This is a 5 tube regenerative receiver for 14MHz band. Building it
filled many nights and I would like to share this nice experience.
As you can see the RF amplifier use a
grounded grid triode. Even if this configuration doesn’t produce
maximum gain it is stable and isolates the detector from antenna. I
tried many tubes. For example 12SF5 produced less than unity gain, 6C5
yield a gain of 2 and 6S9D around 7-8. So finally I have a good reason
to use a “lighthouse” tube. The resonant circuit from anode
is tuned to the middle of the band. At both band edges the gain
decrease to about 70%. I considered that it doesn’t worth to have a
variable capacitor here. The coil uses a T50-6
core. In the cathode there is another coil (L1) that
increases the input impedance of the amplifier. Removing it will cause
a slightly decrease of gain. By the way, the gain is controlled by C1.
The antenna is a mono-band fish-pole. For 20m band, of course.
The detector stage uses a 6J4 tube (6AC7). The L3 coil has a T68-7
core. Material 7 has a better thermal coefficient than material 6 used
in previous stage. The wire has a 1.2mm diameter. Because is so thick
it add mechanical stability and increase Q-factor. It is also
un-isolated. So is easy to move the taping point.
Resistors R6 and R8 should be adjusted only once, to set the limits of
regeneration control. The idea is to limit the voltage variation
obtain with R7 to the minimum required value to have a smooth control.
The detector is supplied from a 150V gas discharge regulator
tube. The C5 and C6 are series/parallel combinations of
silver-mica capacitors in order to obtained desired value. The easiest
way to set the band limit covered by detector is to use a signal
generator. The coil tap is at one wound from the cold end.
The audio preamplifier is built with a 6J8(6SJ7) tube. This is a
build low noise audio amplifier tube. Its main feature is good
between heater and cathode. For example 6J4(6AC7) is not well suited
for audio applications with heater supplied with AC voltage.
The C17 capacitor limits the high frequency gain of the stage. And
there are plenty of sharp noises on a regenerative receiver. Audio
amplifier uses a 6P9 (6AG7) tube. This high gain pentode need a 10k
load according to specs. I found this to be very tolerant actually.
First I use a 60VA toroid transformer. For primary I used the 230V
winding and for secondary winding I just made 4-5 wounds. It sound very
well, the same like when I replace it with a special output transformer.
Bellow you can see a movie with the
working receiver. As you can see it is pretty sensitive, however
selectivity is definitely not its strong point. But consider
that whole selectivity is obtained with a single LC circuit. Tuning
require skill. Actually the movie was made before I figure out that the
regeneration level has to be increased to make the voice more clear.
The regeneration control change also a little the frequency so you
might be tempted to adjust the pitch in this way, but this is wrong.
Also the antenna capacitor should be set to minimum value required to
listen to desired signal in order to minimize strong signals
interference. Overall I was very pleased with the performance
of such a simple radio. Of course, many improvements can be made but
the added complexity will make the radio comparable with a
super-heterodyne design witch is in another class of