Russ Rob

I have studied the works of John Keely for 8 1/2 years, and spent about $600 on old documents. I have received statements from NY Library, The Franklin Institute, Library Of Congress, search services from Detroit, Michigan, and collected majority of old articles photocopies of notations. Another page sent to me has the photo of section of the heavy wall thickness brass tubing with dimensions as cut from behind the walls of Keely machine shop in Philadelphia 1899. Many Keely type reference books were purchased and studied.

He, in 1865 had went to New Jersey machine shop and purchased equipment from man who was bilking money from a doctor. This newsman then drove all the way back to Pa. and to Keely's lab. In the back room was the same equipment as perfectly described in New Jersey. There was a 30" heavy wall thickness sphere high pressure tank capable of 30,000 lbs pressure, a 40HP piston engine that Keely had hooked up to building gas line, and 35,000lb pressure liquid hydraulic screw type fluid pump.

Now, in case you don't know about the demonstrated high pressure generation of moisturized ETHER, here is an example. A rural country homeowner with water pump in his basement , has a separate water tank kept at about 40lbs pressure. There a large air bubble kept at upper inside of tank. Water is normally taken out from the bottom and pumped upstairs. Keely had later kept the sphere tank under the floorboards of first floor. Many knew that he had this, but did not make any notion connection "about it". A valve at top of water tank will RELEASE very small amount "quantity" of air at the same pressure. This is what Keely did with his equipment at 23000lbs. The ID of brass tubing flowed little quantity , but at the very high pressure, and he called it various levels of ETHER disassociation.The gas released was moisturized and had rendered the lengths of tubing cold to touch.
Engineers did not like the pressure gauges as they may have been tampered with, so Keely removed all of them and used another apparatus to prove exact high pressure.  A warehouse weigh scales with vertical cylinder would push upward against negative leverage 14 to one ratio steel arm on pivot end against a 500lb weight hung on end. The long brass tubing across the room attached to very small high pressure vertical tank that contained ID so that the vertical tank volume displaced was matched to the air cylinder internal  volume . The test would then draw very little volume, only as necessary to raise up the arm proving actual 23,000lbs. The scales and cylinder was dismantled to prove no fraud by the engineers. Back in those 1800's very high pressure was not generally known about .

I was especially interested in the demonstration of 15 second generation of 51 inches vacuum for open pan mercury bath with standard  shape manometer as open to air. The displacement of mercury showed level change of both sides of 51 inches. Later, i determined that regular labs without an expensive vacuum source pump, install a special T fitting on sink spout. The center hole produces small vacuum source for lab tests. Keely did the same thing with flowing 23,000lbs air past fitting , causing vacuum suction to change the manometer. reading. The second  floor of Keely shop had carpeting with recurring designs. Upon building inspection of 1899, their were under carpet spring operated valves that could be pressed by foot. Keely kept visitors during energy demonstrations behind a roped area.

Another vertical machine was called The Liberator. It had various tuning forks and a wind up noisy 15 second timer up above. The high pressure thick walled long tank was bracket mounted horizontally to machine. This would also demonstrate 23000lbs pressure when operated, and went to the lever arm tester.
Visitor would open valve at end of tank, pouring in water, so that it flowed out other end. This seemed to satisfy the investors and engineer of no previous charged pressure.  After my several cups of hot chocolate during one winter, I examined the machine photo with magnifying glass and took notes. I had figured out that fraud also. The Keely Motor Company was a $5 million dollar business. He had all the money to purchase expensive parts. The valve at right end of tank was standard type. The other valve at left end was homemade. They both should have been the same.
The next discovery was a weld line on same left end. This means the end was sawed off and welded again. Now....I want you to think about what happens when pressing on the center of your car tire valve with miniature screwdriver. Yes, the air comes out. This is what Keely did when demonstrating the LIBERATOR machine. When turning the left valve, the interior rod moves inward and pushes open the "OTHER" valve on the end of the ENCLOSED interior previous charged pressure tank. (The pretest water had just flowed in between the tanks.) At about the same time, Keely releases the noisy timer , to cover up the air escaping and filling up  in between the tanks. He then proceeded over to the lever machine to show REAL pressure of 23,000lbs. The mounted tank was said to have extra thick steel walls and was why it was extremely heavy, and needed to be permanently attached with strong bracket. ( It was really two steel tanks, and heavy.) At that point of my discovering this method of fraud, and then a couple others, I discontinued the research into Mr. Keely's actions.  Read online ,or purchase photocopies The New York Journal And Advertiser, Jan 29, 1899 pages 12,13, 19. Keely, The Monumental Fraud Of The Century. (Really cool stuff)

This 2008 video is an offshute of the earlier car engine conversions of spark plug electrolysis in Georgia . It is assumed, the same method was applied to the lawn mower engine here. The plastic tub holds the electromagnet putting field over plug wire to retard the timing , for use with hydrogen gas. Also in there is the dual coils relay for use with ONLY the interior 2 inductors and center diode. It is believed the vacuum sealed relay forms RADIANT ENERGY that benefits the firings. RE not escape because of no air, and follows wiring to plug. The 110v AC to Dc inverter (hidden in cement block rear hole )  sends 7 amps peak current as 2.6 amps average DC current through zero ohm Champion spark plug just AFTER the high volts pulse. The Magnetron trigger is aided by an added on, put together Patent TESLA small circuit that increases the Primary back emf , so as the secondary side gives way higher ignition volts, about 24KV as needed. Lawn mower ignition needs to be boosted up as equal to auto ignition coils. Uneven engine runnings during 2005-2008, was allieviated when adding in EGR for the Nitrogen content. Also the Inverter box should be changed to pure sine wave type,- not modified square wave type as commonly used, as not good with inductive electrical loads. The Briggs engine was self running on the ring , dual alternators under flywheel. The small flashlight battery pack for the ignition modification only used milliamps. The large car battery was only there for the starter motor, then completely disconnected. Low power conversion, but 18HP shown as working. The green Inverter in video background was broken and just left in video. That's why man said inverter WAS NOT USED in the 2 successful 2008 You Tube videos of S1R9A9M9.

Man in Georgia converted his 1978 El camino to hydrogen in 2006,, from spark plug electrolysis of water. 400watt inverter to DC with coils inductors for magnetic field applied. . Lots of data on Internet.  7amp peak current as 2.8A average DC current. Timing was changed to - late 34 degrees, as used with Hydrogen gas. He said about 18 MPG on water only. S1R9A9M9 Nathren , and his son, also Nathren worked on it for 8 months to get it running proper. The EGR greatly smoothed out the rough engine running due to the inherent Nitrogen content of exhaust gases applied.  2rnd battery ran the conversion and car alternator charged both batteries with isolater box. They ran it for about 3000 miles and locked it up in security garage. Wife divorced him, and took her car back with the conversion parts removed. He then had Chevy V8 engine on engine stand running large generator for free electricity on water line for work shop. The carb had the main jet redrilled about 33% larger hole to accommodate higher flow of  water.

Long time back I made failed sketches regarding diodes protection of the 400watt inverter that was used on cars conversions in Georgia back at that time. It BUGGED me that the system still was working after 3000 miles on chevy car. My recent clipboard drawing with erasable ink pen finally showed promise. I knew that somewhat similar set up had to be on the Briggs engine. Here goes, — The starter position of car switch provides 12v to ignition coils. The plug gets full 12kv high voltage. When key running on position , there is a ballast resistor that puts 6volts worth of high volts to ignition coil. Now it's 6kv firings. The Georgia schematic on the web forum Waterfuel1978 @ yahoo.com showed only 4 possible diodes  following the Inverter box. That's about 3200V protection. Two in the bridge rectifier1kv + 1KV and 2 in the circuit as 600v each. The bridge rectifier was not shown. Remember , the technicians in shop did all the work, Nathren father was non tech person. He only looked over the shoulders, and reported work done on the web forum. As resistors add up in series to input resistance, spark gap voltages in series also add up to beginning high volts. The Chevy distributor with spinning ROTOR provides the first of 2 spark gaps. Loss of 3KV. So now, there is about 3, 200volts protection against 3000volts ignition final voltage.

The second thing going on is that the spark plug AIR gap of high resistance is way reduced , due to the water solution. Less high volts is needed to fire plug. The project viability has gotten way higher.

The main energy power supply as DC is required to be minimum 100v or higher. Some guys used Inverter 220vAC . This is because the volts input to water solution is WAY REDUCED by 10 TIMES.  220 becomes 22. 110v becomes 11volts. DC low efficiency electrolysis requires minimum 3 volts. (You can't use 12v going to 1.2volts through water solution) Current gets reduced way down , due to tiny electrodes specific area of plug tips. (Not large plates in water)

Zero ohms resistance Champion / Autolite spark plugs pass the needed DC current as 2.8A average or 7 Peak Amps on scope. The circuit resistance by OHMS LAW requires volts divided by certain resistance to get the 2.8A.  The circuit diodes reduce the 110v down, then the 10 to 1 reduction , so as the Relay resistance in series was total 1.2ohms at about 10volts DC. Close enough as 8A, to make Hydrogen gas and electrified water vapor pressure.

The system had been enhanced by the radiant energy provided inside the relay by the high volts against the diode cathode inbetween the 2 inductors. The R.E. Follows the wires to spark plug. One set double coils relay per cylinder. That's why they were called the "Magic relays".

I refer you to the old documents you read long ago about the Engineer            
Who had US Patent on the added diode circuit to regular CDI ignition booster. The main discharge capacitor has LEFT OVER volts charge after firing Primary, so as the balance 50v 1ufd cap discharges balance through microwave blocking diode to high volts side. The loud SNAP at spark plug was aided by what he called the front side Capacitor.  You can also use his Back Side Capacitor as high volts PF capacitor across the spark plug. The PF gets quick first charged by ignition, then when ignition then fires plug, the added energy makes extra snap energy , as called PLASMA SPARK.

I believe that the Briggs engine could have used the other, Tesla added circuit as ignition voltage booster, but would have needed way more diodes protection, that greatly reduced down the final needed low volts, to get the required amps.
I did recent bench tests regarding back side capacitors , liked the 20-30KV Blue ceramic capacitor as 1000PF- .01uf, - good snap noise. This design added is dangerous , and can put you on the floor. This simple method to get Plasma Spark that THEN allows the main amps to go through spark plugs. A plastic tube with insulated wire is necessary to safely discharge that capacitor before putting in the fingers.

As per my previous messages, I found out there are 3 types of Inverter boxes - square wave, modified square wave, and Pure Sine Wave types. Georgia people used the wrong one and always had uneven engines running, but they did run. Electrical loads that include 1.75mh +1.75mh, Inductors must have pure sine wave energy.

To further isolate the main Inverter box from shorting, on the Briggs engine, the return line at bottom of plug goes directly back to the Inverter terminal. The  battery neg, is not connected to engine casting. You don't want 2 grounds together, inverter secondary DC neg and battery neg.  The starter motor has the battery neg only. . The starter is electrically insulated from engine using nylon plastic bolts and washers. Lots of guys long time back had burned out their expensive Inverters with shorts. (neg to neg) Of course you can use 2 separate batteries.

The Hefferlin manuscript was regarding steam to Hydrogen engine using high frequency. Back in those days , 1921-late 1940's, high frequency + high volts was only related to the famous high voltage Tesla coil. The extreme vibrations antagonized the molecules to separate very easily in heated chamber. The modern Patent reference I cited was to show that a particular frequency was relied on to make efficient hydrogen fuel gases from steam. Samuel Leach, 1970's  Patents 4148701, 4113589,4247379,4272345   - special composite coating in chamber inside emits frequency when heated, extracts Hydrogen from steam.  Newsweek April 19, 11976, Sam Leach's box pg78.    
Tesla coils can be made/adjusted for many frequencies output. The secondary tube is made first and accurately adjusted. The Primary is then tuned to to that one. Early models were spark gap and capacitor type. Nowdays they can be solid state firings of secondary. Back issue Electronics magazines have articles.

The S1R9A9M9 Georgia engines conversions had used only AUTOLITE brand spark plugs with solid COPPER ROD center. There are many part numbers, but most are with internal resistor of about 4000 ohms. The AUTOLITE 295 is ZERO ohms, non resistor type,  and not iridium type, as recently tested by me. This validates the Georgia electrical circuit as being able to pass the 7 Peak AMPS Current through spark plugs.


My Horvath Patent power Supply as beefed up CDI for powering the separate water cell is just about done. Fans being hooked up to keep China Inverter transistors / heatsink cooled. The end transformer TR2 with battery cable needs to be wound. 30 turns #12 and 3 turns #6. Photo to be sent later. I have been ordering parts for the aquarium box cell, as fittings gauges, special switches, etc. The ABOVE photo is not the proper Horvath power supply, and was discontinued.  New one to be sent later.

Time to change the wall calander! Happy Holidays!

Amazing Stories v20n06 (1946 09.Ziff Davis)(cape1736)

Invention was kept away from the Germans during WW2. Short tech data published in Sept 1946 Amazing stories magazine, Now on internet free read. Frequency setting was 1.3Mhtz Tesla Coil high frequency, to break up molecules of water to fuel gases in chamber. Modern US Patent was interior coated heat exchanger with steam entry. The coating emits 1.3Mhtz . Engine runs on water to hydrogen. Waste exhaust heat boils water. "Burn water For Fuel"
The Hefferlin Manuscript.

The Meyer Hydrogen system is the 2nd best invention since bench top Cold Nuclear Fusion Heat. HHO is really H+ H+ O-- as charged ions. The gases have more energy total. The Patented Meyer term is Thermo explosive Energy. There were 3 systems. proof of design, concentric tubes water cell, and multi use injectors with electrons extraction, lasers and pulsed high voltage,- at pulse modulation. There are more gases evolved during off time as the on time. See Stan Meyer company manuals as,- Birth of new technology and Water Fuel Cell dealership Manual. Many Patents. The resultant true Meyer cell has thick, tiny, tiny, FOAM rising upwards in cell. All the rest is H2 with water vapor/ and or electrified water vapor large bubbles. Voltrolysis as >1500volts DC in plain distilled water.Deionized water was better. Two inductors resonated with cell and power source. Input secondary pulse through inductor to water capacitor as lowered volts to PREALIGN polarity water molecules. Then high volts DURING OFF TIME with vibration separates with no current, voltage pressure. Diodes make hydrogen, AC makes heated/ water. Very low wattage input. Overunity as huge efficiency. Power for engines and factory blast furnaces.

Horvath Tech Data CDI Circuits Build Project

Patent 3980053 Hydrogen generator was studied by me in great depth, quite a while back in time. I had way earlier, typed out a 2 page corrections of minor errors and items involved , not necessary to add to car engine to run on hydrogen fuel . Fuel upon demand, no gaseous storage. Things were printed as too expensive or particular design manufacturing, to detour others from duplicating the system. I was already aware of these common inclusions in a $5000 Patent. This letter is to inform you of my recent solution to a section of Patent data PROBLEM that was missed by me earlier, as wrong on purpose AND DESIGNED IN,,, the Patent literature. I am presently building the Horvath power supply and water cell tank for duplication, as an average individual would do it without the up above parts/ designs not necessary, but including all of the important EFFICIENCY IMPROVEMENTS as Horvath did. This project was on my mind for years, and quite similar to a beefed up Capacitor Discharge Ignition circuit, as used for powering an improved efficiency Electrolysis cell. He also took data from previous Patents of theirs and incorporated them in his design. It's really the amount of fuel generated, as important, not the efficiency of a power supply!

Here we go:-- A pulse energy into coil has a reverse flow back. The secondary side coil of a transformer follows the primary method the same, except for a ratio change of turns. There can be 60cycles or 20,000 cycles frequency or higher. In general, iron lamination sheets are used for lower frequency range, while Ferrite is on high frequency. As frequency increases, less turns are necessary on primary side. It makes it easier to wind on core. The Patent was correct in that on + off time is required for cell plates, as about/around 50% duty cycle, on the SECONDARY side transformer. This allows the stuck polarized ions to release and rise up in the solution. Horvath said to use iron laminations for TR2, as the secondary side receives what's called Positive Critical wave form as seen on Scope.( High frequency used on low frequency type core.) This was to get a DC pulse for making Hydrogen gas and WITHOUT A Rectifier which would cause loss of 40-70 Watts heat loss, reducing efficiency, the rectifier forward volts drop loss times the current 200amps. This inclusion in Patent was INTENTIONAL TO DETOUR others from duplicating the design as printed. NOT POSSIBLE to get sufficient Inductance in 100-300 microhenry range, at LOW turns, as necessary with the associated capacitor, on iron laminations!!! Ferrite must be used for way higher inductance on low number to turns to be able to get 10:1 ratio for 22A going to 200 A, with heavy battery cable low turns secondary side of transformer.
All you have to do is put fast recovery high frequency diodes/ rectifier on primary side to stop the reverse wave form, on both sides!

The secondary side sees the same thing!!! I have heatsink with flat pack diodes in parallel to increase current rating with balancing resistors .

My rectangle aquarium tank was made with 3/8" thick plates, expecting 6lbs pressure. Presently ordering #6 red cable and gauges. The separate plastic water safety container will have pop off rubber plug, microswitch shuts off oscillator capacitor ground connection, while the flashback arrestor is hand made with brass tube and brass wool, and SS screens. KOH is hard to find. Water solution is done by weight of water + chemical together, as 25% and 75% for total weight solution together.

Recent response came back back from GROKAI.COM assures I was right about how to get secondary side on + off DC properly for water cell with only diode on primary side of transformer TR2. Horvath Patent 3980053 is available to read on line.

Horvath Patents Allowed Engine To Run On Water Improved Electrolysis Efficiency


Horvath Math for designing circuit for hydrogen generator. requires PULSE PEAK input 900w  into primary for 22A 2uf capacitor discharges at 10khz. Use old style scientific calculator for the formulas. I just went through it.  Secondary of TR2 has output diode, so as there being an on + off time for release of stuck ions from plates. I have flat nickel coated plates with flutes, knurled, with N + S magnets on back side. 3/8" acrylic plastic box, 400mesh nylon screen. The capacitor is 2uf as induction stove range type polypropylene capacitor high frequency type. .P type Mosfet paralleled as used for high side switching direct to spark plug, or N channel for regular use as applied to water cell. POH is 25% solution by weight,  meaning gallon is from both together same time -not gallon input water. .22 pulse amps becomes 200 amps pulse to plates. fast recovery heavy amp rectifier, or Horvath just used 3/4" iron laminations block with no diode, to get the critical positive wave out. AS used on input high frequency . China sells open board 1000 watts inverter with option of 330v terminals (to get about 600watts necessary) Can wind your own 2 transformers TR1 + TR, and can build your own switching inverter. L primary = C / I squared divided by E squared.  I = E X square root C / L.   Magnet wire as gauge # circular mills = 750 X I X duty cycle from wires chart. 10khz, .09joules each, 22A peak pulses, 371uh primary, 900w, 2uf, #16 #18 gauge primary, 2gauge secondary, Ratio 10:1 turns, duty cycle secondary .5 , water cell receives 450 Watts. Peak amps or watts is not average watts over the period time, .etc.

Grok  AI

What is average power in watts drawn from input 12v battery required for 900 PEAK pulse watts CDI capacitor into transformer with secondary diode and 50% on time only? Does this assume 450 watts average power to load

To determine the average power drawn from a 12 V battery to supply a CDI (Capacitor Discharge Ignition) system delivering 900 W peak pulse power into a transformer’s primary, with a 50% duty cycle (50% on-time) and a diode on the secondary, and to confirm if this results in 450 W average power to the load, let’s analyze step-by-step.
Key Parameters:

   Peak pulse power: 900 W into the primary during the on-time.
   Duty cycle: 50% (0.5), meaning the power is delivered for half the cycle.
   Battery voltage: 12 V (DC).
   Secondary side: Has a diode, delivering power to the load.
   Question: What is the average power drawn from the battery, and does this setup imply 450 W average power to the load?

Step-by-Step Analysis:
1. Average Power to the Load:

   The primary side delivers 900 W peak power during the 50% on-time. For a pulse transformer, the instantaneous power delivered to the primary is transferred to the secondary (assuming an ideal transformer with no losses).
   The average power delivered to the load on the secondary side is:
   Pavg, load=Ppeak×D=900×0.5=450 W.
   Pavg, load​=Ppeak​×D=900×0.5=450W.
   This confirms that, in an ideal system, the average power delivered to the load is indeed 450 W. The presence of the diode ensures unidirectional current to the load during the on-time, consistent with the 50% duty cycle.

2. Average Power Drawn from the Battery:

   In an ideal system (100% efficiency), the average power drawn from the battery equals the average power delivered to the load, as energy is conserved. Thus:
   Pavg, input=Pavg, load=450 W.
   Pavg, input​=Pavg, load​=450W.
   The CDI system stores energy in a capacitor during the off-time and discharges it into the primary during the on-time to achieve the 900 W peak power. The battery supplies a steady average power to charge the capacitor over the full cycle.

3. Average Current from the 12 V Battery:

   The average power drawn from the battery is:
   Pavg, input=V×Iavg.
   Pavg, input​=V×Iavg​.
   Given V=12 VV=12V and Pavg, input=450 WPavg, input​=450W, the average current is:
   Iavg=Pavg, inputV=45012=37.5 A.
   Iavg​=VPavg, input​​=12450​=37.5A.
   Thus, the battery supplies an average current of 37.5 A to provide 450 W average power, which supports the 900 W peak power during the 50% on-time.

4. Real-World Considerations:

   Real CDI systems have losses due to:
       Transformer inefficiencies (core losses, winding resistance).
       Diode forward voltage drop (typically 0.7–1 V for a standard diode, reducing output power).
       Capacitor charging/discharging inefficiencies.
       Switching losses in the CDI circuit.
   Assuming a typical efficiency of 80% (a reasonable estimate for CDI systems), the input power required to deliver 450 W to the load is:
   Pavg, input=Pavg, loadη=4500.8=562.5 W.
   Pavg, input​=ηPavg, load​​=0.8450​=562.5W.
   The average current from the 12 V battery becomes:
   Iavg=562.512≈46.9 A.
   Iavg​=12562.5​≈46.9A.
   The average power drawn from the battery is thus approximately 562.5 W with 80% efficiency.

5. Answering the Questions:

   Average power drawn from the 12 V battery:
       Ideal case (100% efficiency): 450 W.
       Realistic case (80% efficiency): ~562.5 W.
   Does this assume 450 W average power to the load?:
       Yes, the setup with 900 W peak power and a 50% duty cycle inherently results in 450 W average power to the load in an ideal system, as calculated (900×0.5=450 W900×0.5=450W). This is consistent with the transformer transferring 900 W peak power during the on-time, averaging to 450 W over the full cycle.

Final Answer:

   The average power drawn from the 12 V battery is 450 W in an ideal system, or approximately 562.5 W with 80% efficiency (requiring ~46.9 A average current).
   Yes, this setup assumes 450 W average power delivered to the load, as the 900 W peak power with a 50% duty cycle naturally results in 450 W average power in an ideal transformer syste