I have a vintage radio that I managed to get all circuits and tubes back into original spec. The AM stations tune in fine but I get nothing on FM. I am told this is because it's tuning in "Old FM" - which I can only assume was part of the APEX broadcast band (42mhz - 50mhz)? Or, is this a reference to some band of UHF/VHF? It's not clear to me.

Depending on the answer to the above, might it be possible to put a line-level adapter on the antenna so I can use all the vintage hardware to tune in a few modern stations?

update Model is 1947 Magnavox Georgian. Schematics to come

  • 5
    \$\begingroup\$ A tuning dial often gives a clue to the frequency band by the numbers and scale. If its in the 88 - 108 range, then it is modern and should be compatible with current FM broadcast (although it will be mono, not stereo). \$\endgroup\$
    – glen_geek
    Dec 28 '17 at 2:18
  • 2
    \$\begingroup\$ Modern US FM is in the 88-108MHz range. If your old radio is tuned to the pre-1945 42-50MHz band, you'd need to change some internal components. Simply replacing certain resistors and caps might do the trick -- it wouldn't be optimally engineered for the modern band, but it probably wouldn't need to be. You'd need to find or recreate a schematic, isolate the FM demodulator, figure out which parts controlled the frequency range, and tackle those. \$\endgroup\$ Dec 28 '17 at 2:19
  • 2
    \$\begingroup\$ How exactly have you determined that you have managed to get all circuits "back into original spec" when you don't even know what frequencies it is supposed to tune??? That's substantially more basic specification, than what nominal/bias voltages and signal frequencies should be present at various circuit nodes. In the realm of the practical, I'd hazard that quoting some tube part numbers could yield the vintage and from that the intended capabilities. \$\endgroup\$ Dec 28 '17 at 4:44
  • 1
    \$\begingroup\$ What’s the make/model? \$\endgroup\$
    – Chu
    Dec 28 '17 at 8:33
  • 1
    \$\begingroup\$ Actually, it's very much in the transition period, the proposals to move to 88-108MHz were only announced in 1945. The 42 to 50MHz band was still being used for FM up to at least 1949 albeit only by a few stubborn stations. Lots more info at FM Broadcasting Chronology \$\endgroup\$
    – Finbarr
    Dec 29 '17 at 9:33

Interesting question, I hadn't heard of that band being used for FM radio before (in the UK I think it was used for TV from about 1937).

You might want to build a "downconverter" that would do the job - mix a slice of the modern VHF band with a fixed frequency local oscillator, and filter the mixer output via a 42-50MHz bandpass filter, and feed into your antenna input.

As the FM band is 20MHz wide, and the old one is 8 MHz wide, you would need to select one of three oscillator frequencies - there are crystal oscillator modules readily available with a wide range of frequencies ... let's see if we can find something that works.

Starting at 88MHz, you would need a 46 MHz local oscillator ... maybe we'll find 45MHz, giving 87-95 MHz from the first local oscillator. Then 53 MHz would give 95-103 MHz, and 61 MHz would give 103-111 MHz. You're likely to find 60MHz giving 102-110MHz, and full coverage with a middle oscillator between 52 and 53 MHz. (It wouldn't be a disaster to use 4 oscillators, at 45, 50, 55 and 60 MHz if you had to).


simulate this circuit – Schematic created using CircuitLab

This is just a block diagram, you'll have to work out the details yourself.

Front end bandpass filter : can be borrowed from the input of any FM receiver circuit (and may include an RF amplifier).

Oscillators : should be off the shelf. You may get away with leaving them all connected (via an attenuator network) and only powering the one you want. Farnell for example show 44,52 and 60MHz as available giving full coverage from 86-110MHz.

Mixer: you can either buy a "broadband double balanced mixer" covering at least 40-110MHz, or make using schottky diodes.

Output filter : can be wider than shown (e.g. 30-60MHz would be OK) - or even a 50MHz lowpass filter, and rely on the filtering in your existing radio to do the job. You may be able to omit this entirely.

  • 1
    \$\begingroup\$ A flaw: most mixers will leak the local oscillator into the down-converter's output - it will appear as a very strong nearby signal. High-side injection (45 MHz above the FM band) would be safer. \$\endgroup\$
    – glen_geek
    Dec 28 '17 at 15:45
  • \$\begingroup\$ This is true and creates a problem listening to 88MHz with a 44MHz LO for example; the other suggested oscillators are above 50MHz and less problematic . But high side oscillators may be harder to find. (A double balanced mixer has much lower leakage, but I shouldn't have glossed over the reason for choosing it). \$\endgroup\$ Dec 28 '17 at 15:50
  • \$\begingroup\$ I am giving you the thumbs up. Based on what I discovered which I will detail \$\endgroup\$ Dec 30 '17 at 0:43
  • \$\begingroup\$ After consulting with some very old repair people in the industry I have managed this: There is a mixer on board and an "FM" switch. But apparently there was an auxiliary unit that plugs into the mixer and the "FM" is to activate that unit. That was not at all obvious. I have no idea what this unit looks like or what it's part number is but I am pretty sure I am better off cooking one up on my own along the lines of what you have detailed. \$\endgroup\$ Dec 30 '17 at 0:49

1) Sorry to say so but I do not think you will have much luck. The tuning capacitor is custom made for the frequency band it covers and it is a physical attribute-not just an electrical one. A few other capacitors would need to be changed as well. Even if this was a super-heterodyne circuit where it converted the incoming signal to 10.7 MHZ (to pick out one station out of many), your main issue is the tuning capacitor and connected inductor's.

2) If it is a super-heterodyne circuit the main tuning capacitor also adjust this frequency (local oscillator)as well, so that the difference between the two frequencies is always 10.7 MHZ, and you will have to adjust both tuning capacitor sections to the new frequencies, basically doubling them. If it is NOT a super-heterodyne circuit then it simplifies the tuning issue but of course those ancient radios could not separate channels very well.

3) I do not think you will get far without at least a basic oscilloscope just to see where you have RF and do not, and a basic DVM to verify cathode and anode voltages, and check for open resistors (a common fault with old solid carbon resistors). This is not a place where you poke your finger in to test the circuit.

4) Your best bet might be to shop at pawn shops for vintage radios (88-108 MHZ) and heist the parts you need. They usually had a pointer on a string driven by the tuning wheel, so you need to account for that as well. You may find a spring loaded non-slip gear that actually turns the tuning capacitor. Try to use it if possible, as it made the act of tuning a smooth process.

5) I do not think you will need to change anything in the audio section, other than looking for burnt parts and that all the tubes work. I am familiar with Grundig and Telefunken radios but yours sounds more basic then what they made.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.