I am not a fan of LC VFO's and long ago shifted to the Arduino / Si5351 as a means of having a stable signal source. But some of my pet projects (Shirt Pocket Transceivers) prior to the Digital LO used a crystal switched VXO. Here you could be frequency agile while having a rock solid no drift signal.
My disdain for LC VFO's is simple as there are better tools in the tool kit and it is all the time sink fiddling needed to get you something even close to working. Sure build at least one to say I have been there and done that. But Quickly Move On from there.
One of those better tools is a Super VXO where you get a much wider frequency excursion as you tune through the range.
Some smart dude figured out how to tap into a crystal's inherent inductance and capacitance factors and with external components change those internal elements and in effect move the crystal's frequency of oscillation. Boom a Variable Crystal Oscillator (VXO) -- the old Jumbo Shrimp Paradigm. Variable yet crystal stable.
So what is a Crystal Switched Super VXO and how does it work. N6QW, here, tried adding functionality to make a Super VXO tune an even wider range. I was successful and used two forms of this approach. In simple terms in one variant I built a heterodyne Super VXO.
The 1st method, above, was used in the Shirt Pocket SSB Transceivers where a 12.96 MHz Super VXO (three Crystals in parallel) was mixed in a SA612 with 6 MHz crystals
that were selected as the injection oscillator. Two 6 MHz crystals provide about 60 kHz
of coverage on 20M with a 4.9152 MHz filter. I managed to get this really small, and it worked great! Fortunately, the 60 kHz was in the 14.160 to 14.220 MHz range where I usually hang out.
For those who nitpick everything did you notice the similarity of the 2N3904 Oscillator circuit above and the posting several days ago about the 2N3904 Test Oscillator. Ooops you missed it!
That is a key point as I suspect while all of us want frequency agility, we likely haunt a small area of the band, It is rare on 20M that I operate above 14.300 MHz nor in the CW bands, so already you can see that has loped off a significant amount of spectrum.
The
second form shown below was used in a 17M transceiver that had a 4.9152MHz Crystal
Filter Placing the LO above the incoming meant that the LO had to
operate in the 23 MHz Range where 23.04 - 4.9152 = 18.125 MHz. But the
Bonus there are 11.52 MHz computer crystals that cost pennies. Three such crystals in parallel known as a Super VXO, give a wider frequency
swing than just a single crystal.
Now if you used a diode doubling
circuit you get twice the coverage. Two sets of crystals let me cover
most of the 17 Meter band. I had a second set of crystals made to order for this rig. A panel mounted switch let me select via a
relay which pairs of crystals were being VXO'd. The assembly in the photo shows the crystals and the relay.
Thus, we can use something other than an LC VFO or Arduino to give us agile frequency coverage with a stable signal and typically a small footprint.
The super VXO Crystal Switched LO offers a low-cost way to build high frequency LO's without the drifting mess of an LC VFO. Imagine a stable LC VFO at 19MHz? Not happening!
Computer crystals still are cheap and can be found at the major distributors and it should be noted that crystals in the range ABOVE 10 MHz are better suited for VXO's. The higher the better as larger frequency excursions are possible. If anyone tells you, they created a VXO that tunes the whole 40M band with a 5.2 MHz Filter using a single 1.8432 MHz VXO crystal they are smoking mushrooms! If you can shift a 1.8432 MHz crystal by 1 or 2 kHz you are indeed lucky.
BTW that is also another reason to use crystals in the range of 4 to 12 MHz as devices for a Crystal Filter. If you look at the stability ratings for these computer crystals it is often 50 PPM. Do the math 50 X 25 (MHz) = 1.25 kHz and 50 X 4 = 200 Hz. Just like Size we find Frequency matters.
73's
Pete N6QW