QRP CW Homebrew

Discussion in 'Homebrew and Kit Projects' started by 14FRS3755, Jun 22, 2021.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: l-BCInc
ad: Left-3
ad: L-MFJ
ad: L-Geochron
ad: abrind-2
  1. 14FRS3755

    14FRS3755 QRZ Member

    Hello from France everyone,

    A little intro first, I am making ham devices for my father who's a licenced (F4IHU) ham that has been away from the hobby for more than a decade. He decided to get back and now he's introducing me to the ham world while I help him with all the DIY and eletronic stuffs.

    Now the project for wich I need the help of the electronic gurus among you:
    I'd like to build a simple QRP transceiver for CW in the HF band from scavenged components only (or almost).
    I found a bunch of more or less complex schemes on the web but I don't know how the values of the components are defined so I can't adapt them to what I have.

    What I'd like to do is to give you a list of the major components I have available and we try to define a rig that I could build.
    I know it might be challenging but I'm sure we can do it. Also I know about all those kits you can buy (Pixie, ect...) and I might end up with this but this not the point of this post.

  2. 14FRS3755

    14FRS3755 QRZ Member

    The major components I have right now:

    - a crystal of 20.9845 KHz

    - transistors:
    1x S8550 ; 3x S8050; 1x CJ TL431 ; 2x c945 ; 2x B772; 1x D882; 1x FQP 5N60C ; 1x P16NF06.

    -two variable capacitors (but I don't know there range... ). One is a Shinto with 6 pins from a FM clock.

    - A whole bunch of resistors (and variable resistors), ceramic and chemical capacitors, diodes...

    -A few ferite rings to make coils

    -A few adjustable coils (those with 5 pins, in a metallic case with a colored scew)

    The target output would be around 1 or 2W but if we end up 200mW it would be alreadu a succes!
  3. VK3YE

    VK3YE Ham Member QRZ Page

    The crystal is the hard bit. It is fine for testing (if you're testing a 21 MHz transmitter) but it's outside the amateur band so is not useful in practice.

    You sometimes see amateur band crystals in old computer gear or modems - eg 3.58 MHz, 7.16 MHz, 14.318 MHz etc. But they're inconvenient frequencies for CW contacts.

    But if you can find a 3.58 MHz ceramic resonator then that could be pulled down into a useful CW segment frequency.

    Failing that the best approach is to buy a crystal on ebay - eg 7.023 MHz is a common frequency.

    The easiest way to start is with a basic transmitter. This could be as simple as 1 transistor though 2 or 3 will give more power and better keying.

    The best bands for simple QRP transceievers are either 3.5 or 7 MHz. Higher frequency bands are harder to get contacts on and transistor gain drops.

    If you have a websdr or kiwisdr you could connect it to an antenna and hear yourself. And maybe even make contacts.

    Some simple circuits described on my website at:


    Video below is a 1 transistor tx built without soldering.

    Later on you can add a direct conversion receiver. That can run off the transmitter's crystal oscillator if you keep that running all the time (with a small frequency offset).
    Last edited: Jun 22, 2021
    N8TGQ and US7IGN like this.
  4. 14FRS3755

    14FRS3755 QRZ Member

    Thank you for your answer VK3YE.
    Just to be sure, as 21Mhz (21 to 21.450) is a ham band (here in zone 1), it is because this crystal is a bit under? Isn't there a way to tweak it up a bit?
    I found a few others but still not in the range you specified (1M and 4MHz). My grand father left us a suitcase full
    of crystals, I'll have a look in it to see if there is any of the goo frequancy!
    Last edited: Jun 22, 2021
  5. VK3YE

    VK3YE Ham Member QRZ Page

    Crystals can be moved in frequency but only very slightly, especially for small sized crystals. Yours is too far out of the band to bother with.
  6. 14FRS3755

    14FRS3755 QRZ Member

    So, I found from my grand Pa 5 crystals that could fit my needs. Their frequecies are 7.05 ; 7.075 ; 7.1 ; 7.125 ; 7.15 Mhz.
    There are old FT 243.
    I started to learn a bit more about crystal oscillators, mainly the Collpitt one, but yet I haven't found the formulas to calculate the value of the components
    according to the frequence of the crystal and the desired power output.
  7. VK3YE

    VK3YE Ham Member QRZ Page

    Those frequencies are good to experiment with. In relation to finding activity at least some would be OK in the USA but they are too high for CW in Europe.
    But you could build a crystal controlled double sideband with them. FT243s though aren't so good in VXOs so can't easily be pulled. But there are other ways of vary their frequency involving opening them up and using pencils.
  8. 14FRS3755

    14FRS3755 QRZ Member

    Maybe we can experiment a bit with those. Our project here is really to try to get into CW so I'd rather stay with that. What would be the minimal output power to try to reach North America?
    I saw videos of people doing impressive stuffs with less than a couple of watts. I'm not afraid to climb on local hill, at night if needed!
    Should we rather design a circuit specifically for one of the crystals or is it possible build a "socket" so I can change them at will?
  9. W1VT

    W1VT Ham Member QRZ Page

    A couple watts to a full size dipole on 40 should be enough. 200mW is possible but may require a very good station on the other end.
    US7IGN likes this.
  10. VK3YE

    VK3YE Ham Member QRZ Page

    If all the crystals are all in the one band then it's a good idea to have either a socket or a switch to easily change between the ones you have.

    If it's a multiband rig you will also need to switch other parts of the tx circuit eg the LPF. Better and easier to stick to a single band like 40m.

    Note that very simple 1 transistor rigs may draw high current from the crystal which in extreme cases may break it. Bigger / older crystals
    are more tolerant than smaller crystals.

    This is an example of a damaged crystal.

    You can minimise this risk with a 2, 3 or 4 transistor transmitter. And allow some frequency agility with a VXO which is easier when the crystal is less loaded.

Share This Page