Power Supply Upgrade

17 07 2012

All photos(1/2)
All photos

During the first trial, our langmuir probe told us that our electron beam intensity was fluctuating at 60 Hz.

This is a problem, because one of the main things we are trying to study is the way changes in beam intensity affect potential well depth, so we want a steady intensity. The frequency of the fluctuations suggest that the AC-powered hot cathode is to blame.  I don’t totally understand the details of how a hot cathode running on AC 120v 60Hz translates to this waveform:

data from the langmuir probe displayed on the oscilloscop

The important thing is to prevent it. To do that, I put a full wave bridge rectifier in the power supply. It converts the AC coming from the wall to DC


It has three essential components.

1) The bridge rectifier


This change the AC sine wave into a waveform expressed by the function abs(sin(x)):

Better, but still not steady DC.

2) The filter capacitor


This gets rid of the ripple. you could compare the capacitor to a bucket with a hole in the bottom. Even if I vary the rate at which putting water into the bucket, the rate at which it come out is always going to be more or less the same, provided that it is sufficiently large compared to the volume of water going in.

However, its impossible to get an absolutely perfect DC output with this setup, because the ammount of charge on the capacitor does affect the voltage at which the current comes out.

This 680 uF capacitor takes away enough of the ripple for our purposes:

the output of the power supply when hooked up to a light bulb

3) An isolation transformer


Usually, the diode bridge and the capacitor would be enough, but our AC isn’t coming from the wall, its coming from a grounded auto transformer. this is a problem because the rectifier only works if the AC input is floating. A transformer with an equal number of primary and secondary wingdings accomplishes this without stepping the voltage up or down.

Nest step is to test it in the chamber.

Domenick Bauer

Electron Gun + Polywell Results

8 07 2012

All Photos

Today, we tested the electron gun in tandem with the Polywell core. Right now, our goal is to simply understand effect the e-gun has on the potential well, if it has any effect at all.

We had the pump running since 3AM the night before, and so by about 2PM today, we had a vacuum in the 5 x 10^-5 range.

Not super amazing, but certainly good enough for out purposes.

First, we mounted the power supplies in the rack.

Hooked them up to their proper feed-through pins.

Attached the langmuir probe and the shunt resistor to to the oscilloscope, allowing us to monitor the potential well depth and current going through the coils simultaneously and in real time.

Then we began testing. The coils in the core worked great.

The spike on the top line is the power supply’s capacitor bank discharging. The smaller spike on the lower line is a current induced in the langmuir probe by the sudden appearence and dissappearence of a magnetic field generated by the coils in the core. So the Polywell works.

Then we tried the electron gun, and it didn’t work at all. the filament was glowing, and we were getting high voltage on the accelerator’s feedthrough pins, but no reading on the langmuir probe.

After extensive thinking, speculating, and white-board writing, we decided we had to open the chamber up to see if the connections to the accelerator were right.

Here’s what we found

That’s your problem right there, ma’am

It’s a little hard to see, but a gray plastic piece which connects the core to the feed-through pins was right in front of the accelerator, totally blocking the beam.

Funny how sometimes the causes of problems are so obvious that you don’t even think of them.

Anyway, I got that fixed, tried it again, and got a beam. The readings on the langmuir probe attached to the multimeter were more or less the same as those from the last test. When we attached it to the oscilloscope however, things were a little more complicated.

The beam intensity was not static, but periodic. It fluctuated with a frequency of 60 Hz, pointing to the AC current which powers the hot cathode.

We expected something like this, because the availability of electrons to accelerate fluctuates with the AC powering the hot cathode.

Ideally, our electron beam would be have a perfectly even intensity, because then we could eliminate it as a variable.

Fixing that would involve rectifying the AC, a major upgrade to the power supply, so we decided to leave that for another day and run the experiment.

Here are some of the results

A really good one. The downward spike on the lower line signifies a a potential well. Nice!

Here we see a well, but it’s at the wrong time, it seems to have appeared just after we pulsed the core.

Here’s a strange one in which the Polywell core pulse seems to cause some change in the voltage on the langmuir, but not a well.

This one really demonstrates why the periodic electron beam is such a problem. The top spike came at a moment when the langmuir probe was reading zero. This means that there was no beam when we pulsed the core, so it’s no surprise that we just got an induced current.

The most tantalizing and baffling run we did. The well appears to be extremely deep, greater than 100V, but it’s unclear whether thats credible, because there’s so much other confusing stuff going on.

All in all, our results are little confusing, but good. We were able to create the well, which is a big win, but we weren’t able to do so consistently. In order to really study how the well is affected by tweaking variables, we need consistent baseline well to compare against.

Domenick Bauer

Electron Gun Success

24 06 2012

All photos

Today we tested the electron gun in the chamber, and we detected a negative potential on the Langmuir probe, which means it worked!

Negative nine volts on the Langmuir probe

Fuck yeah!

Here’s what we did

1) Added a faster-acting fuse to the power supply

the new 4A fuse is underneath the black shrink-wrap

We already have a .5 amp fuse to protect the light bulb filament, but this new one  takes less time to actually blow once its current rating is surpassed, so if the cathode arcs to the chamber wall and pulls a large current, this fuse will blow quickly, preventing damage to the chamber.

wide shot of the setup

2) Closed the electron gun assembly in the chamber, connected to feed throughs, and set up the Langmuir probe.

The Langmuir probe is a wire with one end in the path of the electron beam, and the other attached to a multimeter set to volts DC

3) Powered up the vacuum system.

Because there was so much stuff in the chamber, there was also (presumably) a lot of trapped air which leaked out slowly as we pumped down, so the vacuum wasn’t super deep, but it was deep enough for our purposes.

3) Powered up the e-gun.

The cathode immediately started to glow, amd as we turned up the voltage across the cathode, the Langmuir probe started to register a negative potential.

We could not get potential on the Langmuir probe unless we powered up both the cathode and the accelerator, so we concluded that it must be the result of a beam.

There were also a couple of other interesting things we noticed.

Changes in the voltage of the accelerator did not seem to affect the beam intensity. We brought the potential on the accelerator from +500 down to ~+250, and got similar readings on the Langmuir probe.

Changes in the voltage (and current) to the cathode do affect beam intensity. We found that the greatest value we could get on the probe was about -12 volts, using about 90 to 100 volts AC across the cathode. As we kept increasing the cathode voltage/current beyond that, the Langmuir probe started heading towards zero, until the fuse blew.

After this, the Langmuir voltage started to head toward zero.

A little hard to see, that’a 10.59 volts on the Langmuir, and 102.5 volts on the cathode.

We don’t know what is causing this.

Another cool thing we noticed was the effect the electron gun had on the vacuum. Leaving the beam at maximum intensity caused the vacuum meter to show increased pressure. We were literally filling vacuum space with electrons.

Weird to see the the materiality of electrons demonstrated in such a concrete way.

But all that aside, this is a big step for us. From here, getting that electron beam shining into the center of the Polywell shouldn’t be too hard. If we succeed in that and document our results, we will have performed real, original research on the Polywell design. If we can get the potential well deep enough, maybe even do Polywell fusion.

So let me reiterate, FUCK YEAH

Domenick Bauer

Coil Power Supply Upgrade

26 09 2011

All photos.

I completed some nice upgrades to the coils power supply for safety and quality.

I added DIN rail terminal blocks and rearranged the parts to emphasize discharge path:

Added an internal variac:

Added terminal blocks to the trigger board:

These pins on the SCR never soldered well because they are mini-fastons. Fixed:

For safety, I added an AC switch:

I replaced the trigger’s battery with a 9V wall wart:

And added a DIN rail 0.5 A circuit breaker:

CRT in Focus

19 09 2011

All photos.

I finished the CRT power supply and got the CRT running with focus and grid:

As you can see we have an interference pattern. To the eye the beam looks much brighter in the center:

With the focus adjustment I can broaden the beam:

The CRT power supply:


Arduino Controls 30,000 Volts

10 09 2011

All photos.

Today I made arduino control 30,000 volts.

My arduino has 3 channels of analog output 0 to 5 volt.

For testing I used this sin wave generator sketch:



int pwmPin = 9; // output pin supporting PWM

void setup(){

pinMode(pwmPin, OUTPUT); // sets the pin as output


void loop(){

float something = millis() / 1000.0;

int value = 127.5 + 127.5 * sin( something * 2.0 * PI );




This generates a lazy 2 Hz sin wave.

But the output is not really analog, it’s pulse width modulation(PWM):

This tutorial shows how to smooth out  PWM using a low pass filter. My low pass filter used 6kΩ resistor and 4.7 µF @ 45V capacitor.

Here we have the raw PWM output superimposed with the filtered output:

Looks good!

Now we just add the voltage doubling op-amp circuit I made previously, and BOOM:

This shows the source signal and the voltage doubled signal.

Sweet! Now we can control the 30,000 volt glassman power supply.

Here the arduino is sending a slow sin wave to the glassman’s voltage control:

From 2011-09-10

The Glassman’s slew rate is really slow without a load.

Here is the setup:

Terrifying Power

4 09 2011

All photos.

Tonight I really experienced the power of the coil power supply. Whoa.

I’ve been working to increase the coil current from ~1.2kA to ~2.5kA.

Previously I discovered the coil discharge path had more DC resistance than expected.

I rewound the Polywell coils with 16 gauge wire (previously 18 gauge).

The 16 gauge DC resistance is 144 mΩ compared to 227 mΩ for 18 gauge wire.

I beefed up other wires on the coil discharge path (4 gauge):

Lets test the wiring with the dummy coil:

I took the power supply up to 100V… a small test charge…

When I fired, the noise from the coil made me flinch. It was never that loud before.

Lets turn up the power!

300V for second test.

When I fired the coil there was lightning! HOLY CRAP.  Look what happened:

The coil fucking wrapped itself around the transformer (electromagnetic forming). Then it discharged to ground:

So I haven’t measured it yet, but I think we are getting more current to the coils.

Getting Current to the Coils

24 08 2011

All photos.

I am currently repeating an experiment performed by Joe Khachan and Matthew Carr in Sydney, Australia.

Their experiment is written up here:  “The dependence of the virtual cathode in a PolywellTM on the coil current and background gas pressure” ($1.99 pay wall)

Joe and Matt were able to delivery 2.5kA to the coil:

 The coils were driven by a pulsed current power supply that consisted of a 7.5 mF capacitor bank, which could be charged to a maximum voltage of 450 V….A maximum peak current of 2.5 kA was achieved.

We are seeing an effective resistance of ~0.45Ω compared to their 0.18Ω.

We need to lower the resistance and increase the voltage.

I started with raising the voltage. I rewired the coil power supply to use 2 capacitors in series: 0.3mF, 900V

The power supply’s transformer and rectifier only go to 600V (but I pushed them to 800V without issue)

With 800V we get 1300A. More current, but effective resistance increases to 0.61Ω.

OK fine. Lets try lowering the resistance with a dummy coil directly connected to the power supply. 45 turns 6cm diameter. The Polywell coil is same size but 15 more turns.

Here it is connected:

Now we are clearing 2.5kA with 600V!  But look at the strange pulse shape. Hmm.

I thought having the coil so close to the power supply might be a confounding factor. I added 1M of 12 gauge stranded wire to distance the coil:

Revealingly, just adding that 1 meter of wire reduced the current by almost half for same 600V:

So clearly delivering current will be a design challenge.

A note on technique. Based on comments I now ensure probes are perpendicular  to current:

Coil Power Supply: Current Measurements

22 08 2011

All photos.

Today I took a step back and measured the current going through the coils.

Previously I measured 1,200 amps going through the coils with the capacitors charged to 450V.

My setup is a little different now: There are 2 meters of cabling + feedthrough between the coils and their power supply.

For all of these shots, the capacitors were charged to 450V. The shunt resistor shows 100mV across for 100A through. Multiply the voltage by 1000 to get the current in amps.

Surprisingly I’m seeing significant variation of current for identical conditions.

The most current I saw was 1,095 amps:

But with the same conditions I saw this much lower 344 amp current:

The median current reading was around 734 amps:

Disconnected, the coil shows 0.8 ohm of resistance.

I’m rather surprised by this variation. What could be the cause?

I tried some other conditions.

With the capacitors charged to 200V, I got 560 amps of current:

I also tried charging 5 out of 10 capacitors to 450V. With an average current of 500 amps.

Mass Flow Controller Upgrade

18 07 2011

All photos.

Today I upgraded the Mass Flow Controller (MFC):

Previously I used a 9V battery + a potentiometer as a quick hack to control the MFC.

For the sake of fewer parts I wanted to remove the battery from the circuit. The power supply for the MFC provides +15V, which is way over the 0 to +5V I need to control the MFC.

To prevent over voltage I used a TVS diode which shorts to ground above +6V.

It seems to all work! Don’t have to worry about a dead battery now.



The hard part is the craftsmanship needed so it won’t short out. Work in progress:



Get every new post delivered to your Inbox.

Join 630 other followers

%d bloggers like this: