Electrostatic device: Kelvin's Thunderstorm, 'Inline' version

AN IN-LINE WATERDROP ELECTROSTATIC GENERATOR
(c)1996 William Beaty

Also see: LORD KELVIN'S WATER-DROP GENERATOR

Electricity and water do not mix? Wrong! This device creates electrical energy by slowing down water droplets as they fall. It separates the positive and negative charges naturally present in all substances. It gives a high voltage DC output at a current of microamperes. The water exiting the bottom is uncharged, and you could easily feed it to another complete unit stacked in series.

The diagram below depicts a version built from funnels and aluminum jello molds or bundt pans (the ring-shaped kind with the large opening in the center.) However, you could build an enormous version if you were to replace the jello molds with large wooden barrels or 55gal drums with their bottoms cut out. ** bzZAP! **

This thing looks like some sort of particle accelerator. Actually, it is just the opposite. It decelerates charged particles and provides electrical energy, rather than using up electrical energy in order to accelerate particles.

WARNING: Note that this project is an advanced version of the classic Kelvin Thunderstorm device outlined in amasci.com/emotor/kelvin.html. If you're a beginner, then to get the hang of these devices...

...I STRONGLY SUGGEST THAT YOU BUILD THE OTHER, SIMPLER ONE FIRST.

Also, the other article gives hints on making the generator blink a little neon bulb, or make clicks in an AM radio, deflect tissue strips, etc.

                                 Optional rain collector?
   |____                _____|
        \______________/
             \    /
              \  /        Metal              | SIDE VIEW     |
               ||         Funnel             |               |
               ||                            | BILL B's      |
                                             | WATERDROP     |
           __________     Jello Mold         | ELECTROSTATIC |
          / \      / \   (positive inducer)  | GENERATOR     |
         |___|____|___|                      | (Connecting   |
                                             | wires not     |
           __________                        | shown)        |
          / \....../ \    Jello Mold with
         |___|____|___|   metal screen (negative
                          charge collector)
             ______
             \    /
              \  /       Metal 
               ||        Funnel
               ||
           __________
          / \      / \    Jello Mold
         |___|____|___|  (negative inducer)

           __________
          / \....../ \    Jello Mold with
         |___|____|___|   metal screen
                         (positive collector)



        \               /  Bowl to catch
         \             /   water
          \___________/

The diagrams to the left and right show an 'indoors' version. I used plastic Tupperware bowls as the water sources. I drilled a circle of tiny holes in the bottom of each, using a #64 drill bit. This creates a "shower head" effect and greatly increases the output current. Ideally you should connect the water in the two bowls to ground, but since they each supply opposite charge, you can just connect them to each other. Drop the metal end of an alligator cliplead into each.

Note: actual polarity of charges is spontaneously chosen at start-up. The above "positive" and "negative" labels are for convenience, and will end up backwards half the time.

PARTS:

4 - Metal bundt pans or jello molds ( $0.50 each at garage sales!)
2 - Metal funnels, or plastic bowls w/holes poked in the bottom
2 - Cones made from metal window screen
3 - lengths of thin plastic tubing or fishing line, about 3ft long, used to hang the parts of the device from the ceiling.
misc - Tape or glue for the screen and the fish line, alligator clipleads to connect the pans together, old scissors to cut the metal screen.

THE SCREENS

You should shape the metal screens in the collectors to make them either bowl-shaped or funnel-shaped. This will force the water to fall from the center of the screen rather than running down the edge of the charge collector pan. If water drips from the edge of the charge collector rather than from the screen, it will remove charge from the collector. And if the water from the screen forms a long solid stream whose tip extends vertically past the bottom edge of the pan, it will discharge the collector pan and the device will stop working. Only droplets, not streams, should leave the collector ring's central hole.

If you use a simple, pointed, cone-shaped screen, and if all the water falls from the tip in a solid stream, the collector ring discharges itself and the device stops working. I've found that a good shape for the screen is a shallow cone with the tip punched inside-out like this (cross section):

                          _
               \         / \         /
                 \     /     \     /
                   \_/         \_/

The water falls as droplets from the bottom edge. Make the screen cones fairly shallow, so no parts of them extend vertically outside the central hole of the ring pan. The pan acts as an electrostatic shield, and when water touches the inner, shielded part of the pan, *all* of its excess charge will travel to the metal. If the screen is within the sheilded volume, it will extract all the excess charge from each water drop. The screen must remain down inside the donut-hole of the pan.


     _____                           _____
    /     \                         /     \
   |       |           _           |       |   Bundt pan with
   |       |\         / \         /|       |   cone of screen
   |       |  \     /     \     /  |       |   within its hole.
  |         |   \_/         \_/   |         |
  |         |                     |         |
  |- - - - -|                     |- - - - -|

I used black electrical tape on the edges of the screen to hold it in place in the pans. You could tack it with silicone caulk for a more permanent design. Don't forget, the metal screens must make electrical connection to the metal pans. Keep each screen and pan touching together in at least one spot; don't accidentally break their connection with insulating silicone caulk.

CONSTRUCTION:

The entire assembly of bundt pans and funnels can be suspended by fishlines from a hook on the ceiling, or instead you could build some sort of plastic structure to hold the parts.

The upper rain collector can be plastic or metal, or possibly use the downspout from a convenient roof, or a hose, etc. The water supply should be small enough that the funnels don't fill to overflowing.

Whether you use a funnel or a bowl with small holes, you must somehow limit the amount of water flow so that it either comes from the bowl or funnel as single drops, or as a short stream that breaks into drops inside the inducer ring. If the stream is too long and the drops break loose below the inducer ring, the inducer ring won't charge the drops and the device won't work

You can increase the electric output current by fitting some sort of metal "shower head" assemblies onto the ends of the metal funnels. These are sold in garden stores as adapters for watering-can spouts.

If metal funnels are too expensive, you can also try using Tupperware bowls with numerous small holes drilled in their centers. I put six holes in mine, spaced in a circle about 2cm apart to prevent the water streams from sticking together. To connect the plastic bowls to ground, stick a piece of aluminum foil in the water in the bowl and clip your alligator clipleads to the foil. "Ground" can be a water faucet, the screw on an electric outlet, a big sheet of aluminum foil laid on the floor, or even your body if the humidity is high.

The rings and funnels should be one or two inches apart. It's OK to let the end of each funnel extend within the hole of each inducer ring below, as long as the droplets break free of the stream before the stream exits bottom edge of the inducer ring.

To suspend the rings and bowls, I glued three small Plexiglas "ears" on all the rings and funnels. Then I ran three lengths of #20 plastic insulation "spagetti" tubing up the sides, tying it to each "ear," then tying the whole thing to a hook in the ceiling. I could have used fishing line instead. Don't use wire to support the rings, not even if it's insulated wire. The support must be a very good insulator. To speed things up you might try simply tying knots in the right places in the fishline, then glue the knots to the pans with gobs of silicone caulk. (Use tape to hold the lines in place while the caulk sets.) Hang the whole thing from a ceiling hook.

ELECTRICAL CONNECTIONS:

             ______
             \    /  Grounded
         _____\  /   Funnel
        |      ||    w/shower head
        V      ||
               /\
           __________
       __ / \      / \ _________________o
      |  |___|____|___|
      |                                    Output leads
      |                       __________o
      |    __________        |
      |   / \....../ \ ______|
      |  |___|____|___|      |
      |                      |
      |                      |
      |      ______          |
      |      \    / Grounded |
      |   ____\  /  Funnel   |  Connect the bundt pans
      |  |     ||            |  together using bare wires
      |  V     ||            |  and tape, or using alligator
      |        /\            |  clipleads found at Radio
      |                      |  Shack stores
      |    __________        |
      |   / \      / \ ______|
      |  |___|____|___|
      |
      |    __________
      |__ / \....../ \
         |___|____|___|

This device seems to work even in extremely high humidity. Apparently the electrical leakage along the surfaces of the thin plastic fishline is small enough that the device can still work even with that leakage present. Thinner fishing line might work even better, since the thinner the support lines, the less their surface area. The less their surface area, the less the electrical leakage during high humidity.

Any liquid can be used as long as it is slightly conductive. In theory oil will not work because it's an insulator. Too bad liquid mercury is poisonous, mercury would work fine. Tapwater works OK, since impurities make it conductive enough. I doubt that salt water would work better, but it would be something to experiment with. Ultra-pure de-ionized water might not work because it's almost an insulator. For a science fair project you could even try fruit juice. When people ask you why, just say "because acids create mobile ions, leading to higher conductivity, and also because I wanted to make a big, sticky, disgusting mess!" :)

DEBUGGING

Did you build the "classic" Kelvin generator first? If not, then I can't help you. This "in-line" version is an advanced project. It's much more flakey and difficult to make operate. Beginners should go back and try the "four coffee cans" version first.

For further info, debugging, references, and "how it works," see my other article,

Lord Kelvin's Waterdrop Generator
http://amasci.com/emotor/kelvin.html

LIKE THIS ARTICLE? WHY NOT
SUPPORT AMASCI.COM

http://amasci.com/emotor/ikelv.html
Created and maintained by Bill Beaty. Mail me at: