First Results!

Today I detected the first electron trapping in our polywell! I started by mounting everything in the chamber. I decided to use the 3D printed plastic polywell because it can hold much stronger magnetic fields compared to the steel frame polywell.

Plastic polywell model taped to prevent arcing

Positioning everything in the chamber took about two hours. On my first try, the wires became tangled, changing the orientation of the langmuir probe. I had to take everything out and start over.

Initial attempt… Mess of wires dislodged langmuir probe
Reorganized high current feedthrough wires
Better electrical connections
Final setup: Polywell and filaments mounted in chamber

Next I pumped down the chamber and hooked up the oscilloscope. Pressures were 20 mTorr — low enough for initial electron well testing. This should be lower with the diffusion pump but I did not have time to troubleshoot leaks in the main flange. The plastic polywell will also outgas significantly at lower pressures.

The oscilloscope gave me no results at first. It was rolling on a 1 volt scale with no trigger setting. I thought I would be able to see pulses of electron well formation in real time. As I learned later, this was a mistake. Confused by the lack of results, I tried swapping electrical connections and making differential measurements in attempt to avoid grounding issues. Like last time, grounding a lead to the chamber gave noise. Interestingly, the filament supply was also connected to the same ground and induced noice on the oscilloscope when it was turned on (even when it was completely disconnected to the filaments). This shows the grounding issue detailed my last post — an oscilloscope lead cannot be grounded to the chamber because this closes a circuit with significant noise. I made sure to ground my oscilloscope independently of other power supplies for minimal noise.

Dip in voltage occurs when filament power supply is turned on and oscilloscope lead is grounded to chamber. This does not indicate the filaments are actually on, but shows noise emanating from filament supply on chamber.

In my last half hour, I used the trigger setting on the oscilloscope to track capacitor bank discharges. I connected the hot wire and a built-in 50 Ohm resistor to my capacitor bank. While capacitor bank discharges triggered Channel 1, I used Channel 2 to measure the langmuir probe potential. This was successful because it could catch the microsecond pulses of the potential well formation and capacitor bank discharge, unlike running the oscilloscope in real time.

A capacitor bank discharge gives a peak in channel 1 (top line). Here, the filaments were not turned on so no electron well forms.
A with capacitors charged to 140 volts tiggers oscilloscope at 2 volts. Channel 2 is inverted to show positive peak for electron well formation.
Best result yet! Charged capacitor bank to ~160 volts, giving coils ~19,000 amp-turns. This equated to a large spike indicating an electron well on the order of 10+ volts.

Originally, I had channel 2 inverted so it would give a positive spike with the electron well formation. When I undid the invert setting, channel 2 dropped off the map. I changed this in the last few minutes of experimenting and did not have time to look closer.

Where did Channel 2 go?

It feels great to get these results! Tomorrow I will have more time for data collection at a variety of magnetic field strengths. In theory, our polywell coils can hold up to 100kAT (kilo amp-turns), but this will likely explode the plastic and scorch the wires.

1 thought on “First Results!”

Leave a Reply

Your email address will not be published. Required fields are marked *