Half a Rocket Nozzle

The Hydrogen Gas Gun is the middle of three things stacked on top of each other. A Resonant Cavity Assembly sits in water and makes gas. The gun sits on top of that and is supposed to strip the electrons off it and prime it with light. An Optical Thermal Lens sits on top of the gun and burns it — that one has an episode of its own.

It goes by two names in this archive because Meyer gave it two. In Electron Extraction Process the same process is "hereinafter called The Gas Resonant Cavity"; elsewhere it is the Hydrogen Gas Gun. Same hardware, two memos.

How it was built

The archive's replication project measured it off the estate photographs, scaled against the one part whose size is known outright — the LED board is one inch wide. See Photographic Estate Dimensional Analysis - 3D Printable Replication.

• Outer electrode — T-304 stainless, 7.00 in long, 1.000 in OD, 0.940 in ID, 0.030 in wall
• Four windows — 4.00 in long, cut at 90° apart, straight down the tube
• Acrylic lens — cast, 1.025 in ID over that 1.000 in tube: twelve and a half thou of daylight
• Light — four boards of sixteen LEDs, sixty-four in all, about 2.2 W electrical
• Inner electrode — the one dimension the photographs never gave up

The windows run along the tube. In a coaxial line the wall current is axial too, so a lengthwise slot barely disturbs it — that is how a slotted line is built, and it is the right way round whether or not he knew why.

"Resonant cavity" means the C in an LC circuit

A seven-inch coaxial line is a quarter-wave resonator — at 422 MHz, and he drives it from 1 Hz to "1 MHz and more". But he never claims a cavity mode, and the filing says so in plain words:

The overall circuit is characterized as a "resonant charging choke" circuit which is an inductor in series with a capacitor that produces a resonant circuit … Such a resonant charging choke is on each side of the capacitor.

His own Figure 20YIE draws exactly that: the gas resonant cavity 110 with a resonant charging choke either side and a variable one on the return. So the resonance is the choke's:

0.5 in rod in the 0.940 in bore, 7 in, in air	≈ 16 pF
his AU Example I inductor (100 t, 24 AWG, 1 in)	≈ 130 µH
together	≈ 3.5 MHz
with a few millihenries of choke instead	≈ 1 MHz

Which is the band his memo names.

The field, and the dimension nobody has

The usual arithmetic here borrows the water cell's 0.680 in bore and assumes air, gets 2.56 MV/m against air's 3.0, and reads it as a man sitting deliberately at 85 % of breakdown. With the gun's own bore and the gun's own gas:

the bore, measured	0.940 in
the inner electrode	not known
if a 0.5 in rod, at 5,000 V	1.25 MV/m at the rod
air breaks down at	3.0 — so 42 %
hydrogen at about	1.5–1.9 — so 66–83 %
oxygen at about	0.9 — already past it

In the 2:1 fuel mixture, 5,000 V is at or past breakdown. Which means a corona or glow at the rod rather than a quiet capacitor — and a discharge strips electrons for a living. Sweep the voltage in the real gas with a microammeter in series and find where the current stops being capacitive. See Breakdown voltage — Paschen's law.

The circuit nobody discusses

The second half of the machine's name — extraction — is a specific, buildable circuit, drawn and dated 1982. From Hydrogen Gas Gun Technology:

an ELECTRON EXTRACTOR GRID (51) is placed on top of and in space relationship to said Gas Resonant Cavity Structure … allows electron flow to pass to an electrical load (53) … being a typical light bulb or amp consuming device

driven on the opposite phase to the cavity: "As pulse train (61) is gated 'ON,' pulse train (62) is switched 'OFF'", via a variable alternate gate circuit. If electrons come out of the gas, the filament lights. Nobody in forty years has posted what it does.

He also makes a safety claim for it that has never been tested: "The Electron Extraction Process also prevents spark-ignition of said combustible gases travelling through said Gas Resonant Cavity since electron build-up is being prevented."

The three wave-guides

From Taper Resonant Capacitor (ERt), the cavity shapes, in his words:

(35a) Traveling Constant Electrical Voltage Wave by way of linear cylindrical resonant cavity … (35b) Traveling Compressional (concentrating electrical intensity) Electrical Voltage Wave by way of taper cylindrical resonant cavity … (35c) Traveling Expanding Electrical Voltage wave by way of non-linear cylindrical resonant cavity … or any voltage surface combination thereof, each resonant cavity design acting and functioning as a Voltage Wave-guide

A parallel duct, a converging one, and a diverging one. Join the second to the third with a throat between them and you have a de Laval nozzle — the only shape on that list that turns heat into force. He leaves the door open in six words: "or any voltage surface combination thereof".

And he drew the throat three times: Fig 5B of the filing (an "ADJUSTABLE NOZZLE" with a throat ahead of the flame), Fig 6-2 (the cavity tapering to a "NOZZLE PORT"), and Fig 1-22 in Rocket Propulsion ("ENERGY ACCUMULATOR AND NOZZLE", a converging cone). He even drew the expansion — Fig 8-6 shows the flared cavity as an expanding space with "negative dynamic gas pressure", which is the diverging half's physics in his own vocabulary. What nobody has ever drawn is the two as one profile.

Heat is not force — and force needs pressure first

A heat yield is joules and comes out of the chemistry. A force yield is momentum per second, and it exists only if something turns that heat into flow going one way. In Process and apparatus for the production of fuel gas and the enhanced release of thermal energy from such gas he calls the lens assembly a "thrust-nozzle"; the word is right there.

converging duct alone	~14 % of the heat as motion
open out 5:1 after the throat	~60 %
14:1	~73 %, ≈ 2,900 m/s, Isp ≈ 300 s

But those percentages assume a chamber pressure the machine does not have. 5:1 at γ ≈ 1.2 needs a chamber-to-exit ratio of about 31 atmospheres; 14:1 about 100. Meyer's gas feed is regulated, in Hydrogen Gas Fuel & Management System For An Internal Combustion Engine, to "optimally 15 psi" — 2.0 atm absolute, barely enough to choke a bare throat (the critical ratio is 1.77).

At 2 atm a 5:1 bell is grossly over-expanded: the flow separates off the wall and gives less than the plain throat. The missing half is not the shape. It is the combustor and its pressure — and that is the part none of the memos has.

The numbers that do not work

• 115 cc/min at one amp (Resonant Cavity Mode of Operability). Faraday allows 10.45 — an overclaim of 11×. (Same caveat as the rest of the strand: his "one amp" is what he says leaks, not what the supply delivers, so as a verdict it is circular. As a measurement it is the whole game.) See Gas yield and Faraday efficiency.
• 2.5 million barrels of oil per gallon of water (In Summation) is 4.5 per cent of E = mc² — about 301 million times what the chemistry gives.
• The light. His own spec, from the Gas Processor equations: 20 mA at 1.7 V per diode, 64 of them, 2.2 W. A red photon carries 1.88 eV; ionising molecular hydrogen takes 15.43 (80 nm), oxygen 12.07 (103 nm). The best deep-UV LED ever made reaches ~210 nm. And for this gas even near-UV does nothing: H₂, O₂ and N₂ are homonuclear — no dipole, so no vibrational absorption at all and essentially none in the visible. The one thing light does here is make ozone, below 242 nm. (Don Gabel, 2008, holding the hardware: "all these LED lights on it, which he calls laser energy".)
• "Several gigavolts (billionth of a volt)". A gigavolt is a billion volts — the prefix is upside down by eighteen orders of magnitude. He is reading a ladder that counts up through nano, micro, milli, and has grabbed the wrong end of the list.
• "the dielectric value of water … 78.54 ohms" — dimensionless, not a resistance. Though he got it right in the patent, with the citation: "natural water is a liquid which has a dielectric constant of 78.54 at 20 °C", footnoted to the Handbook of Chemistry and Physics, 68th ed. The "ohms" is a slip in one memo. Read the patent first.

The shield, and two memos that disagree

The shield "superheats beyond 3,000 °F", and oxyhydrogen burns at 5,072 — so it is usually called a heat sink. But the gas arriving at it has not burned yet: by his own description it comes through the quenching circuit as unburnt primed gas, near ambient. A wall at 3,000 °F radiating into cold gas is a preheater and a hot-surface igniter, which is what the drawing shows.

The real problem is material. 3,000 °F is 1,650 °C; 304 stainless melts at about 1,420. And Quenching Nozzle says the disc is ceramic and "remains cool" because the flame is projected away from it. The record disagrees with itself, and a thermocouple on the back of it would settle both questions.

Related: He Chose the Losses, Eight Times Too Fast.