Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Help understanding Armstrong's 1924 Superheterodyne Radio Receiver

  1. Jun 28, 2018 #1
    Hi,

    I was wondering if any of you fine people can help me understand a few things about Howard Armstrong and Harry Houke's superhet radio receiver built in 1924. My fist question is that they used a second harmonic and I don't understand why. Where was the second harmonic used and how did it help?

    In addition, I read a description of their design from a Radio News article from May of 1924 where they have a simplified schematic with 4 triodes and then write that it was great because it only used 6! Where are the missing 2 triodes?

    Thank you so much,

    Kathy Screen Shot 2018-06-26 at 3.55.13 PM.png Screen Shot 2018-06-26 at 3.55.13 PM.png
     
  2. jcsd
  3. Jun 28, 2018 #2

    davenn

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    well, it's obviously a schematic, NOT a block diagram ( there are no blocks ( labelled boxes
    with single lines to the next block/box))


    it doesn't look too much different to any other radio schematic of that era , eg .....

    upload_2018-6-29_8-1-39.png


    if anyone would like some really fun learning about a slice of radio history and also where the above diagram came from ......

    https://americanradiohistory.com/UK/Wireless-Age/Wireless-Age-1924-05.pdf

    Dave
     
  4. Jun 28, 2018 #3

    davenn

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

  5. Jun 28, 2018 #4

    davenn

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    after finally finding the article you were referring to
    page 1576 of the magazine

    https://www.americanradiohistory.com/Archive-Radio-News/20s/Radio-News-1924-05-R.pdf

    I strongly suggest that there has been typo's and that it should have said 4, not 6 for the 2 times mentioned in the text

    The article backs up my comments in post #4 that this is the full diagram and not a block diag., as a block diag. is also
    presented in the same article !

    In years gone by, back when I lived in New Zealand, the restoration of old tube radios
    was one of my major interests. I was fortunate to have a several old amateur radio operator
    friends who were well versed in tube operation. One of them even worked for a local business,
    in my home town of Dunedin City, building tubed radios for sale around NZ.

    I had years of fun collecting old radios and restoring them to their former glory as best as possible.
    When I left NZ for Australia in 1999, one of my fellow collector/restorers got all my radio chassis
    and my huge collection of tubes

    Dave
     
    Last edited: Jun 28, 2018
  6. Jul 1, 2018 #5

    tech99

    User Avatar
    Gold Member

    When using a local oscillator at half the required frequency, the second harmonic is generated in the mixer tube and this is used for the frequency conversion.
    The advantages of harmonic mixing are :
    1) The LO is at a different frequency to the antenna tuned circuit so it is not pulled by adjustment of the antenna tuned circuit. If it were at the same frequency, when we adjust the antenna tuning it would severely pull the LO frequency.
    2) The LO energy, being predominantly at half frequency, is not strongly coupled to the antenna, where it would radiated interference for other receivers in the vicinity.
    I should mention that harmonic mixing works very well.
    Discussion: The objective with the early sets was to obtain as much gain as possible, hence the low IF, and this exacerbated the problems I have mentioned. When broadcasting commenced, a few years later, a higher IF was chosen, which together with the invention of special mixer tubes, eased these problems and improved the image rejection.
    The problem faced by Armstrong at the time was the lack of a tube able to provide stable amplification at high frequencies. This came later with the invention of the pentode, and even later, the grounded grid circuit for a triode. Armstrong had previously invented the regenerative detector, which is very sensitive, but it was tricky to adjust and was pulled around by the antenna. In recent times I have made a "proper" version of the regenerative receiver using mechanically solid construction and with pentode RF amplifier to isolate the antenna. This received is extremely stable and sensitive and is better, I am sure, than the superhet shown above. It is able to detect antenna noise and maybe thermal noise.
     
  7. Jul 1, 2018 #6

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    There's a lot of scary looking math that describes mixers. It's worth slogging through the Fourier terms that result from multiplying two signals because it makes it easier to believe they really do work . And it's an eye opener to the genius of early radio engineers.
    Google will return lots of scholarly papers, try a few and find one that suits your own thinking patterns.

    http://www.analog.com/media/en/training-seminars/tutorials/MT-080.pdf
    http://rfic.eecs.berkeley.edu/~niknejad/ee242/pdf/ee242_mixer_fund.pdf
     
  8. Jul 11, 2018 #7
    Thank you everyone who has replied *so much* for taking the time to help me. I thought that the LO energy was at twice the frequency, but it sounds like the LO was half the frequency making the antenna tuned circuit to be at the second harmonic (that makes so much more sense!). So, to put in some numbers, lets say they wanted to tune to a station at 870 kHz, and the IF was 45 kHz, what signal would they mix? 480 Hz (870/2 + 45) or 457.5 ([870+45]/2) or am I hopelessly confused?

    Thanks again for all the help

    Kathy
     
  9. Jul 11, 2018 #8

    tech99

    User Avatar
    Gold Member

    Hi Kathy
    The signal is at 870kHz and the required IF is 45kHz. So we require an LO at either 870+45 = 915 or 870-45=825. So the actual oscillator frequency will be half these values, viz 915/2=407.5kHz or 825/2 = 412.5kHz.
    The input tuned circuit is not involved in obtaining the second harmonic - the non linear action of the mixer tube will do that. Also notice that the input circuit is tuned to 870kHz which is far away from the oscillator. When small adjustments are made to the input circuit, they do not pull the oscillator..
     
  10. Jul 11, 2018 #9
    Thank you! That makes a lot more sense.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted