Homemade Regenerative Tube Radio

There are no microcontrollers in this project. In fact you wont find a single transistor. This classic regenerative tube radio, modeled after an early 20th century homebrew is complete with schematic and additional photos. For those who are not familiar with tube designs and for simplicity, the regeneration circuit can be thought of as feedback though this relation may be argued. Read the rest after the break which includes a crash course in tube operation.

A basic vacuum tube generally consists of a heater, grid, and plate. A current is passed through the heater which as the name suggest, creates heat. This allows electrons to “boil off” under proper conditions. That is the hot heater emits electrons when the grid and plate are positive potential. Opposite charges attract so the electron is attracted towards the grid and plate. The grid has holes very much like a screen door does. Some of the electrons pass through these holes missing the grid and strike the plate. The greater the electrical potential the greater the attractive force in the grid and plate. When an electron is permitted to move from one potential to another an electrical circuit is created. This circuit forms the basics behind vacuum tube operation.

With that said, the project uses a heater powered from two D size batteries. There doesn’t seem to mention of the plate source, though we suspect a few 9 volt batteries wired in series does the trick. Some of the newer commercialized radios (World War II era) operated from considerably higher heater and plate voltage potentials, rendering the common shock hazards associated with tube designs. Tuning is accomplished with an air cap and home made variocoupler. We covered a VFD based regenerative reciever earlier this year.

24 thoughts on “Homemade Regenerative Tube Radio

  1. “There doesn’t seem to mention of the plate source,”

    Grammar aside – I’m stoked about this; I’m not a digital kind of guy – I was born and raised in analog – so this is my thing.

    I know esoteric stuff isn’t the best site fodder, but as long as I get a few of these kinds of things here and there I’ll be happy coming back. I was getting worried for a while…

  2. The ‘vintage electronics’ section of ebay throws up shedloads of suitable tubes for this kind of thing, and google will usually help you with the pinouts if they’re not obvious from looking.

    Quite a lot of smaller tubes can work with a high rail as low as 12 volts for any application that doesn’t need huge gain – headphone amps spring to mind.

    Treat a tube like it’s a FET by giving it nice high impedances to work into and it’ll be your friend – they’re very forgiving as there’s no junction to burn holes in, but if you try dissipating too much power for too long you can still cause overheating… on some tube types you can even burn a hole in the glass envelope.
    (Beam power tetrodes ftw!)

    Don’t ask me how I discovered this. *blushes*

    At a pinch, a car stop & tail bulb with two filaments can be pressed into service any time you need a diode… heat one filament, and stick some potential on the other one, and you’ve got one-way electron-flow. :-)

  3. contrary to what the article says, the grid usually has a negative potential, which repels electrons and cuts off the plate current (the more negative the grid potential is, the more electrons are repelled, the smaller the plate current)

    quite a shame to see technical errors as soon as hackaday goes a little bit further than arduino so-called-“hacks”.

  4. I did something similar a couple of years ago from a Boy’s Life article published in the 1930’s. My variable cap was two pieces of aluminum foil glued to pieces of wood sliding past one another with a piece of cellophane in between. The tube was a type 30, and I think that’s the same one used in this hack; it was one of the earliest low power filament types. If you use two batteries for filament you need to put a silicon diode in series to drop the voltage a bit, because it’s designed for 2V and 3V will eventually burn it out. Not good for a tube that’s survived 80 years before you put it in your hack. I used 3 9V batteries in series for B+.

  5. “@stunmonkey: Correction. People were able to build things before they had the difficult task of *thinking* taken out of the equation. ”

    while i agree to some degree, i think another element here is the fact it is a pure analog project – we don’t see too many of those these days other than some kind of headamp clone.

    good analog design has always been more of an art than digital imo, and i wonder if we may be losing some of that in our modern digital age.

  6. @richard-

    Have you actually *tried* this? It seems as though it should work in theory, but in reality it does not.

    I don’t believe that tail lamp bulbs are evacuated, but might actually be pressurized with something inert (perhaps argon) in order to improve the life of the filaments.

    One time, I heated the glass on one of these bulbs with a torch until the glass softened. Instead of collapsing inward, as I would expect if the envelope contained a vacuum, it actually bubbled outward. When the bubble burst, gas could be heard to escape.

    I have tried numerous tail lamp bulbs and have yet to observe any diode action.

    If you carefully punctured the glass envelope, and evacuated the interior, then, yes, you would surely see some evidence of diode action.

    @richard said:
    >At a pinch, a car stop & tail bulb with two
    >filaments can be pressed into service any time
    >you need a diode… heat one filament, and stick
    >some potential on the other one, and you’ve got
    >one-way electron-flow. :-)

  7. @richard-

    Have you actually *tried* this? It seems as though it should work in theory, but in reality it does not.

    I don’t believe that tail lamp bulbs are evacuated, but might actually be pressurized with something inert (perhaps argon) in order to improve the life of the filaments.

    One time, I heated the glass on one of these bulbs with a torch until the glass softened. Instead of collapsing inward, as I would expect if the envelope contained a vacuum, it actually bubbled outward. When the bubble burst, gas could be heard to escape.

    I have tried numerous tail lamp bulbs and have yet to observe any diode action.

    If you carefully punctured the glass envelope, and evacuated the interior, then, yes, you would surely see some evidence of diode action.

    @richard said:
    >At a pinch, a car stop & tail bulb with two
    >filaments can be pressed into service any time
    >you need a diode… heat one filament, and stick
    >some potential on the other one, and you’ve got
    >one-way electron-flow. :-)
    PS: Forgot to say good post!

  8. no, you really should not run a tube with a positive voltage on the grid (which, contrary to a negative voltage, will generate a significant grid current) – it is very fragile and dissipating any power into it is likely to shorten the tube life.

  9. lekernal, the whole basis of using a triode as a radio detector is that during the half of the RF waveform when it’s positive you are not blocking the flow of current to the plate, and during the half when it’s negative you are. The half when it’s positive is normal and does not damage the tube as the power flowing during the positive cycle is trivial — that’s why you want to amplify it.

  10. @ac7zl
    yes I have – although it was a considerable number of years ago (end of the 60s) so it may have been before they started with the gas-filled bulbs… I always got the best results with the ‘Lucas’ brand bulbs that were very common then – i’m no longer able to drive so I have no idea if the brand even still exists.

Leave a Reply to refujeeCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.