Amps
[Top] [All Lists]

[Amps] HIGH INPUT SWR

To: <amps@contesting.com>
Subject: [Amps] HIGH INPUT SWR
From: "Jim Thomson" <jim.thom@telus.net>
Date: Fri, 20 Dec 2019 10:35:25 -0800
List-post: <mailto:amps@contesting.com>
Date: Thu, 19 Dec 2019 12:30:34 +0000
From: Bill <magoo@isp.ca>
To: amps@contesting.com
Subject: [Amps] HIGH INPUT SWR

<I had a query about my finding regarding the higher than expected input 
<impedance of the pair of GI7B tubes in an amp I have been reworking.? I 
<did some reading and found a post from Frank, W3LPL several years ago 
<regarding the same thing. He discovered that if the input tuning 
<networks are not located very close to the base of the tube the added 
<capacitance of the connecting coax is enough to radically alter the 
<input SWR, resulting in loss of efficiency unless the networks are re-tuned.

<That is the case in this GI7B amplifier-the tuned input is fed through 
<approximately 10" of coax to the tubes, and it appears that this is the 
<source of all the problems with this particular unit.? The elevated 
<input impedance of 150 ohms rather than the expected 100 ohms for two 
<tubes appears to be explained by a flaw in amplifier design.


<73, Bill VE3NH

##  IF..........  EACH  tube  has  150 ohms   input Z,   use  75  ohm  coax  
from  tuned  inputs  to  cathode.

##  what  else  screws  things  up  is  the  stray  C  from  cathode  to  
grounded   grid.   Its a fixed  value..and
easily  measured  with a digital LCR  meter.   That  stray  C  is  directly  in 
 parallel  with  the C2  cap  of  each
of  your  tuned  inputs....via  your  10  inches  of  coax.   Its..free  C2.   
Eimac  + Svetlana  will  list  the  cathode to  grid  C  on  its data  sheets. 

##  To measure  input Z, I  guestimate the input Z....and  design  a   TEMP  
tuned input  PI net  for  40M.   Both C1  +  C2  are  adjustable. 
C1 +  C2  tweaked  for   1:1 input  swr.....  with  amp  driven  full  bore.  
Amp  then  shut off.     MFJ-259B connected to  input side of  40M  tuned input 
...and a 0-250  ohm pot  on 
output  side  of  40M  tuned  input.   Pot  tweaked for  1:1  swr.     Then  
measure pot  with a fluke  87A  ohm meter...  =  input Z. 

##  Once   input Z  is  determined, then  use  GM3SEK   PI   spread sheet  to  
design pi  nets..for each  band.  

##  On  my  HB    GG  Triode amps  I  did  not  want   to  mess   with  9-10  
band switched  pi  nets.   I  used   2 x  broadcast variables   + a tapped 4uh  
coil...  made  from 6  gauge  solid CU  wire.  
Both caps  padded  on  160 M .    6:1  Jackson  Bros  ball  drives on  each  
cap.   1:1  input  swr   from   1.8  to  30  mhz continuously.  

Jim  VE7RF
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>