The Production of Voltage Potentials in Structures:
Embodying Variable Capacitance

by Nicholas A. Reiter

24 June 1996

In May and June of 1996, my associate, Dr. S.P. Faile, and I had been examining spark discharge phenomena from energized large non-inductive plastic and paper capacitors. During the course of routine testing on one of these capacitors, a new, interesting and startling observation was made. We observed that when the device was completely discharged, and then connected to a voltmeter, a potential of over 1 volt DC was produced when the capacitor was flexed or compressed. This production of voltage appears to be linked to the shifting of the capacitive value of the device. Considerable testing has shown that the effect is consistent and is not due to an easily recognized artifact.

First discovery of this effect occured on 9 June 1996. The piece under test was the model 2468 capacitor, built by Dr. Faile from plastic (polyethylene) sheet and aluminum foil plates. In a mechanically non-stressed condition, the value of C for this device was read by an LCR meter at .780 uF. When compressed by a hand push, the alleviation of air gaps and folds caused this value to rise to about .90 uF. An earth grounded Fluke 79 meter was connected to the capacitor, and the leads of the device were shorted for 1 minute. Upon un-shorting, the voltage read across the capacitor rose from 0 to about 40 millivolts in the course of 3 minutes. Such spontaneous low level charging of large value capacitors has, of course, been observed and speculated upon for years. However, while shifting leads, I brushed my hand across the top of the capacitor (lying on a wooden table) and noticed a small jump in voltage on the Fluke meter. By applying a strong push by hand to the body of the capacitor, a voltage of over 500 mV was produced. Later, potentials of over 1 volt were produced by decreasing the impulse period (pushing faster).

The produced voltage appears to rise over the course of the mechanical impulse only. When a steady pressure is achieved, the potential begins to bleed off through the 10 meg-ohm resistance of the meter, and thus drops. If the pressure is taken away, the plastic and metal layers relax, and voltage will rapidly drop back to zero, then actually swing in a negative direction before finally bleeding back to zero again.

In this respect, the voltage production within the capacitors mimics the actions of a piezoelectric transducer. However, it seems unlikely that materials used in the construction of these devices; polyethylene, paper, and commercial aluminum foil, posses any piezoelectric characteristics. To date, three large homemade capacitors have been provided by Dr. Faile for testing. All have exhibited the production of voltage potentials:

Device	Materials Used	C (passive)	C (max compression)	Cm-Cp	V-Max
2443	paper * - Al	.860 uF	1.00 uF	.14 uF	1.2 V
2468	poly - Al	.780 uF	.90 uF	.12 uF	1.0 V
2472	poly - Al	.160 uF	.20 uF	.04 uF	.32 V
* plastic coated meat wrapping paper

A number of experiments have been performed for the purpose of defining the voltage production mechanism. We also seek to assure ourselves that the effect is not due to artifact. A short summary is provided here, of these tests.

1. The 2468 capacitor was wrapped in a sheath of aluminum foil which was earth grounded. Voltage was measured with the Fluke 79, using about a 10 lb. push over 1 second. We read slightly over 1 volt on each of ten trials. Thus, we would conclude that simple electrostatic shielding from the environment does not attenuate the effect to a noticeable degree.

2. The 2468 capacitor was warmed in an oven to about 55C. Voltage produced was NOT noticeably different in magnitude from room temperature trials.

3. A 1 meter wooden beam was used to apply pressure to the 2443 capacitor, placed on a wooden table. Over 1.2V was produced. Thus, we would conclude that the effect is not due to the transference of charges from the human experimenter pushing on the device.

4. Orientation of the capacitor under test does not seem to affect the output voltage.

5. A strong magnet was placed in various locations near the 2443 capacitor while under test. No change in output voltage magnitudes was apparent. From this, we may conclude that the effect is not due to induction, by flexing of the foil plates in the geomagnetic field.

6. Artifact of location seems to be ruled out. The 2443 and 2468 capacitors have both been tested in my house, outdoors on my property (away from power mains), and at an industrial location in Toledo. Readings made at all three locations have been comparable.

7. Artifact of instrumentation seems to be ruled out. I have observed the effect with both grounded and un-grounded meters (Fluke 79, Fluke 77, Kiethley 175) and an oscilloscope (Sencore SC61). Recently, Dr. Faile has informed me that he has confirmed the effect with another home built capacitor, similar to the 2468. A small multi-meter was used for voltage measurement, and about .5V was observed.

8. Current measurements were made on the 2443 using the Kiethley 175. A short period strong push produced a peak current of about 10uA.

If the production of voltages from flexed capacitors is indeed linked to a fundamental quality such as a rapid shift in C, then we should observe the phenomenon with other geometries. The most available device to accomplish this would probably be an air variable capacitor from an RF tuning circuit!

On 15 June, 1996, this hypothesis was confirmed. A 1000 pF (.001 uF) air variable capacitor was procured and tested with the earth grounded Fluke 79. When given a quick spin by hand, the device was observed to produce about 2mV during the period of increasing capacitance (plates meshing). When spun from a full C position to the position of plates completely un-meshed, a similar value was observed in the opposite polarity. Later, this capacitor was mounted in an aluminum box, and coupled to a gear reduced DC motor mounted external to the box. When powered at 12 V, the motor spins the capacitor rotor at 100 rpm.

When the output of the spinning capacitor was connected to an oscilloscope, a low frequency AC signal of about 10mV p-p was noted. When the speed of the drive motor was changed, the amplitude of the output wave changed proportionally.

On 23 June 1996, another arrangement was tried at the suggestion of Dr. Faile. Two non-stressed capacitors were placed such that one could be rapidly switched into a parallel connection with the other, thus instantly adding or subtracting a discreet quantity of capacitance. In this case, no spontaneous voltages were seen to appear. This experiment was performed using non-L capacitors, and also commercial electrolytic capacitors of large value. We would conclude from this that the voltage production effect is based on a MOTIONAL principle of capacitive elements.

One serious factor needed for a proper evaluation of the effect mechanism is the determination of polarity. On both the large flexible capacitors, and the spinning air variables, a voltage polarity is seen which remains quite consistent for a given device. Yet all devices tested thus far are fairly symmetrical in construction. What factor determines + and - ? This question is a primary focal point of current efforts.

We disclose, then, the following:

An effect has been observed which suggests that in the absence of external electrical influences, a device whose geometry contains a value of capacitance will spontaneously develop an electrical potential when this value is rapidly shifted. The magnitude of the voltage produced seems to be proportional to the total change in capacitance, and is inversely proportional to the period of time during which the capacitance changes. This effect has thus far remained consistent for reasonable testing and quantification using easily obtained instruments. Certain aspects of this effect remain unclear, and it is as of yet unknown if a multi-stage mechanism is responsible, or if a fundamental new physical principle is involved.

For whatever the underlying principle behind the observed effect, it seems to be the sort of energy source which is well suited for energy conversion engineering. One possibility would involve a rotating set of thin capacitor plates spun by a motor or engine at perhaps 1000 rpm. An air or vacuum gapped meshing set of rotor plates with a max C shift value of 1 uF at 1000 rpm should produce a pulsed or AC waveform of between 50 and 100 volts RMS.

Another possible application would be in the capturing of periodic wave energy by floating flexible capacitor sheets on ocean or lake surfaces.

Back to top