Replenisher electrostatic machine

A machine that can be considered the first symmetrical rotating electrostatic influence machine was described by Giuseppe Belli his book about electricity published in 1838 [60]. The basic idea of the machine appeared in a paper in 1831 [p61], with hints of it being first constructed by 1826. The same idea was rediscovered later by Varley, in 1860, and by Lord Kelvin (William Thomsom), in 1867 [p92]. Kelvin developed an electrostatic machine to generate charge for electrifiying the ink in a telegraph printer, and used a similar device based on the same idea in electrical instruments to regenerate small charges, called the "replenisher".
The schematic diagram at left shows the structure of the machine. Two curved metal plates P and Q are fixed to a rotating axle O by insulating rods. They rotate inside two larger curved metal plates A and B, also insulated. Springs a and b connect the rotating plates with the outer plates when they are at the horizontal diameter, and springs c and d short-circuit the rotating plates when they are about to leave the outer plates. if there is a charge imbalance between the outer plates, in the case with A being positive and B negative, when the rotating plates are as shown opposite charges are attracted to them. When they continue to rotate and touch the springs a and b, most of the charges in them are transferred to the outer plates by a Faraday ice pail effect, reinforcing the original charges. The cycle then repeats with opposite polarities in P and Q. If the capacitances between the inner plates and outer plates is dominant, each half cycle almost doubles the charges in A and B. In the usual application, one of the plates is connected to an insulated conductor where the voltage is to be kept at certain level. The device is turned manually to restore the charge when the voltage drops due to leakage. The other plate can be left floating, with the conductor cd grounded, or can be grounded, with the conductor cd left floating.

An interesting flat version of the device, at the University of Innsbruck Museum. Another machine, in a demonstration at the Deutsches Museum (from this site).

I build a crude version of this machine by 1973, the first electrostatic machine that I built, that worked quite well but was very small. Im January 2011 I built a larger machine, shown below.

The machine, crank side.

The four plates are made of brass rectangular plates with rounded corners, curved to cylindric sections. The insulating structure is made of high density polyethylene (cutting board). I modified the original structure mounting the springs c and d at the inner sides of the rotating plates, touching fixed contacts. In this way the distance between the inner and outer plates can be constant, maximizing the capacitance between them and the charge generated at each cycle. The distance between the moving and fixed plates can be adjusted by inserting more or less the insulating rods that hold the moving plates in the central block, where screws fix them. The springs are made of nickel-chrome wire, fixed to the plates with adhesive aluminum foil. The wires are folded in loops to avoid sharp exposed points. A crank moves the plates. Two posts with ball electroscopes are mounted connected to the outer plates.Plastic beads terminate the rods to avoid points. Everything is fixed with screws over a laminated MDF base. The machine self-excites easily and never reverts polarity when working. It charges to about ±15 kV before internals sparking limits the voltage, since it has no spark shields to prevent this. This machine is not a powerful generator as built, but is useful in experiments with electrostatics in its original function as a charge replenisher. It is actually being used as a high-voltage bias generator in a certain application. Another view.

The "Replenisher" electrostatic machine