My last entry described my accidental discovery of the PTO for QRP purposes. I breadboarded it and was amazed at the results! I went ahead and built this carefully in an enclosure and the output is wonderful. It’s strong, it’s stable, and it tunes effortlessly over the same range it did before (about 1MHz). The video describes details of the action, and demonstrates the stability of the oscillator by letting you hear it audibly on a nearby receiver.
The fundamental concept and hardware is straightforward. Two nuts are soldered into an Altoids tin providing much-needed grounding for the screw (reduces shift when it’s touched). Also the wire soldered over the screw is pinched firmly at the base to apply constant pressure to the screw to make it hard to turn and therefore more stable while turning. The inductor is a bunch of turns (no idea how many, about a meter of magnet wire) around a McDonalds straw.
Alltogether it’s a simple colpitts oscillator with a MPF102 JFET at its heart, using a 74hc240 CMOS buffer as an amplifier. There’s a voltage regulator in there too.
The result? Pretty darn stable (by CW QSO standards). That’s without any regard to thermal isolation or temperature compensation. I’m quite pleased! I look forward to MUCH more experimentation now that I’m starting to feel good about designing and building simple, tunable, stable oscillators. It’s always hard to nail all 3 in a single device!
4 comments
dave
June 5, 2011 at 7:52 AM (UTC -5) Link to this comment
Nice work!
If you do have stability problems, then you could move the voltage reg out of the case. Those linear regulators can run quite warm, having something dissipating a few hundred or more mW of heat right next to your oscillator is a good recipe for drift.
Scott
June 5, 2011 at 9:33 AM (UTC -5) Link to this comment
This was a dilemma for me because I was previously told to move the regulator INTO the case. The reason was to prevent stray capacitance and inductance causing drift. Previous designs of mine would drift just by waving my hand near the enclosure. I believe this was because the power wires entering the device acted as a bit of a capacitor with the case? I guess it’s possible this was cured with the all-metal case as well, so it’s not a fair decision to make this early…
Dave
June 8, 2011 at 5:38 AM (UTC -5) Link to this comment
If the supply is well decoupled, and fed into the VFO case though a feed-though capacitor then it really shouldn’t “pull” the VFO frequency. Ultimately it comes down to what works for you – if the VFO is stable enough for your liking, then there is no good reason to change the layout.
Best of luck!
Dave
AI4YC
June 8, 2011 at 8:28 AM (UTC -5) Link to this comment
So…. are you bringing it to the next GARS meeting?
Have you played with it on a receiver?