US Patent 7,033,406                25th April 2006                     Inventors: Richard Weir and Carl Nelson





This patent shows an electrical storage method which is reputed to power an electric car for a 500 mile trip on a charge taking only five minutes to complete.  This document is a very slightly re-worded copy of the original.  It has been pointed out by Mike Furness that while a five minute recharge is feasible, it is not practical, calling for cables with a six-inch diameter.  Also, the concept of recharging stations as suggested is also rather improbable as the electrical supply needed would rival that of a power station.  However, if the charging time were extended to night time, then it would allow substantial driving range during the day time.




An Electrical-Energy-Storage Unit (EESU) has as a basis material a high-permittivity, composition-modified barium titanate ceramic powder. This powder is double coated with the first coating being aluminium oxide and the second coating calcium magnesium aluminosilicate glass. The components of the EESU are manufactured with the use of classical ceramic fabrication techniques which include screen printing alternating multi-layers of nickel electrodes and high-permittivity composition-modified barium titanate powder, sintering to a closed-pore porous body, followed by hot-isostatic pressing to a void-free body. The components are configured into a multi-layer array with the use of a solder-bump technique as the enabling technology so as to provide a parallel configuration of components that has the capability to store electrical energy in the range of 52 kWH. The total weight of an EESU with this range of electrical energy storage is about 336 pounds.





1. Field of the Invention

This invention relates generally to energy-storage devices, and relates more particularly to high-permittivity ceramic components utilised in an array configuration for application in ultra high electrical-energy storage devices.



2. Description of the Relevant Art

The internal-combustion-engine (ICE) powered vehicles have as their electrical energy sources a generator and battery system.  This electrical system powers the vehicle accessories, which include the radio, lights, heating, and air conditioning.  The generator is driven by a belt and pulley system and some of its power is also used to recharge the battery when the ICE is in operation.  The battery initially provides the required electrical power to operate an electrical motor that is used to turn the ICE during the starting operation and the ignition system. 


The most common batteries in use today are:

Flooded lead-acid,

Sealed gel lead-acid,

Nickel-Cadmium (Ni-Cad),

Nickel Metal Hydride (NiMH), and

Nickel-Zinc (Ni-Z).  


References on the subject of electrolchemical batteries include the following:

Guardian, Inc., "Product Specification": Feb. 2, 2001;

K. A. Nishimura, "NiCd Battery", Science Electronics FAQ V1.00: Nov. 20, 1996;

Ovonics, Inc., "Product Data Sheet": no date;

Evercel, Inc., "Battery Data Sheet—Model 100": no date;

S. R. Ovshinsky et al., "Ovonics NiMH Batteries: The Enabling Technology for Heavy-Duty Electrical and Hybrid Electric Vehicles", Ovonics publication 2000-01-3108: Nov. 5, 1999;

B. Dickinson et al., "Issues and Benefits with Fast Charging Industrial Batteries", AeroVeronment, Inc. article: no date.


Each specific type of battery has characteristics, which make it either more or less desirable to use in a specific application.   Cost is always a major factor and the NiMH battery tops the list in price with the flooded lead-acid battery being the most inexpensive.  Evercel manufactures the Ni-Z battery and by a patented process, with the claim to have the highest power-per-pound ratio of any battery. See Table 1 below for comparisons among the various batteries. What is lost in the cost translation is the fact that NiMH batteries yield nearly twice the performance (energy density per weight of the battery) than do conventional lead-acid batteries.  A major drawback to the NiMH battery is the very high self-discharge rate of approximately 5% to 10% per day.  This would make the battery useless in a few weeks.  The Ni-Cad battery and the lead-acid battery also have self-discharge but it is in the range of about 1% per day and both contain hazardous materials such as acid or highly toxic cadmium.  The Ni-Z and the NiMH batteries contain potassium hydroxide and this electrolyte in moderate and high concentrations is very caustic and will cause severe burns to tissue and corrosion to many metals such as beryllium, magnesium, aluminium, zinc, and tin.


Another factor that must be considered when making a battery comparison is the recharge time.  Lead-acid batteries require a very long recharge period, as long as 6 to 8 hours.  Lead-acid batteries, because of their chemical makeup, cannot sustain high current or voltage continuously during charging.  The lead plates within the battery heat rapidly and cool very slowly.  Too much heat results in a condition known as "gassing" where hydrogen and oxygen gases are released from the battery's vent cap.  Over time, gassing reduces the effectiveness of the battery and also increases the need for battery maintenance, i.e., requiring periodic de-ionised or distilled water addition.   Batteries such as Ni-Cad and NiMH are not as susceptible to heat and can be recharged in less time, allowing for high current or voltage changes which can bring the battery from a 20% state of charge to an 80% state of charge in just 20 minutes.   The time to fully recharge these batteries can be more than an hour.   Common to all present day batteries is a finite life, and if they are fully discharged and recharged on a regular basis their life is reduced considerably.




In accordance with the illustrated preferred embodiment, the present invention provides a unique electrical-energy-storage unit that has the capability to store ultra high amounts of energy.


One aspect of the present invention is that the materials used to produce the energy-storage unit, EESU, are not explosive, corrosive, or hazardous.  The basis material, a high-permittivity calcined composition-modified barium titanate powder is an inert powder and is described in the following references: S. A. Bruno, D. K. Swanson, and I. Burn, J. Am Ceram. Soc. 76, 1233 (1993); P. Hansen, U.S. Pat. No. 6,078,494, issued Jun. 20, 2000.  The most cost-effective metal that can be used for the conduction paths is nickel.  Nickel as a metal is not hazardous and only becomes a problem if it is in solution such as in deposition of electroless nickel.  None of the EESU materials will explode when being recharged or impacted.  Thus the EESU is a safe product when used in electric vehicles, buses, bicycles, tractors, or any device that is used for transportation or to perform work.  It could also be used for storing electrical power generated from solar voltaic cells or other alternative sources for residential, commercial, or industrial applications.  The EESU will also allow power averaging of power plants utilising SPVC or wind technology and will have the capability to provide this function by storing sufficient electrical energy so that when the sun is not shinning or the wind is not blowing they can meet the energy requirements of residential, commercial, and industrial sites.


Another aspect of the present invention is that the EESU initial specifications will not degrade due to being fully discharged or recharged.  Deep cycling the EESU through the life of any commercial product that may use it will not cause the EESU specifications to be degraded.  The EESU can also be rapidly charged without damaging the material or reducing its life.  The cycle time to fully charge a 52 kWH EESU would be in the range of 4 to 6 minutes with sufficient cooling of the power cables and connections.  This and the ability of a bank of EESUs to store sufficient energy to supply 400 electric vehicles or more with a single charge will allow electrical energy stations that have the same features as the present day gasoline stations for the ICE cars.  The bank of EESUs will store the energy being delivered to it from the present day utility power grid during the night when demand is low and then deliver the energy when the demand hits a peak.  The EESU energy bank will be charging during the peak times but at a rate that is sufficient to provide a full charge of the bank over a 24-hour period or less.  This method of electrical power averaging would reduce the number of power generating stations required and the charging energy could also come from alternative sources. These electrical-energy-delivery stations will not have the hazards of the explosive gasoline.


Yet another aspect of the present invention is that the coating of aluminium oxide and calcium magnesium aluminosilicate glass on calcined composition-modified barium titanate powder provides many enhancement features and manufacturing capabilities to the basis material. These coating materials have exceptional high voltage breakdown and when coated on to the above material will increase the breakdown voltage of ceramics comprised of the coated particles from 3×106 V/cm of the uncoated basis material to around 5×106 V/cm or higher. The following reference indicates the dielectric breakdown strength in V/cm of such materials: J. Kuwata et al., "Electrical Properties of Perovskite-Type Oxide Thin-Films Prepared by RF Sputtering", Jpn. J. Appl. Phys., Part 1, 1985, 24(Suppl. 24-2, Proc. Int. Meet. Ferroelectr., 6th), 413-15. This very high voltage breakdown assists in allowing the ceramic EESU to store a large amount of energy due to the following: Stored energy E = CV2 / 2, Formula 1, as indicated in F. Sears et al., "Capacitance-Properties of Dielectrics", University Physics, Addison-Wesley Publishing Company, Inc.: Dec. 1957: pp 468-486, where C is the capacitance, V is the voltage across the EESU terminals, and E is the stored energy.  This indicates that the energy of the EESU increases with the square of the voltage. Fig.1 indicates that a double array of 2230 energy storage components 9 in a parallel configuration that contain the calcined composition-modified barium titanate powder. Fully densified ceramic components of this powder coated with 100 Angstrom units  of aluminium oxide as the first coating 8 and a 100 Angstrom units of calcium magnesium aluminosilicate glass as the second coating 8 can be safely charged to 3500 V.  The number of components used in the double array depends on the electrical energy storage requirements of the application. The components used in the array can vary from 2 to 10,000 or more.  The total capacitance of this particular array 9 is 31 F which will allow 52,220 W·h of energy to be stored as derived by Formula 1.


These coatings also assist in significantly lowering the leakage and ageing of ceramic components comprised of the calcined composition-modified barium titanate powder to a point where they will not effect the performance of the EESU. In fact, the discharge rate of the ceramic EESU will be lower than 0.1% per 30 days which is approximately an order of magnitude lower than the best electrochemical battery.


A significant advantage of the present invention is that the calcium magnesium aluminosilicate glass coating assists in lowering the sintering and hot-isostatic-pressing temperatures to 800OC. This lower temperature eliminates the need to use expensive platinum, palladium, or palladium-silver alloy as the terminal metal.  In fact, this temperature is in a safe range that allows nickel to be used, providing a major cost saving in material expense and also power usage during the hot-isostatic-pressing process.  Also, since the glass becomes easily deformable and flowable at these temperatures it will assist in removing the voids from the EESU material during the hot-isostatic-pressing process. The manufacturer of such systems is Flow Autoclave Systems, Inc.  For this product to be successful it is mandatory that all voids be removed to assist in ensuring that the high voltage breakdown can be obtained.  Also, the method described in this patent of coating the calcium magnesium aluminosilicate glass ensures that the hot-isostatic-pressed double-coated composition-modified barium titanate high-relative-permittivity layer is uniform and homogeneous.


Yet another aspect of the present invention is that each component of the EESU is produced by screen-printing multiple layers of nickel electrodes with screening ink from nickel powder.  Interleaved between nickel electrodes are dielectric layers with screening ink from calcined double-coated high-permittivity calcined composition-modified barium titanate powder.  A unique independent dual screen-printing and layer-drying system is used for this procedure.  Each screening ink contains appropriate plastic resins, surfactants, lubricants, and solvents, resulting in a proper rheology (the study of the deformation and flow of matter) for screen printing. The number of these layers can vary depending on the electrical energy storage requirements.  Each layer is dried before the next layer is screen printed.  Each nickel electrode layer 12 is alternately preferentially aligned to each of two opposite sides of the component automatically during this process as indicated in Fig.2.  These layers are screen printed on top of one another in a continuous manner. When the specified number of layers is achieved, the component layers are then baked to obtain by further drying sufficient handling strength of the green plastic body.  Then the array is cut into individual components to the specified sizes.



Alternatively, the dielectric powder is prepared by blending with plastic binders, surfactants, lubricants, and solvents to obtain a slurry with the proper rheology for tape casting.  In tape casting, the powder-binder mixture is extruded by pressure through a narrow slit of appropriate aperture height for the thickness desired of the green plastic ceramic layer on to a moving plastic-tape carrier, known as a doctor-blade web coater. After drying, to develop sufficient handling strength of the green plastic ceramic layer, this layer is peeled away from the plastic-tape carrier.  The green plastic ceramic layer is cut into sheets to fit the screen-printing frame in which the electrode pattern is applied with nickel ink.  After drying of the electrode pattern, the sheets are stacked and then pressed together to assure a well-bonded lamination.  The laminate is then cut into components of the desired shape and size.




The components are treated for the binder-burnout and sintering steps.  The furnace temperature is slowly ramped up to 350OC and held for a specified length of time.  This heating is accomplished over a period of several hours so as to avoid any cracking and delamination of the body.  Then the temperature is ramped up to 850OC and held for a specified length of time.  After this process is completed the components are then properly prepared for the hot isostatic pressing at 700OC and the specified pressure.  This process will eliminate voids.  After this process, the components are then side-lapped on the connection side to expose the preferentially aligned nickel electrodes 12.  Then these sides are dipped into ink from nickel powder that has been prepared to have the desired rheology.  Then side conductors of nickel 14 are dipped into the same ink and then are clamped on to each side of the components 15 that have been dipped into the nickel powder ink.  The components are then fired at 800OC for 20 minutes to bond the nickel bars to the components as indicated in Fig.3.  The components are then assembled into a first-level array, Fig.3, with the use of the proper tooling and solder-bump technology.  Then the first-level arrays are assembled to form a second-level array, Fig.4, by stacking the first array layers on top of one another in a preferential mode. Then nickel bars 18 are attached on each side of the second array as indicated in Fig.4.  Then the EESU is packaged to form its final assembly configuration.



The features of this patent indicate that the ceramic EESU, as indicated in Table 1, outperforms the electrochemical battery in every parameter.  This technology will provide mission-critical capability to many sections of the energy-storage industry.



The parameters of each technology to store 52.2 kW · h of electrical energy

are indicated-(data as of February 2001 from manufacturer’s specification sheets).





Ceramic EESU


Weight (pounds)





Volume (cu. inch)





Discharge rate

5% in 30 days

1% in 30 days

0.1% in 30 days

1% in 30 days

Charging time (full)

1.5 hours

8.0 hours

3 to 6 minutes

1.5 hours

Life reduced with deep cycle use





Hazardous materials






This EESU will have the potential to revolutionise the electric vehicle (EV) industry, the storage and use of electrical energy generated from alternative sources with the present utility grid system as a backup source for residential, commercial, and industrial sites, and the electric energy point of sales to EVs.  The EESU will replace the electrochemical battery in any of the applications that are associated with the above business areas or in any business area where its features are required.


The features and advantages described in the specifications are not all inclusive, and particularly, many additional features and advantages will be apparent to one of ordinary skill in the art in view of the description, specification and claims made here.   Moreover, it should be noted that the language used in the specification has been principally selected for readability and instructional purposes, and may not have been selected to delineate or circumscribe the inventive subject matter, resort to the claims being necessary to determine such inventive subject matter.





Fig.1 indicates a schematic of 2320 energy storage components 9 hooked up in parallel with a total capacitance of 31 Farads.  The maximum charge voltage 8 of 3500 V is indicated with the cathode end of the energy storage components 9 hooked to system ground 10.




Fig.2 is a cross-section side view of the electrical-energy-storage unit component.  This figure indicates the alternating layers of nickel electrode layers 12 and high-permittivity composition-modified barium titanate dielectric layers 11.  This figure also indicate the preferentially aligning concept of the nickel electrode layers 12 so that each storage layer can be hooked up in parallel.




Fig.3 is side view of a single-layer array indicating the attachment of individual components 15 with the nickel side bars 14 attached to two preferentially aligned copper conducting sheets 13.





Fig.4 is a side view of a double-layer array with copper array connecting nickel bars 16 attaching the two arrays via the edges of the preferentially aligned copper conductor sheets 13.  This figure indicates the method of attaching the components in a multi-layer array to provide the required energy storage.

Reference No.

 Refers to this in the drawings


System maximum voltage of 3500 V


2320 energy-storage components hooked up in parallel with a total capacitance of 31 Farad


System ground


High-permittivity calcined composition-modified barium titanate dielectric layers


Preferentially aligned nickel electrode layers


Copper conductor sheets


Nickel sidebars




Copper array connecting nickel bars





Fig.1, Fig.2, Fig.3, and Fig.4 of the drawings and the following description depict various preferred embodiments of the present invention for purposes of illustration only.  One skilled in the art will readily recognise from the following discussion those alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles of the invention described here. While the invention will be described in conjunction with the preferred embodiments, it will be understood that they are not intended to limit the invention to those embodiments.  On the contrary, the invention is intended to cover alternatives, modifications, and equivalents, which may be included within the spirit and scope of the invention as defined by the claims.


Preparation of the high-permittivity calcined composition-modified barium titanate powder that is used to fabricate the EESU is explained as follows. Wet-chemical-prepared powders of high-purity as well as composition-modified barium titanate with narrow particle-size distribution have been produced with clear advantages over those prepared by solid-state reaction of mechanically mixed, ball-milled, and calcined powdered ingredients. The compositional and particle-size uniformity attained with a coprecipitated-prepared powder is vastly superior to that with a conventional-prepared powder.  The microstructures of ceramics formed from these calcined wet-chemical-prepared powders are uniform in grain size and can also result in smaller grain size.  Electrical properties are improved so that higher relative permittivities and increased dielectric breakdown strengths can be obtained.  Further improvement can be obtained by the elimination of voids within the sintered ceramic body with subsequent hot isostatic pressing.


High-relative-permittivity dielectrics have inherent problems, namely ageing, fatigue, degradation, and decay of the electrical properties, which limit their application.  The use of surface-coated powders in which the surface region is comprised of one or two materials different in composition from that of the powder overcomes these problems provided that the compositions are appropriately chosen.


Among ceramics, alumina [aluminium oxide (Al2O3)], and among glasses, calcium magnesium aluminosilicate (CaO.MgO.Al2O3.SiO2) glasses are the best dielectrics in terms of having the highest dielectric breakdown strengths and to seal the high-relative-permittivity dielectric powder particles so as to eliminate or significantly reduce their inherent problems.


A glass with a given composition at temperatures below its glass transition temperature range, which is in the neighbourhood of its strain-point temperature, is in a fully rigid condition, but at temperatures above this range is in a viscous-flow condition, its viscosity decreasing with increasing temperature.  The application of hot isostatic pressing to a sintered closed-pore porous ceramic body comprised of sufficient-thickness glass-coated powder will lead to void elimination provided the glass is in the viscous-flow condition where it is easily deformable and flowable.


The wet-chemical-prepared and calcined composition-modified barium titanate powder is accordingly coated with these layers of, first, alumina, and second, a calcium magnesium aluminosilicate glass.  After the first layer has been applied by wet-chemical means, the powder is calcined at 1050OC to convert the precursor, aluminium nitrate nonahydrate [Al(NO3)3.9H2O] to aluminium oxide (corundum) [a-Al2O3].  Then the second layer is applied by wet-chemical means with the use of the precursors in the appropriate amounts of each, and in absolute ethanol (CH3CH2OH) as the solvent, shown in the accompanying table.  After drying, the powder is calcined at 500OC to convert the precursor mixture to a calcium magnesium aluminosilicate glass. It is important that the calcining temperature is not higher than the strain point of the selected glass composition to prevent sticking together of the powder.  The glass coating has the further advantage of acting as a sintering aid and allowing a substantially lower firing temperature for densification of the ceramic body particularly during the hot-isostatic-pressing step.


Another significant advantage of the calcium magnesium aluminosilicate glass coating is that sintering and densification temperatures are sufficiently lowered to allow the use of nickel conductor electrodes in place of the conventional expensive platinum, palladium, or palladium-silver alloy ones.


Preparation of the Calcined Composition-Modified Barium Titanate Powder is Indicated by the Following Process Steps.


A solution of the precursors: Ba(NO3)2, Ca(NO3)2.4H2O, Nd(NO3)3.6H2O, Y(NO3)3.4H2O,

Mn(CH3COO)2.4H2O, ZrO(NO3)2, and [CH3CH(O—)COONH4]2Ti(OH)2, as selected from the reference; Sigma-Aldrich, Corp., "Handbook of Fine Chemicals and Laboratory Equipment", 2000-2001, in de-ionised water heated to 80OC is made in the proportionate amount in weight percent for each of the seven precursors as shown in the most right-hand column of Table 3.   A separate solution of (CH3)4NOH somewhat in excess amount than required, as shown in Table 4, is made in de-ionised water, free of dissolved carbon dioxide (CO2) and heated to 80O-85OC.  The two solutions are mixed by pumping the heated ingredient streams simultaneously through a coaxial fluid jet mixer.  A slurry of the co-precipitated powder is produced and collected in a drown-out vessel.  The co-precipitated powder is refluxed in the drown-out vessel at 90°-95° C. for 12 hr and then filtered, de-ionised-water washed, and dried.   Alternatively, the powder may be collected by centrifugal sedimentation.   An advantage of (CH3)4NOH as the strong base reactant is that there are no metal element ion residuals to wash away anyway.   Any residual (CH3)4NOH, like any residual anions from the precursors, is harmless, because removal by volatilisation and decomposition occurs during the calcining step.  The powder contained in a silica glass tray or tube is calcined at 1050OC in air.  Alternatively, an alumina ceramic tray can be used as the container for the powder during calcining.



Composition-modified barium titanate with metal element atom fractions

given for an optimum result, as demonstrated in the reference: P. Hansen,

U.S. Pat. No. 6,078,494, issued Jan. 20, 2000.

Composition-modified barium titanate with

metal element atom fractions as follows:


Metal Element

Atom Fraction

Atomic Weight


Weight %

























































Calculation of minimum amount of (CH3)4NOH

required for 100 g of the precursor mixture




Wt %

Wt %/FW

Reactant base multiplier

Mol of base required





































[CH3CH(O—)COONH4]2Ti (OH)2












Reactant strong base













Note:        The weight of (CH3)4NOH required is accordingly a minimum of

    (0.738105 mol) (91.15 g/mol) = 67.278 g for 100 g of the precursor mixture.                                               

    Tetramethylammonium hydroxide (CH3)4NOH is a strong base.


Coating of Aluminium Oxide on Calcined Modified Barium Titanate Powder


Barium titanate BaTiO3

FW 233.19

d 6.080 g/cm3

Aluminium oxide Al2O3

FW 101.96

d 3.980 g/cm3



Precursor, aluminium nitrate nonahydrate, as selected from the reference: Sigma-Aldrich Corp., "Handbook of Fine Chemicals and Laboratory Equipment", 2000-2001. Al(NO3)3.9H2O FW 3.75.13


For Calcined Aluminium Oxide (Al2O3) Coating of 100 Angstrom units Thickness on Calcined Modified Barium Titanate Powder  100 Angstrom units = 10-6 cm 1.0 m2 = 104 cm2


area thickness of Al2O3 coating volume (104 cm2/g)(10-6 cm) = 10-2 cm3/g - - - of calcined powder



Al(NO3)3.9H2O (FW 375.13)(2)=750.26


Al2O3 FW 101.96=101.96





For an aluminium oxide (Al2O3) coating of 100 Angstrom units thickness on calcined modified barium titanate powder with particle volume of 1.0 mm3, 39.8 mg of Al2O3 are required per g of this powder, corresponding to 292.848 mg of the aluminium nitrate nonahydrate [Al(NO3)3.9H2O] precursor required per g of this powder.


Coating of Calcium Magnesium Aluminosilicate Glass on Aluminium Oxide Coated

Calcined Modified Barium Titanate Powder







Barium titanate      BaTiO3





Calcium magnesium aluminosilicate (CaO.MgO.Al2O3.SiO2) glass precursors, as selected from the reference: Sigma-Aldrich, Corp., "Handbook of Fine Chemicals and Laboratory Equipment", 2000-2001.


Calcium methoxide            (CH3O)2Ca


Calcium isopropoxide     [(CH3)2CHO]2Ca


Magnesium methoxide       (CH3O)2Mg


Magnesium ethoxide     (CH3CH2O)2Mg


Aluminium ethoxide     (CH3CH2O)3Al


Aluminium isopropoxide     [(CH3)2CHO]3Al


Aluminium butoxide     [CH3(CH2)3O]3Al


Tetraethyl orthosilicate     Si(OCH2CH3)4




Select glass composition, e.g.,


CaO.MgO.2Al2O3.8SiO2  and accordingly the precursors:



Prepare Mixture of these Precursors in Absolute Ethanol (to Avoid Hydrolysis) and in Dry-Air Environment (Dry Box) (also to Avoid Hydrolysis).


Glass Composition: CaO.MgO.2Al2O3.8SiO2 or CaMgAl4Si8O24


1 mol (56.08 g)


1 mol (40.30 g)


2 mol (101.96 g × 2 = 203.92 g)


8 mol (60.08 g × 8 = 480.64 g)



glass FW total 780.98 g/mol

Density of glass: about 2.50 g/cm3


Calcined modified barium titanate powder

Particle volume: 1.0 mm3 or 1.0(10-4 cm)3  = 10-12 cm3;

so there are 1012 particles/cm3 (assumption of no voids)

Particle area: 6 mm2 or (6)(10-4 cm )2 = 6×10-8 cm3;

Particle area/cm3 (no voids):

(6×10-8 cm2/particle)(1012 particles/cm3) = 6×104 cm2/cm3 or 6 m2/cm3.


Then for density of 6 g/cm3, the result is:





For Calcined Glass Coating of 100 Angstrom units Thickness on Calcined Powder:


100 Angstrom units = 10-6 cm 1.0 m2 = 104 cm2


(104 cm2/g)(10-6 cm) = 10-2 cm3/g of calcined powder of glass coating and then




Precursor mixture FW 2756.32 = 3.529

Glass FW 780.98




For a CaMgAl4Si8O24 glass coating of 100 Angstrom units thickness on calcined modified barium titanate powder with particle volume of 1.0 mm3, 25.0 mg of this glass are required per g of this powder, corresponding to 88.228 mg of the precursor mixture required per g of this powder.


   Particle Volume and Area


V particle = a3 for cube

If a = 1.0 mm, V = 1.0 mm3

A particle = 6a2 for cube

If a = 1.0 mm,  A = 6 mm2


Particle coating volume


(6 a2)(t), if t = 100 Angstrom units = 10×103 mm, and 6 a2=6.0 mm2,

then (6.082 m2)(10×10-3 mm) = 60×10-3 mm3 = V coating


Ratio of particle coating volume to particle volume 60×10-3 mm3/1.0 mm3 = 60×10-3 = 0.06 or 6%


With the assumption of no voids and absolutely smooth surface, for an ideal cubic particle with volume of 1.0 mm3 and for a particle coating of 100 Angstrom units thickness, the coating volume is 60×10-3 mm3 or 6.0% that of the particle volume.


Calculations of the Electrical-Energy-Storage Unit's Weight, Stored Energy, Volume, and Configuration.




The relative permittivity of the high-permittivity powder is nominally 33,500, as given in the reference: P. Hansen, U.S. Pat. No. 6,078,494, issued Jan. 20, 2000.


    * The 100 ? coating of Al2O3 and 100 ? of calcium magnesium aluminosilicate glass will reduce the relative permittivity by 12%.

    * K = 29,480

      Energy stored by a capacitor: E = CV2/(2×3600 s/h) = W·h

    * C = capacitance in farads

    * V = voltage across the terminals of the capacitor

      It is estimated that is takes 14 hp, 746 watts per hp, to power an electric vehicle running at 60 mph with the lights, radio, and air conditioning on. The energy-storage unit must supply 52,220 W·h or 10,444 W for 5 hours to sustain this speed and energy usage and during this period the EV will have travelled 300 miles.

      Each energy-storage component has 1000 layers.

      C = eoKA/t

    * eo = permittivity of free space

    * K = relative permittivity of the material

    * A = area of the energy-storage component layers

    * t = thickness of the energy-storage component layers


     Voltage breakdown of the energy-storage components material after coating with Al2O3 and calcium magnesium aluminosilicate glass will be in the range of 1.0×106 V/cm to 5×106 V/cm or higher.  Using the proper voltage breakdown selected from this range could allow the voltage of the energy-storage unit to be 3500 V or higher.

       One hp = 746 W





Capacitance of one layer = 8.854 × 10-12 F / m × 2.948 × 104 × 6.45 × 10-4m2 / 12.7 × 10-6 m


C = 0.000013235 F


With 1000 layers:


C = 0.013235 F


The required energy storage is 

    Et = 14 hp × 746 W /hp × 5 h = 52,220 W·h


The total required capacitance of the energy-storage unit:

    CT = Et  × 2 × 3600 s/h / V2 = 52,220 W·h × 2 × 3600 s/h/(3500 V)2 CT = 31 F


Number of capacitance components required:

    Nc = 31 F / 0.013235 F = 2320


Volume and weight of energy-storage unit:


Volume of the dielectric material:


    Volume = area x thickness x number of layers

                = 6.45 cm2 x 12.72 x 10-4 cm x 1000

                = 8.2 cm3


Total volume = 8.2 cm3 × number of components (2320) = 19,024 cm3

Density of the dielectric material = 6.5 g/cm3

Weight of each component = density × volume = 53.3 g

Total weight of the dielectric material = 53.3 g × 2320 / 454 g per pound = 272 pounds


Volume of the nickel conductor layers:

Thickness of the nickel layer is 1×10-6 m

Volume of each layer = 6.45 cm2×1.0×10-4 cm × 1000 = 0.645 cm3

Density of nickel = 8.902 g/cm3

Weight of nickel layers for each component = 5.742 g

Total weight of nickel = 34 pounds

Total number of capacitance layers and volume of the EESU:

Area required for each component to solder bump = 1.1 inch2

A 12 × 12 array will allow 144 components for each layer of the first array

19 layers of the second array will provide 2736 components which are more than enough to meet the required 2320 components.  The distance between the components will be adjusted so that 2320 components will be in each EESU.  The second array area will remain the same.

      The total weight of the EESU (est.) = 336 pounds

      The total volume of the EESU (est.) = 13.5 inches × 13.5 inches × 11 inches = 2005 inches3 which includes the weight of the container and connecting material.

      The total stored energy of the EESU = 52,220 W·h


From the above description, it will be apparent that the invention disclosed herein provides a novel and advantageous electrical-energy-storage unit composed of unique materials and processes. The foregoing discussion discloses and describes merely exemplary methods and embodiments of the present invention. As will be understood by those familiar with the art, the invention may be embodied in other specific forms and utilise other materials without departing from the spirit or essential characteristics thereof. Accordingly, the disclosure of the present invention is intended to be illustrative, but not limiting, of the scope of the invention, which is set forth in the following claims.















US Patent 1,540,998               9th June 1925               Inventor: Hermann Plauson





Please note that this is a re-worded excerpt from this patent.  It describes in considerable detail, different methods for abstracting useable electrical power from passive aerial systems.  He describes a system with 100 kilowatt output as a “small” system.



Be it known that I, Hermann Plauson, Estonian subject, residing in Hamburg, Germany, have invented certain new and useful improvements in the Conversion of atmospheric Electric Energy, of which the following is a specification.


According to this invention, charges of atmospheric electricity are not directly converted into mechanical energy, and this forms the main difference from previous inventions, but the static electricity which runs to earth through aerial conductors in the form of direct current of very high voltage and low current strength is converted into electro-dynamic energy in the form of high frequency vibrations.  Many advantages are thereby obtained and all disadvantages avoided.


The very high voltage of static electricity of a low current strength can be converted by this invention to voltages more suitable for technical purposes and of greater current strength.  By the use of closed oscillatory circuits it is possible to obtain electromagnetic waves of various amplitudes and thereby to increase the degree of resonance of such current.  Such resonance allows various values of inductance to be chosen which, by tuning the resonance between a motor and the transformer circuit, allows the control of machines driven by this system.  Further, such currents have the property of being directly available for various uses, other than driving motors, including lighting, heating and use in electro-chemistry.


Further, with such currents, a series of apparatus may be fed without a direct current supply through conductors and the electro-magnetic high frequency currents may be converted by means of special motors, adapted for electro-magnetic oscillations, into alternating current of low frequency or even into high voltage direct current.





Fig.1 is an explanatory figure




Fig.2 is a diagrammatic view of the most simple form.





Fig.3 shows a method of converting atmospheric electrical energy into a form suitable for use with motors.





Fig.4 is a diagram showing the protective circuitry.




Fig.5 is a diagram of an arrangement for providing control




Fig.6 is an arrangement including a method of control








Fig.7 shows how the spark gap can be adjusted








Fig.8 shows a unipolar connection for the motor



Fig.9 shows a weak coupled system suitable for use with small power motors






Fig.10, Fig.11 and Fig.12 show modified arrangements





Fig.13  shows a form of inductive coupling for the motor circuit




Fig.14 is a modified form of Fig.13 with inductive coupling.






Fig.15 is an arrangement with non-inductive motor





Fig.16 is an arrangement with coupling by capacitor.





Fig.17, Fig.18 and Fig.19 are diagrams showing further modifications



Fig.20 shows a simple form in which the aerial network is combined with special collectors





Fig.21 shows diagramatically, an arrangement suitable for collecting large quantities of energy.

Fig.22 is a modified arrangement having two rings of collectors






Fig.23 shows the connections for three rings of collectors



Fig.24 shows a collecting balloon and diagram of its battery of capacitors





Fig.25 and Fig.26 show modified collector balloon arrangements.







Fig.27 shows a second method of connecting conductors for the balloon aerials.



Fig.28 shows an auto-transformer method of connection.



Fig.29 shows the simplest form of construction with incandescent cathode.






Fig.30 shows a form with a cigar-shaped balloon.




Fig.31 is a modified arrangement.





Fig.32 shows a form with cathode and electrode enclosed in a vacuum chamber.




Fig.33 is a modified form of Fig.32






Fig.34 shows an arc light collector.



Fig.35 shows such an arrangement for alternating current





Fig.36 shows an incandescent collector with Nernst lamp



Fig.37 shows a form with a gas flame.







Fig.1 illustrates a simple diagram for converting static electricity into dynamic energy of a high number of oscillations.  For the sake of clarity, a Wimshurst machine is assumed to be employed and not an aerial antenna.  Items 13 and 14 are combs for collecting the static electricity of the influence machine.  Items 7 and 8 are spark-discharging electrodes.  Items 5 and 6 are capacitors, 9 is the primary winding of an inductive coil, 10 is the secondary winding whose ends are 11 and 12.  When the disc of the static influence machine is rotated by mechanical means, the combs collect the electric charges, one being positive and one negative and these charge the capacitors 5 and 6 until such a high voltage is developed across the spark gap 7-- 8 that the spark gap is jumped.  As the spark gap forms a closed circuit with capacitors 5 and 6, and inductive resistance 9, as is well known, waves of high frequency electromagnetic oscillations will pass in this circuit.


The high frequency of the oscillations produced in the primary circuit induces waves of the same frequency in the secondary circuit.  Thus, in the primary circuit, electromagnetic oscillations are formed by the spark and these oscillations are maintained by fresh charges of static electricity.


By suitably selecting the ratio between the number of turns in the primary and secondary windings, with regard to a correct application of the coefficients of resonance (capacitance, inductance and resistance) the high voltage of the primary circuit may be suitably converted into a low voltage high current output.


When the oscillatory discharges in the primary circuit become weaker or cease entirely, the capacitors are charged again by the static electricity until the accumulated charge again breaks down across the spark gap.  All this is repeated as long as electricity is produced by the static machine through the application of mechanical energy to it.



An elementary form of the invention is shown in Fig.2 in which two spark gaps in parallel are used, one of which may be termed the working gap 7 while the second serves as a safety device for excess voltage and consists of a larger number of spark gaps than the working section, the gaps being arranged in series and which are bridged by very small capacitors a1, b1, c1, which allow uniform sparking in the safety section.


1 is the aerial antenna for collecting charges of atmospheric electricity, 13 is the earth connection of the second part of the spark gap, 5 and 6 are capacitors and 9 is the primary coil winding.  When the positive atmospheric electricity seeks to combine with the negative earth charge via aerial 1, this is prevented by the air gap between the spark gaps.  The resistance of spark gap 7 is lower than that of the safety spark gap set of three spark gaps connected in series a which consequently has three times greater air resistance.


Therefore, so long as the resistance of spark gap 7 is not overloaded, discharges take place only through it.  However, if the voltage is increased by any influence to such a level that it might be dangerous for charging the capacitors 5 and 6, or for the coil insulation of windings 9 and 10, the safety spark gap set will, if correctly set, discharge the voltage directly to earth without endangering the machine.  Without this second spark gap arrangement, it is impossible to collect and render available large quantities of electrical energy.


The action of this closed oscillation circuit consisting of spark gap 7, two capacitors 5 and 6, primary coil 9 and secondary coil 10, is exactly the same as that of Fig.1 which uses a Wimshurst machine, the only difference being the provision of the safety spark gap.  The high frequency electromagnetic alternating current can be tapped off through the conductors 11 and 12 for lighting and heating purposes.  Special motors adapted for working with static electricity or high frequency oscillations may be connected at 14 and 15.



In addition to the use of spark gaps in parallel, a second measure of security is also necessary for taking the current from this circuit.  This is the introduction of protective electromagnets or choking coils in the aerial circuit as shown by S in Fig.3.  A single electromagnet having a core of the thinnest possible separate laminations is connected with the aerial.  In the case of high voltages in the aerial network or at places where there are frequent thunderstorms, several such magnets may be connected in series.


In the case of large units, several such magnets can be employed in parallel or in series parallel.  The windings of these electromagnets may be simply connected in series with the aerials.  In this case, the winding preferably consists of several thin parallel wires, which together, make up the necessary cross-sectional area of wire.  The winding may be made of primary and secondary windings in the form of a transformer.  The primary winding will then be connected in series with the aerial network, and the secondary winding more or less short-circuited through a regulating resistor or an induction coil.  In the latter case it is possible to regulate, to a certain extent, the effect of the choking coils.  In the following circuit and constructional diagrams , the aerial electromagnet choke coil is indicated by a simple ring S.


Fig.3 shows the most simple way of converting atmospheric electricity into electromagnetic wave energy by the use of special motors adapted for high oscillatory currents or static charges of electrical energy.  Recent improvements in motors for working with static energy and motors working by resonance, that is to say, having groups of tuned electromagnetic co-operating circuits render this possible but such do not form part of the present invention.


A motor adapted to operate with static charges, will for the sake of simplicity, be shown in the diagrams as two semi-circles 1 and 2 and the rotor of the motor by a ring M (Fig.3).   A is a vertical aerial or aerial network.  S is the safety choke or electromagnet with coil O as may be seen is connected with the aerial A.  Adjacent to the electromagnet S, the aerial conductor is divided into three circuits, circuit 8 containing the safety spark gap, circuit 7 containing the working spark gap, and then a circuit containing the stator terminal 1, the rotor and stator terminal 2 at which a connection is made to the earth wire.  The two spark gaps are also connected metallically with the earth wire.  The method of working in these diagrams is as follows:


The positive atmospheric electric charge collected tends to combine with the negative electricity (or earth electricity) connected via the earth wire.  It travels along the aerial A through the electromagnet S without being checked as it flows in the same direction as the direct current.  Further, its progress is arrested by two spark gaps placed in the way and the stator capacitors.  These capacitors charge until their voltage exceeds that needed to jump the spark gap 7 when a spark occurs and an oscillatory charge is obtained via the closed oscillation circuit containing motor M.  The motor here forms the capacity and the necessary inductance and resistance, which as is well known, are necessary for converting static electricity into electromagnetic wave energy.


The discharges are converted into mechanical energy in special motors and cannot reach the aerial network because of the electromagnet or choke.  If, however, when a spark occurs at spark gap 7, a greater quantity of atmospheric electricity tends to flow to earth, then a counter voltage is induced in the electromagnet, which is greater the more rapidly and strongly the flow of current direct to earth is.  This opposing voltage causes the circuit to exhibit a sufficiently high resistance to prevent a short circuit between the atmospheric electricity and the earth.


The circuit containing spark gap 8, having a different wave length which is not in resonance with the natural frequency of the motor, does not endanger the motor and serves as security against excess voltage, which, as practical experiments have shown, may still arise in certain cases.





In Fig.4, spark gap 7 is shunted across capacitors 5 and 6 from the motor M.  This arrangement provides improved over-voltage protection for the motor and it gives a uniform excitation through the spark gap 7.





Fig.5 shows an arrangement for producing large currents which can be used direct without motors, to provide heating and lighting.  The main difference here is that the spark gap consists of a star-shaped disc 7 which can rotate on its own axis and is rotated by a motor opposite similarly fitted electrodes 7a.  When separate points of starts face one another, discharges take place, thus forming an oscillation circuit with capacitors 5 and 6 and inductor 9.  It is evident that a motor may also be connected directly to the ends of inductor 9.





Fig.6 shows how the oscillation circuit may have a motor connected via a variable inductor which opposes any excess voltages which might be applied to the motor.  By cutting the separate coils 9 (coupled inductively to the aerial) in or out, the inductive action on the motor may be more or less increased, or variable aerial action may be exerted on the oscillation circuit.






In Fig.7 the oscillation circuit is closed through the earth (E and E1).  The spark gap 7 may be increased or reduced by means of a contact arm 7b.



Fig.8 shows a unipolar connection of the motor with the aerial network.  Here, two oscillation circuits are closed through the same motor.  The first oscillation circuit passes from aerial A through electromagnet S, point x, inductance 9a to the earth capacitor 6, across spark gap 7 to the aerial capacitor 5 and back to point x.  The second oscillation circuit starts from the aerial 5 at the point x1 through inductor 9 to the earth capacitor 6 at the point x3, through capacitor 6, across spark gap 7 back to point x1.  The motor itself, is inserted between the two points of spark gap 7.  This arrangement produces slightly dampened oscillation wave currents.





Fig.9 shows a loosely coupled system intended for small motors for measuring purposes.  A is the serial, S is the electromagnet or aerial inductor, 9 the inductor, 7 the spark gap, 5 and 6 capacitors, E the earth, M the motor, and 1 and 2 the stator connections of the motor which is directly connected to the oscillator circuit.





Fig.10 shows a motor circuit with purely inductive coupling.  The motor is connected with the secondary wire 10 as may be seen in Fig.11 in a somewhat modified circuit.  The same applies to the circuit of Fig.12.


The circuit diagrams shown so far, allow motors of small to medium strength to be operated.  For large aggregates, however, they are too inconvenient as the construction of two or more oscillation circuits for large amounts of energy is difficult; the governing is still more difficult and the danger in switching on or off is greater.



A means for overcoming such difficulties is shown in Fig.13.  The oscillation circuit shown here, runs from point x over capacitor 5, variable inductor 9, spark gap 7 and the two segments 3a and 3b forming arms of a Wheatstone bridge, back to x.  If the motor is connected by brushes 3 and 4 transversely to the two arms of the bridge as shown in the drawing, electromagnetic oscillations of equal sign are induced in the stator surfaces 1 and 2 and the motor does not revolve.  If however, the brushes 3 and 4 are moved in common with the conducting wires 1 and 2 which connect the brushes with the stator poles, a certain alteration or displacement of the polarity is obtained and the motor commences to revolve.


The maximum action will result if one brush 3 comes on the central sparking contact 7 and the other brush 4 on the part x.  In practice however, they are usually brought on to the central contact 7 but only held in the path of the bridge segments 4a and 3a in order to avoid connecting the spark gaps with the motor oscillation circuit.






As this prevents the whole of the oscillation energy acting on the motor, it is better to adopt the modification shown in Fig.14.  The only difference here is that the motor is not wired directly to the segments of the commutator, but instead it is wired to secondary coil 10 which receives induced current from primary coil 9.  This arrangement provides a good transforming action, a loose coupling and an oscillation circuit without a spark gap.






In Fig.15, the motor is wired directly to the primary coil at x and x1 after the principle of the auto-transformer.  In Fig.16, instead of an inductor, capacitor 6 replaces the inductance and is inserted between the segments 3a and 4a.  This has the advantage that the segments 3a and 4a need not be made of solid metal, but may consist of spiral coils which allow a more exact regulation, and high inductance motors may be used.




The circuits shown in Fig.17, Fig.18 and Fig.19 may be used with resonance and particularly with induction capacitor motors; between the large stator induction capacitor surfaces, small reversing pole capacitors are connected which are lead together to earth.  Such reversing poles have the advantage that, with large quantities of electrical energy, the spark formation between the separate oscillation circuits ceases.


Fig.19 shows another method which prevents high frequency electromagnetic oscillations formed in the oscillation circuit, feeding back to the aerial.  It is based on the well known principle that a mercury lamp, one electrode of which is formed of mercury, the other of solid metal such as steel, allows an electric charge to pass in only one direction: from the mercury to the steel and not vice versa.  The mercury electrode of the vacuum tube N is therefore connected with the aerial conductor and the steel electrode with the oscillation circuit.  Charges can then only pass from the aerial through the vacuum tube to the oscillation circuit and no flow occurs in the opposite direction.  In practice, these vacuum tubes must be connected behind an electromagnet as the latter alone provides no protection against the danger of lightning.


As regards the use of spark gaps, all arrangements as used for wireless telegraphy may be used.  Of course, the spark gaps in large machines must have a sufficiently large surface.  In very large stations they are cooled in liquid carbonic acid or better still, in liquid nitrogen or hydrogen; in most cases the cooling may also take place by means of liquefied low homologues of the metal series or by means of hydrocarbons, the freezing point of which lies between -90oC and -400C.  The spark gap casing must also be insulated and be of sufficient strength to be able to resist any pressure which may arise.  Any undesirable excess super-pressure which may be formed must be let off automatically.  I have employed with very good results, mercury electrodes which were frozen in liquid carbonic acid, the cooling being maintained during the operation from the outside, through the walls.




Fig.20 shows one of the most simple forms of construction of an aerial network in combination with collectors, transformers and the like.  E is the earth wire, 8 the safety spark gap, 7 the working spark gap, 1 and 2 the stator surfaces of the motor, 5 a capacitor battery, S the protective magnet which is connected with the coil in the aerial conductor, A1 to A10 aerial antennae with collecting balloons, N horizontal collecting or connecting wires, from which, a number of connections run to the centre.


The actual collectors consist of metal sheaths, preferably made of an aluminium magnesium alloy, and are filled with hydrogen or helium, and are attached to copper-plated steel wires.  The size of the balloon is selected so that the actual weight of the balloon and its conducting wire is supported by it.  Aluminium spikes, made and gilded as described below, are arranged on top of the balloons in order to produce a conductor action.  Small quantities of radium preparations, more particularly, polonium-ionium or mesothorium preparations, considerably increase the ionisation, and the performance of these collectors.


In addition to metal balloons, fabric balloons which are sprayed with a metallic coating according to Schoop’s metal-spraying process may also be used.  A metallic surface may also be produced by lacquering with metallic bronzes, preferably according to Schoop’s spraying process, or lacquering with metallic bronze powders in two electrical series of widely different metals, because this produces a considerably increased collecting effect.


Instead of the ordinary round balloons, elongated cigar-shaped ones may be employed.  In order also to utilise the frictional energy of the wind, patches or strips of non-conducting substances which produce electricity by friction, may be attached to the metallised balloon surfaces.  The wind will impart a portion of its energy in the form of frictional electricity, to the balloon casing, thus substantially increasing the collection effect.


In practice however, very high towers of up to 300 metres may be employed as antennae.  In these towers, copper tubes rise freely further above the top of the tower.  A gas lamp secured against the wind is then lit at the point of the copper tube and a netting is secured to the copper tube over the flame of this lamp to form a collector.  The gas is conveyed through the interior of the tube, up to the summit.  The copper tube must be absolutely protected from moisture at the place where it enters the tower, and rain must be prevented from running down the walls of the tower, which might lead to a bad catastrophe.  This is done by bell-shaped enlargements which expand downwards, being arranged in the tower in the form of high voltage insulators of Siamese pagodas.


Special attention must be devoted to the foundations of such towers.  They must be well insulated from the ground, which may be achieved by first embedding a layer of concrete in a box form to a sufficient depth in the ground, and inserting in this, an asphalt lining and then glass bricks cast about 1 or 2 metres in thickness.  Over this in turn, there is a ferro-concrete layer in which alone the metal foot of the tube is secured.  This concrete block must be at least 2 metres from the ground and at the sides, be fully protected from moisture by a wooden covering.  In the lower part of the tower, a wood or glass housing should be constructed to protect the capacitors and/or motors.  In order to ensure that the ground lead connects to the water-table, a well insulated pit lined with vitreous bricks must be provided.  Several such towers are erected at equal distances apart and connected with a horizontal conductor.  The horizontal connecting wires may either run directly from tower to tower or be carried on bell-shaped insulators similar to those in use for high voltage electricity transmission lines.  The width of the aerial tower network may be of any suitable size and the connection of the motors can take place at any convenient location.



In order to collect large quantities of electricity with few aerials, it is as well to provide the aerial conductor with sets of capacitors as shown in the two methods of construction illustrated in Fig.21 and Fig.22.  In Fig.21 the set of capacitors 5 is connected between the aerials Z via lead A and an annular conductor from which horizontal run to the connecting points C to which the earth wire is connected.  Fig.22 shows a similar arrangement.


Should two such series of antenna rings be shown by a voltmeter to have a large voltage difference (for example, one in the mountains and one on the plain) or even of a different polarity, these differences may be compensated for by connecting sufficiently large capacitor sets (5, 5a, 5b) by means of Maji star conductors D and D1.   Fig.23, shows a connection of three such rings of collectors are positioned in a triangle with a central set of capacitors.



The capacitor sets of such large installations must be embedded in liquefied gasses or in liquids freezing at very low temperatures.  In such cases, a portion of the atmospheric energy must be employed for liquefying these gasses.  It is also preferable to employ pressure.  By this means, the capacitor surfaces may be reduced in area and still allow the storage of large quantities of energy to be stored, secure against breakdown.  For the smaller installations, the immersing of the capacitors in well insulated oil or the like, is sufficient.  Solid substances, on the other hand, cannot be employed as insulators.


The arrangement in the diagrams shown earlier has always shown both poles of the capacitors connected to the aerial conductors.  An improved method of connection has been found to be very advantageous.  In this method, only one pole of each capacitor is connected to the collecting network.  Such a method of connection is very important, as by means of it, a constant current and an increase in the normal working voltage is obtained.  If, for example, a collecting balloon aerial which is allowed to rise to a height of 300 metres, shows 40,000 volts above earth voltage, in practice it has been found that the working voltage (with a withdrawal of the power as described earlier by means of oscillating spark gaps and the like) is only about 400 volts.  If however, the capacity of the capacitor surfaces be increased, which capacity in the above mentioned case was equal to that of the collecting surface of the balloon aerials, to double the amount, by connecting the capacitors with only one pole, the voltage rises under an equal withdrawal of current up to and beyond 500 volts.  This can only be ascribed to the favourable action of the connecting method.


In addition to this substantial improvement it has also been found preferable to insert double inductances with electromagnets and to place the capacitors preferably between two such electromagnets.  It has also been found that the useful action of such capacitors can be further increased if an induction coil is connected as an inductive resistance to the unconnected pole of the capacitor, or still better if the capacitor itself be made as an induction capacitor.  Such a capacitor may be compared to a spring, which when compressed, carries in itself accumulated force, which it gives off again when released.  In charging, a charge with reversed sign is formed at the other free capacitor pole, and if a short circuit occurs through the spark gap, the accumulated energy is again given back since now new quantities of energy are induced at the capacitor pole connected to the conductor network, which in fact, charges with opposite sign to that at the free capacitor pole.  The new induced charges have of course, the same sign as the collector network.  The whole voltage energy in the aerial is thereby increased.  In the same time interval, larger quantities of energy are accumulated than is the case without such capacitor sets being inserted.




In Fig.24 and Fig.25, two different connection diagrams are illustrated in more detail.  Fig.24 shows a collecting balloon along with its earth connections.  Fig.25 shows four collecting balloons and the parallel connection of their capacitor sets.


A is the collecting balloon made of an aluminium magnesium alloy (electron metal magnalium) of a specific gravity of 1.8 and a plate thickness of 0.1 mm to 0.2 mm.  Inside, there are eight strong vertical ribs of T-shaped section of about 10 mm to 20 mm in height and about 3 mm in thickness, with the projecting part directed inwards (indicated by a, b, c, d and so forth).  They are riveted together to form a firm skeleton and are stiffened in a horizontal direction by two cross ribs.  The ribs are further connected to one another internally and transversely by means of thin steel wires, whereby the balloon obtains great strength and elasticity.  Rolled plates of 0.1 mm to 0.2 mm in thickness made of magnalium alloy are then either soldered or riveted on to this skeleton so that a fully metallic casing with a smooth external surface is created.  Well silvered or coppered aluminium plated steel wires  run from each rib to the fastening ring 2.  Further, the coppered steel hawser L, preferably twisted out of separate thin wires (shown as dotted lines in Fig.24) and which must be long enough to allow the balloon to rise to the desired height, leads to a metal roller or pulley 3 and on to a winch W, which must be well insulated from the earth.  By means of this winch, the balloon which is filled with hydrogen or helium, can be allowed to rise to a suitable height of 300 to 5,000 metres, and brought to the ground for recharging or repairs.


The actual current is taken directly through a friction contact from the metal roller 3 or from the wire or even from the winch, or simultaneously from all three by means of brushes (3, 3a and 3b).  Beyond the brushes, the conductor is divided, the paths being:- firstly, over 12 to the safety spark gap 8, on to the earth conductor E1, and secondly over electromagnet S1, point 13, to a second loose electromagnet having an adjustable coil S2, then to the spark gap 7 and to the second earth conductor E2.  The actual working circuit is formed through the spark gap 7, capacitors 5 and 6, and through the primary coil 9; here the static electricity formed by oscillatory discharges is accumulated and converted into high frequency electromagnetic oscillations.  Between the electromagnets S1 and S2 at the crossing point 13, four capacitor sets are introduced which are only indicated diagramatically in the drawings by a single capacitor.  Two of these sets of capacitors (16 and 18) are made as plate capacitors and prolonged by regulating induction coils or spirals 17 and 19 while the two others (21 and 23) are induction capacitors.  As may be seen from the drawings, each of the four capacitor sets, 16, 18, 21 and 23 is connected by only one pole to either the aerial or to the collector conductor.   The second poles 17, 19, 22 and 24 are open.  In the case of plate capacitors having no inductive resistance, an induction coil is inserted.  The object of such a spiral or coil is the displacement of phase of the induction current by 1/4 periods, whilst the charging current of the capacitor poles which lie free in the air, works back to the collector aerial.  The consequence of this is that in discharges in the collector aerial, the back-inductive action of the free poles allows a higher voltage to be maintained in the aerial collecting conductor than would otherwise be the case.  It has also been found that such a back action has an extremely favourable effect on the wear of the contacts.  Of course, the inductive effect may be regulated at will within the limits of the size of the induction coil, the length of the coil in action being adjustable by means of wire connection without induction (see Fig.24 No. 20).


S1 and S2 may also be provided with such regulating devices, in the case of S2 illustrated by 11.  If excess voltage be formed, it is conducted to earth through wire 12 and spark gap 8, or through any other suitable apparatus, since this voltage would be dangerous for the other components.   The action of these capacitor sets has already been described.


The small circles on the collector balloon indicate places where small patches of extremely thin layers (0.01 to 0.05 mm thick) of zinc amalgam, gold amalgam or other photoelectric acting metals, are applied to the balloon casing of light metal.  Such metallic patches may also be applied to the entire balloon as well as in greater thickness to the conducting network.  The capacity of the collector is thereby considerably strengthened at the surface.  The greatest possible effect in collecting may be obtained by polonium amalgams and the like.  On the surface of the collector balloon, metal points or spikes are also fixed along the ribs.  These spikes enhance the charge collection operation.  Since it is well known that the sharper the spikes, the less the resistance of the spikes, it is therefore extremely important to use spikes which are as sharp as possible.  Experiments have shown that the formation of the body of the spike or point also play a large part, for example, spikes made of bars or rollers with smooth surfaces, have point resistance many times greater than those with rough surfaces.  Various kinds of spike bodies have been experimented with for the collector balloons and the best results were given with spikes which were made in the following way:  Fine points made of steel, copper, nickel or copper and nickel alloys, were fastened together in bundles and then placed as anode with the points placed in a suitable electrolyte (preferably in hydrochloric acid or muriate of iron solutions) and so treated with weak current driven by 2 to 3 volts.  After 2 to 3 hours, according to the thickness of the spikes, the points become extremely sharp and the bodies of the spikes have a rough surface.  The bundle can then be removed and the acid washed off with water.  The spikes are then placed as cathode in a bath containing a solution of gold, platinum, iridium, palladium or wolfram salts or their compounds, and coated at the cathode galvanically with a thin layer of precious metal, which mush however be sufficiently firm to protect them from atmospheric oxidation.


Such spikes act at a 20 fold lower voltage almost as well as the best and finest points made by mechanical means.  Still better results are obtained if polonium or radium salts are added to the galvanic bath when forming the protective layer or coating.  Such pins have low resistance at their points and have excellent collector action even at one volt or lower.


In Fig.24, the three unconnected poles are not connected with one another in parallel.  That is quite possible in practice without altering the principle of the free pole.  It is also preferable to interconnect a series of collecting aerials in parallel to a common collector network.  Fig.25 shows such an arrangement.  A1, A2, A3, A4 are four metal collector balloons with gold or platinum coated spikes which are electrolytically mad in the presence of polonium emanations or radium salts, the spikes being connected over four electromagnets S1, S2, S3, S4, through an annular conductor R.  From this annular conductor, four wires run over four further electromagnets Sa, Sb, Sc, Sd, to the connecting point 13.  There, the conductor is divided, one branch passing over 12 and the safety spark gap 7 to the earth at E1, the other over inductive resistance J and working spark gap 7 to the earth at E2.  The working circuit, consisting of the capacitors 5 and 6 and a resonance motor or a capacitor motor M, such as already described, is connected in proximity around the sparking gap section 7.  Of course, instead of connecting the capacitor motor directly, the primary circuit for high frequency oscillatory current may also be inserted.


The capacitor sets are connected by one pole to the annular conductor R and can be either inductionless (16 and 18) or made as induction capacitors as shown by 21 and 23.  The free poles of the inductionless capacitors are indicated by 17 and 19, and those of the induction capacitors by 22 and 24.  As may be seen from the drawings, all of these poles 17, 22, 19 and 24 may be interconnected in parallel through a second annular conductor without any fear that thereby the principle of the free pole connection will be lost.  In addition to the advantages already mentioned, the parallel connection also allows an equalisation of the working voltage in the entire collector network.  Suitably calculated and  constructed induction coils 25 and 26 may also be inserted in the annular conductor of the free poles, by means of which, a circuit may be formed in the secondary coils 27 and 28 which allows current produced in this annular conductor by fluctuations of the charges, to be measured or otherwise utilised.


According to what has already been stated, separate collector balloons may be connected at equidistant stations distributed over the whole country, either connected directly with one another metallically or by means of intermediate suitably connected capacitor sets through high voltage conductors insulated from earth.  The static electricity is converted through a spark gap, into high frequency dynamic electricity which may be utilised as a source of energy by means of a suitable connection method, various precautions being observed, and with special regulations.  The wires leading from the collector balloons, have up to now been connected through an annular conductor without this endless connection, which can be regarded as an endless induction coil, being able to exert any action on the whole conductor system.


It has now been found that if the network conductor connecting the aerial collector balloons with one another, is not made as a simple annular conductor, but preferably short-circuited in the form of coils over a capacitor set or spark gap or through thermionic valves, then the total collecting network exhibits quite new properties.  The collection of atmospheric electricity is thereby not only increased but an alternating field may easily be produced in the collector network.  Further, the atmospheric electrical forces showing themselves in the higher regions, may also be obtained directly by induction.  In Fig.26 and Fig.28, a form of construction is shown, on the basis of which, the further foundations of the method will be explained in more detail.


In Fig.26, 1,2,3 and 4 are metallic collector balloons, with 5, 6, 7 and 8 their metallic aerial conductors and I the actual collector network.  This consists of five coils and is mounted on high voltage insulators in the air, on high voltage masts (or with a suitable construction of cable, embedded in the earth).  One coil has a diameter of 1 to 100 km. or more.  S and S1 are two protective electromagnets, F is the second safety section against excess voltage, E its earth conductor and E1 the earth conductor of the working section.  When an absorption of static atmospheric electricity is effected through the four balloon collectors, in order to reach the earth connection E1, the current must flow spirally through the collector network, over the electromagnet S, primary induction coil 9, conductor 14, anode A of the audion tube, incandescent cathode K, as the way over the electromagnet and safety spark gap F offers considerably greater resistance.  Owing to the fact that the accumulated current flows in one direction, an electromagnetic alternating field is produced in the interior of the collector network coil, whereby all of the free electrons are directed more or less into the interior of the coil.  An increased ionisation of the atmosphere is therefore produced.  Consequently, the points mounted on the collector balloon, show a considerably reduced resistance and therefore increased static charges are produced  between the points on the balloon and the surrounding atmosphere.  This results in a considerably increased collector effect.


A second effect, which could not be achieved in any other way, is obtained by the alternating electromagnetic field running parallel to the earth’s surface, which acts more or less with a diminishing or increasing effect on the earth’s magnetic field, whereby in the case of fluctuations in the current, a return induction current of reversed sign is always produced in the collector coil by earth magnetism.  Now if a constantly pulsating, continuous alternating field is produced as stated in the collector network I, an alternating current of the same frequency is also produced in the collecting network coil.  As the same alternating field is further transmitted to the aerial balloon, the resistance of its points is thereby considerably reduced, while the collector action is considerably increased.  A further advantage is that positive charges which collect on the metal surfaces during the conversion into dynamic current, produce a so-called voltage  drop in the collector area.  As an alternating field is present, when discharge of the collector surfaces takes place, the negative ions surrounding the collector surfaces produce, by the law of induction, an induction of reversed sign on the collector surface - that is, a positive charge.  In addition to the advantages already stated, the construction of connecting conductors in coil form, when of sufficiently large diameter, allows a utilisation of energy arising in higher regions, also in the most simple way.  As is well known, electric discharges frequently take place at very great elevations which may be observed, such as ‘St. Elmo’s fires’ or ‘northern lights’.  These energy quantities have not been able to have been utilised before now.  By this invention, all of these kinds of energy, as they are of electromagnetic nature and since the axis of the collector coils is at right angles to the earth’s surface, can be absorbed in the same way as a radio absorbs distant radio signals.  With a large diameter of the spiral, it is possible to connect large surfaces and thereby take up large quantities of energy.


It is well known that in the summer months and in the tropics, large radio stations are very frequently unable to receive signals due to interruptions caused by atmospheric  electricity, and this takes place with vertical coils of only 40 to 100 metres in diameter.  If, on the contrary, horizontal coils of 1 to 100 kilometres in diameter are used, very strong currents may be obtained through discharges which are constantly taking place in the atmosphere.  Particularly in the tropics, or still better in the polar regions where the northern lights are constantly present, large quantities of energy may probably be obtained in this way.  A coil with several windings should perform the best.  In a similar manner, any alteration of the earth’s magnetic field should act inductively on such a coil.


It is not at all unlikely that earthquakes and sunspots will also produce an induction in collector coils of that size.  In similar manner, this collector conductor will react to earth currents more particularly when they are near the surface of the earth or even embedded in the earth.  By combining the previous kind of current collectors, so far as they are adapted for the improved system with the improved possibilities of obtaining current, the quantities of free natural energy which are to be obtained in the form of electricity are considerably increased.


In order to produce uniform undamped current oscillations in the improved collector coil, so-called audion high vacuum or thermionic valves are used instead of the previous described spark gaps (Fig.26, 9-18).  The main aerial current flows through electromagnet S (which in the case of a high number of alternations is not connected here but in the earth conductor E1) and may be conveyed over the primary coils in the induction winding through wire 14 to the anode A of the high vacuum grid valve.  Parallel with the induction resistance 9, a regulating capacity of suitable size, such as capacitor 11, is inserted.  In the lower part of the vacuum grid valve is the incandescent filament cathode K which is fed through a battery B.  From the battery, two branches run, one to the earth conductor E1 and the other through battery B1 and secondary coil 10 to the grid anode g of the vacuum tube.  By the method of connections shown in dotted lines, a desired voltage may also be produced at the grid electrode g through wire 17 which is branched off from the main current conductor through switches 16 and some small capacitors (a, b, c, d) connected in series, and conductor 18, without the battery B1 being required.  The action of the whole system is somewhat as follows:-


On the connecting conductor of the aerial collector network being short-circuited to earth, the capacitor pole 11 is charged, and slightly dampened oscillations are formed in the short-circuited oscillation circuit formed by capacitor 11 and self inductance 9.  Because of the coupling through coil 10, voltage fluctuations of the same frequency take place in the grid circuit 15 and in turn, these fluctuations influence the strength of the electrode current passing through the high vacuum amplifying valve and thus produce current fluctuations of the same frequency in the anode circuit.  A permanent supply of energy.  Consequently, a permanent supply of energy is supplied to the oscillation circuits 9 and 10 takes place, until a balance is achieved where the oscillation energy consumed exactly matches the energy absorbed.  This produces constant undamped oscillations in the oscillation circuits 9 - 11.


For regular working of such oscillation producers, high vacuum strengthening tubes are necessary and it is also necessary that the grid and anode voltages shall have a phase difference of 1800 so that if the grid is negatively charged, then the anode is positively charged and vice versa.  This necessary difference of phase may be obtained by most varied connections, for example, by placing the oscillating circuit in the grid circuit or by separating the oscillation circuit and inductive coupling from the anodes and the grid circuit, and so forth.


A second important factor is that care must be taken that the grid and anode voltages have a certain relation to one another; the latter may be obtained by altering the coupling and a suitable selection of the self induction in the grid circuit, or as shown by the dotted lines 18, 17, 16 by means of a larger or smaller number of capacitors of suitable size connected in series; in this case, the battery B1 may be omitted.  With a suitable selection of the grid potential, a glow discharge takes place between the grid g and the anode A, and accordingly at the grid there is a cathode drop and a dark space is formed.  The size of this cathode drop is influenced by the ions which are emitted in the lower space in consequence of shock ionisation of the incandescent cathodes K and pass through the grid in the upper space.  On the other hand, the number of the ions passing through the grid is dependent on the voltage between the grid and the cathode.  Thus, if the grid voltage undergoes periodic fluctuations (as in the present case), the amount of the cathode drop at the grid fluctuates, and consequently, the internal resistance of the valve fluctuates correspondingly, so that when a back-coupling of the feed circuit with the grid circuit takes place, the necessary means are in place for producing undamped oscillations and of taking current as required, from the collecting conductor.


With a suitably loose coupling, the frequency of the undamped oscillations produced is equal to the self-frequency of the oscillation circuits 9 and 10.  By selecting a suitable self-induction for coil 9 and capacitor 11, it is possible to extend operation from frequencies which produce electromagnetic oscillations with a wavelength of only a few metres, down to the lowest practical alternating current frequency.  For large installations, a suitable number of frequency producing tubes in the form of the well known high vacuum transmission tubes of 0.5 kW to 2 kW in size may be connected in parallel so that in this respect, no difficulty exists.


The use of such tubes for producing undamped oscillations, and the construction and method of inserting such transmission tubes in an accumulator or dynamo circuit is known, also, such oscillation producing tubes only work well at voltages of 1,000 volts up to 4,000 volts, so that on the contrary, their use at lower voltages is considerably more difficult.  By the use of high voltage static electricity, this method of producing undamped oscillations as compared with that through spark gaps, must be regarded as an ideal solution, particularly for small installations with outputs from 1 kW to 100 kW.


By the application of safety spark gaps, with interpolation of electromagnets, not only is short-circuiting avoided but also the taking up of current is regulated.  Oscillation producers inserted in the above way, form a constantly acting alternating electromagnetic field in the collector coil, whereby, as already stated, a considerable accumulating effect takes place.  The withdrawal or ‘working’ wire is connected at 12 and 13, but current may be taken by means of a secondary coil which is firmly or moveably mounted in any suitable way inside the large collector coil, i.e. in its alternating electromagnetic field, so long as the direction of its axis is parallel to that of the main current collecting coil.


In producing undamped oscillations of a high frequency (50 KHz and more) in the oscillation circuits 9 and 11, electromagnets S and S1 must be inserted if the high frequency oscillations are not to penetrate the collector coil, between the oscillation producers and the collector coil.  In all other cases they are connected shortly before the earthing (as in Fig.27 and Fig.28).



In Fig.27 a second method of construction of the connecting conductor of the balloon aerials is illustrated in the form of a coil.  The main difference is that in addition to the connecting conductor I another annular conductor II is inserted parallel to the former on the high voltage masts in the air (or embedded as a cable in the earth) but both in the form of a coil.  The connecting wire of the balloon aerials is both a primary conductor and a current producing network while the coil is the consumption network and is not in unipolar connection with the current producing network.


In Fig.27 the current producing network I is shown with three balloon collectors 1, 2, 3 and aerial conductors 4, 5, 6; it is short-circuited through capacitor 19 and inductor 9.  The oscillation forming circuit consists of spark gap f, inductor 10 and capacitor 11.  The earth wire E is connected to earth through electromagnet S1.  FI is the safety spark gap which is also connected to earth through a second electromagnet SII at EII.  On connecting up the capacitor circuit 11 it is charged over the spark gap f and an oscillatory discharge is formed.  This discharging current acts through inductor 10 on the inductively coupled secondary 9, which causes a change in the producing network, by modifying the voltage on capacitor 19.  This causes oscillations in the coil-shaped producer network.  These oscillations induce a current in the secondary circuit II, which has a smaller number of windings and lower resistance, consequently, this produces a lower voltage and higher current in it.


In order to convert the current thus obtained, into current of an undamped character, and to tune its wavelengths, a sufficiently large regulatable capacitor 20 is inserted between the ends 12 and 13 of the secondary conductor II.  Here also, current may be taken without an earth conductor, but it is advisable to insert a safety spark gap E1 and to connect this with the earth via electromagnet S2.  The producer network may be connected with the working network II over an inductionless capacitor 21 or over an induction capacitor 22, 23.  In this case, the secondary conductor is unipolarly connected with the energy conductor.




In Fig.28, the connecting conductor between the separate collecting balloons is carried out according to the autotransformer principle.  The collecting coil connects four aerial balloons 1, 2, 3, 4, the windings of which are not made side-by-side but one above the other.  In Fig.28, the collector coil I is shown with a thin line and the metallically connected prolongation coils II with a thick line.  Between the ends I1 and II1 of the energy network I, a regulating capacitor 19 is inserted.  The wire I1 is connected with the output wire and with the spark gap F.


As transformer of the atmospheric electricity, an arrangement is employed which consists of using rotary pairs of capacitors in which the stator surface B is connected with the main current, while the other A is connected to the earth pole.  These pairs of short-circuited capacitors are caused to rotate and the converted current can be taken from them via two collector rings and brushes.  This current is alternating current with a frequency dependent on the number of balloons and the rate of revolutions of the rotor.  As the alternating current formed in the rotor can act through coils 10 on the inductor 9, an increase or decrease of the feed current in I can be obtained according to the direction of the current by back-induction.  Current oscillations of uniform rhythm are produced in the coil-shaped windings of the producer network.


As the ends of this conductor are short-circuited through the regulatable capacitor 19, these rhythms produce short-circuited undamped oscillations in the energy conductor.  The frequency of these oscillations can be altered at will by adjusting the capacitance of capacitor 19.  These currents may also be used as working current via the conductors II1 and III.  By inserting capacitor 20, a connection between these conductors may also be made, whereby harmonic oscillations of desired wavelength are formed.  By this means, quite new effects as regards current distribution are obtained.  The withdrawal of current can even take place without direct wire connection if, at a suitable point in the interior of the producing network (quite immaterially whether this has a diameter of 1 or 100 km) a coil tuned to these wavelength and of the desired capacity, is firmly or moveably mounted in the aerial conductor in such a way that its axis is parallel with the axis of the collector coil.  In this case, a current is induced in the producing network, the size of which is dependent on the total capacity and resistance and on the frequency selected.  A future possibility is taking energy from the producer network by radio signals as in addition to atmospheric electricity, magnetic earth currents and energy from the upper atmosphere may be tapped.


Of course, vacuum tubes may be used to produce undamped oscillations anywhere spark gaps are shown in the circuits.  The separate large-diameter coils of the producer network may be connected to one another through separate conductors all in parallel or all in series or in groups in series.  By regulating the number of oscillations and the magnitude of the voltage, more or fewer large collector coils of this kind may be used.  The coils may also be divided spirally over the entire section.  The coils may be carried out in annular form or in triangular, quadrangular, hexagonal or octagonal form.


Of course, wires which form guides for the current waves, may be carried from a suitable place to the centre or also laterally.  This is necessary when the currents have to be conducted over mountains and valleys and so forth.  In all these cases, the current must be converted into a current of suitable frequency.


As already mentioned, separate collecting balloons may be directly metallically interconnected a equidistant stations distributed over the entire country, or may be connected by interpolation of suitable capacitor sets by means of high voltage conductors.  The static electricity is converted through a spark gap into dynamic energy of high frequency and could then in that form be used as an energy source after special regulation.


According to this invention, in order to increase the collecting effect of the balloon in the aerial collector conductor or in the earth wire, radiating collectors are used.  These consist of either incandescent metal or oxide electrodes in the form of vacuum grid valves, or electric arcs (mercury or similar electrodes), Nernst lamps, or flames of various kinds maybe simply connected with the respective conductor.


It is well known that energy can be drawn off from a cathode consisting of an incandescent body opposite an anode charged with positive electricity (vacuum grid tube).  Hitherto however, a cathode was always first directly placed opposite an anode, and secondly, the system always consisted of a closed circuit.


Now if we dispense with the ordinary ideas in forming light or flame arcs in which a cathode must always stand directly opposite an anode charged to a high voltage or another body freely floating in the air, or consider the incandescent cathode to be only a source of unipolar discharge, (which represents group and point discharges in electro-static machines similar to unipolar discharges), it may be ascertained that incandescent cathodes and less perfectly, all incandescent radiators, flames and the like, have relatively large current densities and allow large quantities of electric energy to radiate into open space in the form of electron streams as transmitters.


The object of this invention is as described below, if such incandescent oxide electrodes or other incandescent radiators or flames are not freely suspended in space but instead are connected metallically with the earth so that they can be charged with negative terrestrial electricity, these radiators possess the property of absorbing the free positive electrical charges contained in the air space surrounding them (that is to say, of collecting them and conducting them to earth).  They can therefore serve as collectors and have in comparison to the action of the spikes, a very large radius of action R; the effective capacity of these collectors is much greater than the geometrical capacity (R0) calculated in an electro-static sense.


As is well known, our earth is surrounded with an electro-static field and the difference of potential dV/dh of the earth field according to the latest investigations, is in summer about 60 to 100 volts, and in winter, 300 to 500 volts per metre difference in height, a simple calculation gives the result that when such a radiation collector or flame collector is arranged, for example, on the ground, and a second one is mounted vertically over it at a distance of 2,000 metres and both are connected by a conducting cable, there is a voltage difference in summer of about 2,000,000 volts and in winter 6,000,000 volts or more.


According to Stefan Boltzmann’s law of radiation, the quantity of energy which an incandescent surface (temperature T) of 1 sq. cm. radiates in a unit of time into the open air (temperature T0) is expressed by the following formula:


S = R (T4 -T04)  watts per square centimetre


and the universal radiation constant R, according to the latest researches of Ferry, is equal to 6.30 x 10-12 watts per square centimetre.


Now, if an incandescent surface of 1 sq. cm., as compared to the surrounding space, shows a periodic fall of potential dV, it radiates (independent of the direction of the current) in accordance with the above formula, for example at a temperature of 37150 C. an energy of 1.6 kW per square centimetre.  As for the radiation, the same value can be calculated for the collection of energy, but reversed.  Now, as carbon electrodes at the temperature of the electric arc, support a current density up to 60 to 65 amps per sq. cm., no difficulties will result in this direction in employing radiating collectors as accumulators.


If the earth be regarded as a cosmically insulated capacitor in the sense of geometrical electro-statics x, according to Chwolson, there results from the geometric capacity of the earth:


    For negative charging 1.3 x 106 Coulomb For negative potential  V = 10 x 108 volts.


It follows from this that EJT is approximately equal to 24.7 x 1024 watts/sec.  Now if it is desired to make a theoretical short circuit through an earthed flame collector, this would represent an electrical total work of about 79,500 x 1010 kilowatt years.  As the earth must be regarded as a rotating mechanism which is thermo-dynamically, electromagnetically and kinematically coupled with the sun and star system by cosmic radiation and gravitation, a reduction in the electric energy of the earth field is not to be feared.  The energies which the incandescent collectors could withdraw from the earth field can only cause a lowering of the earth temperature.  This is however, not the case as the earth does not represent a cosmically entirely insulated system.  On the contrary, there is conveyed from the sun to the earth an energy of 18,500 x 1010 kilowatts.  Accordingly, any lowering of the earth temperature without a simultaneous lowering of the sun’s temperature would contradict Stefan Boltzmann’s law of radiation.


From this it must be concluded that if the earth temperature sinks, the total radiation absorbed by the earth increases, and further, the rate of cooling of the earth is directly dependent on that of the sun and the other radiators cosmically coupled with the sun.


The incandescent radiation collectors may, according to this invention, be used for collecting atmospheric electricity if they (1) are charged with the negative earth electricity (that is to say, when they are directly connected to the earth by means of a metallic conductor) and (2) if large capacities (metal surfaces) charged with electricity are mounted opposite them as positive poles in the air.  This is regarded as the main feature of the present invention as without these inventive ideas it would not be possible to collect with an incandescent collector, sufficiently large quantities of the electrical charges contained in the atmosphere as technology requires; the radius of action of the flame collectors would also be too small, especially if it be considered that the very small surface density does not allow of large quantities of charge being absorbed from the atmosphere.


It has already been proposed to employ flame collectors for collecting atmospheric electricity and it is known that their collecting effect is substantially greater opposite the points.  It is however, not known that the quantities of current which hitherto be obtained are too small for technical purposes.  According to my experiments, the reason for this is to be found in the inadequate capacities of the collector conductor poles.  If such flame or radiating collectors have no or only small positive surfaces, their radius of action for large technical purposes is too small.  If the incandescent collectors be constantly kept in movement in the air, they may collect more according to the speed of the movement, but this is again not capable of being carried out in practice.


By this invention, the collector effect is considerably increased by a body charged with a positive potential and of the best possible capacity, being also held floating (without direct earth connection) opposite such an incandescent collector which is held floating in the air at a desired height.  If, for example, a collecting balloon of sheet metal or metallised fabric, be caused to mount to 300 to 3,000 metres in the air, and as a positive pole it is brought opposite such a radiating collector connected by a conductor to earth, quite different results are obtained.


The metallic balloon shell which has a large surface area is charged to a high potential by atmospheric electricity.  This potential is greater the higher the collecting balloon is above the incandescent collector.  The positive electricity acts concentratedly on the anode floating in the air as it is attracted through the radiation shock ionisation, proceeding from the incandescent cathode.  The consequence of this is that the radius of action of the incandescent cathode collector is considerably increased and so is the collecting effect of the balloon surface.  Further, the large capacity of the anode floating in the air, plays therefore an important part because it allows the collection of large charges resulting in a more uniform current even when there is substantial current withdrawal - this cannot be the case with small surfaces.


In the present case, the metallic collecting balloon is a positive anode floating in the air and the end of the earth conductor of this balloon serves as positive pole surface opposite the surface of the radiating incandescent cathode, which in turn is charged with negative earth electricity as it is connected to the earth by a conductor.  The process may be carried out by two such contacts (negative incandescent cathode and anode end of a capacity floating in the air) a capacitor and an inductive resistance being switched on in parallel, whereby simultaneously undamped oscillations may be formed.


In very large installations it is advisable to connect two such radiating collectors in series.  Thus an arc light incandescent cathode may be placed below on the open ground and an incandescent cathode which is heated by special electro-magnetic currents, be located high in the air.  Of course for this, the special vacuum Liebig tubes with or without grids may also be used.  An ordinary arc lamp with oxide electrodes may be introduced on the ground and the positive pole is not directly connected with the collecting balloon, but through the upper incandescent cathode or over a capacitor.  The method of connecting the incandescent cathode floating in the air may be seen in Figs.29-33.


B is the air balloon, K a Cardan ring (connection with the hawser) C the balloon, L a good conducting cable, P a positive pole, N negative incandescent cathode and E the earth conductor.


Fig.29 represents the simplest form of construction.  If electric oscillations are produced below on the ground by means of a carbon arc lamp or in any other suitable way, a considerably greater electric resistance is opposed to that in the direct way by inserting an electrical inductive resistance 9.  Consequently, between P and N, a voltage is formed, and as, over N and P only an inductionless ohmic resistance is present, a spark will spring over so long as the separate induction coefficients and the like are correctly calculated.  The consequence of this is that the oxide electrode (carbon or the like) is rendered incandescent and then shows as incandescent cathode, an increased collecting effect.  The positive poles must be substantially larger  than the negative in order that they may not also become incandescent.  As they are further connected with the large balloon area which has a large capacity and is charged at high voltage, an incandescent body which is held floating in the air and a positive pole which can collect large capacities is thereby obtained in the simplest way.  The incandescent cathode is first caused to become incandescent by means of separate energy produced on the earth, and then maintained by the energy collected from the atmosphere.



Fig.30 only shows the difference that instead of a round balloon, a cigar-shaped one may be used, also, a capacitor 5 is inserted between the incandescent cathode and the earth conductor so that a short-circuited oscillation circuit over P N 5 and 9 is obtained.  This has the advantage that quite small quantities of electricity cause the cathode to become incandescent and much larger cathode bodies may be made incandescent.


In this form of construction, both the incandescent cathode and the positive electrode may be enclosed in a vacuum chamber as shown in Fig.32.  A cable L is carried well insulated through the cover of a vessel and ends in a capacitor disc 5.  The cover is arched in order to keep the rain off.  The vessel is entirely or partially made of magnetic metal and well insulated inside and outside.  Opposite disc 5 another disc 6 and on this again a metallic positive pole of the vacuum tube g with the incandescent cathode (oxide electrode) N is arranged.  The negative electrode is on the one hand connected to the earth conductor E, and on the other hand with the inductive resistance 9 which is also connected with the cable L with the positive pole and wound around the vessel in coils.  The action is exactly the same as that in Fig.29 only instead of an open incandescent cathode, one enclosed in vacuo is used.  As in such collectors, only small bodies be brought to incandescence, in large installations a plurality of such vacuum tubes must be inserted in proximity to one another.  According to the previous constructions Fig.31 and Fig.33 are quite self evident without further explanations.




Figs.34-37 represent further diagrams of connections over radiating and flame collectors, and in fact, how they are to be arranged on the ground.  Fig.34 shows an arc light collector with oxide electrodes for direct current and its connection.  Fig.35 shows a similar one for alternating current.  Fig.36 an incandescent collector with a Nernst lamp and Fig.37 a similar one with a gas flame.


The positive pole 1 of the radiating collectors is always directly connected to the aerial collecting conductor A.  In Fig.34, this is further connected over the capacitor set 5 with a second positive electrode 3.  The direct current dynamo b produces current which flows over between the electrodes 3 and 2 as an arc light.  On the formation of an arc, the negative incandescent electrode 2 absorbs electricity from the positive poles standing opposite it and highly charged with atmospheric electricity which it conveys to the working circuit.  The spark gap 7, inductive resistance 9  and induction coil 10 are like the ones previously described.  The protective electromagnet S protects the installation from earth circuiting and the safety spark gap 8 from excess voltage or overcharging.


In Fig.35, the connection is so far altered that the alternating current dynamo feeds the excitation coil 11 of the induction capacitor.  12 is its negative and 13 its positive pole.  If the coil 3 on the magnet core of the dynamo is correctly calculated and the frequency of the alternating current sufficiently high, then an arc light can be formed between poles 1 and 2.  As the cathode 2 is connected to the negatively charged earth, and therefore always acts as a negative pole, a form of rectification of the alternating current produced by the dynamo 3 is obtained, since the second half of the period is always suppressed.  The working circuit may be carried out in the same way as in Fig.34; the working spark gap 7 may however be dispensed with, and instead of it, between the points n and m, a capacitor 5 and an induction resistance 9 may be inserted, from which, a current is taken inductively.


Fig.36 represents a form of construction similar to that shown in Fig.34 except that here instead of an arc lamp, a Nernst incandescent body is used.  The Nernst lamp is fed through the battery 3.  The working section is connected with the negative pole, the safety spark gap with the positive poles.  The working spark gap 7 may also be dispensed with and the current for it taken at 12 over the oscillation circuit 5, 11 (shown in dotted lines).


Flame collectors (Fig.37) may also be employed according to this invention.  The wire network 1 is connected with the aerial collector conductor A and the burner with the earth.  At the upper end of the burner, long points are provided which project into the flame.  The positive electrode is connected with the negative over a capacitor 5 and the induction coil 9 with the earth.


The novelty in this invention is:


(1) The use of incandescent cathodes opposite positive poles which are connected to large metallic capacities as automatic collecting surfaces.

(2) The connection of the incandescent cathodes to the earth whereby, in addition to the electricity conveyed to them from the battery of machine which causes the incandescing, also the negative charge of the earth potential is conveyed, and

(3) The connection of the positive and negative poles of the radiating collectors over a capacitor circuit alone or with the introduction of a suitable inductive resistance, whereby simultaneously an oscillatory oscillation circuit may be obtained.  The collecting effect is by these methods quite considerably increased.















Patent  GB1913,01098           14th January 1914              Inventor: Roy J. Meyers







A rectifier for use with apparatus for producing electricity from the earth consists of mercury- vapour lamps constructed and arranged as shown in Fig.4.  Each lamp comprises two wires 6<1>, 7<1> wound around a steel tube 15 surrounding a mercury tube 11 preferably of copper.  The coil 6<1> is connected between the electrode 14 and the terminal 18, and the coil 7<1> between the terminals 19, 5.  The coils 6<1>, 7<1> are preferably composed of soft iron.




This invention relates to improvements in apparatus for the production of electrical currents, and the primary object in view is the production of a commercially serviceable electrical current without the employment of mechanical or chemical action.  To this end the invention comprises means for producing what I believe to be dynamic electricity from the earth and its ambient elements.


I am, of course aware that it has been proposed to obtain static charges from upper strata of the atmosphere, but such charges are recognised as of widely variant potential and have thus far proved of no practical commercial value, and the present invention is distinguished from all such apparatus as has heretofore been employed for attracting static charges by the fact that this improved apparatus is not designed or employed to produce or generate irregular, fluctuating or other electrical charges which lack constancy, but on the other hand I have by actual test been able to produce from a very small apparatus at comparatively low elevation, say about 50 or 60 feet above the earth’s surface, a substantially constant current at a commercially usable voltage and amperage.


This current I ascertained by repeated tests is capable of being readily increased by additions of the unit elements in the apparatus described below, and I am convinced from the constancy of the current obtained and its comparatively low potential that the current is dynamic and not static, although, of course, it is not impossible that certain static discharges occur and, in fact, I have found occasion to provide against the damage which might result from such discharge by the provision of lightning arresters and cut-out apparatus which assist in rendering the obtained current stable by eliminating sudden fluctuations which sometimes occur during conditions of high humidity from what I consider static discharges. 


The nature of my invention is obviously such that I have been unable to establish authoritatively all of the principles involved, and some of the theories herein expressed may possibly prove erroneous, but I do know and am able to demonstrate that the apparatus which I have discovered does produce, generate, or otherwise acquire a difference of potential representing a current amperage as stated above.


The invention comprises the means for producing electrical currents of serviceable potential substantially without the employment of mechanical or chemical action, and in this connection I have been able to observe no chemical action whatever on the parts utilised although deterioration may possibly occur in some of the parts, but so far as I am able to determine such deterioration does not add to the current supply but is merely incidental to the effect of climatic action.


The invention more specifically comprises the employment of a magnet or magnets and a co-operating element, such as zinc positioned adjacent to the magnet or magnets and connected in such manner and arranged relative to the earth so as to produce current, my observation being that current is produced only when such magnets have their poles facing substantially to the north and south and the zincs are disposed substantially along the magnets.


The invention also comprehends other details of construction, combinations and arrangements of parts as will be fully set forth.





Fig.1 is a plan view of an apparatus embodying the features of the present invention, the arrow accompanying the figure indicating substantially the geographical north, parts of this figure are diagrammatic.


Fig.2 is a view is side elevation of the parts seen in plan in Fig.1

Fig.3 is a vertical section taken on the plane indicated by the line A--A of Fig.2.




Fig.4 is a detail view, partly in elevation and partly in section, showing the connections of the converter and intensifier.








Fig.5 is a transverse section taken on the planes indicated by line 5-5 of Fig.4, looking downwards.


Fig.6 is an enlarged detail fragmentary section illustrating the parts at the junction of the conductors and one of the intensifiers.



Fig.7 is an enlarged detail view partly in elevation and partly in section of one of the automatic cut-outs


Fig.8 is a diagrammatic view of one of the simplest forms of embodiment of the invention.


Referring to the drawing by numerals, 1,1 indicates magnets connected by a magnetic substance 2, preferably an iron wire.  The magnets 1 are arranged in pairs, one pair being spaced beneath the other, and interposed between the magnets are zinc plates 3,3 connected by an iron wire conductor 4.  Suitable insulating supports 5 are arranged for sustaining the respective magnets 1 and plates 3,3.  Each plate 3 is preferably bent substantially into V form, as clearly seen in Fig.1, and the V of one of the plates opens or faces toward the North and the V of the other plate to the South.  I have determined by experimentation that it is essential that the plates 3 be disposed substantially North and South with their flat faces approximately parallel to the adjacent faces of the co-operating magnets, although by experience I have not discovered any material difference in the current obtained when the plates are disposed slightly to one side of North and South, as for instance when the plates are disposed slightly to one side of North and South, as for instance when disposed in the line of the magnetic polarity of the earth.  The same is true with respect to the magnets 1, the said magnets being disposed substantially North and South for operative purposes, although I find that it is immaterial whether the North pole of one of the magnets is disposed to the North and the South pole to the South, or vice versa, and it is my conviction from experience that it is essential to have the magnets of each pair connected by magnetic material so that the magnets substantially become one with a pole exposed to the North and a pole exposed to the South. 



In Fig.1, I have indicated in full lines by the letters 8 and N the respective polarities of the magnets 1, and have indicated in dotted lines the other pole of those magnets when the connection 2 is severed.  I have found that the magnets and zinc plates operate to produce, (whether by collection or generation I am not certain), electrical currents when disposed substantially North and South, but when disposed substantially East and West, no such currents are produced.  I also find that the question of elevation is by no means vital, but it is true that more efficient results are obtained by placing the zincs and magnets on elevated supports.  I furthermore find from tests, that it is possible to obtain currents from the apparatus with the zincs and magnets disposed in a building or otherwise enclosed, although more efficient results are obtained by having them located in the open.


While in Figures 1, 2, and 3, I have shown the magnets and the zinc plates as superimposed, it will be apparent, as described in detail below, that these elements may be repositioned in horizontal planes, and substantially the same results will be secured.  Furthermore, the magnets 1 with the interposed zincs 3, as shown in Figures 1, 2 and 3 merely represent a unit which may be repeated either horizontally or vertically for increasing the current supply, and when the unit is repeated the zinc plates are arranged alternating with the magnets throughout the entire series as indicated below.


A conductor 6 is connected in multiple with the conductors 2 and a conductor 7 is connected with conductor 4, the conductor 6 extending to one terminal of a rectifier which I have indicated by the general reference character 8, and the conductor 7 extending to the other terminal of the rectifier. The rectifier as seen in the diagram Fig.1 may assume any of several well known embodiments of the electrical valve type and may consist of four asymmetric cells or Cooper-Hewitt mercury vapour lamps connected as indicated in Fig.1 for permitting communication of the positive impulses from the conductor 6 only to the line conductor 9 and the negative impulses from conductor 6 on only to the line conductor 10.  The current from this rectifier may be delivered through the conductors 9 and 10 to any suitable source for consumption.


While the said rectifier 8 may consist of any of the known types, as above outlined, it preferably consists of a specially constructed rectifier which also has the capacity of intensifying the current and comprises specifically the elements shown in detail in Figures 4, 5, and 6 wherein I have disclosed the detail wiring of the rectifier when composed of four of the rectifying and intensify in elements instead of asymmetric cells or simple mercury vapour valves.  As each of these structures is an exact embodiment of all the others, one only will be described, and the description will apply to all. The rectifying element of each construction consists of a mercury tube 11 which is preferably formed of glass or other suitable material, and comprises a cylinder having its end portions tapered and each terminating in an insulating plug or stopper 12.  Through the upper stopper 12 is extended the electrode 13 which extends well into the tube and preferably about one-half its length, to a point adjacent the inner end of an opposing electrode 14 which latter electrode extends from there down through the insulation 12 at the lower end of the tube.  The tube 11 is supplied with mercury and is adapted to operate on the principle of the mercury vapour lamp, serving to rectify current by checking back impulses of one sign and permitting passage of impulses of the other. 


To avoid the necessity for utilising a starter, as is common with the lamp type of electrical valve, the supply of mercury within the tube may be sufficient to contact with the lower end of the electrode 13 when current is not being supplied, so that as soon as current is passed from one electrode to the other sufficiently for volatilising that portion of the mercury immediately adjacent the lower end of electrode 13, the structure begins its operation as a rectifier. The tube 11 is surrounded by a tube 15 which is preferably spaced from tube 11 sufficiently for allowing atmospheric or other cooling circulation to pass the tube 11. In some instances, it may be desirable to cool the tube 11 by a surrounding body of liquid, as mentioned below.  The tube 15 may be of insulating material but I find efficient results attained by the employment of a steel tube, and fixed to the ends of the of the tube are insulating disks 16, 16 forming a spool on which are wound twin wires 6’ and 7’, the wire 6’ being connected at the inner helix of the coil with the outer end of the electrode 14, the lower portion of said electrode being extended to one side of the tube 11 and passed through an insulating sleeve 17 extending through the tube 15, and at its outer end merging into the adjacent end of the wire 6’.  The wire 7’ extends directly from the outer portion of the spool through the several helices to a point adjacent to the junction of the electrode 14 with wire 6’ and thence continues parallel to the wire throughout the coil, the wire 6’ ending in a terminal 18 and the wire 7’ ending in a terminal 19.


For the sake of convenience of description and of tracing the circuits, each of the apparatus just above described and herein known as an intensifier and rectifier will be mentioned as A, B, C and D, respectively. Conductor 6 is formed with branches 20 and 21 and conductor 7 is formed with similar branches 22 and 23. Branch 20 from conductor 6 connects with conductor 7’ of intensifier B and branch 21 of conductor 6 connects with the conductor 7’ of intensifier C, while branch 22 of conductor 7 of intensifier C, while branch 22 of conductor 7 connects with conductor 7’ of intensifier D.  A conductor 27 is connected to terminal 19 of intensifier A and extends to and is connected with the terminal 18 of intensifier C, and a conductor 7 connects with conductor 7’ of intensifier D.   A conductor 27 is connected to terminal 19 of intensifier A, and extends to and is connected to terminal 18 of intensifier C, and a conductor 28 is connected to the terminal 19 of intensifier C and extends from the terminal 19 of intensifier B to the terminal 18 of intensifier D to electrode 13 of intensifier B.   Each electrode 13 is supported on a spider 13’ resting on the upper disk 16 of the respective intensifier.   Conductors 31 and 32 are connected to the terminals 18 of intensifiers A and B and are united to form the positive line wire 9 which co-operates with the negative line wire 10 and extends to any suitable point of consumption.  The line wire 10 is provided with branches 35 and 36 extending to the electrodes 13 of intensifiers C and D to complete the negative side of the circuit.


Thus it will be seen that alternating currents produced in the wires 6 and 7 will be rectified and delivered in the form of a direct current through the line wires 9 and 10, and I find by experiment that the wires 6 and 7 should be of iron, preferably soft, and may of course be insulated, the other wiring not specified as iron being of copper or other suitable material.


In carrying out the operation as stated, the circuits may be traced as follows:  A positive impulse starting at the zincs 3 is directed along conductor 7 to branch 23 to conductor 7’ and the winding of the rectifier of intensifier B through the rectifier to the conductor 6’, through its winding to the contact 18, conductor 32 and to the line wire 9.   The next, or negative, impulse directed along conductor 7 cannot find its way along branch 23 and the circuit just above traced because it cannot pass across the rectifier of intensifier B but instead the negative impulse passes along conductor 22 to conductor 7 of intensifier A and its winding to the contact 19 and to conductor 27 to contact 18 of intensifier C, to the winding of the wire 6’ thereof to the electrode 14 through the rectifier to the of the electrode 13 and conductor of intensifier A, electrode 14 thereof and conductor 6’ to contact 18 and wire 31 to line wire 9.


Obviously the positive impulse cannot pass along the wire 20 because of its inverse approach to the rectifier of intensifier B.  The next impulse or negative impulse delivered to conductor 6 cannot pass along conductor 21 because of its connection with electrode 13 of the rectifier of intensifier A, but instead passes along conductor 20 to the wire 7’ and its winding forming part of intensifier B to the contact 19 and conductor 29 to contact 18 and the winding of wire 6’ of intensifier D to the electrode 14 and through the rectifier to the electrode 13 and conductor 35 to line wire 10.  Thus the current is rectified and all positive impulses directed along one line and all negative impulses along the other lie s that the potential difference between the two lines will be maximum for the given current of the alternating circuit.  It is, of course, apparent that a less number of intensifiers with their accompanying rectifier elements may be employed with a sacrifice of the impulses which are checked back from a lack of ability to pass the respective rectifier elements, and in fact I have secured efficient results by the use of a single intensifier with its rectifier elements, as shown below.


Grounding conductors 37 and 38 are connected respectively with the conductors 6 and 7 and are provided with the ordinary lightning arresters 39 and 40 respectively for protecting the circuit against high tension static charges.


Conductors 41 and 42 are connected respectively with the conductors 6 and 7 and each connects with an automatic cut-out 43 which is grounded as at 4.   Each of the automatic cut-outs is exactly like the other and one of the these is shown in detail in Fig.7 and comprises the inductive resistance 45 provided with an insulated binding post 46 with which the respective conductor 6 or 7 is connected, the post also supporting a spring 48 which sustains an armature 49 adjacent to the core of the resistance 45.  The helix of resistance 45 is connected preferably through the spring to the binding post at one end and at the other end is grounded on the core of the resistance, the core being grounded by ground conductor 44 which extends to the metallic plate 52 embedded in moist carbon or other inductive material buried in the earth.  Each of the conductors 41, 42 and 44 is of iron, and in this connection I wish it understood that where I state the specific substance I am able to verify the accuracy of the statement by the results of tests which I have made, but of course I wish to include along with such substances all equivalents, as for instance, where iron is mentioned its by-products, such as steel, and its equivalents such as nickel and other magnetic substances are intended to be understood.


The cut-out apparatus seen in detail in Fig.7 is employed particularly for insuring against high voltage currents, it being obvious from the structure shown that when potential rises beyond the limit established by the tension of the spring sustaining the armature 40, the armature will be moved to a position contacting with the core of the cut-out device and thereby directly close the ground connection for line wire 41 with conductor 44, eliminating the resistance of winding 45 and allowing the high voltage current to be discharged to the ground.  Immediately upon such discharge the winding 45 losing its current will allow the core to become demagnetised and release the armature 49 whereby the ground connection is substantially broken leaving only the connection through the winding 45 the resistance of which is sufficient for insuring against loss of low voltage current.


In Fig.8 I have illustrated an apparatus which though apparently primitive in construction and arrangement shows the first successful embodiment which I produced in the course of discovery of the present invention, and it will be observed that the essential features of the invention are shown there.  The structure shown in the figure consists of horseshoe magnets 54, 55, one facing North and the other South, that is, each opening in the respective directions indicated and the two being connected by an iron wire 55 which is uninsulated and wrapped about the respective magnets each end portion of the wire 55 being extended from the respective magnets to and connected with, as by being soldered to, a zinc plate 56, there being a plate 56 for each magnet and each plate being arranged longitudinally substantially parallel with the legs of the magnet and with the faces of the plate exposed toward the respective legs of the magnet, the plate being thus arranged endwise toward the North and South.   An iron wire 57 connects the plates 56, the ends of the wire being preferably connected adjacent the outer ends of the plates but from experiment I find that the wire may be connected at practically any point to the plate.  Wires 58 and 59 are connected respectively with the wires 55 and 57 and supply an alternating current at a comparatively low voltage, and to control such current the wires 58 and 59 may be extended to a rectifier or combined rectifier and intensifier, as discussed above.


The tests which I have found successful with the apparatus seen in Fig.8 were carried out by the employment first of horseshoe magnets approximately 4 inches in length, the bar comprising the horseshoe being about one inch square, the zincs being dimensioned proportionately and from this apparatus with the employment of a single intensifier and rectifier, as above stated, I was able to obtain a constant output of 8 volts.


It should be obvious that the magnets forming one of the electrodes of this apparatus may be permanent or may be electromagnets, or a combination of the two.


While the magnets mentioned throughout the above may be formed of any magnetic substance, I find the best results obtained by the employment of the nickel chrome steel.


While the successful operation of the various devices which I have constructed embodying the present invention have not enabled me to arrive definitely and positively at fixed conclusion relative to the principles and theories of operation and the source from which current is supplied, I wish it to be understood that I consider myself as the first inventor of the general type described above, capable of producing commercially serviceable electricity, for which reason my claims hereinafter appended contemplate that I may utilise a wide range of equivalents so far as concerns details of construction suggested as preferably employed.


The current which I am able to obtain is dynamic in the sense that it is not static and its production is accomplished without chemical or mechanical action either incident to the actual chemical or mechanical motion or incident to changing caloric conditions so that the elimination of necessity for the use of chemical or mechanical action is to be considered as including the elimination of the necessity for the use of heat or varying degrees thereof.













Pat. Application US 2006/0082,334    20th April 2006    Inventors: Paulo & Alexandra Correa





This patent application shows the details of devices which can produce ordinary electricity from Tesla longitudinal waves.  If these claims are correct (and there does not appear to be the slightest reason for believing that they are not), then implementations of this patent application are capable of producing free electrical power and the importance of this information is enormous.



This invention relates to apparatus for the conversion of mass-free energy into electrical or kinetic energy, which uses in its preferred form a transmitter and a receiver both incorporating Tesla coils, the distal ends of whose secondary windings are co-resonant and connected to plates of a chamber, preferably evacuated or filled with water, such that energy radiated by the transmitter may be picked up by the receiver, the receiver preferably further including a pulsed plasma reactor driven by the receiver coil and a split phase motor driven by the reactor. Preferably the reactor operates in pulsed abnormal gas discharge mode, and the motor is an inertially dampened drag motor. The invention also extends to apparatus in which an otherwise driven plasma reactor operating in pulsed abnormal gas discharge mode in turn used to drive an inertially dampened drag motor.



This is a continuation of application Ser. No. 09/907,823, filed Jul. 19, 2001.



This invention relates to systems for the conversion of energy, inter alia in the form of what we will refer to for convenience as Tesla waves (see below), to conventional electrical energy.



Energy converters that are fed by local or environmental energy are usually explained by taking recourse to the notion that they convert zero point electromagnetic radiation (ZPE) to electric energy.  The ZPE theories have gained a life of their own, as T. Kuhn has pointed out (in his "Black Body Theory and the Quantum"), after emerging from Planck's second theory, specifically from the term  in the new formula for oscillator energy.   In 1913, Einstein and Stern suggested that motional frequencies contributing to specific heat fell into two categories--those that were independent of temperature and those that were not (e.g. rotational energy), leading them to conclude that zero-point energy on the order of  was most likely.   In the second part of their paper, however, they provided a derivation of Planck's Law without taking recourse to discontinuity, by assuming that the value of the ZPE was simply ha.  It is worth noting that Einstein had already in 1905 ("Erzeugung und Verwandlung des Lichtes betreffenden heuristichen Gesichtspunkt",Ann. d. Phys, 17, 132) framed the problem of discontinuity, even if only heuristically, as one of placing limits upon the infinite energy of the vacuum state raised by the Rayleigh-Jeans dispersion law.  According to Einstein, the Rayleigh-Jeans law would result in an impossibility, the existence of infinite energy in the radiation field, and this was precisely incompatible with Planck's discovery - which suggested instead, that at high frequencies the entropy of waves was replaced by the entropy of particles.  Einstein, therefore, could only hope for a stochastic validation of Maxwell's equations at high frequencies "by supposing that electromagnetic theory yields correct time-average values of field quantities", and went on to assert that the vibration-energy of high frequency resonators is exclusively discontinuous (integral multiples of ).


Since then, ZPE theories have gone on a course independent from Planck's second theory. The more recent root of modern ZPE theories stems from the work of H. Casimir who, in 1948, apparently showed the existence of a force acting between two uncharged parallel plates.  Fundamentally the Casimir effect is predicated upon the existence of a background field of energy permeating even the “vacuum”, which exerts a radiation pressure, homogeneously and from all directions in space, on every body bathed in it.  Given two bodies or particles in proximity, they shield one another from this background radiation spectrum along the axis (i.e. the shortest distance) of their coupling, such that the radiation pressure on the facing surfaces of the two objects would be less than the radiation pressure experienced by all other surfaces and coming from all other directions in space.  Under these conditions, the two objects are effectively pushed towards one another as if by an attractive force.  As the distance separating the two objects diminishes, the force pushing them together increases until they collapse one on to the other.  In this sense, the Casimir effect would be the macroscopic analogy of the microscopic van der Waals forces of attraction responsible for such dipole-dipole interactions as hydrogen bonding.   However, it is worth noting that the van der Waals force is said to tend to establish its normal radius, or the optimal distance between dipoles, as the distance where the greatest attractive force is exerted, beyond which the van der Waals forces of nuclear and electronic repulsion overtake the attraction force.


Subsequently, another Dutch physicist, M. Sparnaay, demonstrated that the Casimir force did not arise from thermal radiation and, in 1958, went on to attribute this force to the differential of radiation pressure between the ZPE radiation from the vacuum state surrounding the plates and the ZPE radiation present in the space between them.  Sparnaay's proposal is that a classical, non-quantal, isotropic and ubiquitous electromagnetic zero-point energy exists in the vacuum, and even at a temperature of absolute zero.  It is further assumed that since the ZPE radiation is invariant with respect to the Lorentz transformations, it obeys the rule that the intensity of its radiation is proportional to the cube of the frequency, resulting in an infinite energy density for its radiation spectrum.


What appeared to be the virtue of this reformulated theory was the notion that the vacuum no longer figured as pure space empty of energy, but rather as a space exposed to constantly fluctuating “fields of electromagnetic energy”.


Puthoff has utilised the isomorphism between van der Waals and Casimir forces to put forth the zero-point (ZP) energy theory of gravity, based on the interpretation that the virtual electromagnetic ZP field spectrum predicted by quantum electrodynamics (QED) is functionally equivalent to an actual vacuum state defined as a background of classical or Maxwellian electromagnetic radiation of random phases, and thus can be treated by stochastic electrodynamics (SED).  Whereas in QED, the quanta are taken as virtual entities and the infinite energy of the vacuum has no physical reality, for SED, the ZPE spectrum results from the distortion of a real physical field and does not require particle creation.   Gravity then, could be seen as only the macroscopic manifestation of the Casimir force.


We do not dispute the fact that even in space-absent matter, there is radiant energy present which is not of a thermal nature.   But we claim that this energy is not electromagnetic, nor is its energy spectrum-infinite.  That this is so, stems not just from our opinion that it is high time that Einstein's heuristic hypothesis should be taken as literally factual - in the dual sense that all electromagnetic energy is photon energy and all photons are local productions, but above all from the fact that it is apparent, from the experiments of Wang and his colleagues (Wang, Li, Kuzmich, A & Dogariu, A. "Gain-assisted superluminal light propagation", Nature 406; #6793; 277), that the photon stimulus can propagate at supraluminal speeds and lies therefore well outside of any scope of electromagnetic theory, be this Maxwell's classical approach taken up by ZPE theories, or Einstein's special relativistic phenomenology of Maxwell's theory.  The fact is, that if the light stimulus can propagate at speeds greater than those of light, then what propagates is not light at all, and thus not energy configured electromagnetically.   Light is solely a local production of photons in response to the propagation of a stimulus that itself is not electromagnetic.


It is critical to understand that the implication from this, that - aside from local electromagnetic radiation and from thermal radiation associated with the motions of molecules (thermo-mechanical energy), there is at least  one other form of energy radiation which is everywhere present, even in space-absent matter.  Undoubtedly, it is that energy which prevents any attainment of absolute zero, for any possible local outpumping of heat is matched by an immediate local conversion of some of this energy into a minimum thermal radiation required by the manifolds of Space and Time.  Undoubtedly also, this radiation is ubiquitous and not subject to relativistic transformations (i.e. it is Lorentz invariant).   What it is not, is electromagnetic radiation consisting of randomistic phases of transverse waves.


To understand this properly, one must summarise the differences from existing ZPE theories - and all these differences come down to the fact that this energy, which is neither electromagnetic nor thermal per se, (and is certainly not merely thermo-mechanical), has nevertheless identifiable characteristics both distributed across sub-types or variants and also common to all of them.


Essentially, the first sub-type or variant consists of longitudinal mass-free waves which deploy electric energy. They could well be called Tesla waves, since Tesla-type transformers can indeed be shown experimentally to radiate mass-free electric energy, in the form of longitudinal magnetic and electric waves having properties not reducible to photon energy nor to “electromagnetic waves”, and having speeds of displacement which can be much greater than the limit c for all strictly electromagnetic interactions.


One may well denote the second sub-type by the designation of mass-free thermal radiation, since it contributes to temperature changes - and, as obviously indicated by the impossibility of reaching an absolute zero of temperature, this contribution occurs independently of the presence of matter, or mass-energy, in Space.   In other words, not all thermal radiation can be reduced to vibration, rotation and translation (drift motion) of molecules, i.e. to thermomechanical energy, because the properties of pressure and volume which determine temperature and affect matter, appear indeed to a great extent to be independent from matter, a fact which itself is responsible for the observed catastrophic and unexpected phase changes of matter and has required to this day the insufficient explanation offered semi-empirically by the Van der Waals Force Law.


Finally, the third sub-type may be designated latent mass-free energy radiation - since it deploys neither charge, nor thermal or baroscopic effects, and yet it is responsible for “true latent heat” or for the “intrinsic potential energy” of a molecule.  It is also responsible for the kineto-regenerative phenomenon whereby an electroscope performs a variable charge-mediated work against the local gravitational field.


The common characteristic of all three sub-types of mass-free energy radiation is that they share the same non-classical fine structure, written as follows for any energy unit, where c is any speed of light wave function, and the wavelength  and wave function W are interconnected as a function of the physical quality of the energy field under consideration:


In the instance of longitudinal electric radiation, this takes on the directly quantifiable form:


Wv is the voltage-equivalent wave function corresponding to V,

Pe constitutes the linear momentum corresponding to the conventional q or e,

h is the Planck constant,

 is the Duane-Hunt constant expressed as a wavelength,

 is a wavelength constant; and the sign

 signifies exact equality between an expression in the conventional dimensions of length, mass and time, and an expression in length and time dimensions alone.


In the instance of mass-free thermal radiation (contributing to temperature changes), the transformation obeys Boltzmann's rule (k is now Boltzmann's constant and T is Kelvin-scale temperature):



and in the third instance - of latent mass-free radiation, the transformation obeys the rule:



where  and are frequency functions, being a specific gravitational frequency term, and  being defined as equal to   and   has the value of


If the electric variant of mass-free radiation has a direct quantum equivalence, via the Duane-Hunt Law, none of the three primary aether energy variants possess either the classic form of electromagnetic energy which requires square superimposition of speed of light wave functions c, as c2, or the quantum form of energy, requiring E = .  The critical first step in the right direction may well be attributed to Dr. W. Reich, as it regards the fact that mass-free energy couples two unequal wave functions, only one of which is electromagnetic and abides by the limit c.   We then unravelled the threefold structure described above, and further showed that, in the case of longitudinal electric waves, the postulated equivalence  is merely phenomenological, as these waves are not restricted by the function c in their conveying of electric charge across space.  It can further be demonstrated that all black-body photons are bound by an upper frequency limit (64 x 1014 Hz), above which only ionising photons are produced, and that all black-body photons arise precisely from the interaction of mass-free electric radiation with molecules of matter (including light leptons), whereby the energy of that radiation is locally converted into photon or electromagnetic radiation.  In other words, all non-ionising electromagnetic energy appears to be secondary energy which results locally from the interaction of matter with mass-free electric energy.  It cannot therefore consist of the primary energy that is present in the vacuum, an energy that is neither virtual nor electromagnetic, but actual and concrete in its electric, thermal and antigravitic manifestations.  Lastly, gravitational energy, being either the potential or the kinetic energy responsible for the force of attraction between units of matter, is a manifestation that also requires, much as electromagnetic radiation does, coupling of mass-free energy to matter or to mass-energy.


The Tesla coil is a generator of a mass-free electric energy flux which it transmits both by conduction through the atmosphere and by conduction through the ground.  Tesla thought it did just that, but it has been since regarded instead (because of Maxwell, Hertz and Marconi) as a transmitter of electromagnetic energy.  The transmitter operates by a consumption of mass-bound electric power in the primary, and by induction it generates in the coupled secondary two electric fluxes, one mass-bound in the coil conductor, and the other mass-free in the body of the solenoid.  Tesla also proposed and demonstrated a receiver for the mass-free energy flux in the form of a second Tesla coil resonant with the first.  The receiver coil must be identical and tuned to the transmitter coil; the capacitance of the antenna plate must match that of the transmitter plate; both transmitter and receiver coils must be grounded; and the receiver coil input and output must be unipolar, as if the coil were wired in series.


The generators of mass-free energy with which we are concerned, provide current pulses associated with a dampened wave (DW) oscillation of much higher frequency than the pulse repetition frequency.  A particular problem in recovering the mass-free energy content of such pulses is provided by the dampened wave oscillations. Although in our U.S. Pat. No. 5,416,391 we describe arrangements incorporating split phase motors to recover such energy, their efficiency is a great deal less than what should theoretically be attainable. Other workers such as Tesla and Reich, have encountered the same problem to an even greater degree.


In nineteenth century motor engineering terminology, dynamos capable of producing direct current by continuous homopolar induction were known as “unipolar” generators.  The term “unipolar induction” appears to have originated with W. Weber, to designate homopolar machines where the conductor moves continuously to cut the magnetic lines of one kind of magnetic pole only, and thus require sliding contacts to collect the generated current.  Faraday's rotating copper disc apparatus was, in this sense, a homopolar generator when the disc was driven manually, or a homopolar motor when the current was provided to it. Where the rotating conductor continuously cuts the magnetic field of alternatingly opposite magnetic poles, the operation of a machine, whether a generator or a motor, is said to be “heteropolar”.  Unipolar machines went on to have a life of their own in the form of low voltage and high current DC generators - from Faraday, through Plucker, Varley, Siemens, Ferraris, Hummel, to Lord Kelvin, Pancinoti, Tesla and others - almost exclusively in the form of disc dynamos, but some having wound rotors.


In Mordey's alternator, and in so-called “inductor alternators”, however, homopolar generators were employed to obtain alternating currents, with the use of rotors wound back and forth across the field.  Use of smooth, unwound rotors in AC induction motors (as opposed to AC synchronous motors, such as hysteresis motors) was a later development than homopolar dynamos.  By 1888, Tesla and Ferraris amongst still others, had independently produced rotating magnetic fields in a motor, by employing two separate alternate currents with the same frequency but different phase.  Single phase alternate current motors were developed later, and split-phase motors were developed last.  Ferraris (Ferraris, G (1888) "Rotazioni elettrodynamiche", Turin Acad, March issue.) proposed the elementary theory of the 2-phase motor, where the current induced in the rotor is proportional to the slip (the difference between-the angular velocity of the magnetic field and that of the rotating cylinder), and the power of the motor is proportional to both the slip and the velocity of the rotor.


If an iron rotor is placed within the rotating magnetic field of a 2-phase stator, it will be set in rotation, but not synchronously, given that it is always attracted to the moving magnetic poles with a lag.  But if an aluminium or copper rotor is used instead, it gets “dragged” around by the rotating stator field because of the eddy currents induced in it.  If the aluminium or copper rotor were to rotate synchronously with the stator magnetic field, there would be no induced eddy currents and thus no motor action would result.  The motor action depends, in this instance, upon the presence of asynchronous slip, since the function of the latter is to sustain the induction of those currents in the rotor that are responsible for the motor action of the dragged rotor.  This then is the origin of the term “AC drag motors”.   Once the drag rotor evolved from a cylinder to a hollow cup, they earned the epithet of “drag-cup motors”.   Later, already in the 20th century, the cups were fitted over a central stator member, and the sleeve rotor 2-phase servo motor was born.


Tesla knew that impulse currents as well as CW (constant wave) sinusoidal currents could be used to drive AC motors.  Regarding his invention of a hysteresis motor (which he called a “magnetic lag motor”), he stated: " . . . pulsatory as well as an alternating current might be used to drive these motors . . . " (Martin, T C (1894) "The inventions, researches and writings of Nikola Tesla", Chapter XII, p. 68).  In his search for efficient utilisation of the high frequency DW (dampened wave) impulse currents of his induction coils, Tesla began by employing an AC disc induction motor as shown in Fig.17 of his famous 1892 address (Tesla, N (1892) "Experiments with alternate currents of high potential and high frequency", in "Nikola Tesla Lectures", 1956, Beograd, pp. L-70-71). This consisted of a copper or aluminium disc mounted vertically along the longitudinal axis of an iron core on which was wound a single motor coil which was series wired to the distal terminal of an induction coil at one end, and to a large suspended and insulated metal plate at the other. What was new about this was the implementation of an AC disc induction motor drive, where the exciting current travelled directly through the winding with just a unipolar connection to the coil secondary (under certain conditions, even the series connection to the plate could be removed, or replaced with a direct connection to the experimenter's body): "What I wish to show you is that this motor rotates with one single connection between it and the generator" (Tesla, N. (1892), op. cit., L-70, Tesla's emphasis). Indeed, he had just made a critical discovery that, unlike in the case of mass-bound charge where current flow requires depolarisation of a bipolar tension, mass-free charge engages current flow unipolarly as a mere matter of proper phase synchronisation:



Tesla thought that his motor was particularly adequate to respond to windings which had “high-self-induction”, such as a single coil wound on an iron core.  The basis of this self-induction is the magnetic reaction of a circuit, or an element of a circuit - an inductor - whereby it chokes, dims or dampens the amplitude of electric waves and retards their phase.


For the motor to respond to still higher frequencies, one needed to wind over the primary motor winding, a partial overlap secondary, closed through a capacitor, since "it is not at all easy to obtain rotation with excessive frequencies, as the secondary cuts off almost completely the lines of the primary" (Idem, L-71.).


Tesla stated that "an additional feature of interest about this motor" was that one could run it with a single connection to the earth ground, although in fact one end of the motor primary coil had to remain connected to the large, suspended metal plate, placed so as to receive or be bathed by "an alternating electrostatic field", while the other end was taken to ground.  Thus Tesla had an ordinary induction coil that transmitted this "alternating electrostatic field", an untuned Tesla antenna receiving this "field", and a receiver circuit comprising his iron-core wound motor primary, a closely coupled, capacitatively closed secondary, and the coupled non-ferromagnetic disc rotor.  Eventually, in his power transmission system, he would replace this transmitter with a Tesla coil, and place an identical receiving coil at the receiving end, to tune both systems and bring them into resonance.   But his motor remained undeveloped, and so did the entire receiver system.


Tesla returned to this subject a year later, saying "on a former occasion I have described a simple form of motor comprising a single exciting coil, an iron core and disc" (Tesla, N (1893) "On light and other high frequency phenomena", in "Nikola Tesla Lectures", 1956, Beograd, pp. L-130, and L-131 with respect to Fig.16-II). He describes how he developed a variety of ways to operate such AC motors unipolarly from an induction transformer, and as well other arrangements for "operating a certain class of alternating motors founded on the action of currents of differing phase".  Here, the connection to the induction transformer is altered so that the motor primary is driven from the coarse secondary of a transformer, whose finer primary is coupled, at one end, directly and with a single wire to the Tesla secondary, and at the other left unconnected. On this occasion, Tesla mentions that such a motor has been called a “magnetic lag motor”, but that this expression (which, incidentally, he had himself applied to his own invention of magnetic hysteresis motors) is objected to by "those who attribute the rotation of the disc to eddy currents when the core is finally subdivided" (Tesla, N (1893), op. cit., p. L-130).


In none of the other motor solutions, 2-phase or split-phase, that he suggests as unipolar couplings to the secondary of an induction coil, does the non-ferromagnetic disc rotor motor again figure.   But he returns to it a page later, and indirectly so, by first addressing the disadvantages of ferromagnetic rotors: "Very high frequencies are of course not practicable with motors on account of the necessity of employing iron cores. But one may use sudden discharges of low frequency and thus obtain certain advantages of high-frequency currents-without rendering the iron core entirely incapable of following the changes and without entailing a very great expenditure of energy in the core.  I have found it quite practicable to operate, with such low frequency disruptive discharges of condensers, alternating-current motors."


In other words--whereas his experiments with constant wave (CW) alternating currents, and as well with high-voltage dampened wave (DW) impulses from induction coils, indicated the existence of an upper frequency limit to iron core motor performance, one might employ instead high-current, DW impulses - of high DW frequencies but low impulse rates - to move these motors quite efficiently.  Then he adds "A certain class of [AC] motors which I advanced a few years ago, that contain closed secondary circuits, will rotate quite vigorously when the discharges are directed through the exciting coils.  One reason that such a motor operates so well with these discharges is that the difference of phase between the primary and secondary currents is 90 degrees, which is generally not the case with harmonically rising and falling currents of low frequency.   It might not be without interest to show an experiment with a simple motor of this kind, inasmuch as it is commonly thought that disruptive discharges are unsuitable for such purposes."


What he proposes next, forms the basis of modern residential and industrial AC electric power meters, the AC copper disc motor whose rotor turns on the window of these meters, propelled forward by the supply frequency.  But instead of employing any such Constant Wave input, Tesla uses the disruptive discharges of capacitors, incipiently operating as current rectifiers.   With the proper conditions, e.g. correct voltage from the generator, adequate current from the capacitor, optimum capacitance for the firing rate, and tuned spark-gap, to mention a few, Tesla found that the non-ferromagnetic disc rotor turned but with considerable effort. But this hardly compared to the results obtained with a high-frequency CW alternator, which could drive the disc "with a much smaller effort".   In summary then, Tesla went as far as being the first to devise a motor driven by Tesla waves, that employed a non-ferromagnetic rotor, and whose arrangement encompassed both transmitter and receiver circuits.   For this purpose, he employed a single-phase method in which the signal is fed unipolarly to the winding, placed in series with a plate capacitance.


Tesla also later proposed driving a similar single-phase non-ferromagnetic disc motor from bipolar capacitative discharges through an atmospheric spark-gap now placed in parallel with the main motor winding, and again simulating a split-phase by a closely-wound secondary which was closed by a capacitance.


As Tesla admits, the results of all his AC eddy current motor solutions were meagre and limited by current and frequency problems.  Likewise, the two-phase arrangements proposed by Reich for his OR motor, involving a superimposition of the Dampened Waves of a first phase on a fixed Continuous Wave second phase, require an external power source and a pulse amplifier circuit, and failed to meet Reich's own requirements.


We have previously proposed the use of squirrel cage motors with capacitative splitting of phase to convert the Dampened Wave output of plasma pulsers, but once a Squirrel Cage is introduced, the dampening effect which the non-ferromagnetic copper cage exerts in being dragged by the revolving stator field, is counteracted by the ferromagnetic cylinder of laminated iron, in which the copper cage is embedded, working to diminish the slip and bring the rotor to near synchronism.  This is, in all likelihood, what limits Squirrel Cage motors responding to the DC component of the Dampened Wave impulse, and thus be limited to respond to fluxes of mass-bound charges.  Historically, as we shall see, the obvious advantage of the Squirrel Cage servo motors lay in the fact that, in particular for 2-phase applications, they were far more efficient at performing work without evolution of heat.   Indeed, if the eddy currents in the non-ferromagnetic rotor are permitted to circulate in non-ordered form, the rotor material and stator will heat up rapidly and consume much power in that heating.  This is in fact considered to be a weakness of AC non-ferromagnetic-rotor induction motors.




The present invention is concerned with conversion to conventional electrical energy of the variants of mass-free energy radiation considered above, referred to for convenience as Tesla waves, mass-free thermal radiation and latent mass-free radiation.  The first variant of such radiation was recognised, generated and at least partially disclosed by Tesla about a hundred years ago, although his work has been widely misinterpreted and also confused with his work on the transmission of radio or electromagnetic waves.  The Tesla coil is a convenient generator of such radiation, and is used as such in many of the embodiments of our invention described below, but it should be clearly understood that our invention in its broadest sense is not restricted to the use of such a coil as a source of mass-free radiation and any natural or artificial source may be utilised.  For example, the sun is a natural source of such radiation, although interaction with the atmosphere means that it is largely unavailable at the earth's surface, limiting applications to locations outside of the earth's atmosphere.


According to the invention, a device for the conversion of mass-free radiation into electrical or mechanical energy comprises a transmitter of mass-free electrical radiation having a dampened wave component, a receiver of such radiation tuned to resonance with the dampened wave frequency of the transmitter, a co-resonant output circuit coupled into and extracting electrical or kinetic energy from the receiver, and at least one structure defining a transmission cavity between the transmitter and the receiver, a full-wave rectifier in the co-resonant output circuit, and an oscillatory pulsed plasma discharge device incorporated in the co-resonant output circuit.  The output circuit preferably comprises a full-wave rectifier presenting a capacitance to the receiver, or an electric motor, preferably a split-phase motor, presenting inductance to the receiver. The transmitter and receiver each preferably comprise a Tesla coil and/or an autogenous pulsed abnormal glow discharge device.  The transmission cavity is preferably at least partially evacuated, and comprises spaced plates connected respectively to the farthest out poles of the secondaries of Tesla coils incorporated in the transmitter and receiver respectively, the plates being parallel or concentric.  The structure defining the cavity may be immersed in ion-containing water.  The split-phase motor is preferably an inertially-dampened AC drag motor.


The invention, and experiments demonstrating its basis, are described further below with reference to the accompanying drawings.




Fig.1 is a schematic view of a Tesla coil connected to a full-wave rectifier to form an energy conversion device:


Fig.2 is a schematic view of a Tesla coil connected to a gold leaf electrometer:





Fig.3 to Fig.6 show alternative electrometer configurations:


















Fig.7 to Fig.11 show modifications of the circuit of Fig.1:

















Fig.12 shows apparatus for investigating aspects of the experimental results obtained with the foregoing devices;



Fig.13 is a graph illustrating results obtained from the apparatus of Fig.12:








Fig.14 to Fig.17 show schematic diagrams of embodiments of energy conversion devices:












Fig.18 is a diagrammatic cross-section of an inertially dampened drag cup motor:





Fig.19 is a schematic diagram of a further embodiment of an energy conversion device incorporating such a motor:





Based upon observations of weight loss in metallic matter as induced by exposure to high frequency alternating electric fields, we developed an experimental method to optimise this-weight loss, and from this a device that treats the forces causing weight loss as manifestations of intrinsic potential energy  (or true "latent heat") of the molecules of matter, and converts both "true latent heat" energy present in the neighbourhood of a receiver, and "sensible" heat induced within that receiver, into electric energy which can be used to drive a motor, flywheel or charge batteries.


It is commonly believed that the output of the Tesla coil is ionising electromagnetic radiation.  We have demonstrated that it is not, i.e. that it is neither electromagnetic radiation, nor ionising electromagnetic radiation.  The output of an air-cored, sequentially-wound secondary, consists exclusively of electric energy: upon contact with the coil, a mass-bound AC current can be extracted at the resonant frequency, whilst across a non-sparking gap, mass-free AC-like electric wave radiation having the characteristics of longitudinal waves, can be intercepted anywhere in adjacent space.  Accordingly, the radiation output from such coils is different to electromagnetic radiation.


The basic demonstration that the output of a Tesla coil does not consist of ionising radiation, is that it does not accelerate the spontaneous discharge rate of electroscopes, whether positively or negatively charged.  In fact, in its immediate periphery, the coil only accelerates the spontaneous discharge rate of the negatively charged electroscope (i.e. the charge leakage rate), whereas it arrests the discharge of the positively charged electroscope (i.e. the charge seepage rate falls to zero).  But this dual effect is not due to any emission of positive ions from the secondary, even if it can positively charge a discharged electroscope brought to its proximity.  This charging effect is in fact an artifact, in that metals but not dielectrics are ready to lose their conduction and outer valence band electrons when exposed to the mass-free electric radiation of the coil.


This is simply demonstrated by the apparatus of Fig.1, in which the outer terminal of the secondary winding 6 of a Tesla coil having a primary winding 4 driven by a vibrator 2 is connected to the input of a full-wave voltage wave divider formed by diodes 8 and 10 and reservoir capacitors 12 and 14 (the same reference numerals are used for similar parts in subsequent diagrams).  If the rectifiers employed are non-doped, then the coil appears to only charge the divider at the positive capacitance 10, but if doped rectifiers are employed, the coil will be observed to charge both capacitances equally.  Whereas positive ionises can charge either doped or un-doped dividers positively, no positive ionise can charge a doped divider negatively, clearly demonstrating that the Tesla coil does not emit positive ions.


The basic demonstration that the output of a Tesla coil is not non-ionising electromagnetic radiation of high frequency, such as optical radiation, or of lower frequency, such as thermal photons, is also a simple one. Placement of a sensitive wide spectrum photoelectric cell (capable of detecting radiation to the limits of vacuum UV), wired in the traditional closed circuit manner from a battery supply, at any distance short of sparking from the outer terminal of the coil will show in the dark that the light output from the coil is negligible. This rules out optical radiation at high frequency.  The demonstration that the sensible heat output from the Tesla coil is also negligible will be addressed below.


Our theory proposed the existence of physical processes whereby mass-free electric radiation can be converted into electromagnetic radiation.  Such a process is at work whenever mass-free electric wave radiation interacts with electrons, such as those that remain in the valence bands of atoms.  This mass-free electric energy interacts with charge carriers, such as electrons, to confer on them an electrokinetic energy which they shed in the form of light whenever that electrokinetic energy is dissociated from those carriers (e.g. by deceleration, collision or friction processes).  Such a process is at work to a negligible extent in the coil itself and its usual terminal capacitance, hence the faint glow that can be seen to issue from it, but it can also be greatly amplified in the form of a corona discharge by connecting a large area plate to the output of the secondary, as Tesla himself did in his own experiments, and thus by increasing the capacitance of the coil system.  


Now, what is interesting in this process is that, in the absence of virtually any I2R losses at the plate, and if the plate thus introduced is bent at the edges so that it has no pointed edges, or if it is in the form of a bowl, or in any other manner that precludes sparking at edges and specially corners, and thus enhances the corona discharge, any electroscope, whether negatively or positively charged, now brought close to the plate will show a tendency to arrest its spontaneous discharge rate.  One might say that this is simply the result obtained in a Faraday cage which disperses charge on its outside and electrically insulates its interior, and indeed if an electroscope is placed inside a Faraday cage no amount of Tesla radiation on the outside of that cage, save direct sparking, adversely affects the leakage or seepage rate of the electroscope.  In fact, since the effect of such a cage can be shown to be that of, by itself, inducing arrest of either spontaneous electroscopic discharge, this effect simply remains or is magnified when the cage is bathed by Tesla radiation.  However, a cage constitutes an electrically isolated environment, whereas a plate with or without curved or bent edges does not.  Furthermore, the change observed in the properties of the output radiation from a Tesla coil when certain metal plates or surfaces are directly connected to the outer terminal of the secondary, takes place whilst the capacitance of the coil is increased by the connected plate, and thus the plate is an electrically active element of the circuit - and hence the opposite of an electrically isolated element.


For a long time, we believed that the anomalous cathode reaction forces observed in autoelectronic discharges (atmospheric sparks, autogenous PAGD (pulsed abnormal glow discharge) and vacuum arc discharges) were exclusive to an autoelectronic emission mechanism prompted by a direct potential between discharging electrodes.  Sparking driven by AC potentials could sustain the same forces, but their mutual cancellation over time would not deploy a net force.  In this sense, when a large gold leaf connected directly to the ground (via a water pipe or any other suitable connection) or to another large area plate suspended at some height above the ground, is vertically placed at a sparking distance above the surface of another plate connected to the secondary of a Tesla coil, one would not expect the AC spark to sustain any net force across the gap between the gold leaf and the plate.  In terms of cathode reaction forces, one would expect their cancellation to be simply brought about by the high frequency of the current alternation in the coil, as both leaf and plate would alternate between being the emitting cathode or the receiving anode.  However, this is not what is observed - instead, the gold leaf 16 lifts away from the plate 18 (Fig.2).  If instead, the suspended gold leaf is connected to the coil terminal, and the bottom plate is connected to the ground in the same manner as described above, this also yields the same result.


Even more curious is the finding that this anomalous reaction force deployed by an alternate current of mass-bound charges in the arc, remains present when the sparking is prevented and instead the corona effect is enhanced (by employing a large plate connected to the outer pole of the secondary, and by employing a distance at which sparking ceases), as if the lift itself were the property of the corona underlying the spark channels and not the property per se of the autoelectronic emission mechanism.


By mounting the suspended leaf 16 (41 mg of hammered 99.9996% pure gold) directly at the end of a long dielectric rod 20 balanced at the centre and placed on a light stand over an electronic balance 22, we sought to determine the observed lift of the leaf as weight lost.  Surprisingly, and despite the most apparent lifting motion of the leaf, the balance registered a substantial weight gain, indicating the addition of 1 to 5 mg weight (with the same 14W input to the vibrator stage), independently of whether the leaf was connected to the terminal of the coil or instead to the earth ground via a water pipe.  This suggested to us that, whether formed as a DC or AC spark channel, or whether in the form of a corona discharge, the electric gap develops an expansion force (exactly opposite to a Casimir force) on both electrodes, independently of their polarity, which force is responsible for the observed repulsion.  Yet, this expansion goes hand in hand with an increase in their weight such that some other process is at work in that electric gap.


To examine this problem further, we assembled a different experiment where the gold leaf 16 was suspended between two large metal plates 18 and 24 placed 20 cm apart, and the leaf was not electrically connected to them or to any other circuit, while attached to the dielectric rod employed to suspend it over the electronic balance.  Given that the leaf is suitably and equally spaced from both plates, there is no arcing between it and either plate.  The obvious expectation is that, since the electric field bathing the leaf alternates at high frequency (measured in hundreds of kilohertz), and the corona from both electrodes should equalise and balance any electric wind, no lift should be observed.  In fact, no lift is apparent, but a most curious observation is made: depending upon which orientation is employed for the plates, the gold leaf either gains or loses 4-6% of its weight. This gain or loss is registered for as long as the coil is on.  If the top plate is grounded and the bottom one connected to the different terminal of the secondary, a gain in weight is observed (Fig.3).   If the connections are reversed, an equal weight loss is registered (Fig.4).


Furthermore, in this last instance, if the grounded plate 24 is entirely removed (Fig.5), and only the top plate remains connected to the outer terminal of the secondary, the observed loss of weight continues to occur such that in effect, this reaction can be obtained with unipolar electric fields of high frequency, and it provides a unidirectional force which, once exerted upon metallic objects bathed by its field, can be made to oppose or augment gravity.


Now, these effects can be greatly magnified, in the order of 10-fold, if the same gold leaf is made part of a simple series floating electric circuit where the leaf functions as a large area plate, and is wired in series with a coil 26 which, for best results, should be wound so as to be of a length resonant with the secondary of the Tesla-type coil employed; and this coil is connected in turn to a point antenna 28 upwardly oriented (Fig.6). The entire floating circuit is mounted on the rod 20 and this in turn, is mounted over the sensitive balance.  If both plates are kept as in Fig.3 and Fig.4, the observed weight loss and weight gain both vary between 30% and 95% of the total weight of the leaf.   Again, the gain or loss of weight is registered for as long as the coil is on.


These anomalous findings suggested that, whatever is the nature of the energy responsible for the force observed in that high frequency alternating current gap, any metallic object placed in that gap will experience a force repelling it from the electric ground.  This force will be maximised if the gap frequency is tuned to the elementary or molecular structure of the metallic object.  If the electric ground is placed opposite the actual plane of the earth ground, that force will act in the direction of gravity.  If, instead, the electric ground and the earth ground are made to coincide on the same plane, that force will act opposite the direction of gravity, i.e. will repel the metallic object from the ground.


No such weight alteration was observed with solid dielectrics, for instance with polyethylene and other thermoplastic sheets.


These facts rule out the possibility of a hidden electrostatic attraction force, acting between the plate connected to the different terminal of the secondary and the gold leaf.  Firstly, such an attraction would be able to lift the gold leaf entirely, as is easily observed with the unipole of any electrostatic generator operating with a few milliwatts output with either negative or positive polarity; secondly, the same attraction, if it existed and were the product of an electric force, would surely be manifested independently from whether the experimental leaf was metallic or a dielectric (as again is observed with electrostatic generators).


The results suggest therefore, that whenever a large plate is connected to a Tesla-type coil, it induces in surrounding matter that is not part of its own circuit, a directional thrust which is oriented in a direction which is opposite to the electric ground and, if the electrical ground is on the same side as the surface of the Earth, then a thrust is produced which opposes gravity.


When this thrust is made to oppose gravity, we believe that its effect upon the gold leaf can be compared to the lifting power imparted to the water molecule when it transits from the liquid to the vapour state and which is associated with the increase in internal (or intrinsic) potential “thermal” energy  (See Halliday D & ResnickR (1978) "Physics", Vol. 1, section 22-8, p. 489).  The "specific latent heat" of water (m*L) contains indeed both an expression for the sensible radiant thermal work involving volume and pressure relations:

W = P(VV-VL)   where P = a pressure of 1 atmosphere, and VV and VL are the molar volumes in the vapour and liquid phases respectively,  and an expression for a quantity of "latent" energy () which is associated with the molecule in the more rarefied state. Hence, the relation for the latter with respect to water vapour is:  = mL - P(VV-VL)


We propose that likewise, if a very small portion of the energy of the mass-free electric waves is indirectly transformed by mass-bound charge carriers on that plate into blackbody photons (once those charge carriers shed their electrokinetic energy), the greater portion of those waves are directly transformed in the space adjacent to that plate into the latent energy equivalent to  for the atoms of the surrounding air, and so on, until this process itself is also occurring for the atoms of that gold leaf, thus inducing their non-electrical weight loss and suggesting the existence of a non-thermal "antigravitokinetic" energy term previously unknown to mankind other than as "latent heat" or "internal potential energy".


From this viewpoint, the energy released by any Tesla-type coil to its surroundings, would be tantamount to a radiative injection of "internal potential energy" which would confer on local gas molecules a weight cancellation (a cancellation of gravitational mass occurring in the absence of any cancellation of inertial mass - a process which the inventors theorise is explained by the neutralisation of elementary gravitons), and the same process would be equally at work for metallic solids but not dielectric solids.


Gold vapour also deploys a substantial intrinsic potential energy.  With an enthalpy of vaporisation on the order of HV = 324 kJ mol-1, the molar volumetric work performed by gold vapour at atmospheric pressure at the temperature of vaporisation Tv (2,8560C., i.e. 3,129 degrees Kelvin) is:


W = PVV-L = 23.58 kJ mol.-1  where  VV-L = 0.2327m3.   The intrinsic potential energy of gold vapour is then given by:


 = Hv - W = 300.4 kJ mol.-1  i.e. 12.74 times greater than the volumetric work performed during the phase transition.


It is our contention that this intrinsic potential energy, associated with molecules as their "latent heat", has fine structure that in turn is altered if this energy is released from these molecules and fails to gain a "sensible" thermal form.  What is suggested is that the fine structure of "latent heat" is not electromagnetic and obeys instead the molecular function:


 / NA = n22c n2   where NA is Avogadro's number, the wavelength denoted as n2 is the wavelength-equivalent of the mass of the molecule to which the "latent heat" is associated, obtained by a conversion method proposed in these inventors' theory, and the frequency term  is a non-electromagnetic frequency term, specifically in this case a gravitational frequency function. 


Employing the conversion of Joules into m3 sec-2 proposed by these inventors as being exactly:


1J = 10 NA m3 sec-2, and putting the wavelength n2 down as the wavelength-equivalent of the mass of the gold atom, Au, at 1.9698 m, that frequency term n2 can be obtained as being equal to 2.6 x 10-3 sec-1.


According to the present inventors' theory, the wave function c constitutive of the fine structure of "latent heat" associated with molecules of matter, carries the same wavelength Au and its frequency is given in the usual manner by c/Au = 1.52 x 103 sec-1.  The resultant frequency for the non-Planckian unit quantum of "latent energy" associated with each gold atom at the vaporisation temperature is then obtained by the geometric mean of the two synchronous frequency terms: [(c/Au) n2]0.5 = 624 Hz.  However, this is the signature of that intrinsic potential energy when associated with that gold atom at its vaporisation temperature.  It is not the signature of the energy quantum itself if it is released from that molecule, nor prior to being absorbed (i.e. in transit), at that same temperature.


The fine structure of the same non-Planckian "latent" energy quantum varies to encompass different determinations of the constituent wavelength and frequency functions. The basic relation for the determination of the wavelength of a "latent thermal" energy quantum not associated with matter, but corresponding to one that is, is:


n1 = [ ( / NA) / c]0.666 meters-0.333 seconds0.666


which gives 0.046478 m for the unbound equivalent of the "latent heat" unit quantum of vaporisation associated with the gold atom at a pressure of one atmosphere.   The fine structure of the free quantum is still parallel, as given by:


 / NA = n12cn1


but now notice how the frequency terms have changed value, with the n1 function having the value 4.65 sec-1 and c / n1 yielding 6.48 x 109 sec-1.  The geometric mean of the superimposition of the two frequencies is then:


[(c / n12)n1]0.5 = 173.7 KHz


We contend that it is at this frequency that the atoms of gold vapour absorb "latent heat".


However, this is just the overall scenario of what happens at the temperature of vaporisation of gold.  But at room temperature (e.g. 293 degrees Kelvin), and with respect to processes where there is no sublimation of the atoms of that gold leaf under way (and indeed, once the coil is turned off, the leaf returns to its normal weight), one must infer to a different phase of matter what portion of "latent heat" energy, if any, do the atoms of gold hold in the solid phase lattice.  Assuming the same proportionality between the "sensible" and "latent" thermal energy terms for atoms of gold at room temperature, where the unit thermal energy is NAkT = 2.436 kJ mol-1, we speculate that the gold atom could absorb up to 12.74 times the value of this "sensible" thermal energy, and thus hold NAkT = 31.053 kJ more energy in its own micro-atmosphere.


If this speculation is correct, and employing the above novel methodology, then the mean geometric frequency of the maximal "latent heat" energy quantum of a gold atom at room temperature would be 538 KHz (versus 174 KHz at the vaporisation temperature), and once absorbed its mean frequency mode would reduce to 201.5 Hz (versus 630 Hz once the atom has vaporised).


To test this hypothesis, we employed two different Tesla-type coils having output frequencies of 200 KHz and 394 KHz.  The circuit tested was that shown in Fig.6, and both coils were operated at 50 KV outputs. Whereas the former coil, closer to the 174 KHz marker, could only systematically produce 10mg to 11 mg of weight cancellation in the gold leaf of the floating circuit, the second coil, closer to the speculated 538 KHz marker, could produce 15mg to 35 mg of weight cancellation in the same gold leaf.  The empirical results appear therefore to suggest that our speculation may well be a valid one.


The above-mentioned full wave divider (see Fig.1) can be easily coupled to our autogenous Pulsed Abnormal Glow Discharge technology as described in our U.S. Pat. No. 5,416,391 to form an alternative source of direct current, ultimately powered by Tesla waves, and such a drive can equally be applied to any other vacuum device that can sustain endogenous oscillatory discharges, whether in the PAGD regime or any other pulsatory regime.  For the purposes of experimental and visual determination of power outputs from the divider in question, we have utilised either 2 Torr vacuum tubes operating in the high-current PAGD regime, or 20-100 Torr spark tubes requiring high voltages (2 to 10 KV) for their spark breakdown.  As taught in the above US Patent, the output from the full wave voltage divider can be assessed by the energy spent in driving the tube and the motor, whose rotary speed is proportional, within the limits chosen, to the power input.


Two separate sets of experiments presented in Table 1 below, showed that direct connection of the wave divider to the outer terminal of the coil (set constantly at 6 clicks on the vibrator stage in Fig.1) or to the same terminal but across a large (2 or 3 square feet) plate 30 that increased the capacitance of the secondary (Fig.7), presented the same power output in either case (the effect of the plate is to lower the voltage of the output proportional to the increase in current).   A substantial increase in power output through the divider is observed only when an identically wound Tesla coil is connected in reverse (Fig.8) with the non-common end of its winding 4 not connected, in order to obtain a condition of resonance, and this observed increase is further augmented by now interposing either of the metal plates 18, 24 between the two chirally connected and identical coils (Fig.9).  The increase in plate area appears to have the effect of increasing the output for as long as the plate is isolated between the two chiral image coils.  Throughout these experiments, the input power to the vibrator was fixed at 14W (60 Hz AC).  [Note: ‘Chirality’, or ‘handedness’, is a property of objects which are not symmetrical.  Chiral objects have a unique three-dimensional shape and as a result a chiral object and its mirror image are not completely identical - PJK ].




In our loss of weight experiments described above, we noted that the phenomenon of weight loss by a metallic body placed in proximity of the coil output continued to be observed when only the plate connected to the distal pole of the secondary was retained.  The leaf, although not part of the circuit of the secondary, could however be seen as part of a circuit for the capture of ambient radiant energy, specifically that generated by the coil and, as well, that also possibly picked up, in the process, from other ambient sources. To determine whether the last consideration is a possibility at all, or whether the energy picked up by an analogue of our metallic body or gold leaf in the experiments described above, is entirely a by-product of the energy transmitted by the plate connected to the outer pole of the secondary, we next determined what would happen if the pick-up for the full-wave divider were placed, not at the output from the secondary coil, but from an, in all respects identical, plate (the Receiver plate R, as opposed to the Transmitter plate T) placed a distance away from, and above, the first one.   In other words, the gold leaf is replaced by a receiver plate, and this carries an attached test circuit identical to the test circuit employed to directly assess the coil output.



As shown in Table 2 above, the results of the experiment show that there is no loss of energy picked up at the R plate (Fig.10) when compared to the most favourable situation involving the plate 30 (Fig.9) interposed between the chirally connected coils.  This observation is however not always the case. For best results one should employ iron, gold or silver plates placed parallel to the horizon, with the T plate underneath the R plate.  In fact, if one employs instead aluminium plates and suspends these vertically, one can consistently register a loss of output at the divider when changing the divider input from the T to the R plates.


If however the plate R is connected in turn to a second identical coil, also wired in reverse, and this second coil in turn serves as input to the full-wave divider (Fig.11), then a most curious occurrence takes place - the power output increases considerably (see Table 2), as if the divider circuit had undergone an energy injection not present at the source.  Note that the circuits are in fact resonant, but the energy injection contributing nearly 60-66% (for both plate areas in the previous experiment) of the input that we refer to, is not caused by inductive resonance, since the effect of resonance can be ascribed to the set-up described in Fig.9. The distance between the plates, as well as their orientation with respect to the local horizon system of the observer also appear to matter, best results being achieved at optimal distances (e.g. for 2 square feet plates the best gap, at 43% RH and room temperature, was at least 6 inches).


We tested the possibility that environmental heat produced by operation of the coil might be the source of the injected energy, the plate of the second system acting possibly as collector for the heat present in the gap.  As it turned out, experiments showed repeatedly that in the gap between the T and R plates there was no significant thermal radiation propagating between one and the other.  The more illustrative experiments are those in which we identified where the sensible thermal energy appears, and which involved coupling two cavities: the Transmitter-Receiver gap between plates T and R, and a Faraday cage enclosure 34 (see Fig.12).  The first cavity appears to be much like that of a capacitor: the two identical parallel plates are surrounded by a thick dielectric insulator 32, and a thermometer T2 is inserted half-way through it.  A thermometer T1 is also fixed to the T plate, to measure it’s temperature.   The second cavity is a simple insulated metal cage with a thermometer T3 inserted 2 cm into its top.  Some 2-4 cm above the top of the cage there is placed a fourth thermometer T4, inside an insulated cylinder.


If the Tesla Coil is a source of thermal energy (e.g. IR radiation, microwaves, etc.) we would expect the T plate to be the hottest element from which, by radiation, thermal energy would reach the middle of the first cavity making the next thermometer T2 second hottest, and that the third thermometer T3 inside the second cavity, even if it might initially be slightly warmer than the other two, would, over time, become comparatively cooler than either one of the other two thermometers, despite the fact that the rising heat would still be seen to warm it up over time.  One would expect a similar outcome for the fourth thermometer T4, above the cage.  As shown by Fig.13, where only the temperature differences (T0 - TC0) between the experimental thermometers and the control thermometer reading the air temperature TC0 of the laboratory are shown, the surface of the T plate warms up by 0.10C. at 3 minutes after initiation of the run (closed squares), whereas in the space of the T/R gap a diminutive warming, by 0.050C., is registered after 10 minutes (open circles).  Conversely, the temperature inside the cage, at the top (shaded circles) rises by 0.10C. also by the third minute, and the temperature above the cage itself (shaded squares) rises by a much greater difference of 0.350C., which remains stable after the eighth minute.


These results show that it is not sensible heat that radiates from the T plate.  Instead, some other form of radiation traverses these cavities to generate sensible heat at their metallic boundaries, such that more heat is generated above the R plate (inside the cage) and again above the third plate, i.e. above the top of the cage, than is generated in the T/R gap, i.e. near the T plate.  This clearly shows that the Tesla coil is not a significant source of thermal radiation, and that sensible heat can be detected inside and on top of the Faraday cage only as a further transformation of the radiant energy transmitted across the T/R cavity.


The same experiment also illustrates that, whatever is the nature of the additional environmental energy being injected at the surface of R plate (as shown by Table 2 results above), it is most likely not thermal radiation, at least not energy in the form of sensible heat.  And whatever is the nature of this ambient radiant energy being mobilised by the electric radiant energy transmitted from the T plate, it can produce significant heat inside an enclosure adjacent to plate R.


Since we also know experimentally, that this observation of an ambient energy injection at the R plate or R cage depends upon relative humidity, being most easily observable when the latter is low (<50% Relative Humidity), and being virtually impossible to observe when air is saturated with water vapour, we can infer that water vapour is a good absorber of the electric mass-free radiant energy emitted from the T plate. This strongly suggests that this absorption process is tantamount to increasing the potential intrinsic energy  of the water vapour molecules adjacent to the T plate.  In the absence of significant quantities of water vapour, when the atmosphere is dry, one may speculate that this absorption process is replaced by what one presumes is a parallel process involving the various gaseous molecules of air.  However, either because the air molecules involve molecular species that readily give off this potential energy, as one might speculate is the case with molecular oxygen, hydrogen and nitrogen, or because the air molecules absorb far less "latent" energy (as appears to be the case with inert gases), and therefore there is more of it in the molecularly unbound state (as we explicitly propose as a possibility) and thus available for absorption by the appropriately tuned receiver, the increased  of air molecules conferred by the absorption of the mass-free electric radiation in the T/R gap is transferred to the R conductor together with the latent energy which those molecules already possessed before entering that gap.  Hence the energy injection and its dependency upon the partial pressure of water vapour, which absconds instead with this "latent" energy and succeeds in withholding it from transmission to the R plate.


If the T/R gap can mobilise ambient energy which is neither electromagnetic nor thermal in nature, but which "latent" energy becomes injected into the divider circuit in electric form, the heat (i.e. sensible thermal energy) produced inside and on top of the cage, can also be mobilised electrically as input into the divider circuit.  The obvious place to look for the positioning of the cool junction which could convert sensible heat into electrokinetic energy of mass-bound charges is at the top of the cage, where it is warmest (See top curve of Fig.13 in shaded squares).  This is clearly observed from the results shown in Table 3 below, where the initial temperature difference between the top of the box and the T plate surface was 0.50C., and the top of the box temperature rose by 0.20C. after 2.5 minutes when the divider was connected at the junction, versus 0.350C. when it was not (and the transmitter coil was on).




For the run performed with the naked R cage, the temperature directly above the top of the cage was 24.30C., at the outset, versus the control room temperature of 23.90C.  For the run performed with the insulated R cage exposed directly to the sun at midday, on a cool and clear August day, the temperature directly above the top of the cage was 330C., versus the control air temperature of 18.40C.  The temperature of the cool junction at the top of the cage was 31.90C. while the run was performed.


It is apparent from the data of Table 3, how a second injection of energy has occurred in the apparatus.  If, within the T/R gap, the energy injected appears to be on the order of absorption of "latent heat", at the top of the cage cavity, at the cool junction, the injection is one of radiant "sensible" heat.  Moreover, this secondary energy addition could be further enhanced by placing strong insulation around the whole apparatus or the cage itself, and further so, by exposing the whole apparatus to solar radiation.


We next turned our attention to the T/R gap cavity with the intention of determining whether atmospheric conditions or vacua yield the same or different results. We could not, of course, test the same large area plates as have been employed for the studies undertaken at atmospheric pressures. For the present purpose we employed instead large area electrodes (ca 0.2 ft2) made of high grade stainless steel or even aluminium. Preliminary results showed that these T/R gap tubes, when coupled to the divider circuit, yielded faster pulse rates in the secondary circuit when evacuated than at atmospheric pressure.  The strength of the corona discharge also intensified, as it eventually became replaced by a normal glow discharge.  For purposes of improved spatial capture of (1) the electric mass-free energy radiated from the T electrode and (2) the non-radiant latent thermal energy mobilised by it to be collected electrically at the R plate, an axial cylindrical T electrode was inserted inside a larger concentric cylinder or between two common plates of large surface area (e.g. >100 cm2) functioning as the R electrode(s), in a dielectric container suitable for evacuation (glass, polycarbonate), at a typical distance of at least 3 cm between electrodes, and the entire device was tested at different pressures. 


The secondary circuit connected downstream from the full-wave divider was as shown in Fig.14 (employing an autogenous pulsed abnormal glow discharge, or PAGD, converter circuit), with the PAGD reactor 36 set at 10 Torr (in light of the high-voltage input, which varied between 1,500V and 3,200V) and gave the results presented in Table 4 below.  We should remark also that these pulses charged the charge pack CP through the coupling capacitors 38, bridge rectifier 40 and reservoir capacitors 42, and blocking diodes 44, as expected from the prior art represented by our patents related to PAGD devices.




The effect of the vacuum in the T/R gap tube seems to be dual.  By transforming the corona discharge into a normal glow discharge, it increases the local production of photons (probably associated to the formation and discharge of metastable states in the plasma), and at the same time, increases the pulse rate in the output circuit and thus, in all probability, the energy injected in the T/R gap cavity.  But this did not yet permit us to confirm whether or not it is "latent heat" energy of the plasma molecules which is being tapped at the receiver plate, even if it be plausible in principle that plasmas may effect more efficient transfer of "latent heat" to tuned receivers than atmospheric gases.


The vacuum dependency of the pulse rate of the PAGD reactor employed as example in the secondary circuit downstream from the divider is also rather well marked, with the fastest pulse rates being registered at 1 Torr for the sample run shown in Table 5 below.




It is worth noting here that the illustrated polarity of the wiring of the PAGD reactor tube, as shown in Fig.14, is best for purposes of sustaining regular auto-electronic emission at high voltage.  The reverse configuration, with the centre electrode negative and the plates positive favours instead heating of the cathode and a lapse into a normal glow discharge.


We tested a similar arrangement to that shown in Fig.14 above, but with a PAGD motor circuit (see our U.S. Pat. No. 5,416,391).  A split-phase motor 44 replaces the rectifier and charge pack, and the PAGD reactor is operated at the same pressure of 15 Torr, as shown in Fig.15.  The T/R gap tube tested had a longer plate distance (2''), with one plate now functioning as Transmitter and the other as Receiver.  Note also the different wiring of the PAGD reactor.  The results, as shown below in Table 6, present pulse per second (PPS) and motor revolutions per minute (RPM) curve trends that appear to be analogous and parallel to the well known Paschen curves for breakdown voltage in vacuum - such that the T/R gap performs better either in the atmospheric corona discharge mode, or in the high vacuum normal glow discharge (NGD) mode, than in the low breakdown voltage range of the curve where the discharge forms a narrow channel and takes on the appearance of an "aurora" transitional region discharge (TRD).




These results suggest that plasmas with high lateral dispersion, i.e. formed over large electrode areas (e.g. corona and NGD plasmas) and thus devoid of pinch, are more likely to mobilise electrically, the intrinsic potential energy of the molecular charges than pinch plasmas appear to be able to do (e.g. TRD plasmas). Apparently also, the greater the vacuum drawn from the T/R gap cavity, the more efficient does the transfer of this intrinsic potential energy become, i.e. the mass-bound latent heat, to the electrokinetic energy of the charges circulating in the receiver circuit.  At about 0.06 Torr, this transfer in vacuo is comparable to that observed under atmospheric conditions and thus for a much greater density of molecules.


We investigated whether it Is possible to tap the latent heat energy of water molecules.  It is possible that in the vapour phase they can effectively hold on to their latent energy - but could they give off some of it once closely packed in liquid phase?   To test this hypothesis we immersed the T/R gap in a glass water tank.  The motor employed for these tests was a high-speed 2-phase drag-cup motor (see Fig.18 and associated description), wired in split-phase with two identical phase windings capacitatively balanced, and the galvanised iron plates each had an area of one square foot.  The results are shown in Table 7 below, and clearly indicate that it is possible to tap - within the T/R cavity - the `latent heat` of water in the liquid phase. As observed, immersion of the T/R cavity in water increased the motor output speed 22% (12,117 / 9,888) x 100).  This corresponds to a 50% increase in power output, from 18W at 9,888 rpm to 27W at 12,117 rpm:




Thus the use of ion-containing water or other ion-containing aqueous liquid in the cavity promotes long distance propagation and a greater injection of latent and thermal energies in the receiver circuit.  Such a result is not achieved if the cavity is filled with deionised water.


The preceding results lead therefore to the design of a presently preferred apparatus, based on these findings, for the conversion of mass-free electric energy, "latent heat" energy and "sensible" heat energy into conventional electric energy, as shown in Fig.16, which integrates all of the separate findings and improvements.  The winding 6 of the Tesla coil at the bottom is driven in the usual manner employing a vibrator stage 2 to pulse the primary coil 4.  The outer pole of the secondary 6 is then connected to a circular metal plate T which is one end of an evacuated cylindrical cavity, connected to a vacuum pump or sealed at a desired pressure, or which forms a still containing water or other aqueous solution or liquid.  This cavity constitutes the transmitter/receiver gap, and is therefore bounded by a dielectric envelope and wall structure 32, with the circular receiver plate R as its top surface.  In turn this plate R serves as the base of a conical Faraday cage 34, preferably air-tight and at atmospheric pressure, but which could also be subject to evacuation, which conical structure carries at its apex provisions for a cold junction 45 and any possible enhancement of the same junction by surface application of different metallic conductors that may optimise the Peltier-Seebeck effect.  The output from the cold junction where sensible thermal energy is added to the electrokinetic energy of charge carriers, is also the input to the distal end of the winding 6 of the chiral coil arrangement that sustains resonant capture of all three energy flows ((1) mass-free electric waves of a longitudinal nature, (2) true "latent heat" or the intrinsic (thermal) potential energy, and (3) the thermokinetic energy of molecules, (i.e. "sensible" heat) and, placed in series with the input of the full wave divider 8, 10, feeds the circuit output from the series capacitors 12, 14 grounded at their common tap.  In the T/R gap, the transmitted electric longitudinal wave energy is captured along with any intrinsic potential energy shed by molecules caught in the field.  Within the R element, expanded into an enclosure that guides "sensible" radiant heat, the latter is generated and then recaptured at the cold junction.


The apparatus consisting of the cylindrical T/R gap cavity and the contiguous conical cage is then preferably finished in gloss white and cylindrically enveloped within a matt black container 46 by effective thermal insulation 48, the latter terminating at the height of the bottom disc T.  Apparatus (not shown) may be provided to move the plate T vertically to adjust the T/R gap.


Another alternative embodiment of the apparatus is shown in Fig.17.  Here the circuit driving the apparatus is as we have set forth in our prior patents, which employs an autogenous pulsed abnormal glow discharge tube 50 in the configuration shown, supplied by a battery pack DP through blocking diodes 52 and an RC circuit formed by resistor 54 and capacitor 56 to drive the primary 2 of a first Tesla coil to obtain at the distal pole of the secondary 6 the energy to be injected to plate T in the form of a central electrode of a coaxial vacuum chamber (sealed or not), of which the cylindrical metallic envelope forms the receiver plate R, the latter being placed centrally inside the conical cage 34 and contiguous with its walls and base.  The top and bottom of the coaxial chamber carries suitable insulating discs, preferably with O-ring type fittings.  Again, the apparatus is enclosed in insulation within a cylindrical container 46, and the input into the capture circuit driven from the full wave divider is taken from the cold junction 45 at the apex of the air-tight cage.  The output circuit is similar to that of Fig.15.


We have found however that even when the component values in the motor driver and motor circuits are carefully selected so that these circuits are co-resonant with the dampened wave (DW) component of the motor driver pulses, the motor power output falls well short of that which should theoretically be attainable.  In an endeavour to meet this problem, we replaced the squirrel-cage type induction motor 44 by a drag cup motor of type KS 8624 from Western Electric in the expectation that the low-inertia non-magnetic rotor would allow better response to the Dampened Wave component.  This motor is similar to one of the types used by Reich in his experiments.  Although results were much improved they still fell short of expectations. Replacement of this motor by an inertially dampened motor of type KS 9303, also from Western Electric, provided much better results as discussed below.


Fundamentally, the difficulties we encountered stemmed from the inability of motor couplings to respond efficiently and smoothly, and at the same time, to the pulse and wave components of Dampened Wave impulses: that is, simultaneously to the high-intensity peak current pulses (the front end event), the DC-like component, and to the dampened wave trains these cause, i.e. the pulse tails (or back end event)-or AC-like component.  This difficulty is present even when we just seek to run induction motors from the DW impulses of a Tesla coil, the very difficulty that led Tesla to abandon his project of driving a non-ferromagnetic disc rotor mounted on an iron core bar stator with dampened waves.


We believe that the key to the capture of the mass-free energy flux output in electric form by Tesla transmitters, including any injected latent or thermal energy that have undergone conversion into electrical energy is to employ the tuned, unipolar, Y-fed, PAGD-plasma pulser driven split-phase motor drive we have invented (U.S. Pat. No. 5,416,391) in conjunction with an inertially dampened AC servomotor-generator (see Fig.18): this has a motor shaft 64 which couples a drag-cup motor rotor 60, preferably of aluminium, silver, gold or molybdenum, directly to a drag-cup generator rotor 62 that drives a permanent magnet (PM) flywheel 66, freely rotatable in bearings 67, that provides inertial damping.  The shaft 64, journalled by bearings 61 in the casing of the motor 44, provides a power output through optional gearing 68.  The phase windings of the motor 44 are wound on a stator core 70 having concentric elements between which the rotor or cup 60 rotates.  This structure makes it ideal for the capture of the DW impulses, whether sourced in the transmitter, amplified in the T/R cavity or sourced in the plasma pulser, all in synchrony.  Effectively the motor couples the damping action of the drag-cup sleeve motor rotor, which action, as we have already found for the KS-8624 motors, is quite effective at absorbing the front-end DC-like event, with the inertial damping of the PM flywheel upon the drag-cup sleeve generator rotor, that in turn is quite efficient at absorbing the back-end AC-like wavetrain event.


The KS-9154 motor used by Reich was not an inertial dampened AC drag-cup servomotor-generator.  Had Reich succeeded in overcoming the limitations of his 2-phase OR Motor solution, as we have now shown it is possible to do (by applying the Function Y circuit to the PAGD split-phase motor drive which we invented), his motor would have suffered the same limitations which we encountered with the KS 8624 motor.


Any motor, by itself, has an internal or inherent damping whereby the acceleration only vanishes when the rotor is running at constant speed.  For motors which operate on the basis of the drag principle, where the asynchronous slip is actually constitutive of the motor action, by inducing eddy currents in the rotor, the inherent damping is always more pronounced than for other induction motors.  The damping or braking torque is produced when a constant current flows through a rotating drag disc or cup.


Aside from this inherent braking, dampers can also be applied to servo motors to further stabilise their rotation. They absorb energy, and the power output and torque of the motor is thereby reduced. Optimal operation of servo motors requires both rapid response on the part of the rotor to changes in the variable or control phase, and a stable response that is free from oscillation, cogging and overshooting. The rapid response is assured by employing low inertia rotors, such as drag-cups or cast alloy squirrel-cages, and the overshooting and oscillation are reduced to a minimum by damping or a retarding torque that increases with increasing motor speed.  Typically, in a viscous-dampened servomotor, the damper is a drag-cup generator mounted rigidly on the shaft of the motor rotor, and the generator drag-cup rotates against the stator field of a static permanent magnet field.  The generator develops a retarding torque directly proportional to speed, and the energy absorbed by the damper is proportional to speed squared.  The damping can be adjusted and, as it increases, the same amount of input power yields lower torque and motor speeds. Inertial-dampened servo motors differ from viscous dampened motors in that the permanent magnet stator of the drag-cup generator is now mounted in its own bearings, either in the motor shaft or on a separate aligned shaft, forming a high-inertia flywheel.


This means that, whereas the motor rotor always experiences a viscous damping in viscous-dampened servo motors, in inertial-dampened servo motors the drag cup motor rotor only experiences a viscous damping while accelerating the flywheel, with the damping torque always opposing any change in rotor speed.  Once the flywheel rotates synchronously with the rotor, all damping ceases.  Note that this viscous damping is carried out via the coupling of the drag-cup generator rotor, rigidly affixed to the motor rotor, to the PM flywheel, so that their relative motion generates the viscous torque proportional to the relative velocity.  Use of drag-cup sleeve rotors in inertially dampened servo motors was largely supplanted by squirrel-cage rotors once the latter became produced as cast alloy rotors.  Since inertially dampened motors can be used in open and closed-loop servo applications, and present better stability - even in the presence of non-linearities - and higher velocity characteristics than other induction motors do (Diamond, A (1965) "Inertially dampened servo motors, performance analysis", Electro-Technology, 7:28-32.), they have been employed in antenna tracking systems, stable inertial-guidance platforms, analogue to digital converters, tachometers and torque tables.


The typical operation of an inertially dampened servomotor is as follows: with the reference phase fully excited, the motor rotor -fixedly linked to the generator rotor, as well as the flywheel - remain immobile; once power is applied to the control phase, the motor rotor immediately responds but the flywheel remains at rest. However, as the drag-cup generator 62 is forced to move through the permanent magnetic field of the flywheel, it creates a drag torque that slows down the attached motor rotor proportionally to the acceleration that it imparts to the flywheel that it now sets into motion, thus creating the viscous damper.   As the flywheel accelerates, the relative speed of the motor with respect to the flywheel, as well as the damping torque, decrease until both motor and flywheel rotate synchronously and no damping torque is exercised - at which point the drag on the motor cup exerted by the generator cup is negligible.


The KS-9303 motor is an inertial dampened servomotor but is differentiated with respect to other inertially dampened motors, in that (1) it employs a drag-cup sleeve motor rotor made of aluminium, very much like that of the KS-8624, but with slightly altered dimensions and with a shaft extension for the drag-cup copper generator rotor, and (2) the moving flywheel structure was journalled on a separate, fixed shaft, as already described with reference to Fig.18.  Now, in principle, even application of minimal damping decreases motor efficiency, resulting in diminished torque and speed.  Whether the inertial-dampened motor has a drag-cup rotor, a sleeve rotor or a squirrel-cage rotor, the damping increases the rotor slip.  Laithwaite considers drag-cup motors as being "dynamically inferior to their cage counterparts" (Laithwaite, E R (1957) "Induction machines for special purposes", London, England, p. 323).  If we now add a viscous damping and retarding torque, we should not be able to get much more than a 55% efficiency in the best of conditions.  On the other hand, the inertial damping arrangement described will only abstract or supply energy when the motor rotor is accelerating or decelerating relative to the flywheel.


These drag-cup motors, whether inertially dampened or not, develop a constant torque at constant rpm for a given supply frequency and a suitable phase shift capacitance.  For each frequency the motors respond to, there is an optimum resonant split-phase capacitance, but other values nearby are still suited for operation, and for each value of capacitance, there is an optimum frequency to which the motors respond.  For example the KS-8624 motor responds best at 450 Hz when a 1 microfarad capacitance is employed, responds best at 250 Hz when a capacitance of 10 microfarads is employed, and responds best at 60 Hz, when a capacitance of 100 microfarads is employed.  As the capacitance increases, the resonant CW frequency of the motor is displaced to lower values.  If we fix the capacitance at a value (e.g. 10 microfarads) suitable for testing the frequency response at a fixed voltage of 12 VAC, the observed result for both the KS-8624 and KS-9303 motors show a response distribution of the motor rotary velocity that has an identical peak at 250 Hz for both motors, with the response decreasing to zero smoothly on both sides of the peak.


These results indicate that, when wired as a split-phase motor, the motor rotary velocity varies not as a function of voltage or current, but as a function of frequency when the phase-splitting capacitance is fixed within a suitable range, there being an optimum frequency mode for each value of suitable capacitance, with lower values of capacitance favouring higher frequency modes.  For a given frequency and capacitance, the motor rotary velocity remains essentially constant and independent from voltage and current input, and thus at a plateau.  Torque, in the same circuit arrangement, follows exactly the same pattern as rotary velocity, as a function of input frequency at a fixed potential.  Torque is linearly proportional to rpm in these motors when they are split-phase wired, and rpm linearly proportional to CW frequency, which makes them ideal for experimentation and determination of power output computations.  Moreover, since these are drag machines, the slip itself determines the rotor currents and these are susceptible to tuning such that their retardation and relative position in the field can find resonant modes for varying CW frequency and capacitance.


In the circuit of Fig.17 when using the KS 9303 motor, the inertial damping of the flywheel coupling retards the motor rotor currents sufficiently to allow them to build up torque, with the entire motor assembly serving as the preferred sink for all of the energy, mass-free and mass-bound, captured by the receiving coil circuit with a drawing action established by the motor on the circuit, and providing satisfactory absorption by an inertial damper of the combined, synchronised, dampened wave impulses, those occurring at a low frequency as a result of the firing of the PAGD reactor, and those occurring at a higher superimposed frequency -sourced in the transmitter circuit and picked-up by the receiver plate and coil.  The action of each DW impulse train itself generates two different events: the DC-like auto-electronic-like discontinuity which sets the motor in motion and initiates the rotor currents, and the AC-like dampened wavetrain which supports the consistency of those rotors.  The concentration of current required to kick-start the motor is provided by the DW impulses of the PAGD reactor, whereas, once the motor is in motion, and particularly, once it is stabilised by the flywheel, the cumulative action of the higher frequency DW impulses makes itself felt by accelerating the rotor to an optimum rotary velocity.


For the next series of tests we employed the basic circuit diagram of the improved motor shown in Fig.19. The transmission station is the typical Tesla transmitter with a line-fed, 60 Hz vibrator stage.  At the line input to the first stage, we place a calibrated AC wattmeter (Weston Model 432), and a Beckman 330B rms ammeter in series with the hot lead, we set the vibrator stage for 41 clicks, consuming between 28.5W and 35W, depending upon circumstances yet to be described.  This consumption was confirmed by driving the coil from an inverter powered by a 12 volt battery.  The inverter consumes 2.16 watts, and is 90% efficient. The total consumption from the battery was 42 watts (12V at 3.5A); once the 2.16 watts is deducted and the efficiency taken into account, we obtain the same 36W (vibrator stage at max., i.e. 47 clicks, in this experiment). The T/R gap is adjusted to 3'', and 2 square foot plates are used.  Transmitter and receiver coils are tuned, and so are the plate capacitances, to 250 kHz, also the capacitances of the Function Y circuit connected at the output of the receiving coil.


The rectified voltage and current generated by the transmitter secondary and by the transmitter plate was ascertained with a coil-tuned wave-divider (Function Y) circuit by loading it with different resistive values.  The results constitute a measure of the mass-bound electrical power output directly from the transmitter apparatus. The same method was employed to ascertain the voltage, current and power of the mass-bound charges circulating in the receiving plate and coil circuit.  The results are shown in Table 8 below:




The results indicate that the highest mass-bound power assembled by the secondary transmitter circuit does not exceed 7 watts - and this is directly output from the secondary 26 when the load is 50 Megohm, or from the transmitter plate when the load is 1 Megohm.  The mass-bound electric power emulated by the receiving circuit (plate, coil and Function Y without the plasma pulser circuitry) never exceeds the mass-bound electric power outputted directly by the transmitter, and peaks when the resistive load value (1 Megohm) approaches the pre-breakdown resistance range of the vacuum tube, at 4.72W.  These findings then indicate that when the transmitter circuit is consuming a maximum of 35W, a typical output from the secondary of the transmitter is 7W, and at 3'' of distance within the proximal field of the latter, the pick-up by a tuned receiver will be of the order of 5W of mass-bound current duplicated within the receiving coil.  The loss in the first stage is therefore on the order of sevenfold.


Continuing with the description of the circuit of Fig.19, a 128 cm2 plate area, 6 cm gap PAGD reactor is used, connected as described in our prior art to a high-vacuum rotary pump (Correa, P & Correa, A (1995) "Energy conversion system", U.S. Pat. No. 5,449,989).  Pressure readings were obtained with a thermocouple gauge during the operational runs.  The KS-9303 motors to be tested are then connected to the PAGD reactor in the usual capacitatively-coupled, inverter fashion described in our prior art (Correa, P & Correa, A (1995) "Electromechanical transduction of plasma pulses", U.S. Pat. No 5,416.391).  Their rpm is detected by a stroboscopic tachometer and fed to a Mac Performa 6400 running a motor algorithm program calculating the power output.  Motor measurements were made at five minutes into each run for the unloaded motors, and at ten minutes for the inertially dampened motors.


All experiments were carried out in the same work session. The experimental determination of the continuous rotary power output as a function of the reactor pulse rate confirmed that the improved circuit develops maximum rotary capture of the mass-free energy in the receiver circuit at the lowest rates of pulsation, just as we have previously found for the conversion system of U.S. Pat. No. 5,449,989.  Furthermore, the data showed that even motors of type KS-8624 are able to output power mechanically in excess of the mass-bound power output by the transmitter (7W) or captured by the receiver (5 to a max. of 7W), once the PAGD rate decreases to 1.5 PPS.  Such an anomaly can only be explained by the system having become able to begin capturing the mass-free energy flux in the receiver circuit that we know already is output by the transmitter circuit.  But this excess mechanical power is still less than the power input into the transmitter, and clearly so.  It represents a power gain with respect to the secondary, but a loss with respect to the primary.  The full breadth of the capture of the mass-free electric energy flux circulating in the receiver circuit is not seen until the motors are resonantly loaded because they are inertially dampened.


The KS-9303 motors, once inertially dampened, and thus loaded, are able to recover enough power from the mass-free energy field to develop a mechanical power, not just greatly in excess of the mass-bound power of the secondary, but also greatly in excess of the mass-bound power input to the vibrator stage and the primary, at 28 to 35W.  Once the pulse rate approaches the same 1.5 PPS marker, mechanical power in excess of the mass-bound electric power input to the primary becomes evident, peaking at nearly three times that input.  In fact, the highest output recorded was also obtained with the lowest input to the transmitter circuit, the highest exact coefficient observed in this experiment being 100.8W / 28W = 3.6.  Furthermore, with respect to the secondary mass-bound output, the same mechanical rotary output represents a much greater overunity coefficient of performance, on the order of 14.4 times greater.  This is at least partly the result of the receiver and motor capture of the mass-free electric energy output by the transmitter, and may be partly the result of mass-free energy engrafted by the PAGD regime in the PAGD reactor. 


Reviewing the mechanical power output results as a function of increasing vacuum in the PAGD reactor and at different output power levels, any motor performance below the 5-7W limit of the traditional mass-bound output power of the secondary represents an output mechanical power loss with respect to both the mass-bound secondary output and the mass-bound primary input.  All the results for pressures down to 0.03 Torr fall into this category, and thus represent a very inefficient coupling to the PAGD regime.  Any motor performance between 7W and 28-35W represent a loss with respect to the electrical power input to the transmitter system, but a net gain of power with respect to the mass-bound secondary power output.  None of the non-inertially dampened motors tested were able to perform outside of this area, under the test conditions.  With more efficient primary to secondary couplings in the transmitter station, however, one could advantageously employ these motors alone to extract some of the mass-free power of the secondary or to operate them in enclosed vessels without conventional external electrical connections.


To reach satisfactory levels of recovery of mass-free energy, one must dampen the superimposed DW impulses.  Hence, all results showing outputs in excess of 35W were obtained using the inertially dampened KS-9303 motors, and represent a net overunity power gain over both the power input to the primary and the mass-bound power output by the secondary, or the mass-bound power emulated by the receiver circuitry. This happens when the PAGD pulse rate falls to 2 PPS, with the rotary power output steeply increasing as the rate falls to 1 PPS.


One of the interesting features of the motor circuitry we have proposed is that it can operate with pulsed plasmas in both the TRD and the AGD regions, the least efficient response occurring in the NGD region near the Paschen minimum.  One might think that the voltage depression would allow increased current intensity supplied to the motors, but in fact that is not observed, with the flashing of the NGD yielding erratic oscillations and low values of current.  In keeping with the notion that the TRD plasma is mainly composed of lagging positive ions, whereas the PAGD plasma is mostly an electron plasma, the observed direction of rotation of the motors is opposite in the TRD region to that of the AGD region.  The NGD region therefore marks the depression where the velocity vectors change direction.  In the second or PAGD region, motor operation is very quiet, unlike what is observed in the TRD region.


Part and parcel of the tuning of the circuit components is the selection of the optimum capacitances employed to couple the PAGD reactor to the motor circuit and split the phase to feed the auxiliary winding of the motor.  We have experimented with capacitances ranging from 0.5 to 100 microfarads, and found that best results (for the specific circuit in question - including the characteristics of the transmission), were such that the optimum value of the PAGD coupling capacitance lay near 4 microfarads, and the phase splitting capacitance, near 1 to 4 microfarads, depending upon weather conditions.  In good weather days lower capacitance values can be used, while in bad weather days higher capacitances are needed.  For ease of comparison in demonstrating the need to tune the circuit by employing optimum capacitances in those two couplings (reactor to motor, and motor phase coupling), we employed the same capacitances in both circuit locations.  


A comparison of tests using 1 and 4 microfarad values shows the difference caused by changing those capacitances from their optimum value: across all discharge regions of the pressure range that was examined, the four motors tested, operated with greater motor speeds when the capacitances are set to 4 microfarads rather than to 1 microfarad.  The less efficient performance obtained with 1 microfarad capacitance fits the inverse correlation of pulse power with increasing pulse frequency, such as we have found for the PAGD regime. This is made evident by a comparison of rpm versus pulse rate for the two capacitance values being considered.  They demonstrate the higher pulse rates observed with the lower capacitance, that correlate with the lower motor speeds, and result in lower efficiency of the motor response. The results equally indicate that low capacitance values increase the pulse rate, but if this increase is out of tune with the rest of the circuit values, it results in power waste because it imposes a rate that is not optimum.


We have also determined experimentally that the efficiency of the system is affected by external weather conditions, higher efficiencies being noted on a fine bright day than under poor weather conditions even though the apparatus is not exposed to such conditions.  This may reflect a diminution under poor weather conditions of latent mass-free energy that can be taken up by the system.


The observed high efficiency of circuits including inertially dampened motors indicates that the phenomenon does not reduce to a mere optimum capture of, DC-like pulses produced by the reactor in what is essentially an AC motor circuit.  Effectively, the pulsed plasma discharge deploys a front-end, DC-like pulse, or discontinuity, but this is followed by an AC-like dampened wave of a characteristic frequency (having a half-cycle periodicity identical to that of the front-end pulse) to which the motor circuit also responds.  Moreover, the mass-free electric radiation from the transmitter circuit itself induces, in the receiver antenna, coil and circuit, and in the reactor discharge itself, the train of finer dampened wave impulses responsible, after conversion through the wave-divider, for the mass-bound rectified current which is employed to charge the plasma reactor to begin with.  Serving as trigger of the plasma discharges in the reactor are the DW impulses circulating in the receiver circuit, such that the two different lines of DW impulses, in the receiver circuit (for example 120 PPS for the pulses and 154 kHz for the waves) and from the reactor, are synchronised by interpolated coincidences, since their pulse and wave frequencies are different.  Ideally, these two superimposed DW frequencies are harmonics or made identical.  The receiver stage involves capture of the mass-free electric energy received from the transmitter, duplication of the mass-bound current in the receiver coil, and injection of latent and sensible thermal energy in the T/R gap cavity which augments the emulated mass-bound current.


The mass-bound current is employed to charge the wave-divider capacitance bridge and therefore the reactor.  In turn, the plasma pulses from the reactor are superimposed with the DW impulses from the receiving coil, and together they are coupled to the split-phase motor drive.  Hence the first receiver stage employs the totality of the energy captured in the T/R gap cavity - mass-free electric energy transmitted by the T plate, latent and sensible thermal energy injected at the surface of the R plate - and produces in the receiving coil a mass-bound current comparable to that assembled in the transmitter coil by the action of the primary.   The mass-bound current is stored in the wave-divider bridge and used to drive the plasma reactor in the PAGD region.  Subsequently, the autogenous disruptive discharge that employs a substantial electron plasma generates both a concentrated, intense flux of mass-bound charges in the output circuit, and a mass-free oscillation of its own. The dampened motor is therefore fed directly with (1) the intense mass-bound current output from the reactor; (2) the pulse and wave components of the mass-free electric energy captured by the receiver plate and coil (and matched by conduction through the earth), and which are gated through the wave-divider and the reactor for the duration of the PAGD channel; and (3) any mass-free latent energy taken up from the vacuum by the PAGD event.  Once the motor is set into motion, and is resonantly loaded with an inertial damper, we believe that it will also respond to the much weaker DW impulses captured by the receiver, since these impulses encompass both a DC-like front end - further enhanced by analytic separation through the wave-divider - and a dampened wave at 154 kHz. 


Essentially, the DW impulses that are ultimately sourced in the transmitter - and received unipolarly through the T/R gap - have sufficient DC-like potential (plus all the other requisite physical characteristics, such as frequency) to contribute directly to the motor response, once the motor has gained substantial speed (for they lack the current to set it into motion, one of the contributions from the plasma pulser). This is the case, provided that the motor itself is suited for absorption of both DC-like pulses and AC-like dampened waves, which is precisely the case with motors of the type shown in Fig.18 since the inertia of the flywheel is overcome through homopolar absorption of the dampened oscillations simultaneously in the motor drag-cup rotor and in the generator drag-cup rotor.


We also tested these inertially dampened motors in the traditional DC power supply-driven PAGD circuit we have taught in our previous patents, that is, circuits with an overt HV DC power source, and thus in the absence of any Function Y circuit or transmitter circuit.  Here then, only the DW impulses generated by the PAGD reactor can account for the motor response. The tube employed (A31) had an area of 256 cm2, and a gap distance of 4 cm.  Coupling capacitances employed were 4 microfarads for the inverter coupling, and 1 microfarad for the split phase motor coupling.  The DC power supply delivered up to 1 ampere of current between 150 and 1,000 VDC, and the ballast resistor was adjusted to 215 ohms.  Having determined the basic physical characteristics of the reactor's behaviour in the circuit under consideration, we conducted our experiment in the PAGD region.  We chose a pressure of 0.6 Torr, just off from the Paschen minimum, as we intended to benefit from the lower sustaining voltage which it affords. 


The experiment basically consisted of increasing the sustaining voltage at this fixed pressure in the PAGD regime, and measuring the diverse physical parameters of the circuit and motor response in order to ultimately ascertain the difference between the input electric DC power and the output mechanical rotary power.  We first looked at how the motor rpm response varied as a function of the sustaining voltage (Vs): the results illustrate the importance of starting close to the Paschen minimum in the pressure scale, since the KS-9303 motors reach plateau response (at 17,000 rpm) when the reactor output voltage nears 450V.  Any further increase in potential is simply wasted.  Likewise, the same happened when we measured motor speed as a function of increasing peak DC current, plateau response being reached at 0.1 ADC.  Again, any further increase in current is wasted.   Essentially then, the optimal power input to the reactor when the  output of the latter is coupled to the motor, lies around 45 watts.  This is a typical expenditure in driving a PAGD reactor.  As for pulse rate we once again find a motor response that is frequency proportional in the low frequency range, between 10 and 40 PPS (all pulse rates now refer solely to PAGDs per sec), but once rates of >40 PPS are reached, the response of the motor also reaches a plateau.


The observed increment in speed from 40 to 60 PPS translates only into an increase of 1,000 RPM, from 16,000 to 17,000 RPM.   So, we can place the optimal PAGD rate at ca 40 PPS.  The DC electric power input to drive the PAGD reactor was next compared to the rotary mechanical power output by the inertially loaded motor, driven in turn by the reactor.  This comparison was first carried out with respect to the PAGD rates. The motor response far exceeds the conventional input power, indicating that the whole system can be tuned to resonance such that optimal power capture inside the reactor takes place, the critical limit rate lying at around 60 PPS, when the motor response is firmly within the pulse response plateau.  At this juncture, the break-even efficiency for the measured rates of energy flux over time reach 700% (overunity coefficient of 7), in keeping with the observations and the values we have made in the PAGD conversion system. In the proportional part of the curve, before the plateau is reached, even greater rates of break-even efficiency - up to >1,000% were registered.


These results constitute the first time we have been able to confirm the presence of output energy in excess of break-even over conventional mass-bound energy input in the PAGD inverter system, and the results are comparable to what we have observed and previously reported for the PAGD converter system.  At pulse rates greater than 60 PPS a greater input power results in decreased efficiency, also translated into a noticeable heating of the reactor and motor.   And this is all the more remarkable as experiments we have conducted with inductive tuning of PAGD reactors, or employing PAGD reactors as replacements for the primaries of Tesla coil assemblies, and still, more recently, with the PAGD inverter circuit driving motors, have all shown that it is possible to operate these reactors with minimal mirroring and heating, preserving essentially the cold-cathode conditions and yet focusing the plasma column so that deposition on the insulator is negligible.   It appears that above a certain threshold of optimal efficiency, surplus input energy is just dissipated thermally by both the reactor and the motors.


It should be understood that the above described embodiments are merely exemplary of our invention, and are, with the exception of the embodiments of Figs. 16 to 19 designed primarily to verify aspects of the basis of the invention. It should also be understood that in each of these embodiments, the transmitter portion may be omitted if an external or natural source of Tesla waves is available, provided that the receiver is tuned to the mass-free radiation mode of the source.   For example if solar radiation is available in which the mass-free component has not interacted with the earth's atmosphere (as in space applications), the receiver is tuned to the voltage wave of the mass-free radiation sourced in the sun, e.g. by using a Tesla coil in the receiver constructed to have an appropriate voltage wave close to the 51.1 kV characteristic of such radiation.
















US Patent 5,449,989            12th September 1995             Inventors: Correa, Paulo and Alexandra





This patent shows a method of extracting environmental energy for practical use.  In the extensive test runs, an input of 58 watts produced an output of 400 watts (COP = 6.9).  This document is a very slightly re-worded copy of the original.




An energy conversion device includes a discharge tube which is operated in a pulsed abnormal glow discharge regime in a double ported circuit.  A direct current source connected to an input port provides electrical energy to initiate emission pulses, and a current sink in the form of an electrical energy storage or utilisation device connected to the output port captures at least a substantial proportion of energy released by collapse of the emission pulses.


US Patent References:

3205162    Sep, 1965         MacLean.         

3471316    Oct, 1969          Manuel.

3705329    Dec, 1972         Vogeli. 

3801202    Apr, 1974          Breaux.

3864640    Feb, 1975          Bennett.           

3878429    Apr, 1975          Iwata.   

4009416    Feb, 1977          Lowther.           

4128788    Dec, 1978         Lowther.           

4194239    Mar, 1980          Jayaram et al.   

4443739    Apr, 1984          Woldring.         

4489269    Dec, 1984         Edling et al.      

4527044    Jul, 1985           Bruel et al.       

4772816    Sep, 1988         Spence.           

4896076    Jan, 1990          Hunter et al.     

5126638    Jun, 1992          Dethlefsen.


Other References:

Tanberg, R. "On the Cathode of an Arc Drawn in Vacuum", (1930), Phys. Rev., 35:1080.

Kobel, E. "Pressure & High Vapour Jets at the Cathodes of a Mercury Vacuum Arc", (1930), Phys. Rev., 36:1636.

Aspden, H. (1969) "The Law of Electrodynamics", J. Franklin Inst., 287:179.

Aspden, H. (1983) "Planar Boundaries of the Space-Time Lattice" Lettere Al Nuovo Cimento, vol. 38, No. 7, pp. 243-246.

Aspden, H. (1980) "Physics Unified", Sabberton Publications, pp. 14-17, 42-45, 88-89, 190-193.

Pappas, P. T. (1983) "The Original Ampere Force and Bio-Savart & Lorentz Forces", Il Nuovo Cimento, 76B:189.

Graham, G. M. & Lahoz, D. G. (1980) "Observation of Static Electromagnetic Angular Momentum in Vacuo", Nature, vol. 285, pp. 154 & 155.

Sethlan, J. D. et al., "Anomalous Electron-Ion Energy Transfer in a Relativistic-Electron-Beam-Plasma" Phys. Rev. Letters, vol. 40, No. 7, pp. 451-454 (1978).



This application is a continuation-in-part of U.S. application Ser. No. 07/922,863, filed Jul. 31, 1992 (abandoned), and is also a continuation-in-part of U.S. patent application Ser. No. 07/961,531, filed Oct. 15, 1992, now U.S. Pat. No. 5,416,391.





1. Field of the Invention:

This invention relates to energy conversion circuits utilising discharge tubes operating in the pulsed abnormal glow discharge (PAGD) regime.



2. Review of the Art:

Such discharge tubes and circuits incorporating them are described in our co-pending U.S. patent application Ser. Nos. 07/922,863 and 07/961,531.  The first of these applications discloses discharge tube constructions particularly suited for PAGD operation, and the second discloses certain practical applications of such tubes, particularly in electric motor control circuits.  The review of the art contained in those applications is incorporated here by reference, as is their disclosure and drawings.


It is known that there are anomalous cathode reaction forces associated with the cathodic emissions responsible for vacuum arc discharges, the origin and explanation of which have been the subject of extensive discussion in scientific literature, being related as it is to on-going discussion of the relative merits of the laws of electrodynamics as variedly formulated by Ampere, Biot-Savart and Lorentz.  Examples of literature on the subject are referenced later in this application.




The particular conditions which prevail in a discharge tube operated in the PAGD regime, in which a plasma eruption from the cathode is self-limiting and collapses before completion of a plasma channel to the anode gives rise to transient conditions which favour the exploitation of anomalous cathode reaction forces.


We have found that apparatus utilising discharge tubes operated in a self-sustaining pulsed abnormal glow discharge regime, in a double ported circuit designed so that energy input to the tube utilised to initiate a glow discharge pulse is handled by an input circuit substantially separate from an output circuit receiving energy from the tube during collapse of a pulse, provides valuable energy conversion capabilities.


The invention extends to a method of energy conversion, comprising initiating plasma eruptions from the cathode of a discharge tube operating in a pulsed abnormal glow discharge regime utilising electrical energy from a source in a first circuit connected to said discharge tube, and capturing electrical energy generated by the collapse of such eruptions in a second circuit connected to the discharge tube.





The invention is described further with reference to the accompanying drawings, in which:



Fig.1 shows variation of applied DC current and pulse AC rms currents characteristic of a low current PAGD regime, as a function of decreasing pressure, for a 128 cm2 H34 aluminium plate pulse generator having a 5.5 cm gap length and being operated in the single or plate diode configuration of FIG. 11A, at about 600 V DC.



Fig.2 shows variation of applied DC current and AC rms currents of a high current PAGD regime, as a function of the decreasing pressure, for a device identical to that of Fig.1, and operated at the same potential.



Fig.3 shows PAGD rate vs pulse generator cathode temperature as a function of the time of continuous PAGD operation, for a pulse generator with 64 cm2 plates having a 4 cm gap distance, operated at a DC voltage of 555 (av) and R1 = 600 ohms (see Fig.9).




Fig.4 shows PAGD frequency variation with time, for 18 successive spaced one-minute PAGD runs for a pulse generator with 128 cm2 plates, and a 5.5 cm gap distance, operated at V DC = 560 (av) and R1 = 300 ohms.



Fig.5 shows variation of the PAGD frequency in pulses per minute (PPM) with increasing charge of a PAGD recovery charge pack (see Fig.9), as measured in terms of the open circuit voltage following 15 minutes of relaxation after each one minute long PAGD run, repeated 18 times in tandem, under similar conditions to Fig.4.



Fig.6 shows volt amplitude variation of continuous PAGD at low applied current, as a function of decreasing air pressure, for a 128 cm2 plate area device, gap length = 5 cm; (DC V at breakdown = 860).



Fig.7 shows volt amplitude variation of continuous PAGD at high applied current as a function of the decreasing air pressure, for a 128 cm2 plate area device, gap length = 5 cm; (DC V at breakdown = 860).




Fig.8 is a schematic diagram of a first experimental diode (without C6) or triode PAGD circuit.



Fig.9 is a schematic diagram of a preferred diode or triode PAGD circuit in accordance with the invention.






Fig.10A, Fig.10B and Fig.10C are fragmentary schematic diagrams showing variations in the configuration of the circuit of Fig.9.



Fig.11 is a modification of Fig.9, in which an electromagnetic machine, in the form of an electric motor, is connected into the circuit as an accessory electromechanical arm.



Fig.12 shows a further development of the circuit of Fig.9, permitting interchange of driver pack and charge pack functions.



Fig.13 shows open circuit voltage relaxation curves for battery packs employed in tests of the invention, respectively after pre-PAGD resistive discharge (DPT1 and CPT1), after a PAGD run (DPT2 and CPT2) and after post-PAGD resistive discharge (DPT3 and CPT3).



Fig.14 shows an example of negligible actual power measurements taken immediately before or after a PAGD run, showing both the drive pack loss and the charge pack gain in DC Watts; DP resistance = 2083 ohms; CP resistance = 833 ohms.





Fig.15A and Fig.15B show resistive voltage discharge curves for two separate lead-zero gel-cell packs utilised respectively as the drive and the charge packs; load resistances employed were 2083 ohms across the drive pack (Fig.15A) and 833 ohms across the charge pack (Fig.15B).




Fig.16 shows resistive discharge slopes for a drive pack before and after a very small expenditure of power in providing energy input to a PAGD run; R = 2083 ohms.




Fig.17 shows resistive discharge slopes for a charge pack before and after capturing energy from the collapse of PAGD pulses in the same test as Fig.15; R = 833 ohms.



Fig.18 shows resistive discharge slopes for a drive pack before and after a very small expenditure of power in providing energy input to a PAGD run in a further experiment; R = 2083 ohms.




Fig.19 shows resistive discharge slopes for a charge pack before and after capturing energy from the PAGD run of Fig.18; R = 833 ohms.




Fig.20 shows an example of operational measurements taken videographically during a 10 second period for both the power consumption of the drive pack (PAGD input) and the power production captured by the charge pack (PAGD output); the two values are also related by the expression of percent break-even efficiency.




Fig.21 shows variation of PAGD loaded voltage of a drive pack (in squares) compared with the PAGD charging voltage of the charge pack (in circles), during more than 1 hour of continuous PAGD operation.





The basic PAGD function and the construction of discharge tubes specifically designed for PAGD operation are described in our corresponding co-pending applications Nos. 07/922,863 (the “863” application) and 07/961,531 (the “531” application).  For purposes of the experiments described below four aluminium H34 plate devices (one with 64 and three with 128 cm2 plate areas) and three aluminium (H200) plate devices (one with 64 and two with 128 cm2 plate areas), with inter-electrode gap lengths of 3 cm to 5.5 cm, were utilised at the indicated vacua, under pump-down conditions and with either air or argon (ultra high purity, spectroscopic grade 99.9996% pure) constituting the residual gas mixture.  The pump-down conditions were as described in the “863” application. Some experiments were performed with the tubes under active evacuation, at steady-state conditions, while others utilised sealed devices enclosing the desired residual gas pressures.


The circuit designs utilised in the various experiments to be described are set out further below, and represent further developments and extensions of the circuits set forth in the “531” application.


Test equipment utilised was as follows:


An Edwards (trade mark) thermocouple gauge (TC-7) was employed for the determination of pressure down to 1 micron of mercury (0.001 Torr).


Banks of Beckman (trade mark) rms multimeters 225 and 330 (30 and 100 kHz bandwidths, respectively) were utilised for all current measurements.


Frequency meters capable of discriminating events up to 0.1 nanosecond apart, and having adjustable amplitude windows, were used. Direct analysis on a Tektronix (trade mark) dual-trace, storage scope (Model 549) was also carried out for both parameters.


Split-phase, single-phase and two-phase motors were employed, of the synchronous, induction and universal types, as previously described in the “531” application, in the accessory electromechanical arm that may be coupled to the power producing circuit described in the present application.


Large banks of 12 V, 6 Ah lead-acid gel cells (Sonnenschein (trade mark) A212/6S) were utilised either as power sources (designated as drive packs) or as accumulators of the energy (referred to as charge packs) captured by the test circuits. Charge packs made of rechargeable 9V NiCad or of nominally non-rechargeable C-Zn or alkaline batteries were also utilised.


PAGD emission areas were determined by metallographic examination of a series of craters produced by PAGDs in clean H34 cathodes, under a metallurgical Zeiss (trade mark) standard 18 microscope equipped with an epi-fluorescent condenser, very high power apochromatic objectives and a 100 W mercury lamp.  For best results a focusable oblique source of light (12V halogen) was also added to the incident light.


Following our low and high applied current studies on PAGD production as set forth in the “863” application, we noticed that the AC rms value of the component associated with each abnormal glow discharge pulse varied non-linearly with the magnitude of the applied current.  We originally noted the existence of a current induced shift of the entire PAGD region upward in the pressure scale: while the PAGD regime became more clearly defined as the applied constant DC was increased, the pressure required to observe the PAGD increased two to three orders of magnitude.  In the course of these rarefaction studies we found that, at applied currents of 1mA or less, the rms value of the different AC waveforms associated with the consecutive regimes of the discharge (TRD --> NGDm --> AGD+PAGD) was, by more than half log, inferior to the value of the applied DC current, during the first two regimes (TRD and NGD) and reached a value equivalent to the applied current with the onset of spontaneous PAGD, at pressures < 0.1 Torr (see Fig.1); however, in the downward tail of the PAGD regime (down to 3 x 10-3 Torr), the AC rms current component of each PAGD again decreased to more than half log of the intensity of the applied DC value, in a manner proportional to the log of the decreasing pressure. In stark contrast, at high applied currents of about 500 mA, and aside from the high current-induced upward shift in pressure of the PAGD regime (to the point that the compression of the previous regimes on the pressure scale results in their suppressing, as was the case in the present example), the AC rms component associated with each pulse (see closed circles, Fig.2) is, from onset of the discharge at about 8 Torr, greater in magnitude than the value of the applied current (open circles, Fig.2). Under the conditions described, the distribution of the field current associated with each pulsed abnormal glow discharge approached (on a linear Y axis; not shown) an uni-modal gaussian distribution with the pressure peak at about 1 Torr, and a corresponding observed maximum of 7.5 times. higher AC rms values than the applied DC values.


We have previously described in the “863” application how the PAGD frequency is affected by several factors, namely:

the magnitude of the parallel discharge capacitance,

the value of the negative pressure for the relevant vacuum PAGD range,

the magnitude of the applied potential, the magnitude of the applied direct current,

the inter-electrode gap distance and

the area of the parallel plate electrodes.


In the “531” application we have also described how the wiring configuration (plate diode versus triode) affects the PAGD frequency by adding tungsten auto-electronic emissions from the axial electrode, to those emissions from the plate.  There are other factors which limit the PAGD regime of discharge and have also been discussed in the “863” application. The following data indicates their specific effect upon PAGD frequency.


In the data presented in Table 1, control of the frequency parameter for the circuit shown in Fig.9 is by a ballast resistance R1 within a specific range of interest (about 800-150 ohms, for Table 1 experimental conditions), and this in turn increases the applied current which, at "high current" values (i.e. >100 mA, as for Table 1 conditions), will drive the PAGD frequency up, as previously reported in the “863” application.


Table 2 shows the effect of the progressive displacement of a given frequency, chosen as 200 PPS, with the cumulative pulse count of the same device, in the plate diode configuration.  This displacement of the same frequency (cf. group numbers 1-3 of Table 2) on to higher pressure regions is shown to be promoted by the alteration of the work function of the PAGD emitting cathode, such as this is caused by the cumulative pulse count and resultant crater formation on the electrode surface.  After the first million pulses, the anode facing cathode surface is completely turned over by emission sites, and this corresponds well to the threshold crossed by group 2 of Table 2.  Once the cathode surfaces are broken in, the rates shown in groups 3 and 4 of Table 2, tend to remain constant.


Originally we wondered whether this might be caused by the alteration of the electrostatic profile of the plasma sheaths at the periphery of the envelope, due to the mirroring deposits that result from the sputter of ions and trapped neutral atoms (from air gases or metallic vapour) associated with the auto-electronic emission mechanism (and from further emissions triggered in turn, by secondary ionic bombardment of the cathode with molecular species present in the plasma ball formed over the primary emission site).  However, reversal of the plate polarity (firing the ex-anode as a crater-free cathode) for over a million counts, followed by re-reversal to the original polarity, the entire operation being performed in air as the residual gas substrate, led to the partial recovery of the original work function for as long as the test was run (1.5 x 104 pulses), as shown by a comparison of groups 2, 4 and 5, of Table 2.   From a metallographic examination of the surfaces of plates used solely as anodes, we have also concluded that prolonged PAGD operation has the effect, not only of cleaning the anode surface from surface films and adsorbed gases, as ionic bombardment promoted by electromagnetic induction coils does, but it also does more: it polishes the target surface and smoothes it by a molecular erosive action.  Observations of the surface of reversed cathodes, shows the same smoothing and polishing effects observed in exclusive anodes. Thus the recovery of the PAGD rates promoted by polarity reversal of the plates is not a function of the sputter-promoted mirroring deposits on the envelope wall, but a function of the actual work-function of the emitting cathode.


Another variable that interacts with the PAGD frequency is the molecular nature of the residual gas: Table 3 shows the differential frequency response of air with a halogen quencher, argon, for the same pulse generator employed in the tests of Table 2.  It is apparent that argon obtains much higher rates of AGD pulsation for the same range of negative pressure, for the same "broken in" cathode, than does the air mixture.   All these measurements were taken at cathode support-stem temperatures of 350C.

Time of operation is also a variable affecting the frequency and operating characteristics of the cathode, as it becomes expressed by the passive heating of the cathode, an effect which is all the more pronounced at the higher pressures and at the higher frequencies examined.  Utilising the triode circuit discussed in the next section, the pulse rate of a PAGD generator with 64 cm2 plates can be seen (see Fig.3) to decrease, at a negative pressure of 0.8 Torr, from 41 PPS to the operating plateau of 6 PPS within 15 minutes of continuous operation, as the temperature of the cathode support increased from 190C to about 440C.  As the temperature plateaus at about 510C +/- 10C., so does the pulse rate at 6 PPS, for the remaining 48 minutes of continuous operation.


However, in order to confirm this time-dependent heating effect and threshold, we also performed the same experiment, utilising the same circuit and the same negative air pressure, with twice as large a cathode area (128 cm2, which should take nearly twice as long to heat), being operated for 18 one-minute long continuous periods equally spaced apart by 15 minutes of passive cooling, with the cathode stem always at 19.70C to 210C., room temperature at the start of each period.   The results surprised us, inasmuch as they showed that for a larger area tube which takes longer to heat to the same temperatures at comparable rates of PAGD triggering, one could observe a much earlier frequency reduction (by half, within the first 5 minutes or periods of interrupted functioning) in the absence of any significant heating effect (< 1.50C) of the cathode (see Fig.4).  Repetition of these experiments has led us to conclude that, as shown in Fig.5, the variable responsible for this repeatedly observed reduction in the PAGD frequency, when the PAGD operation sequence is systematically interrupted, is the state of charge/discharge of the battery pack (the charge pack) at the output of the triode circuit in question: the PPM rates in Fig.5 decrease rapidly with the steepest rate of charging of the charge pack and the fastest recovery rate of its open circuit voltage; above a given state of charge, when the open voltage of the charge pack climbs more slowly (> 340 V), in a log fashion, the PPM rate stabilises at its plateau values.


Confirmation of the importance of the charge pack in the PAGD function of the present circuitry here considered, comes from the fact that the size (the number of cells) and the intrinsic capacitance of the charge pack affect the PAGD frequency dramatically (see Table 4): increasing the charge pack size of 29 cells to 31, by 7% leads to a 10-fold reduction in frequency; further increases in the number of charge pack cells extinguishes the phenomenon.   On the upper end of the scale, this effect appears to be tied in to restrictions that it places on the ability of the larger charge packs to accept the discharge power output once the charge pack voltage exceeds the PAGD amplitude potential.  All of these measurements were conducted with the same 128 cm2 plate PAGD generator, at a pressure of 0.8 Torr and in the triode configuration (see Fig.9).


Other factors can also affect the frequency: the motion of external permanent magnetic fields oriented longitudinally with the inter-electrode gap, external pulsed or alternating magnetic fields, external electrostatic or electromagnetic fields, specific connections of the earth ground, and the presence of a parallel capacitative, capacitative-inductive or self-inductive arm in the circuit, such as we have described for our electromechanical PAGD transduction method as described in the “531” application.


Analysis of the modulation of PAGD amplitude is simpler than that of its frequency, because fewer factors affect this parameter:

(1) magnitude of the applied potential,

(2) inter-electrode gap distance and

(3) the negative pressure, as shown in the “863” application, for "low" applied currents.


As the magnitude of the applied potential itself is limited by the gap and the pressure, to the desired conditions of breakdown, the important control parameter for the PAGD amplitude is the pressure factor. This is shown in Fig.6 and Fig.7, respectively for "low" (5 mA) and "high" (about 500 mA) applied currents and for the same plate diode configuration of a H34 Al 128 cm2 plate PAGD generator (5 cm gap), in the simple circuit described in the “863” application; it is apparent that both positive and negative components of the amplitude of these pulses in the oscillograph, are a function of the pressure, but the maximum cut-off limit of our equipment, for the negative component (at 240 volts for the "low" current experiment and at 120 volts for the "high" current), precluded us from measuring the peak negative voltage of these pulses.


However, rms measurements of the pulse amplitude at the plates and DC measurements at the circuit output to the charge pack indicate that the negative component increases with decreasing pressure to a maximum, for a given arrangement of potential and gap distance; no pressure-dependent bell shape variation of the pulse amplitude, as that seen for the positive component at "high" applied currents (Fig.7) is observed with the negative amplitude component.   For the typical range of 0.8 to 0.5 Torr, the rms value for pulse amplitude varies from 320 to 480 volts, for a 5.5 cm gap distance and applied DC voltages of 540 to 580 volts.  PAGD amplitude is a critical factor for the design of the proper size of the charge pack to be utilised in the optimal circuit.


The development of the circuits to be described stemmed from fundamental alterations to the principles implicit in our previous methods of electromechanical transduction of AGD plasma pulses as described in the “531” application.   Whereas this electromechanical coupling (capacitative and self-inductive), utilised directly, energises the AGD pulses inverted from the DC input by the vacuum generator, the purpose of the development that led to the presently described experiments was to capture efficiently, in the simplest of ways, most of the pulse energy in a closed circuit, so that power measurements for the energy transduction efficiency of the observed endogenous pulsation could be carried out. Ideally, comparative DC power measurements would be performed at both the input and output of the system, taking into account the losses generated across the components; this would overcome the measurement problems posed by the myriad of transformations implicit in the variable frequency, amplitude, crest factor and duty-cycle values of the PAGD regime, and necessitated some form of rectification of the inverted tube output.



From the start our objective was to do so as simply as possible.  Early circuits utilising half-wave rectification methods coupled in series to a capacitative arm (for DC isolation of the two battery packs), with the charge pack also placed in series, showed marginal recoveries of the energy spent at the PAGD generator input. Attempts at inserting a polar full-wave rectification bridge led, as shown in Fig.8, to the splitting of the capacitor into capacitors C3 and C5, at the rectification bridge input, and capacitor C4 in series with both capacitors, all three being in a series string in parallel with the PAGD generator.  Under these conditions a DC motor/generator could be run continuously in the same direction at the transversal output (U1 and U2) of the bridge; but if this inductive load was replaced with a battery pack CP (charge recovery pack), either the parallel capacitor C4 had to remain in the circuit, for the diode configuration or, less desirably, a further capacitor C6 could replace C4 and connect one electrode, preferably the cathode C, to the axial member of the discharge tube T, thus resulting in a first triode configuration as actually shown in Fig.8.  Energy recovery efficiencies of the order of 15% to 60% were obtained utilising C6 in this manner, but measurements of the potential and currents present at the output from the rectifier bridge were substantially lower than those obtained using optimal values of C4.  Effectively, under these conditions, much of the power output from the tube was never captured by the output circuit formed by the second, right hand arm of the system and, being prevented from returning as counter-currents to the drive pack DP by diodes D1 and D4, was dissipated and absorbed by the inter-electrode plasma, electrode heating and parasitic oscillations.



Solutions to this problem were explored using the circuit shown in Fig.9, which still maintains the necessary communication link for the quasi-sinusoidal oscillation of the capacitatively stored charges at the input and outputs of the rectification bridge, but integrated the functions of capacitor C4 into the single rectification circuit, in the form of an asymmetric capacitative bridge C7a and C7b placed transversally to the capacitative bridge formed by C3 and C5 and in parallel with the charge pack CP at the output from the rectification bridge D5, D6, D2, D3. 


This second capacitative bridge is so disposed as to have its centre point connected to the anode A through capacitor C5.  If the axial member of the Tube T were to connect to the junction of D2 and D3 instead of at the junction D5-D6, the function of bridge C7a and C7b would be connected to the cathode C through capacitor C3. The capacitative bridge is insulated from the charge pack whose voltage it stabilises, by rectifiers D7 and D8, which also prevent leakage of charge across C7a and C7b.


The anode and cathode oscillations generated by the electrostatic charge transduction through C3 and C5 into the poles of the charge pack are trapped by the transversal transduction of the C7 bridge, at the outputs from the rectification bridge, of which the oscillation has to become split between the bridge inputs into half-waves, for electrostatic transduction and full wave rectification to occur.  In fact, under these conditions, removal of the C7 bridge will suppress the PAGD phenomenon, unless other circuit variables are also altered.  The transversal bridge is thus an essential piece of this novel circuit.  Variations in the circuit as shown in Fig.10 were then studied, the first two being selectable utilising switch S2 (Fig.9).


The presence of the capacitative bridge effectively reduces the dynamic impedance of the charge pack CP so that the output circuit approximates to a characteristic in which it presents a very high impedance to the tube T at potentials below a certain level, and a very low impedance at potentials above that level.


With this modified circuit, more effective recovery of the energy produced by collapse of the PAGD pulses is possible, with more effective isolation from the input circuit utilised to trigger the pulses. Under these conditions, the energy captured by this circuit at the output, is not directly related to that utilised in triggering the pulses from the input.  The attainment of this condition critically depends on the large capacitance of the transversal bridge being able to transfer the output energy from the tube T into the charge pack CP.  Under these conditions, we have found, as will be shown below, that the large peak pulse currents released by collapse of the PAGD pulses released more energy than is used to trigger them, and these findings appeared to tally with other observations (abnormal volt-ampere characteristics and anomalous pulse currents, etc.) associated with the anomalous cathode reaction forces that accompany the auto-electronic emission-triggered PAGD regime.  Experiments so far indicate that the power output can be increased proportionately to the series value of C3, C5 and the two identical C7 capacitors.



The circuit of Fig.10 can be integrated with a circuit such as that disclosed in the “863” application as shown in Fig.11, in which a part of the energy recovered can be shunted by the switch S4 into an induction motor M1 having rotor R, to a degree determined by the adjustment of potentiometer R4 and the value selected for C4.


The circuit of Fig.11 can be further developed as exemplified in Fig.12 to include configurations which provide switching permitting interchange of the functions of charge packs and the drive packs, it being borne in mind that the nominal potential of the drive pack must be substantially higher than that of the charge pack, the former needing to exceed the breakdown potential of the tube at the beginning of a PAGD cycle, and the latter to be less than the extinction potential.



Fig.12 essentially represents a duplication of the circuit of Fig.11, the two circuits however sharing two identical battery packs BP1 and BP2, and being provided with a six pole two way switch, the contact sets of which are identified as S1, S2, S3, S4, S5 and S6.   When the contacts are in position A as shown, battery pack BP1 acts as a drive pack for both circuits, with the upper half (as shown) of the battery pack BP2 forming the charge pack for the upper circuit, and the lower half forming the charge pack for the lower circuit. When the pack BP1 is at least partially discharged, the switch is thrown so that contacts move to position B, which reverses the function of the battery packs thus allowing extended operation of the motors in each circuit each time the switch is thrown.


Based on the manufacturer's data, and using current values within the range of our experimentation as discussed in the next sections, an optimal discharge cycle for a fully charged 6.0 AHr battery pack at 0.300 A draw is 20 hours, as claimed by the manufacturer, and this corresponds to a cycling between 100% (12.83 V/cell open circuit and load start voltage) and < 1% (10.3 V/cell load voltage) of the battery's absolute charge capacity.  Even though the discharge mechanism is a time cumulative process with a log function, the discharge can, within 4 to 5 hour time segments (or periods with 20%-25% of the full range), be regarded as practically linear with time.  This trait, or linearisation of the discharge slope, becomes more marked with advancing age and decreasing absolute storage capacity of the cells.


The proportionality between open circuit voltage and the percentage of residual relative capacity for these cells when new (uncycled and not yet aged) is uniform over 98% of the permissible charge capacity withdrawal.  In practice this translates into a slope that becomes steeper with time, while the absolute storage capacity diminishes.  In turn, this decreasing absolute capacity of the cells results in shorter load discharge times and their further linearisation.


A circuit in general accordance with Fig.9, employed in the studies reported in this and the following sections, utilises a drive pack of 46 12 V Lead acid gel-cells each with a 6.0 Ah rating, and a charge pack with 28 or 29 12 V identical cells.  The charge pack was cycled anywhere from 11.2 V to 12.8 V/cell (open circuit voltages), within the proportional region of the relative capacity slope, to yield a capacity increment in the order of 50% (e.g. from 20% to 70%), anywhere within the range of 2% to 100% of its total charge capacity, assumed for now as invariant.  The charging process, hereinafter referred to as a PAGD run, took about 20-30 minutes under optimal conditions.  The drive pack typically consumed, in the same period of time, 4% to 11% of its initial total capacity , its open circuit voltage typically falling 0.1 V to 0.2 V per cell after a PAGD run, within the open circuit range of 12.8 V/cell (100% relative capacity) and 11.2 V/cell (about 2%).   At the 100% capacity benchmark, the drive pack would theoretically have 20 h x 46 cells x 12.83 V/cell x 0.3 A = 3.5 kWh, and the charge pack, for example, 20 h x 29 x 12.83 V/cell x 0.3 A = 2.2 kWh.   Since the capacity per cell is linear with the open circuit voltage within the proportional range, as claimed by the manufacturer, we projected the open circuit voltage intercepts on the manufacturer's proportional curve in order to determine the residual percentage of the total relative capacity and the standard hours of operation left, from any experimental open circuit voltage measurements.


Three pulse generators (one 64 cm2 and two 128 cm2 plate areas) were employed in these studies; they were operated in PAGD runs at 1-120 pulse/second rates, within a negative pressure range of 0.2 to 0.8 Torr and with applied direct currents of 0.2 to 0.6 A.


Both drive and charge packs utilised cells which were bought new at the same time and had initial charge values of 12.4 to 12.55 V/cell (open circuit).  These batteries are capable of energy densities of 33-35 WHr/Kg.   However, the experiments shown in Table 5 are selected from a series that spanned nearly 12 months, beginning 6 months after purchase; hence, loss of absolute storage capacity by the batteries had occurred in the intervening time, as a function of both age and charge/discharge cycle life.


Measurements of the open voltage of either drive (D) or charge (C) (see column 2, Table 5) packs for 8 separate experiments, all utilising the triode configuration, were performed before (b) and after (a) a PAGD run (see columns 3 and 4), at either 15 or 30 minutes (see column 26) of the open circuit voltage relaxation after a PAGD run was terminated.   Corresponding open circuit voltages per cell are shown in column 5, and the percentages of the predicted total relative charge capacity resulting from the intercepts on the manufacturer's proportional curve are shown in column 6, Table 5.   Equivalent maxima for the theoretical hours of operation left are shown in column 7, the percentage change in relative capacity arising as a consequence of either charge pack charge capture (capacity gained) or of drive pack output (capacity lost) is shown in column 8.   Translating the intercepts into power units yields the values shown in column 9, Table 5, for total kWh left in each pack before and after PAGD production, those shown in column 10 for the actual power gained and lost during the periods of operation (presented in column 12) and those shown in column 13 for the power predicted to be gained or lost per hour of PAGD production.


On the basis of the experimental open voltage values and their intercepts, the predicted net kWh values per hour of PAGD energy production (after deduction of measured losses) and the corresponding experimental break-even efficiencies (where breakeven = 100%) are presented, respectively, in columns 14 and 15.  The PAGD frequency per second is shown in column 11; the number of 12 V cells, in column 16; the tube ID, in column 17; the cathode (and anode) area (s), in column 18; the plate material, in column 19; the input ballast utilised (R1, FIG. 9), in column 20; the size of each capacitor (C3 or C5) of the tube output bridge, in column 21; the size of each capacitor (C7a or C7b) of the transversal capacitative bridge, in column 22; the status of S4 and thus, of the parallel and auxiliary electromechanical arm (see Fig.11), in column 23; the negative air pressure in column 24; the gap distance between the plates, in column 25; and columns 27,28 and 29, show the status of the elements of the switched on parallel electromechanical arm of the circuit--the parallel C4 capacitor, the motor input resistor R4 and the motor revolutions per minute (measured stroboscopically), respectively.


From these figures of Table 5, and utilising the data for the two first examples shown, we calculated the predicted performance of the system based on the open voltage measurements.  In the first example, where the system was run continuously without interruption, the charge pack increased the percentage of its total capacity by 43% (a two-fold increase in capacity) and, during the same period, the driver pack decreased the percentage of its total capacity by 7% (an approximately 10% decrease in capacity relative to the percentage of residual total capacity at the start, i.e. 77%) (cp. columns 6 and 8, Table 5).   Subtracting the predicted initial total energy (0.835 kWh) available to the charge pack before the experimental run (first line of column 9, Table 5) from the predicted total energy (1.823 kWh, second line of column 9) available to the charge pack after the PAGD charge run, gives us the total energy gained by the charge pack: 0.988 kWh (column 10) in 21.5 minutes (column 12) of continuous PAGD performance.


Conversely, subtracting the predicted final total energy (2.4 kWh) available to the driver after the experimental run (fourth line of column 9, Table 5) from the predicted total energy (2.66 kWh, third line) available to the driver before the PAGD charge run, gives us the total energy lost by the drive pack: 0.26 kWh in 21.5 minutes.   If we divide the total available energy gained by the charge pack, by the total energy lost by the drive pack, we obtain a surplus factor of 3.9., or 388% of the break-even point (column 15).  The same values result from dividing the charge pack % of total capacity gain by the drive pack % of total capacity lost, and then down-scaling this value by multiplying it by the typical scale factor for the two packs, 29 / 46 = 0.63 times.


In an analogous fashion, we analysed the results for the second example shown in Table 5.  Here, the charger increased the percentage of its total capacity by 45.5% (a 22.75 fold increase in estimated total relative capacity) and, during the same period, the driver decreased the percentage of its predicted total capacity by 7% (about a 17.5% decrease in capacity relative to the percentage of residual total capacity at the start, i.e. 40%).   By dividing the predicted total available energy gained by the charge pack (0.962 kWh/18 minutes) by the expected total energy lost by the driver pack (0.246 kWh/18 minutes) we obtain a surplus factor of 3.9 times, or 391% of the break-even point.   This corresponds to an interrupted, total sequential run of 18 minutes, each minute-long run being separated by a cooling and voltage relaxation period of 15 minutes before the next run is carried out, at an average PAGD frequency of 61 PPS.


Analysis of the remaining results illustrates how a number of PAGD controlling parameters interact to determine conditions for effective maintenance of a PAGD regime.   The lower gain and higher loss per unit time registered for the third run of Table 5, which results in the lower break-even efficiency of 230% and a smaller net power production rate than before (power estimates of 1.396 kWh/h of PAGD operation vs 2.387 kWh/h, for the second run, Table 5) illustrate, for example, the combined effect of lowering the pressure (0.8 to 0.7 Torr) and running the PAGD continuously (the heating effect), both of which depress the PAGD frequency.   The fourth run of Table 5 identifies the continuous performance of a "broken in" softer grade of aluminium (column 19), having a lower work-function (as determined from the higher PAGD frequency spectrum) than the harder H34 plates of the previous examples, and shows that, despite the series value of the total capacitance being higher (5,333 mF vs 4,030 mF for runs one through three), and despite the higher vacuum (0.2 Torr), the lower work-function results in a higher frequency; however, even though this run registers a predicted higher break-even efficiency (310%) than the previous experiments, these conditions result in a 4 / 5-fold lower estimate of net power produced, when compared to the previous three PAGD runs.


PAGD runs 5 and 6, Table 5, illustrate the effect of switching on the auxiliary electromechanical arm of the circuit shown in Fig.11.   Increasing the amount of charge capacitatively shunted into the electromechanical arm by higher C4 values (column 27), and increasing the current that feeds the squirrel cage induction motor utilised by lowering R4 (column 28), results in a power capture by the charge pack that registers an energy loss (predicted to be 96% efficient, falling short 4% of break-even recovery), as most of the tube output power is spent in the electromechanical arm and its motor effect.   Furthermore, under the conditions of maximum electromechanical action, the drain imposed on the drive pack becomes considerable (see loss in columns 10 and 13), even if the C3 and C5 values are reduced, column 21, Table 5).  These runs also illustrate how the motor appears to function as an electrical induction generator having rpm values much higher than the synchronous values prescribed by the frequency of the PAGD (column 29, Table 5).


The extremely large break-even efficiency of PAGD run 5, Table 5, indicates that with selected values of C4 and R4, it is possible to operate the motor in the auxiliary arm and still accumulate excess energy from the PAGD production in the charge pack.


Runs 7 and 8 illustrate results obtained for 64 cm2 plates, and a shorter inter-electrode gap distance, for two pressures (0.8 and 0.5 Torr), the device being open to a rotary pump manifold in the first instance and sealed from the pump, in the second case.   Despite the lower vacuum, the higher pulse frequency (32 vs 5 PPS) and break-even efficiency (906% vs 289%) registered by run 8 when compared to run 7, are a consequence of the method of run 8, which was interrupted systematically by 5 passive cooling periods, as in the case of run 2, whereas run 7 was continuous.  This again resulted in higher average PAGD frequencies (at lower pressures), a predicted two-fold greater gain and a predicted two-fold smaller loss (columns 13 and 14) for run 8.


Fig.13 shows curves representing the slopes of the open circuit relaxation voltages, which are linear with the log of time elapsed from cessation of discharge, for both drive and charge packs, in the same run 8 set out in Table 5.   The experiment in its entirety consisted of preliminary resistor-loaded measurement discharges and their corresponding open circuit voltages from the moment of cessation of the resistive discharge (illustrated, respectively, by the open squares of DPT1 for drive pack relaxation time 1, and by the open circles of CPT1 for charge pack relaxation time 1), followed by their relaxation rates in the wake of the PAGD production (the hatched squares of DPT2 for drive pack relaxation time 2, and the hatched circles of CPT2 for charge pack relaxation time 2), and finally, by the relaxation rates from the final resistor-loaded measurement discharges (the black squares of DPT3 for drive pack relaxation time 3, and the black circles of CPT3 for charge pack relaxation time 3).   Discharge resistances were 833 ohms for the charge pack, and 2083 ohms for the drive pack in all cases, corresponding to resistors R3 and R2, respectively, of Fig.9.  This methodology will be examined in greater detail below.   It is apparent that, after every load period, be this resistive (CPT1, DPT1, CPT3 and DPT3) or due to PAGD operation (DPT2), the relaxation slope is positive; as shown from slopes CPT1 and DPT1, the log time proportionality of the open circuit voltage relaxation, under these conditions, tends to plateau after about 30 minutes.  The exception to this general behaviour lies in the voltage relaxation slope CPT2, which is negative and reflects the charge accumulation occurring in the charge pack and obtained by capture of energy produced during PAGD operation, triggered by the energy drawn from the drive pack during load time 2.


As a first approximation of electrical power generated and consumed by the energy conversion system of the invention, the previous open circuit voltage method is of significance in showing the basic trends involved in interaction of the operating parameters.  However, in all likelihood, it overestimates the actual values of electrical power consumed and generated, for a variety of reasons.  First, it assumes that the relative capacity scale of the batteries in the drive and charge packs is an absolute charge capacity scale with an invariant maximal charge retention, which it is not; in fact, the absolute charge capacity is itself a variable subject to several factors, such as the cycle life, overcharging or undercharged conditions, cell age, residual memory and the rate of charge and discharge.  Hence, the inference of a uniform time scale on the basis of the open circuit voltage/capacity intercepts may not be warranted.  Finally, it does not integrate the open voltage decrease over time, and utilises the specification load current as the average current over time.


In order to obviate these problems, we resorted to a variety of other measurement methods.  First, we proceeded to compare the closed circuit, preliminary, resistive-load discharge measurements for either charge or drive pack, under conditions of negligible loss of power, as these measurements were statistical means (n = 9) taken, at equal intervals, during the first 90 seconds of the load discharge, and obtained both just before the PAGD production runs (but separated from each PAGD run by an open circuit voltage relaxation of 30 minutes) and just after the runs (but equally separated by a relaxation of 30 minutes).   As an example of the data generated by such an approach, Fig.14 illustrates the shift of the slopes indicating marginal power loss for the drive pack (from the closed squares to the open squares) and those indicating gain of power for the charge pack (from the open circles to the closed circles), in actual total load power values.


Integration of these power measurements over the projected load discharge time, taken from the family of curves generated on the basis of the manufacturer's load voltage over discharge time specifications, led to a direct comparison of the new values, as shown in Table 6, with the values presented in Table 5, for the first three instances introduced.  All values of Table 6 were obtained by resistive measurements of power that entailed a negligible power loss.  Table 6 confirms the fundamental equivalence of runs 1 through 3, as already seen from their corresponding analysis using the open voltage method (see runs 1 to 3, Table 5). This new power estimation method also confirms the lower loss encountered in run 2 utilising interrupted PAGD operation.  While the break-even efficiencies sensibly doubled using this method, the estimates of actual electrical power consumption recovery decreased by a 2 to 3-fold factor. Thus this direct load voltage/amperage measurement method of estimating actual power losses or gains, is a check upon the open voltage method previously utilised.


Direct, instantaneous measurements of the voltage and current characteristics of the PAGD production and capture phenomena being discussed, were also performed during PAGD runs for diverse sets of conditions, including all those described in the two previous sections.   In Table 7 we show these results for two PAGD generators having an identical electrode area (128 cm2) and connected to electrical energy capture circuits of three separate configurations as set forth in Fig.10A, Fig.10B and Fig.10C and column 2, Table 7.  In the configuration of Fig.10C, or double diode configuration, both electrode plates act as cathodes and the axial member as the anode collector (experiments 1-4, for the H220 device and 13-14, Table 7, for the H34 device).   In the configuration of Fig.10B, or triode configuration, one plate acts as the cathode, the axial member as an auxiliary cathode and the other plate as a collector (experiments 5-9, Table 7).  In the configuration of Fig.10A or single (plate to plate) diode configuration, the axial member is disconnected, and the polarity of the plates remain as in the triode configuration (experiments 10-12).  All measurements were taken after 1 minute of PAGD operation of the devices, which were, at the start of each run, at room temperature.   All cathodes had been previously broken in with > 2 x 106 AGD pulses.  The open circuit voltage of the charge pack was, for all cases, at 359 to 365 volts, before each test.  The direct measurements of the PAGD input and output DC voltages and currents were obtained as statistical means of 10 second long measurements, and at no time did the standard error of the plate voltage mean exceed 35 volts.


The air pressure within the tube during these tests is shown in column 3, Table 7, the drive pack DC voltage (X), in column 5, the DC voltage across the plates (Y), in column 6, the drive pack output current (PAGD input current), in column 7, and the drive pack total watts output is shown in column 8. Columns 9 and 10 show the PAGD voltage (PAGD V = (X-Y) / Iav) and the value of the PAGD extinction potential in V/cm.  The recovery co-ordinates (i.e. the PAGD output energy) found at the U1-U2 output (Fig.9), are shown in columns 11 to 13, as the charge pack's E1-E2 input DC voltage, amperage and power watts, respectively.  The calculated resistance of the entire circuit is given in column 14, the registered PAGD frequencies in column 16, and running conditions in columns 17 to 18.   The break-even efficiency obtained by direct comparison of the electrical power figures for the drive and charge packs, respectively, is given in column 15.  This assumes, for purposes of a generalisation of power production rates over time, that the quasi-instantaneous, direct measurements here obtained can be translated to outputs obtained per unit time, and thus into direct Watt-hour measurements.


Data from runs 1 through 4 demonstrate that, at these PAGD frequencies, there is no difference between using fast switching (32 nanoseconds) MUR 860 diodes, or regular 40HFR-120 silicon diodes, in the rectification bridge of the electrical energy capture circuit, and that the PAGD frequency varies as a function of decreasing air pressure.


Runs 5 to 14 show that, in general, for the same tube, the single and double diode configurations are the most efficient, for the same pressure, the diode configuration typically yields some 1.5 to 2 times larger break-even efficiencies (cp runs 10-11 and 13-14, with runs 5-9, Table 7).  The largest accumulations of power are also registered in the diode mode(s).  This trend appears to be a function of the much lower cathodic work-function of the aluminium plates, than of the tungsten of the axial member utilised as an auxiliary cathode in the triode configuration.   A feature of the data from these 14 different runs is the consistent excess power outputs (column 15, Table 7) and their narrower range (218 to 563%), when compared to those observed with the previous two methods of experimental analysis.


Run 12, Table 7, shows that the switching on of the electromechanical arm can be performed without entailing a power loss in the PAGD capture circuit, as previously found for run 5, Table 5, utilising the open circuit voltage method.   In fact, with C4 = 8 microfarads and R4 = 500 ohms, the AC induction motor behaves as an electrical flywheel (e.g. 2800-3000 rpm for 10 PPS inputs), while the electrical energy capture circuit still registers a sizeable excess electrical power production (compare runs 11 and 12, Table 7).  Runs 13 and 14 illustrate how the charge pack's state of charge and its inherent capacitance affects both the PAGD frequency and the power producing efficiency of the entire system: as the charge pack is reduced from 29 to 19 cells, the PAGD generator adjusts by reducing its frequency logarithmically and, while the charge pack input current is greater than before, the drive pack loss becomes still larger and the break-even efficiency much lower (by >1/2, from 563% to 228%).  This is because the circuit must translate the naturally larger PAGD amplitude into a larger surplus of output current, and in this process becomes less efficient.


If the first measurement method employed (the open circuit method) had to make too many theoretical assumptions about the system's performance under load conditions and hence about its effective charge capacity, the second approach still had to suppose an invariant discharge time and thus an invariant absolute charge capacity on the part of the battery systems (charge packs) employed for capture which it approximated by an operation of integral calculus.  With the third method described above, theoretical assumptions were avoided except that, in these measurements, the actual performance of a given battery in terms of time, time of delivery and time of capture, was also ignored; no account is taken of the time-dependent modulation of the PAGD frequency, as effected by certain of the parameters analysed, namely the charge pack state of charge, the method of sequencing the PAGD runs (continuous vs interrupted) and its concomitant heating effects, and the state of charge (load voltage and current capacity) of the drive pack.   A simple, non-negligible, resistive measurement of power lost by the drive pack, and an identically non-negligible measurement of the power gained by the charge pack, for the same experiment and the same singular time of PAGD production, were performed repeatedly to corroborate the previous three approaches. For this purpose, all experiments were designed as a continuous series of sequential phases:


1) Before a PAGD run, a resistive discharge was measured across either pack over periods of 1 to 3 hours (utilising the DP and CP resistances previously reported in the open voltage section) and followed by a 15 to 30 minute open circuit voltage relaxation;


2) Then, the PAGD runs were performed, either continuously or as interrupted, composite sequences, and the corresponding open circuit relaxation voltage(s) were measured, after the cessation of the integral PAGD run;


3) Finally, resistive discharge measurements, obtained under identical conditions to those recorded before the PAGD run, were carried out for either pack, followed by concomitant battery voltage relaxation rate measurements.


Under these experimental conditions, exact power measurements could be taken from an analysis of the actual battery discharge curves before and after the PAGD run.   Based on a comparison of the curve trends of the pre-run resistive discharge of the drive pack with those of the post-run resistive discharge, the effective power drawn (DeltaEc) from the withdrawable power capacity of the drive pack incurred during a PAGD run, was ascertained.  This represents the power consumption during the run, and the experimental value thus recorded constitutes the actual power figure that must be matched for break-even to occur.  Hence, the break-even value equals, by definition, the electrical energy input to the system.  Similarly, a comparison of the charge pack pre-run and post-run resistive discharge curve trends identified the effective power (DeltaErho) added to the withdrawable capacity of the charge pack.  This quantity represents the electrical energy recovered during the run.   The relation for the two quantities is expressed by the break-even efficiency equation:


% = DeltaErho  / DeltaEc x 100


If the break-even efficiency is less than 100%, then the apparatus registers a net loss in electrical energy in the CP with respect to the DP.   Conversely, if the efficiency exceeds 100%, then there is a net gain in electrical energy in the CP, as compared to that lost in the DP.   For purposes of this analysis, a limit to the minimum withdrawable capacity was placed, from experiment and in agreement with the load current curves of the manufacturer, at 115 W for the driver pack (average current of 0.250 A, minimum current of 0.230 A), and at 90 W for the charge pack (average current of 0.375 A, minimum current of 0.334 A), as a function of both their total cell size (respectively, 46:29) and the difference in the resistive loads employed for the discharge measurements. All cathodes had been broken in, as described before.


The results obtained with this fourth method, for six selected experiments with three diverse types of devices (using different electrode plate areas, gap lengths, and electrode work-functions), configured both in the triode or the (single) diode (e.g. Fig.10B) arrangements, at the indicated pressures, are presented in Table 8. In all cases, a net excess of combined battery pack charge, expressed as electrical watt hours, is registered (columns 8 and 10, Table 8) and the break-even efficiencies are all >100% (column 10).  Experimental groups 1 and 2 again demonstrate that, for the same cathode, the interrupted PAGD sequence method of group 2 (1 minute of PAGD function, followed by a 15 minute relaxation, and so on) yields a higher break-even efficiency because of the lower losses registered with this minimal plate heating method (column 10, Table 8).   Group 3 of Table 8, shows that the PAGD power production efficiency is also higher for a lower work-function cathode material (H220 vs H34), being subjected to PAGD auto-electronic conditions at a 4-fold lower pressure than the control groups 1 and 2; however, the lower pressure depresses the frequency and, together with the interrupted PAGD sequencing method, it also lowers the loss, causing an actually much larger break-even value than registered for the previous two groups.   Groups 4 and 5 exemplify the dual effect of lowering both the plate area and the gap distance: the former affects the PAGD event frequency, whereas the latter affects the PAGD amplitude, and thus the capture efficiency of the charge pack.  Despite a cathodic work-function practically and operationally identical to that of groups 1 and 2, these smaller plate area and shorter gap devices utilised in groups 4 and 5, yield 3- to 6-fold lower net power outputs, as well as lower break-even efficiencies, than the former groups, at the same pressure. Finally, group 6 exemplifies the results obtained for the plate diode configuration, where the frequency is lower (no triggering role for the axial member), and a higher loss leads to the lower break-even efficiency, comparable to that of the lower area and shorter gap groups 4 and 5.


In order to verify the discharge curve lengths employed in these analyses and experimentally establish the actual charge capacity of the battery packs, calibration resistive discharges, between the maximum charge state and the minimum limits chosen, were performed for each pack with their respective discharge resistances R2 and R3 (see Fig.9).   These discharge calibration curves were plotted for half maximal charge values shown in Fig.15A and Fig.15B, and from the curve produced, we have determined the total half-charge capacities of each battery pack to be 1.033 kWh (100%=2.066 kWh) for the drive pack and 660 WHr (100%=1.320 kWh) for the charge pack.   Based upon the corresponding maximal (100%) capacity values, we determined the actual percentages of the relative charge capacities shown in column 5, Table 8, which correspond to the experimental values obtained.   We also noted that the curves plotted showed two quite distinct time linear slopes, the slope of the delivery of power per time unit steepening very markedly at the approach to the limits of the permissible withdrawable capacity, occurring at 115 W into R2, and 90 W into R3.


The pre-PAGD run and post-PAGD run, drive and charge pack discharge curves corresponding to groups 3 and 6, respectively for triode and plate diode configurations, in Table 8, are shown in Fig.16 (drive pack) and 17 (charge pack), for group 3, and in Fig.18 (drive pack) and Fig.19 (charge pack), for group 6.  In all cases, the open symbols represent the pre-PAGD run discharge curves, whereas the closed symbols represent the post-PAGD run discharge curves.


As a further check on these values, a videographic, millisecond analysis of the singular power simultaneities occurring at both ends of the system (drive and charge packs) was performed for various 10 second samples of diverse PAGD runs.   A typical example is shown in Fig.20, which is a sample of the PAGD run designated as 6 in Table 8.   While the drive pack DC wattage spent as input to PAGD production varied from 36.6 to 57.82 watts, by a factor of 1.6 times, the DC wattage entering the charge pack as captured PAGD output varied more pronouncedly by a factor of 2.7 times, from 146.4 to 399.6 watts (all meters were in the same selected ranges of voltage and current) with the semi-periodic, intermittent character of each singular emission, though within specific, ascertainable ranges for both amplitude and current outputs.


Assimilation of the singular behaviour of the PAGD in this sample, by a statistical treatment of its variation (with n = 64), indicates that the operational break-even efficiency observed during this sampled period lies at 485.2% +/- 18% with projected 48.3Wh drive pack loss and 221.7Wh charge pack gain. This matches rather closely the observed 483% break-even efficiency, and the 37.7Wh loss as well as the 182.2 kWh gain for the overall PAGD run reported in group 6 of Table 8, and indicates how close are the values obtained by the operational and extensive non-negligible resistive discharge power measurement methods employed.


Finally, an example of the correlation between the drive pack PAGD load voltage and the charge pack PAGD charging voltage, as a function of the duration of the intervening PAGD run between resistive discharge measurements, is shown in Fig.21, for the PAGD run corresponding to group 4 of Table 8.


Using the same pulse generator with H200 Al 128 cm2 plates, in a double diode configuration, and the same circuit values (but with CP = 23 cells), three experiments were conducted at different PAGD frequencies, as a function of varying air pressure.   Analysis of driver pack losses and charge pack gains by the extensive load discharge measurement method, as described before, led to the determination of the gross and net gains (respectively, without and with losses included) per pulse, in milliwatt-hour, for each frequency, as well as of the gross and net power gains per second of PAGD operation. The results are shown in Table 9.  Even though the gross and net gains of power per pulse were observed to increase with decreasing frequency, the gross power gain per unit time increased with increasing frequency.  However, this last trend does not necessarily translate into a higher net gain per unit time, because the losses in the driver pack (not shown) also increase significantly with PAGD frequency.   These losses are in all probability related to more energy retention by the plasma at higher frequencies when plasma extinction becomes incomplete.   We expect net gains to reach optimal thresholds for any given type of circuit configuration set of values and pulse generator dimensions.


Certain additional observations made during experiments with the double diode configuration of Fig.10A may assist in understanding of the invention.


1) Replacing residual air with argon gas leads to higher PAGD frequencies, as noted by us when utilising a 128 cm2 H200 AC plate pulse generator in the double diode configuration (V = 575).   At 1 Torr, the pulsation rate went from 20 PPS in air to 1300-1400 PPS in argon.   With 29 12V cells in the charge pack, input currents ceased to flow into it.   Under these conditions, the tube potential across the plates decreased and the drop across the input resistor increased.   The value of E (= V/d) became smaller (gap size = 3 cm from plate to axial anode collector), as the extinction voltage decreased.


2) With frequencies of 400 PPS, the currents flowing into the charge pack fell to zero.  Replacing a fast-recovery type HFR 120 (1200v, 40A) diode bridge by a type MUR 860 (600v, 8A) diode bridge had no effect. When the amplitude of plate potential oscillations falls below the potential of the charge pack, there is also a tendency to produce arc discharges.  For output currents from the vacuum pulse generator to enter the charge pack, the number of cells must be reduced so that the potential of the charge pack is low enough to admit the transduced currents.   A reduction from 29 to 23 cells allowed currents of 250 mA to enter the CP, and further reduction to 19 cells doubled these currents (per polarity arm).


3) Our observations show that it suffices under these conditions (CP of 19 cells) to increase the vacuum, so that the frequency decreases, and the plate potential and the charge pack input currents increase.  At 0.1 Torr, the currents reached 1A DC per plate, and at 0.05 Torr, 2A DC


The interconnection between these factors indicates that the extinction voltage is a function of the PAGD frequency: the higher the PAGD frequency, the lower the extinction voltage, until empirical (in distinction from predicted) VAD field values are reached.   As a consequence, the start voltage of the charge pack must be adjusted, by varying the number of cells composing it, so that it lies below the lowest extinction voltage of the PAGD, for any given geometry and gap distance.


Secondly, as the ion plasma is made more rarefied, the frequency of the emissions decreases, but the peak values of the output voltage and current per pulse increase.   The slower the PAGD and the more rarefied the atmosphere, the higher is the output energy produced by the system relative to the input energy.


Autographic analysis of PAGD-induced cathode craters in H34 plates was performed, and their average inner diameter and maximum depth were determined.   Similar studies were performed for PAGD-induced craters in Alzak (trade mark) plates.  The secondary craters characteristically found in Alzak plates, along fracture lines irradiating from the main crater, are absent in H34 plates; instead, in H34 plates, one observes a roughened surface surrounding the emission crater, quite distinct from the original rough aspect of the pulled finish of these hardened aluminium plates.   Also, unlike the Alzak main craters, the H34 craters often have a convex centre occupied by a cooled molten metal droplet, whereas the Alzak craters had a concave, hollowed out aspect.  Eventually, as the pitting resulting from PAGD cathodic emissions covers the entire cathode, the metallic surface gains a very different rough aspect from its original appearance.  In this process, craters from earlier metal layers become progressively covered and eroded by subsequent emissions from the same cathode.   Altogether different is the surface deposition process occurring at the anode; here, the surface appears to become more uniform, through the mirroring and possibly abrasive actions of cathode jets.  Macroscopically, with increased periods of PAGD operation, the anode surface appears cleaner and more polished.


With the data obtained by the metallographic method of crater measurement, we estimated the volume of metal ejected from the cathode, by assuming that the crater represents a concavity analogous to a spherical segment having a single base (1/6pi x H [3r2 + H2], where H is the height of the spherical segment and r the radius of the sphere), while disregarding the volume of the central droplet leftover from the emission.  The following are mean +/- SEM crater diameters (D), crater depths (H) and maximum volumes (V) of extruded metallic material for two types of aluminium cathodes, Alzak and H34 hardened aluminium, subject to a high input current PAGD:


1.  Alzak: D -0.028 cm +/- 0.003; H -0.002 cm +/- 0.0002; V - 6.2 x 10-7 cm3


2.  H34: D -0.0115 cm +/- 0.0004; H -0.0006 +/- 0.0001; V - 3.1 x 10-8 cm3


Accordingly, utilising plates composed of either material with 3 mm of thickness, and thus with a volume of 38.4 cm3 per plate and considering that only 2/3rds of the cathode shall be used (a 2 mm layer out of the 3 mm thickness), the total number of pulses per plate total (TLT) and partial (PLT) lifetimes is theoretically:


1.  Alzak: TLT: 6.2 x 107 pulses; PLT: 4.1 x 107 pulses;


2.  H34: TLT: 1.2 x 109 pulses; PLT: 8.1 x 108 pulses.


Typically, an H34 device can produce about 0.25 kWh per 10,000 pulses.   The corresponding value for a PLT is thus a minimum of 1.0 MWh/Alzak cathode and of 20 MWh/H34 cathode.   As the cathode for each combination is only 66.7% consumed, the vacuum pulse generator may continue to be used in a reverse configuration, by utilising the other plate in turn as the cathode; thus, the estimated minimal values become, respectively, 2.0 MWh/Alzak pulse generator and 40 MWh/H34 pulse generator.  The same rationale applies for the double diode configuration of Fig.10C.


We have created a two-ported system for the production of the singular discharge events which we have previously identified in the “863” application as an endogenous pulsatory abnormal glow discharge regime where the plasma discharge is triggered by spontaneous electronic emissions from the cathode.  We have examined the functioning of this two-ported system in order to determine what were the electrical power input and output characteristics of a sustained PAGD regime.  Despite the wide (10-fold) variations in net power and break-even efficiencies measured by the four different methods employed (open voltage measurements, time integration of negligible power measurements, operational power measurements and real time non-negligible power measurements), all methods indicate the presence of an anomalous electrical transduction phenomenon within the vacuum pulse generator, such as can result in the production at the output port of electrical energy measured and directly captured which is greater than would be anticipated having regard to the electrical energy input at the input port.   With the most accurate of the methods employed, we have found typical PAGD power production rates of 200 WHr/hour of PAGD operation, and these may reach >0.5 kWh/h values.


The discrepancies between the methods utilised have been extensively examined in the preceding section. Our systematic approach demonstrates that the most frequently employed method of measuring the charge capacity of batteries by the open voltage values is the least reliable approach for the determination of the actual net power lost or gained by the battery packs used in the system: when compared to all three other methods, it overestimates net power consumed and produced by up to 10 fold, as well as distorting the break-even efficiencies, particularly at the extremes of operation. All this results from the grossly diminished (50-60% of manufacturer's theoretical estimate) effective charge capacity of the lead acid gel cells employed, as determined experimentally from Fig.18 and Fig.19, when compared to the theoretical maximal charge capacity values that serve as scale for the open voltage measurements. In other words, the effective energy density of the batteries during these experiments was in fact approximately half of the manufacturer's estimated 30 WHr/kg.


Under these actual conditions of battery performance, the third and fourth methods (respectively, operational and real-time non-negligible power measurements) of power consumption and production proved to be the best approach to measure both PAGD electrical power input and output, as the results of both methods matched each other closely, even though the former is a statistical treatment of simultaneous events and the latter is a real time integration of their cumulative effects.  The second method is clearly less reliable than either the third or the fourth methods, and this stems from the fact that the power consumption slopes of negligible resistive discharges not only are very different from the quasi-steady state discharge slopes (beginning at >5 - 15 minutes) of extensive resistive discharges, but also their proportionality may not reflect the real time proportionality of equivalent prolonged resistive discharges.


The main advantage of the fourth method is that it effectively takes into account the actual time performance of the batteries comprised by the overall PAGD production and capture system we have described.   As such, the method may have the main disadvantage of reflecting more the limitations of the batteries employed (their high rate of degradation of the absolute value of total effective charge capacity, and limited efficiency in retaining charge derived from discontinuous input pulses) than indicating the actual power output.   There are a number of possibilities for fine tuning of the system introduced by the present work, beginning with the utilisation of secondary batteries or other charge shortage or absorption devices that have less variable or more easily predictable actual charge capacity. 


In this respect, there are two major shortcomings to the batteries used to form the drive and charge packs; (1) their significant memory effect and (

2) their design for constant, rather than discontinuous, DC charging.


Recently developed Nickel Hydride batteries are an example of an electrostatic charge-storage system that lacks a substantial charge memory effect, and their experimental batteries are being developed currently for higher efficiency intermittent charging methods.   Electrostatic charge retention systems having better energy densities, better charge retentivities and insignificant memory effects will probably be more efficient at capturing and holding the energy output by the circuit. In practical embodiments of the invention, effectiveness in charge utilisation will be more important than measurability, and any device that will use the energy effectively whilst presenting an appropriate back EMF to the system may be utilised.


The effect of the performance characteristics of the drive and charge packs is only one amongst many parameters affecting operation of the invention.   As shown by our extensive investigation of the diverse PAGD phenomenon the recovery of energy from it by electromechanical transduction as in the “531” application, or electrostatic capture as described above, the factors involved in modulating the frequency, amplitude and peak current characteristics of the PAGD regime are complex.   Manipulation of these factors can improve electrical energy recovery, or reduce it or even suppress PAGD.  We have so far noted numerous factors that affect PAGD frequency and some amongst those that also affect the PAGD amplitude. Aside from these factors, the circuit parameters of the output port portion of the circuit, in addition to the nature and chemical characteristics of the battery cells already discussed, the charge potential of the charge pack, the characteristics of the rectifiers in the recovery bridge in relation to the period of PAGD super-resonant frequencies, and the effective values of the parallel and transversal capacitance bridges can all influence the results achieved.   Certain factors however have a radical effect on PAGD operation, such as the gap distance and the charge pack potential.  


Too small a gap distance between the cold emitter (cathode) and the collector will result in an increasing reduction in energy recovery.   The potential presented by the charge pack must be less than the voltage amplitude developed by the PAGD, as specified by a given gap distance at a given pressure.   Too large a charge pack size with respect to PAGD amplitude and the gap length will preclude PAGD production or result in extremely low PAGD frequencies. In brief, the energy absorption rate and the counter potential presented by the charge pack or other energy utilisation device are important factors in the operation of the circuit as a whole, and should either be maintained reasonably constant, or changes should be compensated by changes in other operating parameters (as is typical of most power supply circuits).


Since our test results indicate that the electrical power output of the circuit can be greater than the electrical power input to the circuit, the circuit clearly draws on a further source of energy input.   Whilst we do not wish to be confined to any particular theory of operation, the following discussion may be helpful in explaining our observations.  These observations have been discussed in some detail so that the phenomenon observed can be reproduced, even if the principles involved are not fully understood.


In the “863” and “531” applications we have identified a novel, cold-cathode regime of vacuum electrical discharge, which we have termed the pulsed abnormal glow discharge (PAGD) regime.   This regime, which occupies the abnormal glow discharge region of the volt-ampere curve of suitable discharge tubes, has the singular property of spontaneously pulsing the abnormal glow discharge in a fashion which is coming from the tube and its circuit environment that constitutes a vacuum pulse generator device, when it is operated under the conditions which we have identified.   In fact, when stimulated with continuous direct current, in such conditions, such a circuit responds with spontaneous abnormal glow discharge pulses that enable effective segregation of input and output currents.


We have demonstrated electrically, metallographically, oscillographically and videographically, how the pulsed discontinuity results from a self-limiting, auto-electronic cathode emission that results in repeated plasma eruptions from the cathode under conditions of cathode saturated current input.  The auto-electronic triggering of the PAGD regime is thus akin to that of the high-field emission mechanism thought to be responsible for vacuum arc discharges (VAD regime).   However, under the PAGD conditions we have defined, this mechanism is found to operate in the pre-VAD region at very low field and low input average direct current values, with very large inter-electrode distances and in a self-limiting, repetitive fashion.  In other words, the PAGD regime we have identified has mixed characteristics: its current versus potential (abnormal glow) discharge curve is not only distinct from that of a vacuum arc discharge, but the electrical cycle of the PAGD regime itself oscillates back and forth within the potential and current limits of the abnormal glow discharge region, as a function of the alternate plasma generation and collapse introduced by the discontinuous sequencing of the auto-electronic emission process. Accordingly, the intermittent presence of the abnormal glow, as well as the observed segregation of the current flows, are due to the diachronic operation of these spontaneous cathode emission foci.  The micro-crater and videographic analyses of the PAGD have demonstrated the presence of an emission jet at the origin of each pulse, a phenomenon which VAD theory and experiment has also identified.   Metallic jets originating at the cathode spots of VADs have been known to present velocities up to, and greater than 1000 m/sec.


In light of the above, the energy graft phenomenon we have isolated would have to be operated, at the micro-event scale, by the interactions of the cathode emission jet with the vortex-formed impulse-transducing plasma in the inter-electrode space.   Several aspects can be approached in terms of the complex series of events that constitute a complete cycle of operation, on a micro-scale.  There are interactions within the cathode, interactions at the cathode surface, interactions between the emission jet and the plasma globule close to the cathode, and finally, interactions of the resulting electron and ion distributions in the inter-electrode plasma, within parallel boundaries.


In general, in the presence of an electrical field, the distribution of potential near the cathode forms a potential barrier to the flow of electronic charge, as this barrier is defined by the energy that the most energetic electrons within the metal (the Fermi energy electrons) must acquire before freeing themselves from the cathode surface potential, to originate an emission jet.  Before any free electrons become available for conduction in the space adjoining the cathode, they must cross the boundary posed by the potential barrier. With a weak applied field, classical electron emission from a metal can only occur if an energy practically equal to the work-function of the metal is imparted in addition to the Fermi energy.  Under thermionic conditions of emission, the heating of the cathode provides the needed energy input.  However, the cold-cathode Fowler-Nordheim quantum-field emission theory predicted the existence of a finite probability for an electron to tunnel through the potential barrier, when the applied field is high.  Cold-cathode electron emissions are thus possible, under these conditions, at practically Fermi energy levels, as the high field would catalyse the tunnelling through the potential barrier by narrowing the barrier width for the Fermi energy electrons.  The exact localisation of the emission would then depend on the randomised fluctuations of high fields at the cathode, which were produced by positive space charges sweeping in proximity to it.


For most purposes, this theory has been the working hypothesis of the last 60 years of field emission studies, which have centred upon the VAD mechanism, despite the fact that observed field gradients are evidently inadequate to explain breakdown as a function of the theoretical high field mechanism. The Fowler-Nordheim theory has therefore suffered major revisions and additions, mostly to account for the fact that it postulates, as a condition for cold-cathode field emission in large area electrodes, the presence of enormous fields (>109 V/m) and extremely low work functions, neither of which are borne out by experimental VAD investigations. Some researchers have found that the breakdown responsible for the VAD field emission is promoted by Joule heating and vaporisation of microscopic emitter tips, and that this requires a critical current density (1012 A/cm2), while others emphasised that this explanation and these thresholds did not hold for large area emitters and that a space charge effect of concentrating the ion distribution near the cathode promoted breakdown under these circumstances, when the field reached a critical value; large field enhancement factors (more than a thousand-fold) have been postulated to explain the discrepancy between theoretical predictions and experimental findings regarding the critical breakdown field values, and others have demonstrated how this critical field value effectively varies with work-function and electrode conditioning.


The PAGD regime and its self-extinguishing auto-electronic emission mechanism stands as an exception to the high field emission theory as it currently stands with all its modifications, especially given that in this phenomenon we are confronted with a cathode emission that spontaneously occurs across the large gaps in large plate area pulse generators, at very low field values (down to <1 x 104 V/m), as shown above and in the “863” application.   Moreover, a Fowler-Nordheim plot (in the form Log10 (I/V2) versus 1/V) of the PAGD volt-ampere characteristic exhibits a positive slope, rather than the Fowler-Nordheim negative slope characteristic of VAD field emission.   However, current density values obtained from correlations of autographic analysis of the cathode with an analysis of event-oscillogram (peak pulse currents), indicate that the PAGD current density J may reach values of 105 to 107 A/m2 during the emission process (the larger Alzak craters have an associated lower J value), values which, at the upper end, do not reach the 109 A/m2 current density threshold required by the Fowler-Nordheim theory.   Considering these two distinct observations with regards to field strength and current density, we have to admit the existence of a low field, large area cold-cathode auto-electronic emission endowed with high current densities, which is not predicted by current field emission theory.


Unlike the typical VAD regime, the PAGD is neither a high frequency oscillation, nor does it occur in a random fashion.  It constitutes a semi-regular, quasi-coherent, periodic energy transduction which cycles between cathode drop limits that are higher by a factor of 2 to 15 than typical vacuum arc cathode drops. The intermittent cathode emission responsible for the low frequency, pulsed behaviour of the abnormal glow, is also self extinguishing and self-starting, under the conditions we have defined.   Furthermore, we have also identified a novel and unexpected dependency of the periodic pulse rate upon the cathode area. This indicates the presence of field emission control parameters heretofore unsuspected.  It is likely that field fluctuations of the polarised pre-breakdown field is responsible for eliciting the particular localisations of the auto-electronic emission foci, as well as what imparts, in a lens-like fashion, the distorted field energy needed for electron surface release. In this sense, external, electrical or magnetic field fluctuations (e.g. motion of static charges or of constant magnetic fields) induced by us at pre-breakdown potentials, provoked PAGD emissions and breakdown at these levels.


In general, VAD studies have shown that, for large area electrodes, microgeometry, adsorbed gas layers and gas impurity contents of the cathode play a role in modulating field emission.   In our PAGD studies, the interactions at the cathode surface and across the cathode potential drop are clearly modulated by:

(1) the nature of residual gases, as shown by our air vs Argon studies;

(2) their pressure,

(3) electrode conditioning,

(4) work-function and

(5) cumulative pulse count, amongst others.


The plasma, in leak-controlled or low pressure PAGD devices, has both residual gas and metallic vapour substrates. In devices initially closed at high to very high vacua (diffusion pump pressures), the major residual substrate, whose presence increases with time of operation, is the metallic vapour released from the cathode and not impacted on to the envelope walls or the anode.  It has been previously shown for externally (magnetically or electrostatically) pulsed plasma accelerators, that the amount of residual gas or vapour left in the inter-electrode space diminishes with increasing number of consecutive discharges and a growing amount of electrode-insulator absorption of gas.  The effect of such removal of residual gas or vapour is to decrease the vacuum of a sealed envelope.   With high vacuum sealed PAGD generators we have observed that prolonged operation and sputter-induced mirroring of the envelope causes a progressive disappearance of the discharge, as the voltage potential needed to trigger it also increases.  At the thermocouple, low frequency pulsed abnormal glow discharges can also be seen to increase the vacuum significantly.  These results suggest instead the presence of a pumping mechanism in the PAGD which is somewhat analogous to that of sputter ion pumps, where collision of ionised gas molecules with the cathode is responsible for the sputtering of cathode material that either combines with the gas substrate (`gettering` action) or `plasters over` the inert gas molecules on to the anode (a process known as `ion burial`). These are the two basic pressure reducing actions of sputtered getter atoms, in ion pumps.


However, in ion sputter pumps, the initiation of the cycle is a function of the presence of high velocity electrons in the high field plasma of the glow discharge, which are necessary to ionise the gas substrate molecules; also, the getter material typically has a high work-function for field emission.  Hence, the sputtering is due to the secondary impact of plasma positive ions at the cathode, after plasma ionisation has occurred in the inter-electrode space.  Altogether different is the mechanism of spontaneous, primary electron emission from the cathode, which is characteristic of the low field PAGD: here, the sputtering is caused by the electronic emission itself and attendant metallic vaporisation processes.  By artificially confining the firing foci to a part of the cathode, we have shown in the single diode configuration how the PAGD induced sputtering is associated with the cathode auto-electronic emission mechanism, rather than with the abnormal cathode glow per se, given the localisation of sputtering on to the emission region of the plate, despite its overall cathode glow saturation.


These observations would thus seem to corroborate the hypothesis of a progressive vacuum increase with the cumulative number of emitted pulses, were it not for the fact that experiments performed with leak controlled devices (reported here and in previous studies) show that, when the negative pressure is maintained by balanced leak admission of air or argon, pulse rates still decrease with cumulative pulse count, and do so neither as a function of an increase in vacuum, nor as a function of envelope mirroring (unless this is so extensive as to establish envelope conduction), but rather as a function of processes (generally referred to as conditioning) inherent to the electrodes, specifically, to the cathode.   We have further shown that, for such altered emitter states, the pressure of the vessel must be increased, not because of an increasing vacuum (precluded by the controlled gas leak), but because of the effect that residual gases may have in modulating the low field PAGD emission.


PAGD electrode conditioning is a cathode-dominant process resulting from the cumulative emission of high numbers of pulses by a cathode, and has been shown to be a factor independent of the nature and pressure of the residual gas and partially reversible only by operation with reversed plate polarity, unlike reports of copper cathode-dominant conditioning.  It is thought that electrode conditioning and the accompanying increase in VAD breakdown potential are due to the progressive adsorption of residual gases, though cathode-dominant conditioning processes, such as subjecting the vacuum gap to consecutive discharges, have been shown to correlate the decrease in plasma impulse strength with electrode outgassing of absorbed or adsorbed gases.   Moreover, given the pitting action of crater formation at the cathode by the PAGD regime, and, as we shall see below, the metallic plating of the anode, the PAGD cathode-dominant process of conditioning we have observed with respect to decreased pulse frequency and increase in potential, suggests that the apparent increase in cathode work function is not due to gas adsorption or absorption.   These processes are more likely to occur on the plated anode.  It is likely that, given the observed PAGD pressure reducing effect caused by the cathodic jet, a certain outgassing of the cathode is in fact occurring during PAGD function.


One might also expect that the anode, if plated by sputtering atoms, would increase its gas content in the formed surface film.   However, controlled leak experiments suggest instead that some other type of alteration of the cathode work function is occurring, which is, as we shall examine below, independent of the adsorbed gas state of the electrodes, as well as independent of the PAGD ion pump-like effect.  Nonetheless, even at the level of the anode, the PAGD sputtering action may have contradictory effects: it may impact inter-electrode gap molecules on to the collector, as well as release, by ionic bombardment and vaporisation, gases adsorbed to, or contaminating the anode.   If we assume that gas adsorption by impact on the collector is the predominant mechanism, one could explain the increase in the number of breakdown sites per unit time, as observed by us for a re-reversed cathode, if the number of PAGD breakdown sites depended on the quantity of adsorbed gases, e.g. oxygen, on the cathode being tested.  Recovery of the cathode work-function would depend on the electronic charge recovery of the positively charged, adsorbed or occluded gas layer at the cathode- either by reversal or as a function of time of inactivity.


The surface film theory of “electrical double layer formation at the cathode” in fact contended that, low field flash over is a photocathodic effect dependent upon the presence of a glowingly positively polarised gaseous film at the cathode; this film would lower the cathode emissivity by decreasing the field between the cathode surface and the leading edge of the cathode glow, across the cathode drop.  However, even though the surface film theory of “electrical double layer formation at the cathode” predicts the lowering of the emission breakdown potential and the increase in flash over rate when the electrodes are reversed - as the anode would have acquired a surface charge capable of affecting the breakdown potential, it acknowledges nevertheless, that the anodic surface charge hardly explains the observed intensity of the polarisation effects.


Moreover, non-reversed, conditioned cathodes retained their lower PAGD frequencies in a time-independent manner, for as long as reversal was avoided (excluding a PAGD frequency recovery effect due to plate cooling, which may be as short as 15 minutes).  PAGD conditioning was independent of idle time and increased with cumulative pulse count.  Moreover, the AGD pulses are not UV photocathodic Townsend discharges, liberating secondary electrons via positive ion impact at the cathode.  Nor could photocathodic emissions generate currents of the magnitude observed in the PAGD.  Lastly, the PAGD discharge and breakdown thresholds appear to be unaffected by UV, though they may be somewhat depressed by visible light, and the emission mechanism in the PAGD is the primary process.


Removal or flattening of protuberances and tips from the emitting cathode by the action of the discharge, is a process also thought to play a role in hardening the cathode or increasing its field emission work-function. However, this explanation may not be adequate for the PAGD emission process, if we consider our metallographic findings of a smoothing action of the discharge at the collector. In fact, it would appear that the flattened, smoother, plated, mirrored and cleaner surfaces subjected to PAGD bombardment are the explanation for the observed increased emission ability of re-reversed cathodes: mirrored Alzak surfaces emit at higher frequencies than do dull H34 and H220 surfaces; new, polished surfaces emit at a higher frequency than do pitted, broken-in surfaces; anode surfaces, never before utilised as cathodes but subjected to prolonged PAGD action, emit at higher frequencies when employed as cathodes, than do new, identical cathode surfaces; and ex-cathodes, employed for prolonged periods as anodes, regain a higher emission frequency upon re-use as cathodes.  The better PAGD emission performance of smoother cathodes, compared with the worse VAD emission performance of the same, when pitted cathodes (lacking protuberances) are used, requires explanation.


Rakhovsky has put forth a VAD model for cathode spots, that distinguishes between Type I spots (quickly moving spots, far from steady state and responsible for crater formation), and Type II spots (quasi-stationary and near steady-state, but leaving an itinerant track with no sign of crater formation).   Whereas the former would obey the Fowler-Nordheim requirement for high fields (>109 V/m), the latter could hardly be expected to do so with typical arc voltage drops in the order of 10 V.   Once again, autographic analysis of the PAGD emission aspect indicates mixed characteristics: the PAGD cathode spot is a hybrid.  It behaves as an intermittent instability that leaves single (e.g. in H34) or clustered (e.g. in Alzak) craters, which are both qualities of Type I cathode spots; and it exists under low field conditions (<105 V/m), with cathode drops of 20 to 150 V, in a quasi-coherent mode, leaving an itinerant track of successive craters when operating at the higher frequencies, all of which are properties approaching those of a VAD Type II cathode spot.


Furthermore, the macroscopically visible metal sputtering (due to the explosive action of the PAGD emission phenomenon) occurring at the upper end of the permissible DC current input scale, and the presence of large solidified molten metal droplets in and around the craters, suggest models which have been proposed for explosive electronic emission.   Explosion models propose that the creation of a residual plasma ball in front of a microprotuberance provokes the large potential drop at the prospective emission focus and sufficiently high resistive and Nottingham heating to reach >107 A/cm2 current densities during the explosive consumption of these microemitters.  Whether the explosive action associated with cathode spots is an auxiliary effect that applies solely to the vaporisation of the emitting microprotrusion, or an integral emission and vaporisation explosive process, it does not appear that it can be restricted to high-field VAD Type II cathode spots, given that it can be equally made to occur with the low field PAGD hybrid cathode spot, and be macroscopically observed.   Indeed, in the plate diode configuration, it is easy to visualise the metallic particle explosions that surround and accompany the plasma jets, near to upper current limit conditions. However, if we are to assume that any of these models apply to the emission mechanism, we would, in all likelihood, have to conclude that the PAGD initial emission sites must be sub-microscopic (100 to 10 nm), rather than microscopic.


Resolution limits to our own metallographic examination of the smoothing action of the PAGD discharge on the collector would thus have precluded us from detecting formation of such sub-microscopic protrusions, as well as their presence in a “soft” cathode and thus infer their disappearance from a pitted, hardened cathode; but if the disappearance of such sub-microprotuberances were responsible for the observed alteration of cathode work function, one would also thereby have to postulate the existence of a mechanism for microroughness regeneration (e.g.. tip growth) at the anode, in order to explain the observed increased emission upon cathode re-reversal.   Furthermore, this regeneration would have to be actively promoted by operation with reversed polarity, and this is problematic.  Focusing of the distorted or magnified field upon alumina inclusions on pure iron electrodes has been demonstrated to degrade breakdown voltage for field emission, but the effect was greater for larger microscopic particles.   If we were to apply this concept to our work, it would require the existence of unmistakably abundant microscopic heterogeneities in the quasi-homogeneous electrode surfaces employed, which we did not observe; on the contrary, their absence suggests that either the microroughness responsible for the low field PAGD emission is sub-microscopic, or that the field distortion responsible for eliciting the PAGD is independent of the presence of these protuberances.  This last possibility must be taken all the more seriously, in light of the fact that PAGD functioning is able to cover the entire surface of an emitter with craters.


Whereas the discharge potentials observed in the PAGD have been shown to be relatively independent of the kind of gas present, there is a gas effect in the PAGD phenomenon, particularly in what concerns its frequency, observed when the same “run down” cathode was capable of much higher emission rates when exposed to argon, than to air.   Utilising the technique of bias sputtering, it has been demonstrated that the number of charge symmetric collisions (dependent upon sheath thickness d and the ion mean free path) in the plasma sheath, which are responsible for lower energy secondary peaks in ion energy distribution N(E), at pressures of 0.2 Torr, is substantially greater in argon than in argon-nitrogen mixtures, and thus that, under these conditions, mostly Ar+ and Ar++ ions impact the negatively biased electrode.  In non-equilibrium RF discharges, greater ion densities have also been attained with argon, than with air.  With respect to field emissions, one would expect a gas effect only with regards to changes on surface conditions, though such studies have shown contradictory effects of argon upon cathode work function.


In light of the foregoing, and given that the PAGD is an emission discharge and not a sputtering discharge per se, in the strict sense, we can conceive of the role of inert gas atoms in increasing, as compared to air or nitrogen, the ion energy density distribution at the PAGD cathode spot interface with the cathode surface emitter, and thus elicit increased emission rates from the cathode, by pulling electrons from the metal via the field effect.   While this is consistent with the concept of focused distortions of space-charge field fluctuations inducing localisation of the emission foci, the argon effect can be observed in the PAGD regime over the entire range of the Paschen low vacuum curve, and into Cooke's mid to high vacuum curve, at low fields and without negative biasing.   Thus, it is not simply a high pressure (nor a gas conditioning) effect, even if the gas effect in question applies to the description of a local pressure rise at the emission site/cathode spot interface, which may play a role in enhancing the local field.


Considered together, the PAGD emission-derived sputtering, the observed metallic plating of the anode and the explosive aspect of the discharge, suggest the presence of a jet of metallic vapour present in the discharge and running, contrary to the normal flow of positive ions, from the cathode to the anode.  This jet appears to have properties similar to the high speed vapour ejected from the cathode in a VAD, as first detected by Tanberg with his field emission pendulum (Tanberg, R. (1930), "On the cathode of an arc drawn in vacuum", Phys. Rev., 35:1080) In fact, the VAD high field emission process is known to release, from the cathode spot, neutral atoms with energies much greater than the thermal energy of the emission discharge. This anomalous phenomenon brings into play the role of the reported cathode reaction forces detected in vacuum arc discharges (Tanberg, as above, also Kobel, E. (1930), "Pressure and high vapour jets at the cathodes of a mercury vacuum arc", Phys. Rev., 36:1636), which were thought to be due to the counterflow of neutral metallic atoms, from the cathode on to the anode (charged metallic ions are normally expected to target the cathode).   In absolute units of current, this current quadrature phenomenon has been shown to reach, in the VAD regime, proportions of the order of 100 x I2 (see also the Aspden papers referenced below).


Early interpretations attributed this to the cathode rebounding of <2% of gas substrate-derived plasma positive ions hitting the cathode and being charge-neutralised in the process, but having kept most of their thermal energy.  Tanberg held instead that the counterflow of neutral particles responsible for the cathode reaction force was cathode derived, effectively, that it constituted a longitudinal interaction acting in the direction of the metallic arc jet.   However, even though secondary high energy distributions of neutral atoms emanating from the cathode do not have thermal energies, their modal distribution does (Davis, W. D. and Miller, H. C. (1969) J. Appl. Phys., 40:2212) furthermore, the major anomalous atomic counterflow that accompanies the high-energy electron flow toward the anode, was shown mass spectrographically to consist predominantly of multiply ionised, positively charged ions of cathode metal, rather than neutral atoms.  If this made it easier to abandon the primacy of the rebounding model, it was now more difficult for field emission theorists to accept and explain the observed high energies (ion voltages in excess of the discharge voltage drops) and the high ionisation multiplicity associated with these counter-flowing positive ions.


This field of investigation has indeed been one of the mounting sources of evidence suggesting that there is something amiss in the present laws of electrodynamics.  The anomalous acceleration of counter-flowing ions, and the energy transfer mechanisms between high speed or “relativistic” electrons and ions in a plasma (Sethion, J. D. et al, "Anomalous Electron-Ion Energy Transfer in a Relativistic-Electron-Beam-Heated Plasma" Phys. Rev. Letters, Vol. 40, No. 7, pages 451-454), in these and other experiments, has been brilliantly addressed by the theory of the British physicist and mathematician, H. Aspden, who first proposed a novel formulation of the general law of electrodynamics capable of accounting for the effect of the mass ratio factor (M/m') in the parallel (and reverse) motion of charges with different masses, (Aspden, H. (1969) "The law of electrodynamics", J. Franklin Inst., 287:179; Aspden, H (1980) "Physics Unified", Sabberton Publications, Southampton, England). The anomalous forces acting on the counter-flowing metallic ions would stem from their out-of-balance interaction with the emitted high speed electrons, as predicated by the electrodynamic importance of their mass differential.   This results in a fundamental asymmetry of the plasma flow between electrodes, localised on to the discontinuous interfaces of the plasma with the electrodes, namely, in the cathode dark space and in the anodic sheath: on the cathode side, electrons act upon ions, as the emitted electrons having less than zero initial velocities, drift against the incoming ion flux and in parallel with the ion and neutral counterflows; on the anode side of the discharge, positive ions flowing toward the cathode confront mainly the incoming counterflow of positive ions and neutral atoms, as the high speed electrons have abnormally transferred their energy to counter-flowing, high speed, cathodic metal ions.  An out-of-balance reaction force thus results at the cathode, to which the leaving metallic atoms impart a force of equal momentum but opposite direction, a force which is added to the cathode momentum generated by impacting, normal flowing positive ions.


Moreover, Aspden confirmed theoretically the fundamental contention of Tanberg's experimental findings that an electrodynamic force will manifest itself along the direction of the discharge current flow, and thus, that the atomic counterflow is a metallic jet.   Aspden further demonstrated that this asymmetry of plasma discharges does not imply any violation of the principles of conservation of energy and charge equivalence, given that there will be no out-of-balance force when such anomalous forces are considered in the context of the whole system of charge which must, perforce, include the local electromagnetic frame itself.   Such discharges must be viewed as open-energy systems, in balance with their electromagnetic environment: their apparatuses may constitute materially closed or limited systems, but they are physically and energetically open systems. Current work on Aspden's formulation of Ampere's Law indicates that both classical electromagnetism and special relativity ignore precisely, in circuits or in plasma, the longitudinal interactions that coexist with transverse ones.  Standing longitudinal pressure-waves, of a non-electromagnetic nature, have been previously shown in plasma electrons, which did not conform to the Bohm and Gross plasma oscillation mechanism (Pappas, P. T. (1983) "The original Ampere force and Bio-Savart and Lorentz forces", I1 Nuovo Cimento, 76B:189; Looney, D. H. and Brown, S. C. (1954) "The excitation of plasma oscillations" Phys. Rev. 93:965)


The present theoretical approach to the novel regime of electrical discharge which we have isolated in specially designed devices, and to its mixed glow-arc characteristics, suggests that a similar, out-of balance current quadrature phenomenon occurs in the discharge plasma during the low field, auto-electronic emission-triggered PAGD, and is responsible for the observed surplus of energy in the experimental system described in this report.  Clearly, all the evidence we have adduced indicates that there is a powerful longitudinal component to the emission-triggered PAGD, i.e. that the discharge pulses characteristic of this pre-VAD regime are longitudinally propelled jets of cathode-ejected high speed electrons and high speed ions.  We have performed experiments, in the PAGD regime of operation, with very thin axial members that bend easily when placed in the path of the discharge, or with Crooke radiometer-type paddle-wheels, and both show the presence of a net longitudinal force in the plasma discharge acting in the direction of the anode, which confirms the magnitude of the atomic counterflow (ionised and neutral) present during the PAGD, very much like Tanberg's pendulum did for the VAD.


These observations also tally with the explosive action of the emission mechanism, such as we have examined it above. In this context, two aspects of the PAGD are remarkable: the fact that a phenomenon akin to field emission occurs at low field values, for large area electrodes across large gaps, and the conclusion that the PAGD must deploy an excessively large counterflow of, in all probability, both ionised and neutral cathodic particles.   The observation of ion current contributions to the cathode current on the order of 8 to 10%, in VADs, can hardly apply to the PAGD mechanism responsible for the anomalous currents and counterflows observed.   Hence, we should further expect that the characteristically intermittent, or chopped current regime of the PAGD, is a major factor in the generation of disproportionately high energy longitudinal pulses and in allowing our system to capture most of the electrical energy output from the device.  In all probability, field collapse at the end of discharge favours the nearly integral collection of the plasma charge, and ensures the transduction of most of the plasma energy of the pulse (blocked, as it is, from flowing back through the input port to the drive pack) to the output port, through the parallel, asymmetric capacitance bridge that interfaces with the charge recovery reservoir (the charge pack). Collapse of the field of the discharge may also be a contributing factor to the anomalous acceleration of ions, and to the observed anode plating effect.


It is equally possible that such abnormally large longitudinal pulses may never be observable, for a given arrangement and scale, above threshold frequencies of the oscillation; we have, in this sense, presented data that indicates that for a given geometry, above specific PAGD frequencies, the capture of surplus energy decreases steadily in efficiency until it ceases altogether, for a given arrangement.   The point at which this surplus begins to decrease coincides with the setting in of frequency-dependent irregularities in the discharge sequence and, most importantly, it coincides with a reduction of the peak pulse current for each PAGD pulse. We have further remarked that increasing the PAGD frequency above the zero surplus point, for a given arrangement, by manipulating any of the frequency control parameters, provokes the slippage of the PAGD into a full fledged VAD regime, while input currents greatly increase and output peak currents greatly decrease (to comparable peak input levels of 10 to 15A).  


The transition between the two modes of emission-triggered discharge, PAGD and VAD, thus appears to be tied in to adjustable thresholds in the frequency of the emission discontinuities; in this sense, it is rather likely that the plasma field collapse plays a major role in regularising and optimising the anomalous energies of field emissions, as in the PAGD regime.   At low frequencies of low field emission, the emission regime is highly discontinuous, diachronic and regular, for it has time to fully extinguish the discharge; hence the PAGD singularity, in which the phases of each discharge pulse are well defined and sequential.   Above a given high frequency, when ion and electron recombination will happen more often, before each can be collected at the electrodes, the stream of emitted discontinuities merges into a noisy, randomised continuum, where simultaneous emissions become possible and the plasma field no longer has time to collapse and fully resolve the longitudinal pulses.  Any anomalous energy generated is then minimised and trapped in the plasma body and, in these conditions, the VAD regime eventually sets in.  Such model would easily explain why the high field VAD experiments performed to date have never detected such extraordinarily large anomalous forces.


On the other hand, the quasi-coherent aspect of the discharge suggests that the vacuum gap, in functioning during the PAGD regime both as an insulator and as a conductor with capacitative and self-inductive properties, is periodically altered by large and intense polarisations which are resolved by the discrete emission of longitudinal pulses from the cathode.   It is possible that these non-linear oscillations resulting from sudden depolarisation of the vacuum gap by high-speed explosive emissions elicited at the convection focus of the distorted field, might be in resonance or near resonance with the external circuitry, but the most apparent effect of increasing the capacitance in all bridge members is to increase the jet current and the transduced current flowing into the charge pack.  The PAGD amplitude variation also presents, after the large negative discontinuity, a growing oscillation at very high resonant frequencies, which are typical of inductive chopping currents in a VAD, before extinction occurs.   Unlike the VAD inductive case, in the absence of any coils other than the wire wound resistors, the PAGD relaxation oscillations which follow each pulse only extinguish the discharge when the voltage potential of the amplitude curve rises above the applied voltage, just as the plasma potential drops the most.


Given the entirely non-inductive nature of the external circuit utilised in many instances, the inductive properties in evidence are those of the vacuum device itself.   It also suggests that, in the absence of any need of an applied external magnetic field for the PAGD discharge to occur coherently, it is possible that the magnitude of the currents generated produces by itself a significant self-magnetic field.   Thus, we cannot rule out the possibility of a self-organisation of the plasma discharge, which may, in Prigogine's sense, constitute a dissipative structure (Prigogine, I. and George, C. (1977), "New quantisation rules for dissipative systems", Int. J. Quantum Chem., 12 (Suppl.1):177).   Such self-ordering of the PAGD plasma jet is suggested by the experimentally observed transition of these pulses from the current saturated limit of the normal glow discharge region, into the PAGD regime, as a function of increasing current: smaller foci of discharge can be seen to discontinuously agglutinate into larger emission cones, or into jets with a vortex-like appearance, when the input current reaches a given threshold.


It is possible that, under these conditions, the distribution of the charge carriers and their sudden fluctuations may render any steady-state plasma boundary conditions ineffective and provoke a singularity in the discharge mechanism; this non-linear behaviour, together with any self-magnetic effects, might provide radial coherence of the plasma flow along the longitudinal path of the discharge.   This concept is akin to what has been proposed for periodically fading-away solution structures referred to as “instantons”, that represent self-organising transitions between the two states of a system.  The PAGD may well be an instance of an instanton type structure bridging the open, or conductive, and the closed, or insulating, states of the vacuum gap.   An analytical formulation of the problem of the plasma flow from the cathode spot to the anode, which would take into account the self-magnetic and self-organising properties of the PAGD plasma channel, would be extremely difficult, given the out of balance longitudinal force, its abnormal energy transfer and associated counterflow, as well as the competition between collisional and inertial exchanges.


The plating observed at the anode most likely results from the impact of counter-flowing ions (and possibly neutral atoms), whereas the pitting of the (locally molten) cathode results from the emission of vaporised metallic material and electrons, as well as, secondarily, from bombardment by incident positive ions.  The first action smoothes the surface by mirroring it (deposition of cathode-derived atoms) and abrading it, whereas the latter smoothes it in places by rounding concavities and by forming molten droplets upon local cooling, while simultaneously roughening it on the crater peripheries.  One might think that this cathode roughening should lower the work function and facilitate the discharge, but the facts indicate that just the opposite must be happening in view of changes in the PAGD according to the nature and state of the cathode surface.  The observed alterations of electrode work function for PAGD low field emission must thus be related to the molecular and charge effects of these different actions at the two electrodes.  It appears that for large parallel plate electrodes, the PAGD low field emission is modulated by the nature and, most likely, by the molecular structure of the metallic surface layer of the emitter.


We have thus devised a system for the capture, as electricity, of the energy of anomalously energetic longitudinal pulses sequentially triggered by spontaneous emissions of high-speed electrons and ions generated from low work function cathodes, during the low field and singularly mixed PAGD regime of electrical discharge in vacuo.   To confirm the above interpretation of the anomalous flux in the observed PAGD phenomenon, the cathode jet composition, as well as time-dependent and usage-dependent changes occurring in the tubes, with diverse sealed negative pressures and after submission to prolonged PAGD operation, must be analysed by mass-spectroscopy.  In any event, the excess energy present in the anomalous counter-flowing force appears to stem from a discharge mechanism that effectively pulls high speed electrons and constituent atoms out of a metal surface, at low fields and with high current densities, and is modulated by a complex multiplicity of parameters.


The system described appears to transduce efficiently the observed non-linear longitudinal pulse discontinuities of the plasma field, under conditions of current saturation of the cathode, because the self-extinguishing and self-limiting properties of the discharge allows the energy from the collapse of the discharge to be captured.  The particular design of the circuitry, which couples a rectification bridge to the asymmetric bridge quadrature of large capacitances, placed at the output of the PAGD generator, permits effective capture.   Our findings constitute striking evidence for Aspden's contention of a need to revise our present electrodynamic concepts.   The dual ported PAGD discharge tube circuits which we have described are the first electrical systems we know of which permit effective exploitation of anomalous cathode reaction forces and allow for the recovery of electrical energy from systems exhibiting this effect.  Any apparent imbalance in the electrical energy input to the system and withdrawn from the system by its operator must be considered in the context of the entire continuum in which the system operates, within which it is anticipated that accepted principles of energy balance will be maintained.


Moreover, the energy conversion system of the invention has substantial utility as an electrical inverter accepting direct current, and providing one or more of a direct current output at lower voltage and higher current, variable frequency input to alternating current motors, and, by suitable combinations of discharge tube systems, more flexible DC-to-DC conversion systems.


As an alternative to the batteries used in the experiments described, a DC power supply may be utilised or, more advantageously from the viewpoint of entailing less transformation losses, a DC generator to provide the electrical energy input to the system.   As a DC motor can be run directly from the rectified output of the circuit of Fig.9 at El-E2, in place of a battery charge pack, DC motor/generator sets of suitable characteristics (in terms of back E.M.F. and circuit loading) can be used to charge the batteries of the drive pack, utilising the rectified PAGD output to drive the DC motor component of the set.   This provides a simple, one battery pack solution, where the PAGD input and output circuits are electrically separated by the DC motor/generator interface: the drive pack is simultaneously being discharged to drive PAGD production, and charged by the DC generator output which, in turn, is being driven by the electromechanical transformation of the rectified PAGD output that would typically accrue to a charge pack in the experiments already described.  The main limitations to such an arrangement lie in the efficiency of the motor and generator transformations utilised.


A pulsed DC source could be used to provide input to the circuit if suitably synchronised, but care is needed not to interfere unduly with the auto-electronic mechanism of the field induced cathode emissions.








































Patent US 5,590,031     31st December 1996     Inventors: Franklin Mead & Jack Nachamkin






This patent shows a system for converting Zero-Point Energy into conventional electrical power.




A system is disclosed for converting high-frequency zero-point electromagnetic radiation energy to electrical energy.  The system includes a pair of dielectric structures which are positioned near each other and which receive incident zero-point electromagnetic radiation.  The volumetric sizes of the structures are selected so that they resonate at a frequency of the incident radiation.  The volumetric sizes of the structures are also slightly different so that the secondary radiation emitted from them at resonance, interferes with each other producing a beat frequency radiation which is at a much lower frequency than that of the incident radiation and which is amenable to conversion to electrical energy.   An antenna receives the beat frequency radiation. The beat frequency radiation from the antenna is transmitted to a converter via a conductor or waveguide and converted to electrical energy having a desired voltage and waveform.



US Patent References:    

3882503    May., 1975        Gamara 343/100.

4725847    Feb., 1988         Poirier               343/840.

5008677    Apr., 1991         Trigon et al.       342/17.







The invention relates generally to conversion of electromagnetic radiation energy to electrical energy, and, more particularly, to conversion of high frequency bandwidths of the spectrum of a type of radiation known as ‘zero-point electromagnetic radiation’ to electrical energy.


The existence of zero-point electromagnetic radiation was discovered in 1958 by the Dutch physicist M. J. Sparnaay. Mr. Sparnaay continued the experiments carried out by Hendrik B. G. Casimir in 1948 which showed the existence of a force between two uncharged parallel plates which arose from electromagnetic radiation surrounding the plates in a vacuum.  Mr. Sparnaay discovered that the forces acting on the plates arose from not only thermal radiation but also from another type of radiation now known as classical electromagnetic zero-point radiation.  Mr. Sparnaay determined that not only did the zero-point electromagnetic radiation exist in a vacuum but also that it persisted even at a temperature of absolute zero. Because it exists in a vacuum, zero-point radiation is homogeneous and isotropic as well as ubiquitous.  In addition, since zero-point radiation is also invariant with respect to Lorentz transformation, the zero-point radiation spectrum has the characteristic that the intensity of the radiation at any frequency is proportional to the cube of that frequency.  Consequently, the intensity of the radiation increases without limit as the frequency increases resulting in an infinite energy density for the radiation spectrum.   With the introduction of the zero-point radiation into the classical electron theory, a vacuum at a temperature of absolute zero is no longer considered empty of all electromagnetic fields.  Instead, the vacuum is now considered as filled with randomly fluctuating fields having the zero-point radiation spectrum.   The special characteristics of the zero-point radiation which are that it has a virtually infinite energy density and that it is ubiquitous (even present in outer space) make it very desirable as an energy source.   However, because high energy densities exist at very high radiation frequencies and because conventional methods are only able to convert or extract energy effectively or efficiently only at lower frequencies at which zero-point radiation has relatively low energy densities, effectively tapping this energy source has been believed to be unavailable using conventional techniques for converting electromagnetic energy to electrical or other forms of easily usable energy. Consequently, zero-point electromagnetic radiation energy which may potentially be used to power interplanetary craft as well as provide for society's other needs has remained unharnessed.


There are many types of prior art systems which use a plurality of antennas to receive electromagnetic radiation and provide an electrical output from them.   An example of such a prior art system is disclosed in U.S. Pat. No. 3,882,503 to Gamara.   The Gamara system has two antenna structures which work in tandem and which oscillate by means of a motor attached to them in order to modulate the radiation reflected from the antenna surfaces.  The reflecting surfaces of the antennas are also separated by a distance equal to a quarter wavelength of the incident radiation.   However, the Gamara system does not convert the incident radiation to electrical current for the purpose of converting the incident electromagnetic radiation to another form of readily usable energy.   In addition, the relatively large size of the Gamara system components make it unable to resonate at and modulate very high frequency radiation.


What is therefore needed is a system which is capable of converting high frequency electromagnetic radiation energy into another form of energy which can be more readily used to provide power for transportation, heating, cooling as well as various other needs of society.   What is also needed is such a system which may be used to provide energy from any location on earth or in space.




It is a principle object of the present invention to provide a system for converting electromagnetic radiation energy to electrical energy.


It is another object of the present invention to provide a system for converting electromagnetic radiation energy having a high frequency to electrical energy.


It is another object of the present invention to provide a system for converting zero-point electromagnetic radiation energy to electrical energy.


It is another object of the present invention to provide a system for converting electromagnetic radiation energy to electrical energy which may used to provide such energy from any desired location on earth or in space.


It is another object of the present invention to provide a system for converting electromagnetic radiation energy to electrical energy having a desired waveform and voltage.


It is an object of the present invention to provide a miniaturised system for converting electromagnetic radiation energy to electrical energy in order to enhance effective utilisation of high energy densities of the electromagnetic radiation.


It is an object of the present invention to provide a system for converting electromagnetic radiation energy to electrical energy which is simple in construction for cost effectiveness and reliability of operation.


Essentially, the system of the present invention utilises a pair of structures for receiving incident electromagnetic radiation which may be propagating through a vacuum or any other medium in which the receiving structures may be suitably located. The system of the present invention is specifically designed to convert the energy of zero-point electromagnetic radiation; however, it may also be used to convert the energy of other types of electromagnetic radiation. The receiving structures are preferably composed of dielectric material in order to diffract and scatter the incident electromagnetic radiation. In addition, the receiving structures are of a volumetric size selected to enable the structures to resonate at a high frequency of the incident electromagnetic radiation based on the parameters of frequency of the incident radiation and propagation characteristics of the medium and of the receiving structures. Since zero-point radiation has the characteristic that its energy density increases as its frequency increases, greater amounts of electromagnetic energy are available at higher frequencies. Consequently, the size of the structures are preferably miniaturised in order to produce greater amounts of energy from a system located within a space or area of a given size. In this regard, the smaller the size of the receiving structures, the greater the amount of energy that can be produced by the system of the present invention.


At resonance, electromagnetically induced material deformations of the receiving structures produce secondary fields of electromagnetic energy therefrom which may have evanescent energy densities several times that of the incident radiation. The structures are of different sizes so that the secondary fields arising therefrom are of different frequencies. The difference in volumetric size is very small so that interference between the two emitted radiation fields, and the receiving structures at the two different frequencies produces a beat frequency radiation which has a much lower frequency than the incident radiation. The beat frequency radiation preferably is at a frequency which is sufficiently low that it may be relatively easily converted to usable electrical energy. In contrast, the incident zero-point radiation has its desirable high energy densities at frequencies which are so high that conventional systems for converting the radiation to electrical energy either cannot effectively or efficiently so convert the radiation energy or simply cannot be used to convert the radiation energy for other reasons.


The system of the present invention also includes an antenna which receives the beat frequency radiation. The antenna may be a conventional metallic antenna such as a loop or dipole type of antenna or a rf cavity structure which partially encloses the receiving structures. The antenna feeds the radiation energy to an electrical conductor (in the case of a conventional dipole or comparable type of antenna) or to a waveguide (in the case of a rf cavity structure).  The conductor or waveguide feeds the electrical current (in the case of the electrical conductor) or the electromagnetic radiation (in the case of the waveguide) to a converter which converts the received energy to useful electrical energy.  The converter preferably includes a tuning circuit or comparable device so that it can effectively receive the beat frequency radiation.  The converter may include a transformer to convert the energy to electrical current having a desired voltage.  In addition, the converter may also include a rectifier to convert the energy to electrical current having a desired waveform.





Fig.1 is a plan view of the receiving structures and antenna of a first embodiment of the system of the present invention with a schematic view of the conductor and converter thereof and also showing the incident primary and emitted secondary electromagnetic radiation.


Fig.2 is a front view of the receiving structures, antenna and waveguide of a second embodiment of the system of the present invention with a schematic view of the converter thereof and also showing the incident primary and emitted secondary electromagnetic radiation.



Fig.3 is a perspective view of the receiving structures, antenna and waveguide of the second embodiment shown in Fig.2 with a schematic view of the converter thereof and also showing the incident primary and emitted secondary electromagnetic radiation.




Fig.4 is a front view of the substrate and a plurality of pairs of the receiving structures and a plurality of antennas of a third embodiment of the system of the present invention with a schematic view of the conductor and converter thereof and also showing the incident primary and emitted secondary electromagnetic radiation.




Fig.5 is a top view of some of the components of the third embodiment of the system of the present invention showing two of the plurality of pairs of receiving structures and two of the plurality of antennas mounted on the substrate.




Fig.6 is a diagram of a receiving structure of the system of the present invention showing an incident electromagnetic plane wave impinging on the receiving structure and illustrating the directions of the electric and magnetic field vectors thereof.




Fig.7 is a diagram of a spherical co-ordinate system as used in the formulas utilised in the system of the present invention.





Fig.8 is a graph showing an imaginary rho parameter plotted against a real rho parameter illustrating the values thereof at resonance as well as values thereof at other than resonance.



Fig.9 is a graph showing a portion of the graphical representation shown in Fig.8 illustrating the real and imaginary rho values at or near a single resonance.


Referring to the drawings, a first embodiment of the present invention is generally designated by the numeral 10. The system 10 includes a first and second means for receiving 12 and 14 incident electromagnetic radiation 16. The means for receiving 12 and 14 are preferably a pair of spherical structures 12 and 14 which are preferably composed of a dielectric material. Alternatively, the spheres 12 and 14 may be cubical structures or any other suitable shape.  The spheres 12 and 14 may be mounted on a suitable foundation by any suitable mounting means (not shown), or spheres 12 and 14 may be suspended from a suitable foundation by any suitable suspension means (not shown).  The spheres 12 and 14 are preferably composed of a dielectric material.  The dielectric spheres 12 and 14 scatter and concentrate electromagnetic waves.  At very sharply defined frequencies, the spheres 12 and 14 will have resonances wherein the internal energy densities can be five orders of magnitude larger than the energy density of the incident electromagnetic field driving the spheres 12 and 14.  At resonance, the electromagnetic stresses, equivalent to pressures proportional to the energy density, can cause material deformation of the spheres 12 and 14 which produce a secondary electromagnetic field.  The spheres 12 and 14 are preferably positioned proximal to each other, as shown in Fig.1.   Although the proximity of the spheres to each other will adversely affect the resonances, the very high "Q"s of the isolated-sphere resonances results in such adverse affect being relatively small. However, the proximity of the spheres 12 and 14 allows the spheres to interact electromechanically which increases the magnitude of the secondary radiation emitted from them.


The electromagnetic radiation incident upon the spheres 12 and 14 which drives the spheres to resonance is preferably zero-point radiation 16.  However, other types of electromagnetic radiation may also be used to drive the spheres 12 and 14, if desired.


The effect of a dielectric sphere such as 12 or 14 on an incident electromagnetic radiation such as a plane wave thereof is shown in Fig.6.  The plane wave propagates in the z axis direction and is diffracted by the sphere 12 resulting in scattering thereof.  This scattering is commonly known as Mie scattering.  The incident radiation wave has an electric vector component which is linearly polarised in the x axis direction and a magnetic vector component which is linearly polarised in the y axis direction.


An electromagnetic wave incident upon a structure produces a forced oscillation of free and bound charges in synch with the primary electromagnetic field of the incident electromagnetic wave.  The movements of the charges produce a secondary electromagnetic field both inside and outside the structure.  The secondary electromagnetic radiation comprising this secondary electromagnetic field is shown in Fig.1 and designated by the numerals 18 and 20.   An antenna which is shown simply as a loop antenna but may also be a dipole or any other suitable type of antenna, is also shown in Fig.1 and designated by the numeral 22.  The non-linear mutual interactions of the spheres produces interference between the secondary electromagnetic radiation 18 and 20 produces a beat frequency radiation 24 which is preferably at a much lower frequency than the primary radiation 16.  It is this beat frequency radiation 24 which is desired for conversion into electrical energy because it preferably is within the frequency range of rf radiation which may be converted into electrical energy by generally conventional systems.  Thus, the radiation 24 received by the antenna 22 is fed via an electrical conductor 26 to a means for converting the beat frequency radiation 24 to electrical energy.  This means for converting is designated by the numeral 28 and preferably includes a tuning capacitor 30 and a transformer 32 and a rectifier (preferably a diode) 34.  Instead of including the capacitor 30, transformer 32 and rectifier 34, the converter 28 may alternatively include an rf receiver of any suitable type.


The resultant field at any point is the vector sum of the primary and secondary fields. For the equations that follow, the structure receiving the incident plane wave is a sphere of radius a having a propagation constant k1 positioned in an infinite, homogeneous medium having a propagation constant k2. The incident plane wave propagates in the z axis direction and is as shown in Fig.6.  The spherical co-ordinate system used for the vector spherical wave functions is shown in Fig.7.


Note:  As this patent contains so many non-standard keyboard characters, the remainder of this document is produced using direct images of the original text.


















































US Patent  4,936,961           June 26, 1990             Inventor: Stanley A. Meyer





Please note that this is a re-worded excerpt from this patent.  It describes one of the methods which Stan used to split water into hydrogen and oxygen using very low levels of input power.




It is an object of the invention to provide a fuel cell and a process in which molecules of water are broken down into hydrogen and oxygen gases, and other formerly dissolved within the water is produced. As used herein the term "fuel cell" refers to a single unit of the invention comprising a water capacitor cell, as hereinafter explained, that produces the fuel gas in accordance with the method of the invention.






Fig.1  Illustrates a circuit useful in the process.



Fig.2  Shows a perspective of a "water capacitor" element used in the fuel cell circuit.




Figs. 3A through 3F are illustrations depicting the theoretical bases for the phenomena encountered during operation of the invention herein.




In brief, the invention is a method of obtaining the release of a gas mixture including hydrogen on oxygen and other dissolved gases formerly entrapped in water, from water consisting of:

(a) Providing a capacitor, in which the water is included as a dielectric liquid between capacitor plates, in a resonant charging choke circuit that includes an inductance in series with the capacitor;

(b) Subjecting the capacitor to a pulsating, unipolar electric voltage field in which the polarity does not pass beyond an arbitrary ground, whereby the water molecules within the capacitor are subjected to a charge of the same polarity and the water molecules are distended by their subjection to electrical polar forces;

(c) Further subjecting in said capacitor to said pulsating electric field to achieve a pulse frequency such that the pulsating electric field induces a resonance within the water molecule;

(d) Continuing the application of the pulsating frequency to the capacitor cell after resonance occurs so that the energy level within the molecule is increased in cascading incremental steps in proportion to the number of pulses;

(e) Maintaining the charge of said capacitor during the application of the pulsing field, whereby the co-valent electrical bonding of the hydrogen and oxygen atoms within said molecules is destabilised such that the force of the electrical field applied, as the force is effective within the molecule, exceeds the bonding force of the molecule, and hydrogen and oxygen atoms are liberated from the molecule as elemental gases; and

(f) Collecting said hydrogen and oxygen gases, and any other gases that were formerly dissolved within the water, and discharging the collected gases as a fuel gas mixture.


The process follows the sequence of steps shown in the following Table 1 in which water molecules are subjected to increasing electrical forces. In an ambient state, randomly oriented water molecules are aligned with respect to a molecule polar orientation.

They are next, themselves polarised and "elongated" by the application of an electrical potential to the extent that covalent bonding of the water molecule is so weakened that the atoms dissociate and the molecule breaks down into hydrogen and oxygen elemental components.

Engineering design parameters based on known theoretical principles of electrical circuits determine the incremental levels of electrical and wave energy input required to produce resonance in the system whereby the fuel gas comprised of a mixture of hydrogen, oxygen, and other gases such as air were formerly dissolved within the water, is produced.




Process Steps:

The sequence of the relative state of the water molecule and/or hydrogen/oxygen/other atoms: