This rainy weekend I decided to replace the old National XR-50 slug
tuned coils in the input pi-network tuned circuits of my 25 year-old
monoband 3-1000Z amplifiers with powdered iron toroids. The primary
advantage of this upgrade is somewhat improved isolation between the
input tuned circuit (the toroids are self-shielding) and the plate Pi-L
tuned circuit.
This proved to be an exceptionally easy upgrade. I thought readers
would be interested in the procedure and the final component values. I
found no reason to use adjustable capacitors, mine are all fixed value
500 volt dipped silver micas.
After removing to bottom cover of each amp, I soldered a temporary 62
ohm carbon composition resistor from the filament to ground, then swept
the input VSWR to the amp with an AEA SWR-121. No filament or DC
voltages were applied to the amp during this test. After noting the
VSWR curve, the old XR-50 slug tuned coil was removed and a new powered
iron coil was installed with a few excess turns more than I expected to
need. I then removed turns until the input VSWR curve was close to
the original curve with the XR-50. At this point the 62 ohm resistor
was removed and the amp was reinstalled for a "smoke" test. In each
case, the VSWR was close to the original VSWR, and one or two trials
spreading or squeezing the turns on the toroid quickly brought the VSWR
to essentially 1:1 across the entire 40, 20, 15 and 10 meter bands.
Only 80 meters required a compromise setting so that the VSWR is below
1.3:1 from 3500-3900 kHz, which is good enough to not require use of the
tuner on my FT-1000MPs.
As many builders of G-G amps have discovered, the efficiency of a G-G
amp is significantly better if the input tuned circuit is located
immediately adjacent to the tube socket. Even six inches of RG-58
between the tuned circuit and the tube socket significantly reduces
efficiency, especially on 10 meters. My monoband amps have essentially
identical efficiency, regardless of band. Any difference in output is
related primarily to tube emission. My best tubes produce 1500 watts
output with 3500 volts of plate voltage and 80 watts drive. My weakest
tubes require 90-95 watts of drive for 1500 watts output.
I placed a .02 uf mica blocking capacitor between the exciter and the
input of the pi-network tuned circuit; rather than on the output
(filament side) of the tuned circuit. This allows the pi-network output
capacitor to be connected from filament to ground directly on the
3-1000Z socket. An additional .02 uF mica capacitor is connected
between the 3-1000Z filament pins to keep them at the same RF voltage.
Final input tuned circuit pi-network values:
Band Input C (exciter side) Powdered Iron toroid
Output C (tube side)
80 1820 17 turns #22 on a T68-2 (red)
1500
40 910 14 turns #22 on a T50-6
(yellow) 750
20 470 10 turns #18 on a T50-10
(black) 360
15 300 8 turns #18 on a T50-10
(black) 240
10 240 6 turns #18 on a T50-10
(black) 180
(I haven't done my 160M 3-1000Z amp yet...)
73
Frank
W3LPL
donovanf@erols.com
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|