VFO coils

[G4COE]

Finally started to sink in, I think I grab it at last... thanks for your patience.

Think my mistake wasn't thinking about the length but just the diameter, if that changes mighty I should have thought about it's length.... lol, I never thought of train lines!

Dave

[AB9LZ]

I built a vfo using stacked t-50-6's (epoxied together then mounted with a nylon bolt / washer combo) . Happy to say that it's one of my most stable analog creations. I built a thick (0.64) box to mount the board and the tuning capacitor to insure that they don't move in relation to each other, which helps considerably.

73 m/4

[G4COE]

Just curious what the config. was.... the one I've seen was the Hartley with the tapped gate, this one uses two T68-6 cores stacked, thought I'd give it a whirl.

My concoctions have always been Hartley or Colpitts ossy's tuning no more than about 500Khz wide.

My early days ossy's used to drift like the clappers they seem to stabilize as you get older...!

Dave

[N0WVA]

Ive found it helps to mount the coil a little distance away from the PC board that contains the active components. You dont want heat from the transistors getting into the coil via conduction. Mount all zeners or regulators and associated circuitry on the outside of the vfo box. The idea is to isolate all the critical components from any heat production whatsoever.

[KO6WB]

Just curious what the config. was.... the one I've seen was the Hartley with the tapped gate, this one uses two T68-6 cores stacked, thought I'd give it a whirl.

My concoctions have always been Hartley or Colpitts ossy's tuning no more than about 500Khz wide.

My early days ossy's used to drift like the clappers they seem to stabilize as you get older...!

Dave

Take a look back in the publications and see if you can find the Vackar circuits. These were supposed to be the best VFO circuits for stability. None of the DC voltage passed through the tuning coil. It's reported to drift in the 10Hz or less range. I'm not sure if that's really a fact or not but it sure sounded good to me. Almost every VFO circuit I've ever build was of a Colpitts variation and they always worked. Drift wasn't bad at the time. I always put my VFO's in the oven at about 150-165 degrees F. Then put them in service. They still work today and the drift is low. Those were also the days when you could get N750 capacitors and those helped.
Have fun
73
Gary

[WB5WSV]

Ham Radio Magazine's Ham Notebook Vol 1 discusses Vackar VFOs and provides a circuit for a version of a Seiler VFO called a "synthetic rock." It was designed by W3JHR and is supposed to be very stable. It used an ARC-5 transmitter oscillator circuit as the basis, was very popular and used all over the world. I'm going to build one, as soon as I have the 300 or so other projects I now have in work completed.

Robert
WB5WSV

[G4COE]

I do have the Synthetic rock ossy somewhere, it's in one of my ARRL h/books.

I was reading somewhere there's something to be had using a NE602 as a oscillator, claiming much stabler than discreet devices because the article stated the oscillator portion uses some kind of RF feedback control along with temperature stabilizer.

Whenever I used a Gilbert cell mixer I've always seem to find the oscillator quite reliable whether Hartley or Colpitts config, they're not much cop for a strong front end mixer though... but they do work!

Have a couple of ceramic formers out of a BC221 heterodyne frequency marker, the smallest is about 1" dia. though, in my case it's for a 5 - 5.5Mhz mixer vfo config.

Dave

[KM1H]

Ive built many FET and bipolar oscillators and the best core Ive found is the 7 Mix. I also use whatever wire size fills about 75% of the core with about double spaced between turns which lowers the coupling C. For stability I attach the coil to the board with a nylon screw/nut and flat washer.

Carl

[G4COE]

KM1H.
I can see that spacing the winding will lower it's self capacity, same if the ends are not too close. Surely, winding with double spacing of the wire diameter will limit the inductance that one requires since the cores have a fixed Al and core size, I suppose you could could always use a high C to L ratio to attain resonance, the wire you don't want too thin either, maybe I'm missing something in your approach.

I do think there's a lot of myth about building VFO's such as this circuit or that circuit, I'd say it's about 50% on the components and 50% on how it's built. In the day's of tubes where you had heat from the valve, thus heating components by conduction as well as convection guess some circuits would perform better than others.

I often wonder how much the core heats up due to RF heating or it's transistor for that matter, since the current being drawn is usually peanuts. Perhaps I'll try controlling it's frequency with variable temperature.... only joking!

Dave

[KM1H]

My approach is to pick a core diameter that allows the inductance I want. High C also offers higher Q and stability but its not mandatory for general use. Many QRPers use #28 wire and any RF resistance caused heat is too much. A transistor with well above the circuits power requirements minimizes that issue.

Carl