Build a Working Fusion Reactor Model: the Real Life Arc Reactor




About: I'm an enginnering student who loves to build stuff!

I've always been fascinated by fusion: cool looking devices that hold the promise of one day providing green power to each one of us, while glowing like the one in Iron Man's suit. So, I started building models of them, precisely of the Fansworth IEC (Inertial Confinement Fusion) kind.

The fancy name just means that nuclei of atoms are accelerated and them collide with each other. If they're fast enough, the magic happens and we get helium and energy! The energy output in this design is lower than the amount required to make fusion happen, but it could be used as a neutron source and is the easiest one to build.

This model is different because despite not being actually fusing anything, it could do so by supplying deuterium gas. Besides, it was built with low cost materials, no expensive stainless steel chambers here. The most expensive part is the vacuum pump. Now, enough talking, let's get our hands dirty!


Step 1: Materials

First, a warning: this project involves high voltage, in the order of tens kV, so be careful. You could get a big shock or be killed in case of heart problems, I'll take no responsibility.

But if you are brave and know what you're doing, let's proceed. You will need:

  • About 40cm of thin wire with a high melting point, I used 0.5mm tungsten from ebay
  • 3/8" brass hose barb with male thread (like the one in the picture)
  • Olives glass jar (choose one with thick glass, otherwise it could implode under vacuum)
  • An O'ring with the same diameter of the glass jar opening (about 6cm)
  • Some 3/8" rubber tubing to connect the vaccum pump to the jar
  • About 5cm of 3/8" copper tubing
  • At least 10cm of HV wire (I got mine from old TV flybacks)
  • A TV flyback transformer (prefer DC type, as DC current is used for real fusion, but AC will produce the same visual effect, actually it may even glow more!)
  • A ZVS driver (you could buy or make one). There are even kits with both zvs and flyback online.
  • Silicone adhesive to seal any leaks (not shown here)
  • Some ordinary power cord wire
  • Some scrap metal to make the electrodes. I'll be using a piece of thin aluminum and two 5.6cm old speakers
  • A glass marble or test tube rubber stopper (both up to 2cm diam.) to make the central grid (this is not mandatory, but will help when coiling the wire). I used a 1.3cm diameter rubber stopper.
  • A vaccum pump (I'm using one for refrigerator service)


  • A drill (like a Dremel) with bits for glass and metal
  • Ruler, scissor and marker
  • Hammer and plier

Step 2: Prepare the Lid Electrode

Using a marker, mark the shape of the speaker on the piece of scrap metal (here, I'm using aluminum), then cut it.

Remove the magnet, coil and everything else but the armature of the speaker, then put the metal circle on top of it, you may glue it with silicone, super glue or whatever you need, just don't cover the top.

Now, grab the copper tube and put it on the speaker hole. If it doesn't fit, dont'worry, smash it a little with a plier and once it's inside, smash it a little again.

Step 3: Prepare the Bottom Electrode

First, mark and cut a circular piece of metal just like before. But before putting it on the speaker, go get the regular copper wire and tie it to the speaker just like shown in the pictures. After tying, you may put the metal.

Step 4: Bottom Electrode and Top O'ring

Now, with your drill and the proper glass drill bit, drill a hole in the middle of the jar bottom, test the size to match the wire's, it souldnt't be too loose.

Take the bottom electrode and pass the wire through the hole, just like the picture (there's an older electrode there, but it's the same idea). Seal it with silicone adhesive and wait, it'll take some time to cure.

While you wait, grab the o'ring and put it on top of the jar, the seal it with silicone too,there can be no leaks!

Leave the jar resting, note that some adhesives may require several hours begin to cure.

Step 5: The Almighty Grid!

Using the rubber stopper, glass marble or some cilinder you have lying around, make 3 circles with the tungsten wire. Use a plier to bend the wire and polish out any sharp parts. For one of them, you may leave some extra wire to attach to the HV cable.

With all three circles, after attaching the HV cable to one of them, arrange the other two so it takes the shape of a hollow sphere. Avoid using solder as it can melt during operation. Try to do it like you see in the pictures, and isolate any bare part of the HV cable with tape and silicone, only the grid should be exposed.

When the grid is ready, drill a hole in the middle of the jar side wall and put it from the inside, just like we did for the bottom electrode. The grid should stay close to the middle of the jar (like in the o'ring picture). After these steps, apply silicone to seal any leak.

Step 6: The Lid and Final Assembly

Almost ready! Use your drill to make a hole in the lid and screw the hose coupler into it. Then, screw the copper tube electrode on the coupler thread. The electrode end should match the jar wall mark (if your jar has one) or be at the same distance from the grid as the bottom electrode.

Put the lid on the jar, it shoud fit tightly, if that happened, that's it, our little monster is finished!

Step 7: Machine Food: the Power Supply

The power supply is also very important. Whether you choose to buy the zvs kit or build your own,
keep calm and follow the images. The ZVS input voltage should be at least 12V and is rated to a 36V maximum. However, I like to take some risk and run my homemade version on about 50V! You may try it too, I just can't guarantee it will work forever (or work at all).

Step 8: Have Fun!

Ladies and gentleman, we may now officialy start our reactor!

Connect the supply ground to the grid and the positive terminal to the two side electrodes.

Turn on the vaccum pump for a few seconds, then turn on the power supply. As the vacuum increases, the plasma between the electrodes will become a glowing ball in the grid, just like a star in a jar.

This beautiful plasma ball is the region within fusion happens in actual reactors, and even in our model it is extremely hot. The acceleration and deceleration of charged particles in the plasma also emits radiation in a wide spectrum that ranges from infrared (heat) to X-rays, and even though we just see visible light, you should keep exposition time low just to be safe.

To finish, I'd like to share this site called where you will find lots of information regarding amateur fusion.



    • Arduino Contest 2019

      Arduino Contest 2019
    • Fandom Contest

      Fandom Contest
    • Colors of the Rainbow Contest

      Colors of the Rainbow Contest

    16 Discussions


    3 months ago

    ZVS driver are nice to making any high freq and high voltages circuit...thanks...


    Question 9 months ago

    How much does this cost on avg?


    Reply 2 years ago

    Hi, sorry for the delay, but thanks for the compliment!


    2 years ago

    AH! I thought I recognized the application.. One minor caviet, the plasma it generates, is also putting off lots of X-Ray bandwidth. Only operate this short-term, and reduce exposure. Also, be extra careful, with the jar and the vacuum pump. the source of what I recognize this from, was an old issue of MAKE: Magazine, called 'A Nuclear Fusor', where it also warns of the glass imploding from the vacuum, and the X-Ray exposure. Still, Great job!

    3 replies

    Reply 2 years ago

    Hi! That's a great article on how to build this fusor. I built my version
    trying to keep it as simple and low cost as possible. Glass implosion
    is indeed a problem, and I broke a few in my first tries using flat
    glass and a cooking pan :p This olive jar version never imploded, but it
    is a thing to keep in mind. I don't have a way to detect x-ray, but
    there's probably some being generated at higher input voltages. I've updated the instructables with your advices ;)


    Reply 2 years ago

    if you had some phosphors and shielded it from light and other less powerful electromagnetic waves you might be able to create a x ray detector. By the way you should try manipulating the plasma with magnets.


    Reply 2 years ago

    Good ideas! I'll try to make a radiation detector from an old CRT monitor. It should work provided that the phosphor don't degrade under air.


    2 years ago

    Great idea for a science class too, which may inspire some kid out there to beat the big guys at solving our energy problems:) You never know! Cool project and I enjoyed the lesson too:)

    1 reply

    Reply 2 years ago

    Thank you! It would be great if that happened haha

    yash yash

    2 years ago

    this is not a fusion reactor like this but a cathode ray tube as fusion reactor needs a lot of pressure and its threshold energy is too high which can only provided by nuclear fission ie atom bomb!!!


    Reply 2 years ago

    Hey, it's mainly for learning and fun! You can think of this as a proper introduction to the fusion world. Just like we have "Hello world" for programming languages.


    2 years ago

    Thanks!! It makes me very happy to see that you both liked it.

    Anirudh Ralhan

    2 years ago

    That's SOOOOOOO AWESOME and relatively easy as well. Absolutely Love it. Favourited and Voted.


    2 years ago

    That's neat!