HEAT TRANSFER DEVICE

The invention relates to a device for transferring heat and a method of controlling such a device, the device comprising a stator chamber containing: a liquid; an input heat exchange surface; an output heat exchanger; and a rotor arranged to be rotated by vapour bubbles.

Skip to: Description  ·  Claims  · Patent History  ·  Patent History
Description
FIELD OF THE INVENTION

The invention relates to a device and method for transferring heat.

BACKGROUND

A thermosyphon is a heat transfer device in which heat is inputted at the bottom of a pipe or loop, and heat is extracted at the top, so as to drive circulation of a liquid which transfers heat from the bottom to the top. Variations include two phase thermosyphons (in which the liquid boils at the heat input end and the rising bubbles of vapour help drive the liquid convection), and actively pumped thermosyphons (which have a pump in the circuit). The latter is familiar as a central heating system. A passive thermosyphon requires gravity to operate, and unless it has pumped assistance the output heat exchanger must be above the input.

A heat pipe is often described as a two phase thermosyphon where liquid in a tube boils, thereby removing local heat, and the vapour condenses some distance away, giving up its heat. The liquid condensate flows back to the input end by gravity and or capillary means through very narrow passages. Capillary flow frees the heat pipe from requiring gravity to operate, so that a heat pipe can operate with the input higher then the output, and a heat pipe will operate in zero gravity in space. However, heat pipes are limited in size and can dry out (leading to a loss of capillary action) if they get too hot.

Thermosyphons and heat pipes are generally designed to operate at low internal pressures because this reduces the boiling point of the liquid employed. However thermosyphons can be operated as systems open to the atmosphere and with many different combinations of liquids and pressures according to the application and boiling point required.

These technologies are of great interest because heat density in computer chips, photovoltaics, IGBTs and LEDs is rapidly increasing, posing heat removal problems. Two phase, or boiling, heat transfer is up to 1000 times faster than heat transfer through copper, which is the best performing easily available material. However, in two phase transfer, as the rate of heat input rises, Critical Heat Flux is reached when the production of bubbles becomes so general over the input surface that the bubbles coalesce to form an insulating blanket on the surface and heat transfer drops dramatically. This is a problem that limits the maximum performance of all boiling heat transfer devices including heat pipes and two phase thermosyphons. The situation can be improved by pumping liquid across the heat input surface or by liquid jets impinging on the heat input surface. These techniques force liquid on to the surface and dislodge the bubbles. However, pumped solutions add cost, complexity and noise as well as requiring external power.

The present invention improves boiling heat transfer in heat pipes and thermosyphons. Thus, thermosyphons can be used in applications where previously only heat pipes were used.

SUMMARY OF THE INVENTION

The invention is defined in the appended claims.

By way of introduction, devices according to embodiments of the invention are self-powered by the lifting force of bubbles from boiling. The liquid in which a bubble is immersed exerts a force on the bubble equal to the weight of liquid displaced. Because the gas in a submerged bubble is at equal pressure to the surrounding liquid it will exert the same force on a surface that prevents it rising as the surrounding liquid exerts on it. This enables work to be extracted from bubbles formed by boiling a liquid. The energy to boil the liquid may come from any suitable heat source such as fluids that are required to be cooled, solar and other radiation, waste heat, and chemical combination or disassociation.

In operation, the rotor, which may be of any suitable shapes and dimensions including conveyors rotating round two or more axes, is powered by the buoyancy force of the bubbles themselves, and, optionally, by the power that can be extracted from bubble growth. Devices can be optimised for transferring heat or generating power or a combination, depending on the requirement of the application.

A further advantage of the embodiments, described in more detail below, is that more power is produced than is required to operate the device (i.e. more than is required to cause rotation of the rotor). Such excess power can be extracted mechanically from the rotor and used to drive mechanical devices such as fans (for example, a fan can be driven by the rotor and used to force air flow through the condenser), or can be used to drive a generator so as to produce electrical energy. This enables the device to fulfil requirements in the field of recycling waste heat, harnessing low grade heat, solar powered electrical generation and improving the efficiency of engines.

As will be described by reference to the Figures, embodiments of the invention firstly provide a means to increase thermal transfer in thermosyphons by using a self-powered rotor to increase turbulence and to scrape or sweep, without surface contact, the vapour bubbles created by boiling heat transfer off the input surface as the bubbles form.

Secondly, there is provided a means of power generation by harnessing the buoyancy of bubbles and/or pressure increase from the formation of bubbles. Mechanical power take-off can also be used to drive a fan for the forced circulation of air through the fins of the condenser heat-exchanger and/or a heat source.

Thirdly, there is provided an improved method of extracting work from vapour bubble buoyancy in a liquid, using an Archimedes screw. Embodiments include improvements to Archimedes screws, including radially and axially tilted vanes which increase volumetric capacity, and blocking members which enable a reduction in the size of the central core thereby increasing volumetric capacity.

DESCRIPTION OF FIGURES

FIG. 1a shows a heat transfer device according to an embodiment of the invention.

FIG. 1b shows a rotor for a heat exchanger, the rotor having one-way liquid jet valves.

FIG. 1c shows an improved rotor having curved/spiral vanes.

FIG. 1d shows a disc rotor arranged to rotate close to a side wall of a stator chamber.

FIG. 2 is a detail view of a rotor and stator housing for a heat exchanger, incorporating ducts for bubble-propulsion of liquid.

FIG. 3 shows a heat transfer device arranged to cool a 3D semiconductor chip and having two horizontal-axis rotors.

FIG. 4a shows a heat transfer device having a vertical axis rotor propelled by liquid jets.

FIG. 4b is a top-down view of the heat transfer device of FIG. 4a.

FIG. 5a shows a heat transfer device employing a vertical axis screw-type rotor

FIG. 5b shows a heat transfer device employing an inclined Archimedes screw as a rotor.

FIG. 6a shows a vertical-axis rotor having a central passageway for the passage of fluids.

FIG. 6b shows a cross-section view through the rotor and housing of FIG. 6a.

FIG. 7 shows a screw-type rotor device having a tapered rotor.

FIG. 8a shows a pressure regulating bellows-type device for regulating the pressure within a heat transfer device

FIG. 8b shows an alternative pressure regulating device having a spring and piston.

FIG. 9 shows an improved Archimedes screw having blocking members so as to allow a reduction in the size of the central core of the screw.

FIG. 10 shows a conventional Archimedes screw vane,

FIG. 11 shows the vanes of an Archimedes screw, the vanes having a radial skew.

FIG. 12 shows the vanes of an Archimedes screw, the vanes having an axial skew.

FIGS. 13 to 16 show an Archimedes screw having radially skewed vanes.

FIGS. 17 to 20 show an Archimedes screw having axially skewed vanes.

DETAILED DESCRIPTION

There follow, by way of illustration, various embodiments and features of the invention. These are not intended to define the invention, which is instead defined in the appended claims. It will be clear, however, to persons skilled in the art, that numerous combinations and variations can be applied in particular fields without departing from the claimed invention.

As shown in FIG. 1a, a first horizontal rotational axis embodiment of a heat transfer device 100 that is optimised for transferring heat from hot spots in electronic devices comprises a stator chamber 110 containing a closely fitting rotor 120 and a suitable charge of liquid. A stator duct 130 is provided on the upwardly moving (in use) side of the rotor and is fluidly coupled to a heat exchanger 140 which may be remote from the rotor. A second duct 150 fluidly couples (or connects) the heat exchanger 140 to the downwardly moving (in use) side of the rotor. The rotor periphery is provided with cells 160 (also termed “compartments” or “pockets”), which, for example, may be created by radial vanes 165.

In operation, an input heat exchange surface 170 at or near the bottom of the device is bonded to an element or heat source 175 (for example, a computer chip) which is required to be cooled. A suitable pressure and fluid charge is established in the device 100, and heat is transferred from the element to be cooled, via the input heat exchange surface 170, to the liquid in the device 100. The input heat from the element to be cooled causes the liquid inside the stator chamber 110 to boil, creating bubbles, which are trapped in the rotor cells 160, which rise due to gravity, and whose upward buoyancy rotates the rotor 120. The bubbles then escape from the rotor cells 160 after their surrounding rotor cell 160 has rotated to an upwardly facing position, and the bubbles then rise up the upward duct 130 to the heat exchanger 140 where heat carried in the vapour bubbles is transferred to the heat exchanger 140 and out of the stator chamber 110, and the bubbles condense into liquid. Condensed liquid descends from the heat exchanger 140 under the action of gravity, is carried by the downward duct 150, and continuously charges the rotor cells 160 with cooled liquid which is denser than the heated liquid and vapour bubbles in the upward duct 130. The downward thrust adds further impetus to the rotor.

Simple horizontal axis versions of the invention (where the rotor axis is horizontal, or substantially horizontal, as shown in FIG. 1a) comprise a rotor 120, of any suitable length, with projecting vanes 165 that create cells 160 which are arranged around the circumference of the rotor 120. As a cell 160 moves over a heat input surface (or input heat exchange surface) 170, the vane 165 dislodges (or “scrapes”) bubbles from the heat input surface 170 and the cell 160 is gradually filled with bubbles, imparting a lifting force, due to buoyancy, on the rotor 120. As the rotor turns further, such that the rotor vane 165 approaches the horizontal orientation, the leverage of the bubbles exerts an increasing rotational force on the rotor. Rotor work surplus to the scraping effect is generated and can be applied to driving a connected pump, enabling the condenser 140 to be placed below the evaporator (or heat exchange surface 170), or applied to generating electrical power or any other device or process requiring rotary power. Thus a thermosyphon according to embodiments of the invention can be used in applications which were previously the preserve of heat pipes where the condenser 140 is below the evaporator 170. The scraping (or sweeping, or brushing) of bubbles from the surface of the input heat exchange surface 170 assists with removal of the bubbles from the heat exchange surface 170, thereby avoiding the possibility of a blanket of insulating vapour bubbles forming. Thus, such a 2-phase cooling device can be operated over a wider range of heat input rates (“extending the curve”), and undesirable effects such as “kettling” are avoided or reduced.

In other embodiments, as shown in FIG. 1c, the rotor 120 can incorporate curved or spiral vanes 1065, which curve such that the outer tip 1066 of each vane 1065 is rotationally retarded with respect to the radially inner part 1067 of the vane 1065. This arrangement allows bubbles 181 to be trapped within the cells 160 for longer (in terms of rotational procession of the rotor), being released only after or near the point where the vane outer tip 1066 passes the rotor 120 centreline 185 (on the non-driving side), such that the bubbles impart greater work to the rotor 120. FIG. 1d shows an embodiment where a rotor 120 is arranged to rotate close to a side wall of the stator chamber 110, with cells 160 passing over a heat exchange surface 170 (also termed an evaporator), and the stator chamber wall optimally arranged at an angle of about 20 degrees to horizontal.

Suitable bearings on which the rotor 120 rotates are plastic, stainless steel or glass ball or roller bearings. The rotor and stator can be made from metals such as aluminium, or alternatively from plastics or ceramics or any suitable material having similar characteristics. The evaporator is preferably made from a thermally conductive material such as copper or aluminium. The condenser is arranged with fins made of a thermally conductive material such as copper or aluminium.

In a prototype of an embodiment it has been found that at high rotor speeds there is insufficient time for all the bubbles to leave a cell as it approaches top dead centre and that a large upward duct opening and a suitable cell depth are required to give time for all the bubbles to leave. A large size of the duct opening is preferable for assisting bubbles to leave the cells 160, but duct opening size may be balanced against the desirability of avoiding the kinetic energy of down-flowing liquid being dissipated by leakage across to the upward duct 130 (it has been found that this can be substantially avoided by keeping at all times at least two full cell widths of stator wall between the upward duct 130 and the downward duct 150).

Additionally, in embodiments, it has been observed that some bubbles stick to the rotor. This is advantageous when bubbles stick to the axial ends of the rotor, as it reduces liquid drag between the ends of the rotor and stator wall. However, it is not advantageous when bubbles stick within a cell, and are carried past the centreline 185 of the rotor, such that they are carried over into the next cycle (into the downwardly moving half of the rotor). To maximise efficiency, therefore, the rotor may have a bubble-repellant surface or coating within the cells 160, and a vapour or bubble attracting surface or coating at the axial ends of the rotor vanes 165. The ends may also be provided with spiral grooves to assist in maintaining a layer of gas between the rotor 120 and the end wall of the stator chamber 110 (also termed “housing”).

In other embodiments, additional features described below such as non-return valves 180 may be added to increase torque. In simple horizontal devices, as shown in FIG. 1a, the liquid displaced by the pressure of growing bubbles 181 escapes the cell 160 in all directions through the gaps between the rotor 120 and casing 110. In more highly engineered and tight fitting devices this liquid can be channelled through non-return valves 180 as shown in Figure 1b, or flow biasing ducts 230 (as shown in FIG. 2) as jets of liquid 185, as described above.

Additional power, beyond the buoyancy force, that can be extracted from bubble growth in devices according to the invention, arises when a discrete cell on a rotor passes a heat input surface. When the fluid within the cell is heated, vapour forms and pressure within the cell rises (since the rotor is arranged to fit closely within the stator housing such that fluid flow from each cell is controlled and optimally kept to a minimum). This pressure rise can optionally be harnessed in two ways: first, by connecting each cell on a rotor to the following cell through non return valves 180 (as shown in FIG. 1b), or flow biasing ducts (diodic fluid valves); and second, by providing one or more ducts leading from a lower portion of a stator wall adjacent to the input heat source 175.

As shown in FIG. 1b, first, by connecting each cell 160 on a rotor to the following cell through non return valves 180, or flow biasing ducts (diodic fluid valves). A cell non-return valve 180 may be created by having a vane that is radially increasingly compliant in the trailing direction, or comprises a rigid vane of reduced radial length or width that supports a trailing compliant member so that the compliant member can only bend backwards. As the vapour bubbles accumulate in a cell 160 they cause the pressure in the cell to increase (at least some cells are shrouded, e.g. by the housing, also termed stator chamber, thus preventing pressure from leaking away without being directed through the valves). This pressure forces liquid through the valve 180 or duct into the following cells, creating a reaction force on the rotor and also displacing the rotor vis-a-vis the fluid mass surrounding it. Pressure can be released from an un-shrouded cell, away from the evaporator, thus avoiding circulation and equalisation of pressure around the rotor. This process allows the device to consistently start rotating in the designed direction and with heat input from directly below the rotor. A device relying on bubble lift alone may rely on heat input and boiling past the bottom dead centre of the rotor (i.e. to one side of the rotor centreline 185), or if (as shown in FIG. 1c) a rotor with curved vanes is employed, heat input can be provided directly below the rotor.

The second method by which the pressure rise in the cell can be harnessed is by providing one or more ducts 230 (as shown in FIG. 2) leading from a lower portion of a stator wall adjacent to the heat input 170 (also referred to as the evaporator, or boiler). Each duct is below the level of liquid in the cell, for the time which the cell 160 is open to the duct, and as pressure in the cell grows, liquid is forced into the duct and may be used for cooling or other purposes elsewhere. In particular, the pumped flow may be used in cooling computer chips with 3D architecture, where it is advantageous to pump fluid through micro channels within the chip to achieve the required heat removal.

In further embodiments, as shown in FIG. 2, ducts 230 in the stator wall, below the liquid level in a cell 160 (i.e. below the level at which bubbles collect in a cell 160), may be used to harness the pressure fluctuation as each cell 160 passes the ducts, so as to pump liquid through micro-channels in the target cooling area (element to be cooled 175). In embodiments, a plurality of smaller micro-channels lead away from each duct, such that the duct acts as a header ‘tank’ for the micro-channels. Because the fluid is liquid and pumping force fluctuates with the passing of each cell, the diameter of the duct 230 from the stator wall to the header entry to micro channels can be relatively large (compared with the micro-channels). The science of minimising pressure drop from the header to the micro-channels, and even (equal) flow through each micro-channel is non-trivial, and yet is important if even cooling of the device to be cooled is to be achieved. If header losses (pressure loss in the transition from the duct to the micro-channels) prove relatively high, and/or uneven distribution of flow between micro-channels proves problematic, multiple micro ducts may instead be provided (in some embodiments), so as to directly connect the cell 160 to the ducts and/or micro-channels within the element to be cooled (e.g. a chip) 175.

To maximise the efficiency of the pressure pumping feature, in an embodiment a layout (as shown in FIG. 3, which may be used, for example, for 3D chip cooling) has a vertically oriented multi-layer chip (or any other layered heat source which is to be cooled) 310 sandwiched between the heat input areas of two heat transfer devices 100 similar to that illustrated in FIG. 1a. Devices of this type take heat from the outer layers of the chip, but also pump liquid via a duct 330 under the chip 310 and up vertical micro channels 340 within the chip 310 to cool the inner layers. Such micro channels 340 may be of varying section and extended with a channel extension 350 to vent vapour and liquid at or near the top of the condensing heat exchanger 140. This venting of vapour can provide an additional pumping force as any bubbles expand. If heat is inputted from the chip 310 to one vertically divided side of each of the heat transfer devices 100 (e.g. to the right of the centreline 185 shown in FIG. 1a), it is advantageous to close the other vertically divided side of each of the cells 160 with a shroud or shrouds, and thereby reduce leakage of pressure that might otherwise be usefully directed to driving liquid through the micro channels 340. Other methods of utilising the pressure fluctuation include driving diaphragm pumps via the duct 230.

Embodiments having rotors 120 with cells 160 and non-return valves (otherwise known as jet valves) 180 can also have a vertical axis of rotation, as in the embodiment shown in FIG. 4. This means that the axis of rotation has at least a vertical component, as well as possibly also a horizontal component. In such embodiments, the top of the rotor 410 is closed such that cells 160 are shrouded on their upper side, and after rotating through a heat input zone 170 each cell 160 releases its bubbles to an upward duct 430 through a first port (or cut-out) 440 in the stator wall. Similarly, a second port (or cut-out) 450 in the stator wall connects the cell 160 to a liquid return downward duct 460. This type of embodiment can also, for example, be applied to a conveyor type rotor (not shown) that rotates round two or more pulleys and can move heat horizontally.

Additional work can be extracted from vertical axis celled rotors by stacking more rotors on top of the first to extract further work from the rising bubbles. In this case the rotors may be mechanically linked.

Vertical axis devices as shown in FIG. 4 have partially shrouded cells and the rotor is rotated by pressure rise causing jetting through valves into the following cell. Other vertical axis embodiments of the invention as shown in FIG. 5a (where the rotor axis is vertical, or substantially vertical) have a spiral rising duct and the rotor is rotated by the pressure of bubbles rising up the spiral.

In the embodiment shown in FIG. 5a, there is provided a substantially vertical stator chamber 510 containing a close fitting cylindrical rotor 520 or conical rotor and filled with a suitable fluid. Such embodiments are advantageous for use in situations where heat input is effected via relatively large surface areas, since heat can be inputted around the whole circumference of the rotor 520. The rotor 520 has one or more spiral strips (also termed “projections” 530 projecting from the rotor surface and extending from the bottom of the rotor to the top of the rotor. The strips 530 may be provided with turned down outer edges or seals 540 to retain bubbles under the spiral strip 530. The strips 530 may also be provided with one or more downwardly projecting fins 550 to catch and slow bubbles, thereby increasing the torque passed to the rotor 520, and also helping to prevent formation of long sausage-like bubbles through which vapour may rise too rapidly. A conical rotor can be arranged such that it has a larger diameter towards the top of the rotor, so that as bubbles rise through the rotor, their expansion is accommodated in the progressively larger volume of the rotor as its diameter increases with height. Thus, additional work can be extracted from the bubble expansion. Alternatively, instead of a conical screw, a parallel sided screw can be used but with a member of decreasing cross-section (in the direction of bubble rise) can be inserted into the space between the vanes, thus effectively providing increasing volume between the vanes in the direction of bubble rise, so as to provide for bubble expansion. Alternatively, multiple screws of increasing sizes can be stacked to allow for expansion, and these can be mechanically coupled.

There is further provided heat exchange means (e.g. a heat exchanger 140) in the stator chamber 510 wall, above the rotor, at or near the top of the rotor, by which means heat can be transferred outside of the heat transfer device 100. There is further provided one or more ducts 560 external to or within the stator chamber 510, or internal to the rotor 520, through which cooled fluid can return from the heat exchanger 140 to the bottom of the rotor 520.

In operation, the internal pressure in the device (100) is lowered by external pressure reducing means (such as an external vacuum pump) such that, when placed in contact with a heat source (e.g. a surface or fluid flow that is required to be cooled), the fluid in the device 100 boils at the heat interface (i.e. at the portion of the stator chamber 510 which is in contact with the heat source). The use of a vacuum lowers the temperature at which boiling takes place.

The heat interface is preferably as close to the base of the rotor as possible, but the device 100 can operate, albeit with reduced efficiency, if the heat interface extends partially up the rotor. The bubbles of vapour, which are produced by boiling of the fluid in the device, rise and are deflected by the spiral (or spirals) 530, imparting a turning force on the rotor 520. The bubbles rise up each spiral 530 until they encounter a downward projection 550 and are restrained from rising until following bubbles merge with them to create a bubble large enough for a portion to escape from under the projection 550.

A downwardly pointing conical rotor can be employed and has the advantage that any bubbles escaping between a spiral 530 and the stator wall 510 are trapped again by the spiral above, rather than moving vertically up the wall and so potentially avoiding being trapped by the spiral above. Such a conical rotor has a larger diameter towards the top of the rotor. A further advantage of such a conical rotor is that as bubbles rise through the rotor, their expansion is accommodated in the progressively larger volume of the rotor as its diameter increases with height. Furthermore, as diameter increases, so does torque exerted on the rotor by the bubbles.

Thus, additional work can be extracted from the bubble expansion. In other embodiments, multiple vertical-axis rotors are stacked on top of each other. In such embodiments, each successive rotor optionally has a larger diameter than the one below.

Another embodiment shown in FIG. 5b employs an Archimedes (Archimedean) screw as the rotor 520, inclined at an approximately 45 degree angle (although other angles between 0 degrees and 90 degrees could be used). This embodiment offers improved resistance to bubbles travelling unimpeded up the spiral, since in use the bubbles become trapped in compartments between the rotor vanes 530. The screw housing or “cylinder” 505 extends downwards so as to cover the area close to the heat source and thereby capture the majority of the vapour bubbles. The bubbles impart a force on the rotor, causing the rotor to turn, and allowing the bubbles to rise. The vapour bubbles then travel to the condenser 140, where they give up heat and condense. The condensate returns to the reservoir 510 via return pipe 560, the system being sealed such that the screw remains filled with liquid apart from the vapour bubbles rising through it.

Devices according to embodiments can be chained, with the heat exchanger (condenser) of a first device acting to provide heat to the heat input source (evaporator) of the next device in the chain. Each device can be arranged with appropriate liquid and internal pressure so as to maximise efficiency of each device, each device operating at a different temperature range. Thus, devices can be “compounded”. Multiple devices according to embodiments can be deployed in exhaust systems and boilers etc., in which embodiments it is advantageous if they share a heat exchanger 140.

All devices according to this invention can produce work (mechanical energy) which may be used within the device (for example for driving a pump to provide pressurised flow for hydraulic actuation) or transferred outside the device by mechanical or magnetic means. Also, for the highest heat transfer rates it may be necessary to provide additional power to the rotor to drive it at speeds that bubble growth alone cannot provide. In those embodiments it should be noticed that as long as there is gas in the cells on the upwardly moving side of the rotor, there will be a torque supplied by the liquid on the downward side, reducing the input work to the rotor. Furthermore, the greater the height of the liquid column the greater the torque supplied. The rotor can be used to power an electrical generator for generating electrical power, or a mechanical fan for forcing air flow through fins of the condenser and/or a heat source, among other uses. Cooling the condenser 140 can help to improve device efficiency.

In a yet further embodiment shown in FIG. 6, employing a vertical rotation axis rotor with a spiral, there is provided a substantially vertical stator housing 610 comprising an outer stator cylinder 615 and an inner stator cylinder 616 with end closures to define a hollow cylindrical chamber 620 in which a close fitting rotor 630 is rotate-ably mounted. The inner cylinder 616 of the stator housing 610 defines an inner cylindrical duct 640 for passage of a fluid to be cooled. Such an embodiment is thus optimised for cooling a flow of fluid internal to the rotor. The rotor 630 is provided with one or more spiral projections 650 to define spiral passages up the rotor 630. A duct 660 is provided to join, via a heat exchanger 670 (which can, for example, be external), a fluid inlet 680 at the lower end of the stator housing 610 with a fluid outlet 690 at the top end.

In operation, the stator housing 610 is filled with liquid at a suitable pressure, which may be externally controlled. Hot fluid moves through the inner duct 640 and heat is transferred from the hot fluid through the stator wall 610 to liquid in the stator chamber 620, causing it to boil and produce vapour bubbles. The bubbles rise and produce a lifting and turning force on the rotor 630, which rotates. Because the rotor is arranged to be close to the heat exchange surface, as the rotor 630 rotates, the spirals 650 brush, scrape or sweep the bubbles off the surface of the stator housing 610, to prevent insulation of the heat exchange surface (between the hot liquid in the inner duct 640 and the liquid in the stator chamber 620) by the bubbles. Rotation of the rotor causes pumping of fluid in a circuit via fluid outlet 690, the duct 660, and the heat exchanger 670 where the fluid is cooled and then returns to the stator liquid inlet 680. The spirals optionally have turned down edges and vapour restraining projections as described above. Further optional variations also have external spirals 655 to the rotor so that fluid flow past the exterior surface of the stator may also be cooled and thus provide a more compact cooling device.

Turning to FIG. 7, an embodiment adapted for operation in low gravity condition comprises a stator chamber 710 containing a close fitting rotor 720 and a quantity of fluid. The rotor comprises a first axial end section 730 with radially separated cells 740 at the periphery (around the circumference of the rotor first axial end 730). The cells 740 have one-way valves or flow biasing openings (as described above with reference to FIGS. 1 and 2) connecting each cell 740 to the next. A second axial end 750 of the rotor 720 (which, in embodiments, optionally has a tapering section) terminates in a heat exchange volume of the stator chamber 710, the heat exchange volume and stator chamber forming together a heat exchanger 760. The rotor 720 has an interior duct 770 connecting the second axial end 750 to the first axial end 730. The stator wall at the first axial end 730 has a cut-out or passageway 780, at least partially opposite the interior of one of the cells 740, to provide a connecting duct 790 between: the interior duct 770 of the rotor; and the cells 740 of the rotor (as rotation of the rotor 720 brings the cells 740 into position adjacent to the cut-out 780, either one or two cells will be fluidly coupled by the cut-out 780 to the rotor interior duct 770, depending on the exact orientation of the rotor 720). The second rotor section 750 has one or more spiral projections 705 on its surface that conform closely to the inner surface of the stator chamber 710. The projections 705 extend beyond the second axial end 750 to become scrapers (for scraping bubbles off the inner surface of the stator housing 710) in the heat exchanger 760. On the rotor 720, the spiral projections 705 define spiral ducts 7100 that lead from the stator heat exchanger 760 volume to a stator wall projection 7110 that sealingly separates the first section 730 of the rotor from the second section 750 of the rotor. The projection 7110 is provided with an opening 7120 further round in the direction of rotation than the cut-out 780 so that it will connect the spiral ducts 7100 with a cell 740 after the cell 740 has ceased to be in communication with the cut-out 780.

In operation, the internal pressure of the device is first set to a level suitable for the fluid employed and the temperatures of the source and sink. Heat transfer through the stator chamber wall 710 at a heat input area heats and boils liquid in the rotating cell 740 for the time being over the heat input area. The heat input area is in the region of the larger end of the rotor. The boiled off vapour is forced to the radially inner part of the cell 740 by pressure from the liquid which is centrifuged outward by the rotation of the cell. Rising vapour pressure from increased boil off in the cell forces a liquid jet through one way valves between the cells 740, or through flow biasing passages on the trailing edge of each cell 740. This produces a reaction thrust on the rotor 720. The passage of fluid into each following cell will increase the pressure in each following cell with similar but diminishing reaction effect. The reaction thrust on the rotor causes it to rotate. Rotation brings the cell under the cut out 780 and the still slightly pressurized vapour leaves the cell 740 and moves up the rotor interior duct 770 to the stator heat exchanger 760. Here the vapour is discharged against a stator vane 7130 which is set at an angle to assist in establishing a vortex of the same direction of rotation as the rotor 720 in the heat exchanger 760 volume, and the vapour is cooled and condenses.

One or more spiral scrapers 706 attached to the rotor 720 are used to gather liquid condensate into the rotor spiral ducts 7100. The scrapers 706 also have the effect of further encouraging a vortex in the heat exchanger 760 volume. The vortex moves liquid to the scraped inner surface of the stator housing 710 by centrifugal force.

If the heat exchange surface is tapered down in the axial direction away from the celled end of the rotor, the rotating vortex of vapour will act on the condensate on the tapering surface to drive it towards the rotor. This may be sufficient to allow the rotor spirals to be dispensed with unless it is desired to have the heat exchanger at some distance from the rotor. In this case surplus rotor work may be used to operate a fan to improve the vortex or be used to power a pump as described above.

Having entered a rotor spiral 7100 the liquid is moved to the other end of the spiral, which in some embodiments is shrouded, and the liquid is injected through the projection opening 7120 into the most recently vapour discharged cell 740 which has by now rotated under the opening 7120.

All the above devices may be operated at any suitable internal pressure to suit the target temperature and fluid selected, but will usually be designed for low internal pressure. A method of facilitating manufacture and providing both pressure control and increased surface for transferring heat out of the above-described type of thermosyphon is to use a bellows device as shown in FIG. 8a.

In such a bellows device 800, adapted to the present purpose of regulating pressure within the stator chamber 110, a rigid flat plate 810 is bonded to a thin sheet 820 that has concentric or other suitable ribbing to make the sheet flexible. A compression spring 830 (biasing member), which is bimetallic so as to change rate with temperature, and which is arranged between the plate and the sheet, in operation pushes the sheet 820 away from the plate 810, increasing the internal volume. The internal volume is connected to the internal volume of the stator chamber 110, so that the bellows device 800 can regulate the internal pressure of the stator chamber 110. The bellows device 800 may have a second ribbed sheet instead of the plate. In other embodiments, a second bi-metallic spring 840 external to the device is also used, to adjust the internal volume and thereby the pressure, according to ambient temperature. In other embodiments, other means such as an external screw 850, to compress the internal spring 830 or allow it to move out, can be employed for controlling the internal pressure.

For assembly, the internal spring is compressed by external force, reducing the internal volume, and air evacuated through a suitable duct. A valve in the duct allows retention of the vacuum until the device is attached to a heat transfer device (e.g. a thermosyphon) such as that described above. The heat transfer device is charged with liquid before use, and the bellows device is attached, inducing a vacuum in the heat transfer device, and sealing both devices. The spring 830 is released and the plate 810 and sheet 820 are forced apart by the spring, lowering the internal pressure. When used on a heat transfer device, such as a thermosyphon, this device produces the required vacuum in the thermosyphon, acts as a pressure controller and also as a heat exchanger with the ribs acting as fins. Thus the device can provide a self regulating fin type heat exchange surface of improved capacity.

Another pressure regulating device 805, attachable to the stator chamber 110, for regulating the internal pressure inside the stator chamber is shown in FIG. 8b. The regulator device 805 has a body 809, within which is a spring 808 and piston 811, the spring acting against the piston 811. A diaphragm seals the piston against the body 809 (although a sliding seal could be used instead of a diaphragm). In use, the spring acts against the piston to increase the space inside the body, thereby lowering the pressure. The spring is optionally bi-metallic so as to provide a temperature-dependent pressure. An opening 812 connects the interior of the body to the interior of the stator chamber 110, thereby communicating lowered pressure with the inside of the stator chamber 110. The spring pressure and body volume can be adjusted so as to provide the appropriate pressure inside the stator chamber 110.

The above-described pressure regulator devices 800,805 additionally accommodate the change in volume required within the stator chamber 110 when vapour bubbles grow (since vapour occupies greater volume than liquid).

As described, certain embodiments employ an Archimedes screw as the rotor. Archimedes screws have been used for lifting water and for power generation from low head water sources. In these devices force is produced by the lifting force of displacement of liquid and this is combined with the rotational speed to produce power. Such devices have a power to volume ratio of the same order as a large wind turbine tower, however it would be desirable for this ratio to be increased. It is a further purpose of this invention to increase the useful displacement of an Archimedes screw, to increase the lifting force.

Traditional water power Archimedes screws have been found by experiment and analysis (Chris Rorres-Journal of Hydraulic Engineering January 2000 pages 72-80) to have a maximum fill ratio of 60% of available volume when operating to lift water. This is less than half the total volume of the screw because the volume of the central cylindrical core is 25% of the total volume.

The following two factors have been considered for improvements to conventional Archimedes screws:

1) increasing the screw volume available to be filled with fluid (liquid and gas). This is restricted in optimised conventional Archimedes screws by the central cylinder to which the vanes are attached and which takes up about 25% of the cross-sectional area of the screw.

2) increasing the lifting force of the screw by increasing the ratio of gas to liquid within the Archimedes screw.

Considering the above factors, the volume available for fluid can be increased by decreasing the central core cross-sectional area. However, the central core surface acts to separate pockets of fluid or particulate solids, and enables them to be moved up the screw. Furthermore the reason the conventional inner cylindrical core is so large is that its diameter determines the level at which fluid overflows into the next compartment. Any reduction in cross-sectional area of the central core must be achieved without decreasing the ability to separate liquid and gas. It has been further realised that more of the total volume could be filled with fluid if the point at which the fluid escapes into the next compartment could be: lowered if operating on gas flowing upwards; or raised if operating on water flowing downwards. By way of illustration, in an Archimedes screw operating on gas flowing upwards, if a vertical barrier for preventing gas escaping from each compartment to the next is extended downwards, i.e. lowered, then the core size can be reduced without allowing gas to escape from one compartment to the next (which would otherwise result in a loss of torque). Similarly, in an Archimedes screw operating on water flowing downwards, if a vertical barrier for preventing water escaping from each compartment to the next is extended upwards, i.e. raised, then the core size can be reduced without allowing water to escape.

In order to solve this problem with reference to an Archimedes screw operating on ascending gas or vapour, it has been further realised that the liquid in each compartment is acting as a rolling valve between the gas packages as they rise up the screw. The size of this water valve is limited on the upside by spillage downwards as in a water screw pump, however the limit on it being small is firstly that it must prevent unimpeded gas flow upwards and that there must be room through the valve gap for the gas to circulate upwards as the screw turns. Gas can move rapidly and with low drag through a small gap, compared to fluid. Reducing the operational size of the water valve will also reduce any work required to lift the water valve to the top of the screw.

In the light of the above analysis, an improved Archimedes screw has been produced, as follows:

An embodiment of a screw architecture that fulfils the requirement of minimizing the water valve (central core) size comprises a tilted screw 900 with one or more flights of vanes 910 (of optionally narrow pitch), mounted on and around a central core 920 (which can be relatively small compared to conventional Archimedes screws), as shown in FIG. 9. Extending from the central core 920 there are provided blocking members 930, regularly spaced both radially around the core 920 and axially along the length of the core 920. These blocking members extend between each pair of vanes, reaching axially downwards towards the lower end of the screw 900 from the trailing side of each upper vane 910. Each blocking member 930 extends substantially radially outwards from the core 920, with the outer edge 940 of each blocking member 930 oriented at an angle that results in the outer edge 940 of that member 930 being substantially horizontal when that member 930 points radially downwards (the outer edge 940 of each blocking member 930 extends from near the tip 955 of the upper 950 of the two flights of vanes 950,960 between which it lies, towards the core 920 and towards a radially inward part 965 of the lower vane 960). In other words, the outer edge of each blocking member extends from a radially outer end 955 of the upper vane 950 of a pair of adjacent vanes, towards a radially inner position and towards the lower vane 960 of the pair, such that the radially outer edge 940 of the blocking member closes with a liquid surface 941 to isolate a gas portion 980 between adjacent vanes (when the blocking member is oriented maximally downwards in use). In use, the screw is inclined typically at 30 to 60 degrees from the horizontal.

In operation the screw 900 rotates and at any time at least one blocking member 930 outer edge 940 is extended into the surface of the liquid valve 970 (the liquid which is trapped between the two flights of vanes 950,960) so as to provide a seal, isolating a portion of gas 980, and thereby preventing gas escaping from one compartment (also termed “pocket” or “cell”) into the next compartment. The vane tips and/or blocking member outer edges (or tips) 940 are optionally angled radially or axially or otherwise shaped to reduce agitation as each member 930 enters the liquid 970. Clearly it is important (although not essential) to have the minimum number of vanes 910 that allows a desirably small liquid valve 970, without the blocking members 930 agitating the liquid surface so much that gas 980 can pass. An additional factor is that, because of the curvature of the cylinder 905, at the bottom of the cylinder in each compartment, a decrease in liquid depth brings a proportionally smaller decrease in surface over which the liquid rolls (and a corresponding decrease in drag). This means that for ever smaller decreases in drag, more blocking members 930 are required for sealing. A number of members between 6 and 12 per revolution has been found to provide the optimum for most liquids, although a greater or smaller number of members can be used, albeit with increased complexity or drag. In other embodiments, the vanes 910 are supported only by the blocking members 930 which are arranged radially around the rotational axis of the screw 900 and connect between each vane 910 such that a central core 920 is not required.

The above description concerns separation of gas portions in compartments between vanes. However, by reversing certain aspects, the design can be adapted in other embodiments to separate portions of liquid instead. For example, if instead the outer edge of each blocking member extends from a radially outer end of an upper vane of a pair of adjacent vanes, towards a radially inner position and towards the lower vane of the pair, when in use the blocking member is oriented maximally upwards, the outer edge of the blocking member is able to isolate a liquid portion between adjacent vanes (and by analogy the blocking member closes with a gas surface). Depending on the screw inclination when in use (such a screw can be mounted at various inclination angles in use), the radially outer edge of the blocking members are optimally shaped so as to match (i.e. close with) the liquid/gas surface boundary (which remains substantially horizontal, regardless of screw inclination), when the screw is suitably rotated in use (to either the most upwardly position for liquid separation, or the most downwardly position for gas separation).

The top and bottom turns (at the top and bottom of the screw 900) may have truncated and/or tapering blocking members, since for the first and last turns, the blocking members 930 otherwise agitate the liquid and add drag without an increase in lift. The screw can optionally be shrouded by a co-rotating tube 906 within the screw cylinder 905. The cylinder 905 optimally extends over the evaporator such that the majority of bubbles are captured by the end of the screw, and/or optionally a duct 911 connects the evaporator 170 to the cylinder 905.

Further enhancement to screws enclosed in co-rotating tubes 906 can be made by reducing drag from the liquid between the co-rotating tube 906 and the (outer) screw cylinder 905 (also termed “housing”). In this aspect of the invention shown in FIG. 9, a seal (which is, for example a circular washer) 907 is arranged at the top of the cylinder between the co-rotating tube 906 and cylinder 905, so as to trap gas between the co-rotating tube 906 and the cylinder 905, and thereby to reduce drag. For example, the seal 907 in this embodiment is fixed to the screw cylinder 905 and the inner portion of the seal 907 rests on top of the co-rotating tube 906. At the start of operation, the interior of the screw 900 is flooded with liquid. Vapour from an evaporator 170 is conducted into the device via duct 911, or by virtue of the cylinder 905 extending over the evaporator 170. In the present embodiment, the vapour is not ducted into the co-rotating tube, but advantageously rises to the upper wall of the cylinder 905 and up the gap between the container 905 and co-rotating tube 906. The vapour is prevented from escaping at the top of the cylinder 905 by the seal 907, and as the vapour accumulates it pushes the liquid surface down the cylinder 905 until the vapour begins to escape into the screw (inside the co-rotating tube 906), which it fills and causes the screw to rotate, as described elsewhere. Because the pressure of liquid on the top side of the seal 907 is similar to the vapour pressure underneath, the seal can be preloaded to form an effective rotating seal that is lubricated by the operating fluid. As the liquid at the top of the device is not in communication with the liquid at the bottom (they are isolated by the seal), a by-pass duct 560 is provided to connect the two ends of the device (although in embodiments with a co-rotating tube 906 and without seal 907, liquid can return to the bottom of the screw via the clearance between the co-rotating tube 906 and the cylinder 905, therefore no separate duct 560 is required in those embodiments). The seal 907 thus reduces the drag, the heat loss through the outer container and the charge of fluid required in the device, which reduces warm-up time. Clearly the seal 907 can be implemented in a variety of ways and can be fixed to the cylinder 905 or the top of the co-rotating tube 906, or even not be fixed to either. The relative rotational speed of the seal and sealing surface contacted by the seal can be reduced by narrowing the neck of the co-rotating tube 906 in the vicinity of the seal 907, thus reducing frictional losses.

In all embodiments having a spiral screw rotor, including those using vertical screw rotors and/or Archimedes screw rotors, the pitch of the screw can optionally be arranged to increase in the direction of bubble travel, such that the expansion of bubbles is accommodated. For example, it has been found that with a 3 metre high screw, the bubbles expand to approximately 5× their volume at the top compared with at the bottom of the screw (although rotor can optionally be increased in the direction of bubble travel, such that the rotor is cone-shaped. Alternatively, a combination of these features can be applied in embodiments. Alternatively, the number of rotor vanes can be increased or decreased by changing the spiral from a single spiral to a double or triple spiral at various points along the length of the spiral screw rotor, thereby changing the volume between each rotor vane. By accommodating the expansion of bubbles, further work is imparted to the rotor, thus increasing efficiency.

A further improvement to Archimedes screws will now be described.

In its conventional form, an Archimedes screw consists of one or more vanes (sometimes also referred to as ‘blades’) twisting around a central shaft. This is enclosed within an outer cylinder. A cross-section through the screw perpendicular to the axis of rotation shows that this conventional design is characterised by a radial blade profile (FIG. 10). The maximum volumetric flow rate and the associated geometric parameters of this type of screw have previously been characterised (“The turn of the screw—optimal design of an Archimedes screw”, Rorres, 2000).

The volumetric flow rate is an important factor as it influences the cost and size of a device for a particular application. Two design modifications are presented here which result in increased volumetric flow rate compared to the conventional screw. For clarity, the two design improvements have been illustrated schematically for a single blade. They can also be applied to screws with multiple blades.

It is clear that when the axial and radial skew angles are both zero, the conventional screw geometry is obtained as shown in FIG. 10. Using a non-zero value for one or both of the axial and radial skew angles can increase the volumetric flow rate of the screw compared to the maximum value possible with a conventional screw of the same outer radius. This is achieved because of an increase in the volume of fluid held between each pair of screw vanes. Although the total volume (i.e. the volume of liquid plus the volume of gas) between each pair of screw vanes is substantially unchanged, the volume of liquid, or fluid, able to be held in each compartment, between the screw casing and the “tip-over” level (the “tip-over” level is the fill level in a compartment at which fluid begins to overflow the central section, or core 1340, of the screw—from one compartment into the next), is increased or reduced (the ratio of gas to liquid is changed). Volumetric efficiency and/or torque are thereby improved. These novel screw geometries therefore represent an improvement over the conventional design.

FIG. 11 shows an embodiment, having what is referred to as ‘radial skew’, where the vanes 1320 are radially inclined, i.e. a cross-sectional view of the vanes, taken across the end of the cylinder, shows that each vane section 1330 is inclined with respect to a radius from the axis of rotation 1310. Experiment has shown that improved volumetric efficiency is achieved when the vane sections 1330 are aligned at an angle (shown in the cross-section) tending towards a tangent to the circumference of the core 1340.

FIG. 12 shows an embodiment, having what is referred to as ‘axial skew’, where the vanes 1320 are axially inclined (i.e. a cross-sectional view of the vanes, taken along the length of the cylinder and across a diameter of the cylinder, would show each vane section inclined with respect to a normal to the cylinder surface). The result of this vane orientation, when viewed as a cross section across the end of the cylinder, as shown in FIG. 14, is that the vane sections 1330 appear curved (viewed end-on to the cylinder).

Both radial skew and axial skew features result in an improvement in volumetric capacity (the amount of fluid conveyed for a given rotor size and for a single rotation). The increase in fluid volumetric capacity increases torque produced by liquid travelling down the screw and/or by vapour bubble travelling up the screw. For a particular application, the design of a screw can thus be optimised so as to maximise the volume of liquid conveyed per revolution of the screw. Volumetric flow rate is the product of the volume per revolution and the rotational speed of the screw. Thus, for the same size and rotational speed, these improvements allow a higher volumetric flow rate. Both radial skew and axial skew can be used alone or combined, and can also be combined with the above-described blocking member embodiments so as to further improve performance. It has also been found that a greater number of vanes per unit length increases the volumetric efficiency, although a limit is anticipated where additional drag from additional vanes cancels out the increase in torque resulting from increased volumetric efficiency.

FIGS. 13 to 16 (radially skewed screw 1500) and 17 to 20 (axially skewed screw 1700) show further views of the two improved Archimedes screw designs, including end-on and side-on views and two 3D views for each. In both cases (radial skew and axial skew) the illustrated ratio of inner radius to outer radius is 0.5, the screw pitch equals the outer radius, and a skew angle (radial or axial respectively) of 45 degrees has been used to illustrate the general case, although the optimum geometry is almost certainly different and will depend on the application (including depending on the angle of inclination of the screw).

Further, one or more heat exchange devices can be nested inside another heat exchange device.

In operation, various methods of control may be used to control the rate of transfer of heat through a heat transfer device according to the above-described embodiments. The rate of heat transfer can be controlled by one or more of the following methods:

a) controlling the level of heat input to the system. This gradually affects the whole system.

b) changing the level of vacuum or pressure, which immediately affects the whole system by changing the boiling and the saturation point of the liquid and vapour.

c) changing the level of heat transfer out of the system, by changing the level of insulation at various points around the system.

d) by changing the size of the system, for example closing off parts of the condenser 140 or removing some liquid from circulation.

e) by restricting circulation between parts of the system, in particular by reducing the flow of liquid from the condenser 140 to the heat input (evaporator) 170.

f) by actively introducing hot water from the heat input 170 to higher parts of the screw 910, which results in a large increase in vapour generated within the screw 910. This can be done via a hollow central tube 640.

g) by reducing or increasing air flow over the condenser 140.

h) by using mechanical output from the rotor to rotate a fan for forced air flow through the condenser, and by controlling the fan speed or the fan's proximity to the condenser.

The features of the various aspects and embodiments can optionally be combined in any combination.

Here follow, by way of example, some more complex applications of the invention as applied to cooling and power generation (this list is non-exhaustive).

Whereas the advantages of recovering energy from low grade heat such as engine exhausts and warm air exhausted by air conditioning systems are well known, there has been little attention paid to the advantages of more rapid cooling of exhaust streams. Conventional internal combustion engines exploit the difference between the internal pressure generated and exhaust back pressure. Lowering back pressure increases power output. However, achieving this by shortening the exhaust tends to work against correct timing of pressure wave reflections within the exhaust, which improve power by scavenging exhaust gas from the cylinder at an optimal time during the cycle. Exhaust system lengths are optimised for a particular engine speed band. Outside that band, reflection timing may seriously degrade engine performance. The timing of pressure wave reflections is a product of exhaust length and the speed of sound in the exhaust. A method of rapidly reducing the temperature of the exhaust gases and so reducing the back pressure would be beneficial if it was sufficiently controllable to also control the speed of sound in the exhaust. The current device removes heat rapidly and controllably and could fulfil the specified need, perhaps by using the engine's existing liquid cooling system as a first heat sink for high temperatures and ambient air as a second heat sink for second stage cooling. Generating electrical power from the cooling process would be an added benefit.

Similarly in the field of gas compression, gas is discharged at high temperatures and cooled in heat exchangers, incurring a pressure drop. More rapid heat exchange would enable a smaller pressure drop in the exchanger by shortening the required heat exchange passage length and lower the back pressure. This would lower the pressure ratio across the compressor and reduce the compressor work consumed to achieve the required compression. Thus, parasitic losses are reduced.

Reducing pressure drop would also be advantageous in refrigeration systems, air conditioning systems, desalination and other industrial processes. In a typical domestic fridge the condenser is 12 to 18 metres of narrow bore finned tubing with many bends. This imposes a pressure drop which consumes about 10% of the system's energy input. As pressure drop is proportional to length, cooling the flow over a length of 1.2 metres instead of 12 metres would reduce the system's energy consumption by 9% from this effect alone. This would result in a reduction in power consumed by the compressor, and a reduction in motor heat generated. In hermetic domestic systems this introduces a virtuous circle of a reduction in motor heat, thus reducing the inlet temperature rise incurred from cooling the motor with inlet gas, thus leading to lower compressor work requirement and so on. Another advantage of reducing the condenser length is that it significantly reduces the refrigerant charge, lowering cost and increasing system response. Again, work can be recovered from the transfer of heat by a rapid cooling device according to the present invention.

Further, in refrigeration systems, a fan can be driven from a rotor 120 of a heat transfer device 100 as claimed, with the heat source being the compressor heat. A fan has been found to provide a useful effect wherever there is approximately a 15 degree temperature difference between the evaporator 170 and the condenser 140. The fan can be used to drive air flow through the condenser of the refrigeration system, in use, thereby improving the heat transfer from the refrigeration system condenser. Heat can also be removed from the hermetic compressor container. Both of these effects can lead to improved refrigeration efficiency.

In industrial electric motors a power failure, trip or shut down, can lead to temperature excursions because the air cooling fan is part of the motor (and therefore rotates with it). A quick restart with high current draw causes high temperature damage to the winding varnish. Lengthy cooling is therefore mandated before a motor is restarted, and this can result in considerable production loss on complex process lines. Self powered devices according to the invention continue to provide cooling as long as there is heat flow above the designed temperature, thus enabling processes to be re-started more quickly. Parasitic losses (due to electrical resistance which generally increases with temperature) are thereby also reduced.

A further application for such a heat transfer device is for cooling concentrator-type photo-voltaic cells, for example in domestic installations, where the efficiency of the cells can be increased by cooling them, and at the same time hot water can be produced by exchanging heat from the heat exchanger (condenser) 140 to water contained in a domestic hot water tank. Similarly, such a heat transfer device is useful for cooling high power LEDs.

An application of the invention in the field of mechanical power that fulfils a domestic heating requirement. Domestic radiators are large because of the heat surface area required to transfer heat to air by purely convectional means. Such radiators have to be even larger to transfer heat from the low temperature water produced by heat pumps. The rate of heat transfer slows as the room temperature and water temperature converge. Slow response and warm-up may result from the use of standard radiators. This problem is being overcome by the addition of electrically driven fans to increase flow of air. However this requires electrical cabling, supply and controls to turn the fan on and off.

Aspects of the present invention, called a fan system in this application, can be applied as a solution by making part of the radiator into the container for an evaporator for a two phase vacuum driven mechanical and electrical power producing system.

In a favoured embodiment the fan system is closed and filled with water under vacuum. The evaporator is in thermal contact with the hottest part of the radiator, which is where the hot water enters the radiator. The system is provided with a fan driven by magnetic means by preferably an Archimedes type screw according to various aspects of the invention. Optionally the system is also provided with a generator to convert a fraction of the power output into electricity.

In operation, hot water reaches the radiator soon after the heating system starts circulating water. Heat transfers to the fan system and heats up its charge of water. When the water boils the rising vapour causes the rotor to turn, which in turn causes the fan to turn via a coupling (preferably a coupling having no rotating seals, for example a magnetic coupling). The fan turns, causing an air flow through the radiator and its ducting, and thus produces improved heat transfer. Some of the power transmitted by the coupling may be used to generate electricity and power temperature sensors for wireless transmission with a central controller, and thereby allow control of the radiator valves to adjust the room temperature. Because the condenser of the fan system also acts as a radiator there is additional improvement in heat transfer to the room. The fan system optionally has a separate air path from that of the main body of the radiator, although in some embodiments the condenser 140 of the fan system could be the sole heat exchanger with the room, such that maximum energy is available for rotating the fan.

Thus the benefits of a fan driven radiator can be obtained without use of externally supplied electrical power or cabling and controls. As such radiators are inherently safe because of their low temperature and low pressure design, they are eminently suitable for installation in bathrooms, where an electric fan might be a hazard. Furthermore the addition of a self-powered fan would also improve the performance of electric oil-filled radiators. The addition of a fan would also be helpful in cooling PCs. Industrial applications where cooling must be guaranteed in the event of power failure are also possible.

Some further examples of domestic applications follow:

a) The heat transfer device (also termed a buoyancy engine in the examples below) can operate on solar power to produce both electricity and hot water during the day. Given a reasonable amount of insulated hot water storage, generation can continue after sundown by generating from stored hot water. When this is exhausted, burning wood or fossil fuels can keep the engine supplied with heat.

b) In far northern winters the engine can operate on a low boiling point liquid and generate electricity by using the below freezing heat from permafrost or the sea etc to generate, by using the even lower night air temperature as a sink. The electricity can be used to power a heat pump to provide hot water.

c) The buoyancy engine can operate on the exhaust heat from a domestic boiler to provide electricity to run the boiler controls and fan, thus freeing the boiler from grid dependence. The same engine can switch to operating on solar during the day when the boiler is off.

d) A small device can operate off waste fridge heat (from the heat exchanger of a refrigerator unit) to charge a battery, which in turn is used for mobile phone charging.

Further, control can be exercised over pressure in the system to enable mechanical power to be extracted after the heat source has been removed. Since there is a large amount of energy stored in the hot liquid (e.g. water) after the heat (solar or other source) is no longer available, if the pressure is gradually reduced then power generation can continue due to continued boiling of the liquid at lower temperatures under the lower pressure. Although the embodiments described above have been described as operating under a lowered pressure or vacuum, by the selection of different liquids having different boiling temperatures, it is possible for such a system to operate under raised pressure, for example during the day when input heat is highest (when the sun is shining on a solar collector acting as heat source). Then, at night, when only a lower temperature heat source is available (such as a tank of hot water which was heated during the day when there was an excess of heat provided by the solar collector), the system pressure can be reduced so as to allow the liquid to boil at a lower temperature. It is also possible to enhance this effect by deliberately increasing the volume of water in the system, or within an auxiliary (or “side”) tank (used for storing excess heat collected during the day), should there be additional heat available which is not being used for generation at the time or should the value of stored energy be high. An example is where the engine is sized to produce say 100 W during the way and this requires 1000 W of heat for an engine of 10% efficiency. However, additional heat is available particularly at the peak of the day and this is used to store hot water. When the sun goes down, this second, auxiliary, tank of hot water is circulated and used as a heat source to allow generation to continue. Pressure inside the engine can be reduced as the temperature of the heat source decreases, thus lowering the boiling point and allowing further power generation. This can eliminate the need for a battery in stand-alone off-grid systems, which is a major cost and source of complication for off-grid power systems.

Some further examples of small industrial applications follow:

a) Wireless sensors can be powered or re-charged by very small buoyancy engines. These would be particularly effective in or on industrial or central heating ducts where solar is not practicable and where the system may remain cold for months. Buoyancy engines would start generating as soon as heat flowed, overcoming the problem of discharged batteries.

b) Buoyancy engines can recover exhaust energy from distributed power generation units such as Diesel generators.

c) Buoyancy engines can recover energy from warm exhaust from air-conditioned buildings systems.

d) Buoyancy engines can recover energy from compressors and refrigerators.

Some further examples of large Industrial applications follow:

a) Vertical versions of buoyancy engines can be built into industrial chimneys so as to recover power, in mechanical and/or electrical form.

b) Concentrator solar systems in deserts require large amounts of cooling water. The hot water can be stored to produce more valuable night-time electricity. As the desert night air cools, the engine can continue to run on the progressively lower heat levels in the stored water of the plant.

c) Heat removal in data centres is becoming a major problem. Buoyancy enginess can utilize the heat from chillers and refrigeration plants (e.g. from their heat exchangers) to produce recycled energy.

d) There are many uses in water treatment where Archimedes screws are widely used for lifting water. Sewers and septic tanks are at a temperature of approx 15-20° C. all year round, from the warm water released and bio-decay. This heat can be applied to generation. In sewage farms Buoyancy Engines (BEs) can be used to directly drive lifting Archimedes screws, driving distributors and agitators. In this application, cooling water from electric motors and biogas generators can be piped at, say 60° C., to the BEs to stir and distribute sewage.

e) Mine cooling. Like the London Underground, mines can have uncomfortably high air temperatures. Like heat pipes, BEs can transfer heat from evaporator to condenser at supersonic speed over long distances. Thus they have the capacity to generate power at the workface, while simultaneously cooling the working environment.

Claims

1. A heat transfer device having a housing, the housing containing an evaporator and a rotor, the rotor rotationally drivable by bubbles produced when a liquid is heated by the evaporator, the rotor further arranged proximate to a surface of the evaporator so as to aid removal of bubbles from the surface.

2. A device according to claim 1 wherein the removal of bubbles is by scraping, brushing or turbulence.

3. A heat transfer device having a housing, the housing containing an evaporator and a rotor, the rotor rotationally drivable by bubbles produced when a liquid is heated by the evaporator, wherein the rotor is further arranged to have successive compartments for containing bubbles and each successive compartment in the direction of bubble travel has increased volume so as to accommodate bubble growth.

4. A device according to claim 3, wherein the rotor has multiple rotor sections which are coupled together, each successive rotor section having larger compartments than the preceding rotor section.

5. A device according to claim 3, wherein the rotor is arranged to provide a mechanical drive output and/or an electrical power output.

6. A device according to any preceding claim 3, wherein the rotor has a plurality of at least partially radial vanes around its circumference, so forming cells.

7. A device according to claim 6 wherein the rotor comprises a one-way valve arranged to allow fluid displaced by bubble growth to flow out of a rotor cell in a direction contrary to the direction of rotor rotation.

8. A device according to claim 7, wherein the rotor rotational axis is substantially horizontal and the rotor vanes are curved so as to trap bubbles in the cells for an extended range of rotor rotational angles in operation.

9. A device according to claim 8 wherein the rotor axis is inclined from about 0 degrees to about 45 degrees from horizontal.

10. A device according to claim 8, wherein the housing comprises a downward duct to deliver condensate near the descending side of the rotor.

11. A device according to claim 6, wherein the rotor rotational axis is substantially vertical.

12. A device according to claim 11 wherein the rotor axis is inclined from about 0 degrees to about 45 degrees from vertical.

13. A device according to claim 11, wherein the housing incorporates shrouding around the rotor, providing first and second ports for bubble release and condensate return, respectively.

14. A device according to claim 13 wherein the ports communicate with upward and downward ducts for conveying bubbles from the evaporator to the condenser, and from the condenser to the evaporator, respectively.

15. A device according to claims 1, wherein the rotor comprises a screw having a vertical component to its axis of rotation, and the compartments are formed between screw vanes.

16. A device according to claim 15 wherein the rotor screw comprises downwardly facing projections for capturing bubbles.

17. A device according to claim 15, wherein the rotor is an Archimedes screw.

18. A device according to claim 17 wherein the screw rotor has an axial skew, such that the screw vanes extend from the axis of rotation in a direction, which when viewed side-on in cross-section through a diameter of the screw, the direction is inclined from perpendicular.

19. A device according to claim 17, wherein the screw rotor has a radial skew, such that the screw vanes extend from the axis of rotation in a direction which when viewed end-on in cross-section is inclined from a radius.

20. A device according to claim 17, wherein the screw rotor further comprises a plurality of blocking members radiating from the axis, the outer edge of each member extending from a radially outer end of a first vane of a pair of adjacent vanes, towards a radially inner position and towards a second vane of the pair, the outer edge of each blocking member arranged so as to close with a liquid/gas surface, when the blocking member is oriented maximally down/up in use, so as to isolate a gas/liquid portion between adjacent vanes in use.

21. A device according to claim 20 wherein the number of blocking members per screw rotation is between about 6 and 12.

22. A device according to claim 15, wherein the rotor screw has an expanding cross-sectional area in the direction of bubble travel.

23. A device according to claim 22 wherein the expanding cross-sectional area is achieved by increasing the screw pitch, increasing the screw outer diameter, decreasing the screw inner core volume, and/or decreasing the volume of a projection located between the screw vanes.

24. A device according to claim 15, wherein the rotor screw has a co-rotating tube around it.

25. A device according to claim 24 wherein a rotating seal is arranged between the co-rotating tube and the housing of the screw.

26. A device according to claim 1, wherein the rotor comprises an internal duct.

27. A device according to claim 1, wherein the rotor has a bubble-repellant coating.

28. A device according to claim 1, wherein the housing comprises an upward duct for conveying bubbles from the evaporator to a condenser.

29. A device according to claim 1, wherein the housing comprises a downward duct to carry condensate from the condenser to the evaporator.

30. A device according to claim 29 wherein the downward duct is positioned and angled so as to direct condensate onto the rotor in the direction of rotation.

31. A device according to claim 1, wherein the housing is arranged to shroud at least one rotor cell at a position close to the evaporator.

32. A heat transfer device having a housing, the housing containing an evaporator and a rotor, the rotor rotationally drivable by bubbles produced when a liquid is heated by the evaporator, wherein the rotor has a plurality of radially arranged cells, the housing and rotor are arranged to close at least one cell when the cell is rotated adjacent to the evaporator, and the cell has a one-way valve arranged to allow fluid displaced by bubble growth to flow out of the cell in a direction contrary to the direction of rotor rotation.

33. A heat transfer device having a housing, the housing containing an evaporator and a rotor, the rotor rotationally drivable by bubbles produced when a liquid is heated by the evaporator, wherein the rotor has a plurality of radially arranged cells, the housing and rotor are arranged to close at least one cell when the cell is rotated adjacent to the evaporator, and the housing has a duct positioned adjacent to the evaporator to allow fluid displaced by bubble growth to flow out of the cell.

34. A device according to claim 33 wherein the duct incorporates a one-way valve.

35. A heat transfer device according to claim 32 when used for cooling exhaust gas from an internal combustion engine.

36. A device according to claim 35 wherein multiple devices are arranged in series, and optionally wherein successive devices are arranged to operate at lower temperatures and/or exhaust gas pressures.

37. A device according to claim 35, wherein the device incorporates a screw rotor having increasing cross-sectional areas between screw vanes in the direction of gas travel.

38. A heat transfer device according to claim 33, wherein the rotor is arranged to drive a fan for increasing air flow over a heat source.

39. A device according to claim 38 wherein the heat source is a domestic hot water radiator, refrigeration condenser, refrigeration compressor, LED, photo-voltaic cell, solar collector, heat exchanger or any suitable source of heat having a higher temperature than the housing.

40. A heat transfer device according to claim 33, wherein the rotor provides mechanical power and electrical power using electrical generating means.

41. A heat transfer device according to claim 33 further comprising a pressure regulator device, the regulating device comprising a spring and piston, or a bellows arrangement, and optionally wherein the spring is bi-metallic.

42. A heat transfer device according to claim 33 further comprising a vacuum pump.

43. A heat transfer device according to claim 33 further comprising a rotor shroud arranged direct bubbles into the rotor.

44. An Archimedes screw having vanes formed from a spiral which is arranged around a rotational axis within a cylinder, the screw further comprising a plurality of blocking members radiating from the axis, the outer edge of each member extending from a radially outer end of a first vane of a pair of adjacent vanes, towards a radially inner position and towards a second vane of the pair, the outer edge of each blocking member arranged so as to close with a liquid/gas surface, when the blocking member is oriented maximally down/up in use, so as to isolate a gas/liquid portion between adjacent vanes in use.

45. An Archimedes screw having vanes formed from a spiral which is arranged around a rotational axis within a cylinder, the screw further comprising a plurality of blocking members radiating from the axis, the outer edge of each member extending from a radially outer end of a first vane of a pair of adjacent vanes, towards a radially inner position and towards a second vane of the pair.

46. An Archimedes screw according to claim 45 wherein the first and second vanes are respectively upper and lower in use, such that the outer edge of the member closes with a gas/liquid surface to isolate a gas portion between adjacent vanes when the member is oriented maximally downwards in use.

47. An Archimedes screw according to claim 45 wherein the first and second vanes being respectively lower and upper in use, such that the outer edge of the member closes with a gas/liquid surface to isolate a liquid portion between adjacent vanes when the member is oriented maximally upwards in use.

48. An Archimedes screw according to claim 45, wherein the number of blocking members per screw rotation is between about 6 and 12.

49. An Archimedes screw having vanes, wherein the vanes have a radial skew, such that the vanes extend from the axis of rotation in a direction which when viewed end-on in cross-section is inclined from a radius.

50. An Archimedes screw having vanes, wherein the vanes have an axial skew, such that the vanes extend from the axis of rotation in a direction, which when viewed side-on in cross-section through a diameter of the screw, the direction is inclined from perpendicular.

51. (canceled)

52. A method of controlling the transfer of heat through a heat transfer device, the heat transfer device having a housing containing a condenser and an evaporator arranged to boil a liquid by heat input, the method comprising controlling the rate of transfer of heat by carrying out one or more of the following steps:

a) controlling the level of heat input to the system;
b) changing the level of vacuum or pressure in the housing;
c) changing the level of heat transfer out of the system;
d) changing the size of the system by closing off parts of the condenser or removing some liquid from circulation;
e) restricting circulation between parts of the system, in particular by reducing the flow of liquid from the condenser to the heat input (evaporator) 170;
f) actively introducing hot water from the evaporator to higher parts of the rotor, thereby increasing vapour generated within the rotor; and
g) reducing or increasing air flow over the condenser.
Patent History
Publication number: 20130213613
Type: Application
Filed: Jun 20, 2011
Publication Date: Aug 22, 2013
Inventors: John Philip Roger Hammerbeck (London), Keith Robert Pullen (London), Matthew Gordon Read (London)
Application Number: 13/805,311
Classifications
Current U.S. Class: With Pressurizing Means Or Degassifying Means (165/104.27); 416/223.00R
International Classification: F28D 15/00 (20060101);