So I have finally decided to build some sort of a Tesla coil, starting with an HFVTTC.
Whether or not this was a good starting point, I have no idea, but the schematic is rather straightforward.
My design was based on this one by uzzors2k:
The difference was mainly that I used a different tube, and handled the g2 power differently.
I am for now using GU50 Soviet pentode, while he used a way more powerful GK-71.
As for the G2 supply, I am currently using rectified mains (filtered) which is put across g2 and the cathode instead of a large resistor which just connects G2 anode potential.
This leads to a low G2 voltage of only about 110V. I plan to fix this by using a CW voltage multiplier in the future.
Note: In a further iteration, this was no longer an issue since the circuit was driven with way higher voltage anyway.
For another change, I also used a different capacitor in the feedback which is only about 5kV and its capacity is way too large. This should not be much of an issue unless I short out the jar lid™ capacitor while running at high powers, but I do plan to replace it.
I also think that I should up the number of turns on the L1 (first) inductor, as that seems to have worked better for others.
Updates1
Both the resonator coil and the choke have been rewound:
- Resonator: ~23 turns
- Choke: ??? I don’t know, I eyeballed it.
I have also split them off so they are no longer on the same coil form. I specifically 3D printed a 5cm wide 7cm high form for the choke.
I have also pushed way higher voltage into the setup which helped the initial arc to stay ignited.
Some Results
For now, I haven’t been able to get much, but I plan to improve the output with the aforementioned ideas. Resolved in the section above.
Here are some arcs produced by this setup, from what I’d guess they are maybe 10cm+ at best and the tube only starts redplating at very high powers. Since this run I somehow managed to “kill” the setup, it still works but the arc doesn’t grow as much with the voltage.
I will probably try rebuilding it for it to be a bit cleaner and handle the high voltages a bit better, but for now I’m a bit clueless about what happened since the setup didn’t change at all. I am still quite happy with the results I was able to get. Here is a video of the coil running
I was also able to get some neat images of what is seemingly a neon bulb being excited in an interesting way.
Despite being filled with neon (for the most part) giving a deep orange color in the EMF of my standard SSTC, it gives off a blue glow here. This could be caused by some mixed-in argon or mercury vapor (no droplets seem to be forming inside the glass).
Here’s a picture with the tube being somewhat spaced from the “antenna”:
And here it is placed right next to the “antenna” where you can see some of the neon orange starting to show at the contact point:
This causes lots of interference; a radio far away across the house is unusable.!
Hopefully, I can get some more arcs (over several cm) out of this circuit soon™ and post some updates here.
Sidenote
A fun way to start a project is to realize some of your components are unusable.
For me, I found out one of my MOT has died and that my microwave diode is dead as well.
This led me to make an abomination of a diode (which somehow didn’t explode yet) out of 1N4007 diodes.
As for the MOT, I just used a different one after realizing trying to save it is futile.
-
Updates ↩