I have an RCA AM broadcast Battery/mains radio that I
still use.
Works well. The multivoltage battery though bad supplied the
authentic looking cover for a box I'd made using uses NiCds
to run a switchmode supply to provide the A,B and C voltages
required. Runs for hours on that too.
I'm suprised it needs a separate C supply. UK sets all used self-bias by
this time, I think. C (grid bias) batteries were not used for radios much
after the 1930 in the UK.
Myself as well. The design of this radio is early 50s. I have a similar
circuits from the 1960 RCA tube manual that require only A and B(HT) voltages.
However this radio had a battery based on destructive examination and markings
that provided 7.5V, 1.5V and 90V. So the likely case is "C" voltage or as
Battery radios in the UK typically used 4 valves :
DK92 or DK96 pentagrid frequency changer
DF91 or DF96 IF amplfier
DAF91 or DAF96 detector diode/1st audio (pentode)
DL91 or DL96 audio output.
Each had a 1.5V filamanet, excpet for the output pentode which had a 3V
filament with a centre tap. The first valves listed had 50mA filaments,
the second had 25mA ones.
Such sets generally either put all 5 filamenet sections in parallel,
running them off a 1.5V battery (often 2 or 4 cells in parallel) or put
them in series, running them off a 7.5V battery.
I think all of them by this point used self-bias.
yet unexamined posibility, a seperate heater supply
for the audio output tube.
Possible, although a 7.5V filament is not common. And having 5
signal-stage valves with their filaments in series (giving 7.5V) would
not be common either.
It would be interesting to know what's going on in this set. Can you get
a schematic? Being an American set, it's not in Poole and Molloy :-(
I must try the
circuit that was in Elektor a couple of months back to get
90V from low-voltage batteries.
Unfamiliar, Elektor is not seen here in USA.
It's one of the better electronics magazines available over here. It has
some interesting projects, and they've now started making microcontroller
source code available for most, but not all, of them.
The circuit in question was a pretty standard inverter circuit, running
at about 50Hz (not a typo, they wanted to keep RF noise down), feeding a
normal mains transformer used in reverse. Looks like it should work with
no problems.
-tony