VK2ACY - G5RV coupler

related wiki pages: Antennas, Propagation

My G5RV  Dedicated  Coupling  Unit  for  80 / 40 / 20M

To my  knowledge  (despite  the  reams  and  terrabytes  already  expended  on discussion / argument / tutorial)  re. the G5RV,  nobody  has  ever  designed  a  dedicated  coupling  unit  just for  this  often  misunderstood  antenna. After exhaustively  confirming  absolutely NO difference between my G5RV and (full-sized half wave centre and end fed) dipoles on 80 & 40M at my QTH, the decision to retain my G5RV as primary antenna for these bands prompted my designing such a  unit  with  fixed  settings, providing  replicable  swr  readings  ( without even  the  need  for  me  to  check  each  time  used ). My prototype  used  all  secondhand  parts,  namely:

One HF toroidal balun core,   6 metres of  0.63  or  0.71mm enamelled  winding  wire,   8 inches of  3/4inch  diameter  wooden  dowel  rod,  12 inches  of  hookup  wire,  6 inches  of  thin  tinned  wire,  three  small  alligator  clips – white,  black  and  green,   2 only 680pf  250V ceramic / styroseal  or  similar  capacitors, one  SO-239 socket,  two  insulated  banana  terminal / binding  posts,  4 small  brad  tacking  nails,  2 small self  tapping  screws,  and  an  old  MacKintosh’s  Chocolates & Toffees  candy  tin  with  a  handy  swing  lid. ( 8 x 5 x 2 _ inches  W x D x H )

The attached  pictures  show  what  I  did  and  this  is  the  order  in  which  I  did  it:

a.Mounted the  SO-239  socket  in  the  middle  of  the  rear  panel  of  the  “toffee  tin”  and  then  mounted  each  of  the  insulated  banana  terminal / binding  posts  either  side  of  it.

b.Cut two  2 ft  pieces  of  the  enameled  winding  wire,   and  wound  10 turns  of  each  parallel  to  each  other  (ie: side  by  side)  tightly  around  the  balun  core,  ensuring  each  of  the  pairs  of  turns  tightly  follows  the  outside / inside  surface  of  the  “doughnut”. I spaced  the  windings  about  2.5mm  apart,  ensuring  that  the  10 (pairs of)  turns  covered  about  three  quarters  of  the  doughnut’s  entire  circumference.

c.Left about 2 inches  of  winding  wire  at  each  end  of  the  windings  before  cutting  off  the  excess. Identified the  “start”  and  “end”  of  one  winding  (primary  winding),  trimmed  these  close  to  the  required  length,  and  (using  a  sharp  knife)  scraped  enough  enamel  off  the  wire  to  permit  tinning with  solder  and  soldering  directly  to  the  SO-239  connections  (the  “start”  of  the  primary  winding  soldered  to  the  centre  pin,  and  the “end”  of  the  same  winding  soldered  to  the  chassis / shield  lug.)

d)Similarly scraped  and  solder-tinned  the  “start”  of  the  other  winding (secondary  winding)  leaving  it  ready  for  soldering  later  as  required.  Also  scraped  and  solder-tinned  the  “end”  of  the  secondary  winding  leaving  only  enough  wire  so  that  folded  once  over  on  itself  it  will  form  a  0.5cm  “tab”  which  a  crocodile  clip  could  manage  to  grip  firmly  onto.  This  “end  of  winding”  tab  will  form  one  balun  connection  which  is  selected  by  moving  the  appropriate  “crocodile  clip”  when  using  the  antenna  on  the  40 & 20M  bands.  I  labeled  this  connection  to  the  balun:  “Full  Winding  Tap  for  40 & 20M.”

e)Counting four  whole  turns  back  from  this  0.5cm  “end  of  winding”  tab,  a  small  area  of  enamel  is  scraped  off  the  top  of  the  same  wire  which  forms  the  secondary  winding  to  enable  solder-tinning  and  a  small  1 cm  tab  of  wire  to  be  also  soldered  on  at  that  point.  This  1cm  tab  will  similarly  form  another  balun  connection  which  is  selected  by  moving  the  appropriate  “crocodile  clip”  when  using  the  antenna  on  the  80M  band.   I   labeled   this   connection   to   the   balun:   “Half   Winding  Tap  for  80M.”

f)Turning my  attention  to  winding  the  required  7 uH  tapped  inductors,  I  hammered  two  nails  into  the  dowel  rod,  each  half  an  inch  either  side  of  the  exact  middle  of  it’s  8inch  length, leaving  only  about 3mm  or  so  of  the  top  of  each  nail  exposed.  I  then  hammered  one  of  the  remaining  nails  exactly  two and  a  half  inches  further  out  towards  the  left-hand end  of  the  dowel  rod,  from  the  left-hand  nail.  Then  I  similarly  hammered  the  last  remaining  nail  exactly  two  and  a  half  inches  further  out  towards  the  right-hand  end  of  the  dowel  rod,  from  the  right-hand  nail.    I  found  that  these  nails  formed  convenient  anchoring  posts  for  the  enameled  wire  about  to  be  wound  onto  the  dowel  rod.

g)I then  cut  a  2metre  length  of  enamelled  wire  and  scraped  an  inch  or  so  of  enamel  off  one  end  to  permit  anchoring  by  wrapping  twice  around  the  left-hand  nail  closest  to  centre  of  the  dowel  rod.  After  tinning / soldering  it  in  place,   I   then  proceeded  to  tightly  wind  the  wire  around  the  dowel  at  a  pitch  of  15 turns  per  inch,  for  a  total  of  37 turns,  arriving  just  near  the  outer  left-hand  anchoring  nail.    I  temporarily  held  the  end  of  the  winding  in  place  with  tape  while  I  cut  the  wire  (leaving one  inch  for  scraping / tinning,  anchor – wrapping  and  soldering  around  the  nail  itself.)

h)Once this  was  done  I  proceeded  to  repeat  the  exercise  for  the  inductor  to  be  wound    on  the  right-hand  side  of  the  dowel  rod,  winding  carefully  and  making  sure  that  the  direction  of  the  winding  was  the  same  (ie:  not  counter-clockwise  with  respect  to)  the  first  winding  I  had  made.  These  two  identical  inductors  would  eventually  be  “crocodile  clipped”  in  series  with  each  leg  of  my  G5RV ‘s  feedline,  depending  on  the  frequency  band  in  use.

i)Counting from  the  very  centre  of  the  dowel  rod  (where  the  “starts”  of  each  inductor’s  windings  are  located)   I   carefully  marked  with  texta  at  the  8, 17, 18, 19 & 20 turn  points                   on  each  inductor,  and  proceeded  to  scrape  a  small  area  of  enamel  off  the  wire  at  each  point  to  permit  solder-tinning  and  attaching  a  small  connection  “tab”  (using  very  short  pieces  of  the  thin  tinned  wire  doubled  over)  to  which  a  crocodile  clip  could  grip  firmly  onto.

j)At this  point  I  breathed  a  sigh  of  relief  because  I  knew  that  the  unit  was  JUST  ABOUT  FINISHED !!!!!  and  proceeded  to  mount  the  dowel  rod  on  which  I  wound  the  inductors  by  screwing  the  two  small  self-tapping  screws  through  the  sides  of  the  tin  case  and  firmly  into  the  ends  of  the  dowel  rod  to  prevent  it  spinning  or  moving.  The  rod  was  aligned  so  that  the  nail  heads  and  inductor  taps  were  all  facing  the  rear  of  the  unit.

k)I then  cut  a  short  2 inch  piece  of  insulated  hookup  wire  and  soldered  one  end  to  the  inside  of  the  insulated  banana  terminal / binding  post  on  the  left  hand  side  of  the  rear  panel,  and  the  other  end  to  the  black  crocodile  clip.  This  could  then  be  clipped  to  the  appropriate  winding  “tap”  on  the  inductor  closest  to  it,  for  the  band  in  use.

l)Similarly, another  short  2 inch  piece  of  insulated  hookup  wire  was  soldered  to  the  insulated  banana  terminal / binding  post  on  the  right-hand  side  of  the  rear  panel,  and  the green  crocodile  clip  fitted  to  the  far  end,  eventually  allowing  connection  to  the  appropriate  inductor  taps  closest  to  it.

m)ALMOST FINALLY – taking  one  of  the  680pf  capacitors,  one  lead  of  which  was  soldered  to  the  left-hand  inductor’s  “start”  of  winding  connection  (at  the  central  anchor  nail),   the  other  lead  of  the  capacitor  was  then  soldered  to  the  balun’s  secondary  winding “start”  connection  (as  mentioned  above  in  item  “d.” )

n)The other  680pf  capacitor  was  similarly  soldered  to  the  right-hand  inductor’s  “start”  of  winding  connection (at  the  central  anchor  nail).  A  short  one  inch  length  of  insulated  hookup  wire  with  the  white  crocodile  clip  soldered  to  the  far  end  was  then  soldered  to  the  capacitor’s  other  lead.  This  could  then  be  clipped  to  the  appropriate  “Full  Winding”  or “Half  Winding”  balun  connection  required  for  the  frequency  in  use.

FINALLY – after extensive  testing  including  SWR  measurements  and  on-air  comparisons,  I  was  able  to  sit  down  and  make  the  following  chart  to  remind  me  of  the  “crocodile  clip” settings for  each  band  of  interest:

80M Band:   White  crocodile  clip  connected  to  “Half  Winding”  Tap  on  balun  transformer.

Black and  green  crocodile  clips  connected  to  respective  inductor  taps  as  per the following  centre  frequencies: 3.550 Mhz use  both  taps  at 20 turns 3.580 Mhz   “      “        “       “   19 turns 3.610 Mhz   “      “        “       “   18 turns 3.640 Mhz   “      “        “       “   17 turns

(Higher centre  frequencies  available  by  tapping  appropriately  at  the  ascending  rate  of  30kHz  per  turn.  Worst  case  measured  SWR  was  1.5:1  in  between  adjacent  turn/taps.)

-

40M Band:   White  crocodile  clip  connected  to  “Full  Winding”  Tap  on  balun  transformer.

Black and  green  crocodile  clips  connected  to  respective  inductor  taps  at  the 8 turns position (covers  whole  of  40M  band)

20M Band:   White  crocodile  clip  connected  to  “Full  Winding”  Tap  on  balun  transformer.

Black and green  crocodile  clips  NOT  connected  to  inductor  taps,  but  rather: on 20M only  these  are  each attached  directly  to the  capacitor  closest  to them by clipping onto  the capacitor’s  lead  which  is NOT soldered  to  the  inductor  rod.

So how  exactly  does  my  G5RV  Dedicated  Coupler  actually  work ?

Very simply  in  fact – so  much  so  that  I  didn’t  feel  the  need  to  draw  a  circuit  diagram  for  the  completed  unit  ( I  could  have  built  another  one  in  the  same  amount  of  time !!! )

The connection  from  the  SO-239  socket  is  fed  directly  to  the  primary  winding  of  the  balun  transformer. The balun  performs  the  necessary  “balancing”  in  order  to  run  the  G5RV  as  a  true  balanced  doublet,  and  also  performs  the  necessary  step-down  impedance  transformation  when  the  “Half  Winding” Tap  is  selected  during  use  on  80M. From there,  the  two  balanced  balun  connections  each  feed  through  a  coupling  capacitor  and  then  an  inductor  winding  on  their  way  to  the  antenna  feedline  connections.

As the  Q  factor  of  the  inductors  is  a  medium-range  figure  on  80M,  the  useable  bandwidth  is  limited  to  about  60kHz,  so  a  number  of  taps  are  provided  (at  20, 19, 18 & 17 turns)  to  reduce  the  inductance  as  higher  frequencies  in  the  band  are  used. The use  of  larger  diameter  inductors  (say  4 inch  dia.)  would  have  negated  this  requirement,  but  made  for  a  somewhat  cumbersome – sized  unit  to  say  the  least – not  to  mention  greatly  increased  cost.

On 40M  the  G5RV  is  a  very  broadbanded  antenna,  so  the  one  set  of  inductor  taps (at 8 turns)  more  than  satisfactorily  provides  a  suitable  match  without  the  need  to  change  as  higher  frequencies  are  used  further  up  the  band.

On 20M  the  capacitors  and  inductors  are  by-passed  and  the  balanced   feedline  connections  simply  “clipped”  onto  the  output  connections  from  the  balun  operating  in  1:1  mode. Measured swr  right  across  the  whole  20M  band  was  under  1.5 to 1  when  operating  the  G5RV  in  this   true  balanced  mode  (as  the  reactive  coax  inductance  does  not  come  into  play  as  it  did  prior  to  my  using  this  unit)

So there  you  have  it. An afternoon's work and  absolutely  NO  financial outlay, resulted in an SWR of better than 1.4:1 across 80/40/20 and the knowledge that my G5RV is now operated as a true  balanced  doublet,  with  consistently  up  to  1 S point  improvement  particularly  on  80M.



73 from  Wayne P.     VK2ACY      email:  pickmey@yahoo.com

P.S:  For  the  record,  my  G5RV  was  installed  as  a  flat-top  at  a  height  of  25ft  and  constructed  with  51ft  of  1.25mm  enamelled  wire  per  leg,  fed  with  28ft  of  solid  300 ohm  TV  type  ribbon  feeder  directly  to  the  coupler  situated  on  the  ground,  and  7.5metres  of  RG-58  coax. from there  to  the  station’s  operating  position. Estimated maximum  power  rating  of  the  coupler  built  by  me  is  thought  to  be  around  200watts PEP.

Note:  Feel  free  to  use  or  pass  on  this  information  as  desired  (except  for  commercial  purposes  or  profit)  however  NO  responsibility  accepted  for  any  consequences  arising. Please email  me  should  you  make  use  of  this  info  as  the  results  of  my  other homebrew – friendly endeavours  will  be  shared  depending  upon  received  feedback.