This is the story of my quest to build an all-tube VFO for 26.0 to 26.5MHz, for use in a 21MHz SSB/CW superhet transceiver with 5.0688MHz intermediate frequency (IF). Construction efforts below were assisted by advice and commentary by G4OEP, K3IMW, K7HKL and VK6JY.

An oscillator needs to be reasonably stable - the frequency should not drift too severely as the temperature changes. This is more challenging with a valve VFO than a solid state oscillator. Even with solid state, it would be very difficult to build a stable enough LC VFO at 26MHz. The best approach is to build a lower frequency VFO and mix it with a stable crystal oscillator to get the desired output. Right from the start, my aim was to build an LC-VFO for 6.0-6.5MHz, mix it with a 20MHz crystal oscillator, and select the 26MHz sub with suitable filtering.

	1. Sub-mini valve VFO First attempt is to build a stable 6MHz VFO using a type CK512AX sub-miniature tube, originally used in hearing aids (idea from Andy G4OEP). This ultimately didn't work because the particular tube is unsuitable for this application. Read more...
	2. Mixer VFO mk1 The next attempt uses 12AT7 crystal oscillator and VFO; 6BE6 mixer, and EF91 buffer. It produces a 26.0-26.5MHz VFO that works, but the heat build-up in the box was very high and the drift was 50-70kHz per hour! Read more...
	3. Mixer VFO with insulated tank components The same circuit as 2, but now with the tank components insulated, and the tubes in the open for better ventilation. This greatly reduced the drift, and temperature compensation of the VFO tank components reduced the drift to 50Hz/hour after the initial warm-up. Read more...
	4. Mixer VFO using battery valves To reduce the warm-up drfit, this mixer VFO now uses low power battery valves - DF96 VFO, 1T4 crystal oscillator, DK91 mixer, DAF91 buffer. It works and the drift is within 50Hz/hour, but not really repeatable. Also shown here, voltage regulator tube experiments. Read more...
	5. DF96 VFO and logger The next rebuild abandoned the idea of insulation, which seems to slow down drift but not eliminate it, intending to put a DF96 VFO and buffer inside the box. However it became more about a spin-off project to develop a frequency-counting data-logger, to make measurements easier and automated...Read more...
	6. 6088 submini-tube VFO One reader kindly sent me a pair of 6088 pencil tubes to experiment with, and they seem much more suitable than the CK512 I tried before in Andy G4OEP's circuit. Now at last, I managed to realise my dream of many years, to try out Andy's sub-miniature valve VFO. Read more...
	7. Dual-6088 Franklin VFO Another dream, was to try out a Franklin VFO. Here I did so using the pair of 6088's, and had no trouble making them oscillate nicely. The Franklin oscillator topology isolates the tuned circuit from the load and capacitance variation of the active device and is supposed to be very stable. Read more...
	8. BC221 Tuning capacitor I found a very nice air-variable tuning capacitor in the junk box, which seems to have been designed for the BC221-AK frequency meter. It has a 100:1 worm-gear reduction drive and seems perfect for this VFO project.Read more...
	9. Boxed Franklin VFO Since the dual-6088 VFO is my favourite, and seems the most stable, of any of those so far tried on this page, here I am boxing it up and using the BC221 tuning capacitor. The resulting oscillator covers 6.0-6.5MHz and I was able to watch it stay within +/-2Hz for 30 minutes.Read more...