An Easy To Build (Hollow State) VFO

Discussion in 'Homebrew and Kit Projects' started by N2EY, Jan 3, 2014.

ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: l-BCInc
ad: Left-2
ad: abrind-2
ad: L-Geochron
ad: Left-3
  1. N2EY

    N2EY Ham Member QRZ Page

    I tried it years ago. I found the W2YM circuit to be superior, at least for amateur-band use.

    The W2YM circuit is a high-C Colpitts with the oscillator run at low B+, followed by a cathode follower for isolation. The result is very low drift and excellent mechanical stability.

    On thing I have found to be true in VFOs is that the mechanical design is as important if not more so than the electrical. It's all about "the weakest link".

    73 de Jim, N2EY
  2. SM0AOM

    SM0AOM Ham Member QRZ Page

    Usually the mechanical design features "win" in VFO stability, as the electrical
    properties, especially in the narrow amateur bands, are not so decisive.

    However, the electrical properties may have influence in especially low-noise oscillator designs. I recall a German write-up in DUBUS from the 70s about how high-stability oscillators are influenced by their coupling mechanisms and noise figure.

  3. G3EDM

    G3EDM Ham Member QRZ Page

    @N2EY as you know I already have most of the parts.... It sort of screams BUILD ME.


    I think it's this one and therefore pretty much exactly the one Jim is describing.
    It is rapidly moving up my to-do list after I went on the air for the first time recently and discovered the exigencies of crystal control. For those who have never experienced it, it basically means that you are in a straitjacket of calling CQ and waiting for a response. "Hunt and pounce" is out of the question because it is virtually impossible, with crystal control, to zero-beat your transmitter on another ham's signal and modern receivers have such high selectivity that just "being in the same frequency neighbourhood" does not work. Very few hams listen for responses on either side of their own signal any more.

    (I also need to adapt my little transmitter so it can actually accept VFO control. That is under way!)

    73 de Martin, G3EDM
    Last edited: Oct 6, 2021
  4. G3EDM

    G3EDM Ham Member QRZ Page

    @N2EY Jim I think you asked me about it in another thread. In the parts photo in the previous post, if you look at that coil in front of the filter choke, it is Miniductor-style but already mounted on a rectangular perspex bar with a hole at either end for mounting posts. It may have to be shortened I suppose but that should be relatively easy to do.

    The Millen right-angle drive coupler on the right is because I was never enamored of the "tube sticking out of the back" scenario. So I envisaged building it sideways (a standard VFO layout but side to side instead of front to back), having the tube stick out of the side instead of the back (and covered with a perforated box), and mounting the dial at the right-hand end, driven by the right-angle coupling.

    But in my new shack where everything has to be vertically arranged, it may be that the more classic design with the tube sticking out the back may be better! Or I could just rotate the variable capacitor to face the front -- I don't know why I thought it essential to mount it in a particular direction.

    I have not started planning the chassis yet. The simplest solution, which you alluded to elsewhere, is fastening a steel and an aluminium Bud/Hammond chassis, with one of them upside down on top of the other. I may yet go that route. Otherwise, I might do it the hard way and use very thick aluminium panelling, cut in rectangles and fastened together with angle brackets. I have used that construction method for light-duty projects (such as cabinets) successfully in the past.

    We have a great online service here (surely also similarly available in America) where you can have aluminium pre-cut for you at whatever specs y0u provide. It might sound like cheating ... but I'm not that keen on sawing 1/8" aluminium stock with my bare hands....

    73 de Martin, G3EDM
    Last edited: Oct 6, 2021
  5. SM0AOM

    SM0AOM Ham Member QRZ Page

    I would suggest that you take Jim's suggestions seriously, and I also would like to add my own. Some decades of hard-earned experience, with the perhaps most hard-earned being "homebrewing" decent-sounding VHF/UHF transmitter frequency generation during the early 70s, have provided me with these thoughts;

    • Always build the VFO as a separate unit from both the transmitter and power supply
    • Strive for a good shielding of the VFO tuned circuit
    • If possible, operate the VFO on a submultiple of the transmitter output frequency
    • Make sure that all buffers operate without grid current
    • Make the output level from the VFO unit adjustable using variable screen voltage for the last stage
    For your application, I would suggest an approach as follows:

    1. Build a power supply for your future higher powered transmitter, with a selection of VR stabilised voltages also available. In the PS, put in a number of output octal sockets as well as binding posts for making voltages available for tests and experiments (you have to decide if the PA HT+ should be available on binding posts). Make the HT+ and filament supplies individually switched

    2. Use a die-cast box as a chassis for the VFO, with the tuned circuit inside, and the tubes on top. Use an internal partition to isolate the VFO and buffer/multiplier tuned circuits from each other

    3. Find a suitable dial with a 6:1 reduction. Calibration is quite unimportant, only the lower band edge really needs to be marked, if the fundamental tuning range is limited to 3500-3600 kHz (which I suspect will be entirely sufficient)

    4. Use a high-slope pentode as buffer/multiplier, but there is a compromise between slope and power dissipation, perhaps the 6AH6 is one of the better

    5. Convert the input stage of the existing 5763 transmitter to a grid-plate oscillator, which is better adapted to external VFO input
    Now to some testing and integration aspects:

    In order to test and evaluate the VFO, you certainly need a better receiver.
    Some form of crystal calibrator is also needed, a suggested one is a transistor oscillator using a 3500 kHz band-edge crystal. The VFO can then be zero-beat against the calibrator with the receiver in AM mode, and stability characteristics determined even if the receiver itself drifts somewhat. However, a 100 kHz calibrator has other applications.

    If the VFO is built well-shielded and its power leads filtered it can often be operated without turning off the oscillator itself even during listening periods.

    G3EDM likes this.
  6. N2EY

    N2EY Ham Member QRZ Page

    It may not be so easy. I'd recommend using the other length and cutting to length before mounting on a piece of perspex. There should be enough for both the VFO coil and the buffer coil.

    Also consider reducing the tuning range. If your only interest is CW there's no need to cover 3500 to 4000 kHz.

    The most important consideration is mechanical rigidity. I don't think that right angle drive has anti-backlash gearing, which means it's NOT what you want for a tuning mechanism!

    G3EDM and AJ4GQ like this.
  7. G3EDM

    G3EDM Ham Member QRZ Page

    Good point. I haven't yet operated on 80m and 20m but it looks like a tuning range of (maximum) 150 kHz should be fine for CW on 80/40/20, so 3500 to 3650.

    I do that sort of thing all the time. For instance, when drilling front panels, have them screwed on to the chassis already, then drill through both at the same time. If you don't do it that way, it's awfully easy to get the hole(s) mis-aligned....

    At the very least, the pilot holes should be drilled that way.

    73 de Martin, G3EDM
  8. G3EDM

    G3EDM Ham Member QRZ Page

    There seem to be two schools of thought about this. My Elmer at school had a huge power supply (and modulator) from which he operated essentially all of his ham equipment. A single PSU may make sense in my shack because of limited space (although there is now at least the possibility of getting more space, we will see). But it would need a pretty large chassis including a lot of empty space for future expansion.

    A few years ago I did obtain a large number of both octal and 11-pin male and female connectors, both inline and chassis-mounted.

    73 de Martin, G3EDM
  9. SM0AOM

    SM0AOM Ham Member QRZ Page

    There are as many "schools of thought" as there are radio amateurs... In many aspects, we are as the medical profession, we cannot agree about anything :).

    However, a 100-150 W input twin power supply (+750/350/255/105/-100 V from one, +300/150/105/-100 V from the other) using modern electrolytics and silicon rectifiers does not need to be larger than about 180x300 mm "floor space", with the VR tubes and the power transformers above and the filter chokes + electrolytics below "floor level".

    Regarding VFO tuning circuit values, I went through the bandspread equations for a 3490-3610 kHz range, and with all other component values unchanged, a variable capacitor of about 25 pF range will give this frequency variation with a 5.5 ┬ÁH coil.

    Since you have ceramic coil forms with about 25 mm diameter a better coil with better mechanical stability can be made by space-winding 19 turns of 0.8 mm wire, which gives a Q of about 300.

    No expense should be spared on the variable capacitor.
    It should be double ball-bearing and preferably be made of brass.

    Wingrove & Rogers (Polar) and Jackson made a series of very good capacitors that were part of the RF26/27 tuners.

    (available from

    G3EDM likes this.
  10. K6BSU

    K6BSU Ham Member QRZ Page

    I have the original 1949 "whole" article by Jiri Vackar. It wastes a lot of words describing the basic oscillator types before actually getting down to the nitty-gritty describing the Vackar circuit. The paper is too long to post here. If anyone wishes a copy, contact me by e-mail and I can mail a printed copy.

Share This Page