ad: Mircules-1

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

    N2EY Ham Member QRZ Page

    KB1WSY reminded me of this, so I finally found the photos....

    ----

    GEDC1249.JPG
    ----

    In QST for October, 1960, W2YM described an 80 meter VFO using the Colpitts circuit and a 12AU7A tube. The article was aptly titled "Stability with Simplicity". However the unit described was meant for use with W2YM's homebrew rig, and could not drive the typical MOPA rigs of the day directly.

    In QST for February, 1962, W1ICP described "An Easy To Build VFO", based on the W2YM article. This unit added a 6AU6 buffer/multiplier and external power supply, and used a simpler mechanical design. The resulting unit could drive almost any hollow-state transmitter of the times.

    In the late 1970s I found myself in need of just such a VFO, and built a version of the W1ICP design. I used the PA capacitor from an ARC-5 transmitter rather than the dial/capacitor combo W1ICP and W2YM used. Besides having the ARC-5 capacitor on hand, it provides a much slower tuning rate, for easy zero-beating.

    It was tossed together rather quickly, and worked so well that it was left as-is rather than being prettied up. The extra holes in the bottom chassis are because it was reused from a hamfest-find.

    This little VFO was used for several years, driving several different transmitters: a Johnson Viking II, a homebrew 40 watt 6AG7/807 MOPA, and finally a 6AG7 - 2x807 150 watt rig. Then I started building transceivers, about 1981, and the VFO was stowed away. The knobs, tubes and dial were "borrowed" but it's still pretty much all there.

    Here's the front view

    evfofront.jpg

    Knob on the right is buffer tuning, switch is for spotting, the dial disk goes under the big nut and the tuning is on the far left.

    Here's the works inside:

    evfotop.jpg

    Underneath is a relay, which is more convenient for station control than W1ICP's switching arrangement:

    evfobottom.jpg

    And a rear view:

    evforear.jpg

    There's a top plate which adds to the shielding and rigidity of the unit (first pic). The top plate is 1/8 inch thick aluminum, and the bottom chassis is steel, resulting in a really solid unit.

    This VFO was so easy to build, and worked so well, that I wonder why anyone interested in hollow-state bothers with crystals or anything less. My version isn't the neatest - it was actually built on a kitchen table while living in a small apartment.

    Not sure what I'm going to do with it. It's been sitting for 30 years now....
     
  2. G3EDM

    G3EDM Ham Member QRZ Page

    Seriously? From my reading, I've been led to believe that it's quite hard to get anything even close to the stability of Xtal control....

    Over the past year I've amassed the parts for that very same VFO that you built those several decades ago.... I hope to build it soon and it will be interesting to see how it performs in today's environment.

    (By "soon" I mean sometime in the next year or two.)
     
  3. W7UUU

    W7UUU Principal Moderator Lifetime Member 133 Administrator Volunteer Moderator Platinum Subscriber Life Member QRZ Page

    In my 40 years as a ham, I've built 3 VFOs using tubes.

    I hope that by my death they will all zero-beat for a period of 5 minutes

    :)

    Dave
    W7UUU

    NOT REMOTELY doubting you Jim.... tried my best but only built HF sliding whistles - not VFOs...
     
    KA9P likes this.
  4. G3EDM

    G3EDM Ham Member QRZ Page

    Dave, how can you possibly doubt Jim on this subject? (And I know that you're not doubting him!)

    I've spent much of the past couple of years patiently amassing parts for That VFO:

    VFO_parts.JPG

    Sliding Whistle? We'll see! (And if you have the last laugh ... so be it!).
     
  5. W7UUU

    W7UUU Principal Moderator Lifetime Member 133 Administrator Volunteer Moderator Platinum Subscriber Life Member QRZ Page

    Good for you WSY.... I have EVERY confidence in you.

    You're an inspiration here

    Dave
    W7UUU
     
  6. G3EDM

    G3EDM Ham Member QRZ Page

    Seriously though, it does appear (from the vintage literature) to be quite hard to get the mechanical/electronic stability required to replace crystal control with a VFO, given the constraints from the 1960s. However if you look at Jim's build (or the ARRL builds in the photos from QST) it does seem *possible* with strong mechanical and component stability.

    So here I am, a middle-aged would-be ham but with the beginnings of a proper workshop, especially as concerns metal-working. I have floor-mounted drill press, a lot more money than teenage 1960s hams, and a lot more patience. So perhaps I can make something better than a sliding whistle. Plus, if not, what's the big deal? Just leads to more experimentation.

    UUU, I'm extremely flattered that you could derive any inspiration from me, given that I'm still not quite on the air, after 42 years.... You, and EY, are the guys and gals I look up to at this stage. Along with lots of others here on QRZ and eHam.
     
    Last edited: Jan 3, 2014
  7. KL7AJ

    KL7AJ Ham Member QRZ Page

    The first VFO I ever built was for a homebrew 160 meter rig. It was definitely EASIER to build a stable hollow state VFO than a transistor one...not that my tube one was anything to write home about. The real key is to build everything like a brick outhouse. Thermal drift seems to be much less of a problem than "cabinet warp tuning."

    Eric
     
  8. N2EY

    N2EY Ham Member QRZ Page

    Yes, seriously. I gave up on crystal control when I got my Advanced in 1968 and never looked back.

    Crystal control is simple, particularly when those old articles were written. Ham-band FT-243 (or equivalent) rocks were everywhere, the grid-plate and modified-Pierce circuits using a 6AG7, 6CL6 or 12BY7 are sure-fire, and you just built to the schematic and done. More than that, in those days crystal control was required by law for Novices, so almost all simple transmitters, and many not-so-simple ones, were designed for crystal control.

    Even today, it can make sense to follow some of the old ways. For example, my old MOPA rigs could all do crystal control - this made them easy to test and get working.

    Here are some things I've learned about making good stable hollow-state VFOs:

    1) Start with good solid mechanical design. Everything has to be RIGID and SOLID. The W1ICP VFO uses an aluminum chassis upside-down on a steel chassis, with a top plate and lots of hardware. The W2YM VFO is a small box, with internal brace and extra-thick top and bottom plates. The chassis are small, adding to the rigidity. Everything is bolted down so it cannot move. Wiring is carefully planned, rigid and short.

    Many VFO designs don't do the rigidity thing anywhere near enough. Thus the problems.

    2) Add good thermal design. Note that in both the W2YM and W1ICP VFOs the tubes stick out the back of the box - they are not buried inside. The total power dissipation in the VFO circuit is small - the VFO runs at only 75 volts on the oscillator (there's a typo in the W1ICP schematic; the lower VR tube is a 0C2 not 0A2). The VFO is a separate unit, kept away from the heat of the power supply and transmitter.

    When you see a VFO design with the whole thing on one chassis, using relatively high-power tubes and such, the heat dissipation is bound to be high - and the drift.

    3) Use the best parts available for the job. BC-type capacitors are OUT; their plates are aluminum. The ARC-5 variables and capacitors from the LM and BC-221 frequency meters are the way to go; they are much higher quality than almost anything found in most ham gear - and they're cheap and common, even today. Plus a lot of mechanical headaches go away when you use a variable with a built-in gear-drive mechanism. Don't be afraid to use big parts!

    I have found that 3011 and 3015 Miniductor is good stuff if it's in good shape. Ceramic form coils too. The red dipped silver mica caps are excellent. APC air trimmers - be sure to use ones with brass plates, not aluminum, in the VFO. Ceramic insulation, not bakelite, in the tube sockets and tuned circuits.

    4) Only now do we think about the schematic. (Too many hams think that if they have a schematic they have a complete design. That's just not the case with VFOs!)

    Besides the keep-the-power-down concerns, stability requires eliminating things like bandswitches. Keep the VFO frequency down and the capacitance up (note how high C the W2YM VFO is - hundreds of uuF and only a microhenry or two, of 80 meters!). Reach the higher HF bands by heterodyning, not multiplying.

    Use regulated voltages for the oscillator, and a buffer stage. Use tubes with the A suffix if available (12AU7A is better than 12AU7).

    Some folks claim VFOs can be keyed, but I've found it's much better to key the later stages and let the VFO and a buffer stage run when in transmit mode. This means no QSK unless you do the heterodyne-VFO thing.

    The best hollow-state VFO designs use the heterodyne principle, One good combo is to have the VFO on about 5 MHz and a crystal oscillator on about either 9 or 12.5 MHz. 9 minus 5 gives you 80/75, 12.5 minus 5 gives you 40, 9 plus 5 gives you 20. Such VFOs require even more careful design in the mixer, which is beyond this discussion, but I've done that stuff too. With a heterodyne VFO, the oscillators can run all the time, even in receive, and you key the mixer.

    5) Do test runs to determine drift. For an 80 meter VFO, one way to do this is to listen to WWV on 15 MHz and tune the VFO to 3.75 MHz. Adjust the VFO for zero beat (fourth harmonic). Any drift is multiplied four times, and WWV doesn't drift. For a VFO in the 5 MHz range, second harmonic to WWV on 10 MHz or third harmonic to 15 MHz.

    6) Have reasonable expectations. Even the best tube designs need some warmup - 15 minutes to a half hour is normal. This means drift testing takes time, because the thing has to be allowed to cool off completely between runs.

    Yes, I learned most of this the hard way. You don't have to.
     
  9. N2EY

    N2EY Ham Member QRZ Page

    Some more notes on VFO stability and overall design:

    7) Do NOT pull plates from variable capacitors! Yes, the Ancient Ones sometimes did that, but there's no reason to do it today. If the range of a variable is too great, it can be reduced by a series capacitor, or tapping down on the coil, or both, depending on the circuit.

    8) Linearity of a dial can often be controlled by careful design. For most ham bands, a straight-line-capacitance (SLC) variable will give nearly a nearly linear dial (see W2YM design). If you have a straight-line-wavelength (SLW) capacitor, such as from an ARC-5 transmitter, a series capacitor can be used to linearize the dial.

    9) Resonant frequency calculations are made easy by the formula LC = 25,330/f squared. Using a spreadsheet, you can dial in all sorts of combinations.

    10) When doing stability checks, be sure to check sensitivity to heater voltage changes. The W2YM circuit is quite immune to them, but some other circuits can be sensitive. What seems to be drift may actually be frequency shift from line voltage fluctuations.

    11) As K9STH points out, it is a very good idea to run the VFO from its own dedicated power supply, rather than "stealing" power from the transmitter. Voltage regulation is a key reason. I would add that the VFO power supply should be on a separate chassis, well-filtered and regulated, and oversized for the job (improves regulation).

    12) Watch out for unsuspected causes of drift. Iron-core RF chokes can be problematic - I try not to use any ferrite or powdered iron in VFOs. (The slug in an ARC-5 transmitter VFO coil compartment is OK, though). I don't use mica compression or ceramic trimmers in VFO tuned circuits, and I use oversized carbon comp resistors.

    13) The lower the VFO frequency, the lower the drift. It's not a linear function, either - the same design on 40 will have more than twice the drift it had on 80.

    14) Be sure to do a "short circuit" test on the VFO output. What this means is that shorting the output of the VFO system should not cause frequency shift. It's an indication of how well the buffer stages are working. Many cases of chirp and drift are really a matter of the VFO reacting to the varying load of a keyed transmitter.
     
  10. N2EY

    N2EY Ham Member QRZ Page

    What designs did you use, Dave?

    ---

    Let's see...in 46 years, I've built at least 8 VFOs. Learned a lot along the way. (See above posts). Perhaps the biggest lesson is that VFOs aren't magic; they just take care and patience.
     

Share This Page

ad: w5yi