather  small  effect. Consequently
the  plotting  of   grid  voltage  as  a  function  of  plate  voltage  and   plate
current,  as  is  done  on  standard  data  sheets  on  pentodes,  does  not give
the  most  significant  data  on  the  pentode.   However,   plotting   the   char-
acteristics of pentodes on  the  basis  of  their  screen  voltage  shows  directly
the importance of the  rate  of  change  of  plate  current  with  screen  voltage.
It also shows the importance of the dependence.

  The  screen  characteristic  curves  are  plotted  for  a  ratio  of   plate-to-
screen voltage (e_b/E_c2)  equal  to  two.  Values  of  i_b,  and  g_m1  for  ratios  of
plate-to-screen voltage  other  than  two  may  be  obtained  by  the   X_p   cor-
rection  curve  included  in  the  upper  left-hand  comer  of  each  data   sheet*.
 

*note - The correction curves are obtained by plotting the curves X_p = i_b/I_p and X_c2 = i_c2/I_p as a function of e_b/e_c2, where I_p is the value of i_b, where e_b/e_c2 = 2. A series of these correction  curves  may  be  prepared and  averaged, with possibly  a little extra weight being given to the contours obtained for bias voltages near zero.  The  resulting relations show the variations of X_p above e_b/e_c2 = 0.5 to within 3 - 5 %  of  the  true value, or much closer than can be expected from the tube  itself.  The  values  of  X_c2 take the same form from tube to tube, but may  differ  in  overall  magnitude  by  from  5 - 25 % with average tubes.

2

< Previous - Table of Contents - Next >

Copyright 2008 for Phyllis K. Pullen, M.D.,
by Robert J. Legg