US 4,455,826, Thermodynamic machine and method, names Stellan Knoos as inventor and AGA AB of Sweden as original assignee. It describes a regenerative gas machine in which the two ends of the regenerator are pushed hotter than the hot space and colder than the cold space, a technique the patent calls "pressure ratio intensification" (PRI). It belongs in this dossier as background: nearly two decades before Kenneth Rauen's patent, it describes a regenerator end running colder than the engine's own cold space, and it does so while placing its performance inside the Carnot limit. Like any patent, it records what the inventor proposed, not a tested result.
The document and its history
The record gives application number US06/406,257, filed on 1982-08-09 and granted on 1984-06-26, with 50 claims. Knoos assigned the application to AGA Aktiebolag (effective 1982-07-26), and AGA assigned the patent back to Knoos, then at an address in La Jolla, California, with an effective date of 1984-09-28.
The patent was short-lived. The legal events list a maintenance fee reminder mailed on 1988-01-26 and a lapse for failure to pay maintenance fees on 1988-06-26; an entry dated 1988-09-13 records the expiry with an effective date of 1987-06-26. The listing labels the status as expired for fee-related reasons and notes that this status is an assumption, not a legal conclusion.
Three earlier patents are cited: US3115014 (Little Inc A, 1963), Knoos's own US3698182 (1972) on a hot gas engine or gas refrigeration machine, and US4389844, a two-stage Stirling engine from Mechanical Technology Incorporated. The classifications place it among closed-cycle hot-gas plants with combined cycles of the Vuilleumier type.
The problem Knoos set out to solve
The specification divides regenerative machines into two families. Stirling machines build large pressure differences across a power piston and deliver work. Constant-volume machines, of which the Vuilleumier device is the stated example, only shuttle gas between chambers at different temperatures and act as heat pumps. Knoos argues that the second family is attractive because, with almost no pressure difference across its moving parts, it avoids the sealing and loading problems of Stirling machines and can be built very large.
The weakness, according to the patent, is that such machines depend heavily on the ratio of absolute temperatures across the regenerator, on dead volume and on regenerator quality. Below a ratio of about 2 to 1 the heat handled per unit of machine is described as usually too low to be acceptable. The patent puts the pressure swing of a conventional machine at a ratio of typically 1.1 or less when the temperature ratio is 1.7 or less. That rules out the moderate temperatures available from solar collectors, waste heat and ambient sources, which are the applications the patent has in view.
How pressure ratio intensification is said to work
The main example (FIG. 1) is a solar-driven refrigeration and heat pump unit. One housing holds a hot chamber, a cold chamber and an intermediate chamber, with a hot displacer and a cold displacer cycling gas through a regenerator that connects to all three.
Knoos adds two things. The regenerator is lengthened at each end by a "superheating" and a "supercooling" extension, and those extensions open into the two ends of a separate compressor/expander chamber, labelled hh and cc, containing a double-ended displacer with its own drive. Because the displacer is double-ended, the two volumes always change exactly 180° out of phase, and the machine as a whole stays at constant volume.
The timing is the essence of the method. The added displacer sends gas into the cc end when system pressure is at its minimum and into the hh end when pressure is at its maximum. The patent says the phase is chosen so that the pressure-volume loop in hh heats that chamber above the hot chamber, while the loop in cc cools that chamber below the cold chamber. FIG. 5 shows the two loops as small and roughly mirror images: a counterclockwise one for superheating, with a negative pressure-volume integral, and a clockwise one for supercooling.
The effect is therefore paid for. The specification attributes the heating of the hot end to cyclic net compression work done on gas in the superheating chamber, and it describes the drive work as small compared with the heat loads. At steady state, it says, the net superheating and supercooling only make up for thermal losses in the extensions. The claimed benefit is indirect: a larger temperature ratio along the whole regenerator makes regenerator imperfection matter less and raises the apparent pressure ratio from 1.1 to between 1.2 and 1.3 to 1, which enlarges the area of the pressure-volume diagram and so the output per cycle.
Numbers and variants in the specification
- FIG. 1 temperatures: hot input at 230° to 290° C from parabolic tracking solar collectors, heat rejected at the intermediate level at 40° to 60° C, and a refrigeration load at -10° to +30° C.
- Regenerator extensions: preferably above 99% thermal efficiency, obtained for example by stacking fine mesh screens with typical diameters of 0.02 to 0.08 mm.
- Inverse heat pump (FIG. 7): input at 40° to 60° C from flat plate collectors, a hot output lifted to approximately 95° C for water heating, and a combined reciprocating displacer-regenerator made of four regenerator sections.
- Thermal transformers (FIGS. 8 to 10): local units drawing on a 150° C superheated-water line and an ambient source, or pairs of units shifting heat to a temperature better suited to long-distance transmission.
- Other hardware (FIG. 11): separate PRI displacers in line with the main displacers, and a radial-flow regenerator with ceramic walls.
- Stirling engine (FIG. 12): the same extensions and displacer added to a single-cylinder engine fed by a source at 300° C or lower, where the patent says the pressure-volume diagram is otherwise too thin to be useful.
What the patent does and does not establish
The patent stays within conventional thermodynamics. Coefficients of performance above 1.0 appear only in the ordinary heat pump sense, and the thermal gain of its transformers is described as "limited by Carnot-process values". For the Stirling variant it says only that overall efficiency "could" also improve.
It reports no test of any of this. The temperature ranges are given as examples, the pressure ratios as typical values, and the figures as diagrammatic or qualitative. No built machine, measured coefficient of performance or measured power density is described.
How it bears on the Rauen dossier
Rauen's patent, summarised on US Patent 6,698,200 (2004): Rauen's Efficiency Thermodynamic Engine, the Proell Effect and Proposed Refinements to the Second Law, also centres on constant-volume displacement of gas through a regenerator and also asserts gas colder than the regenerator's cold end, which it calls self-refrigeration. The two documents account for that cooling differently. Knoos produces it with an additional driven displacer and books the work that displacer does. Rauen's patent attributes it to the Proell Effect within a single displacer stroke and builds on it the argument, examined on Superclassical Heat Engine Claim, that the Carnot efficiency can be exceeded.
The Knoos patent is therefore useful context, not support. It shows that a regenerator end colder than the cold space was an engineering proposal in 1982 with no claim to exceed Carnot, and it says nothing about whether the engine on Rauen Superclassical Heat Engine behaves as claimed. The later documents listed as citing US 4,455,826 are mostly solar power, heat pump and Stirling patents; US 6,698,200 is not among them.
Related
- US Patent 6,698,200 (2004): Rauen's Efficiency Thermodynamic Engine, the Proell Effect and Proposed Refinements to the Second Law
- Superclassical Heat Engine Claim
- Rauen Superclassical Heat Engine
- Rauen
- Rauen Claims vs. Evidence: An Archive Dossier Without Test Data
Source notes & attribution
- https://patents.google.com/patent/US4455826