The present invention relates to a combined heating and cooling system to be used primarily for buildings and the like in which a primary source for the energy requirements of the system is the sun's radiation.

The present invention is the outcome of a systematic effort to minimize, for a given performance, the cost of heating and cooling systems utilizing the sun's thermal energy, supplemented by that of the earth's atmosphere.

DESCRIPTION OF THE PRIOR ART

While there have been prior attempts to utilize the sun's radiation to heat and cool buildings by employing the absorbing surface of solar collectors, the prior attempts have been ineffective and inefficient. This is particularly true in the case of planar solar collectors which are of particular importance because they are most readily capable of mass production and therefore less expensive to produce.

The cost effectiveness of planar solar collectors as a power source to heat and cool buildings and the like is dependent on a large number of diverse factors. However, prior efforts have failed to address themselves to five very important factors which substantially increase the effectiveness of planar solar collectors for this purpose. These factors are:

(1) maintaining the absorbing surfaces of the collector at a uniform temperature;

(2) utilizing the same working fluid to remove thermal energy from the solar collector and to release it to the medium to be heated when the temperature of the collector's absorbing surface is both above and below the temperature of the medium to be heated;

(3) utilizing the energy absorbed by the solar collector and the same working fluid for both heating and cooling;

(4) reducing the mass flow rate of the working fluid per unit of heat absorbed in, and removed from, the collector; and

(5) utilizing a collector module configuration which is suitable for both large and small scale installations.

Furthermore, prior art systems fail to utilize essentially the same equipment to extract heat from the earth's atmosphere to supplement the heat obtained from the sun's radiation.

A more detailed discussion of prior attempts to utilize solar energy in heating and/or cooling systems appears hereinafter in connection with the detailed description of the various features of the present invention.

SUMMARY OF THE INVENTION

The present invention relates to improved systems and methods for heating primarily buildings and their hot water supplies, and for cooling buildings, by utilizing free sources of thermal energy. The invention contemplates utilization of the solar energy intercepted by a planar solar collector which may be supplemented by the thermal energy of the atmosphere. The latter energy is, in one configuration, absorbed by the fluid flowing in the same evaporator as the one used (in the collector) to absorb solar heat.

The working fluid used throughout the system belongs to the class of volatile fluids referred to as refrigerants.

For heating applications, the same refrigerant is used throughout the system and thus heat exchangers between two different working fluids, such as water and a refrigerant or two different refrigerants, are always eliminated. In applications including cooling, two different refrigerants may be used in some cases, depending on the design parameters of a particular installation. The preferred embodiment described in the present disclosure outlines the preferred embodiments for the case when the same refrigerant is used throughout.

The complete system of the present invention uses five different thermodynamic cycles: three alternative cycles for heating; and one cycle alone, or two different cycles simultaneously, for cooling. The five cycles can be referred to generically as follows:

(i) a Rankine-type transfer cycle

(ii) a solar-assisted, vapor-compression heating cycle,

(iii) an (atmospheric) air-assisted, vapor-compression, heating cycle,

(iv) a Rankine power cycle, and

(v) a vapor-compression cooling cycle.

The names used to describe cycles (i) through (v) have been chosen solely so that they can contribute to the clarity of the description of the present invention. Cycles (i) through (iii) are used for heating, cycle (v) alone for cooling, and cycles (iv) and (v) simultaneously for cooling.

The Rankine-type heat transfer cycle is used when the saturated-vapor temperature of the refrigerant in the solar collector is significantly higher than the temperature of the medium to be heated. The cycle has been so named because it resembles a Rankine power cycle without superheat in which the engine has been omitted. More specifically, the working fluid absorbs heat in the solar collector primarily by evaporation, releases it to the medium to be heated primarily by condensation, and is returned, in its condensed form, to the solar collector by a pump.

The solar-assisted, vapor-compression heating cycle is used when the saturated vapor temperature of the refrigerant in the solar collector is below the temperature of the medium to be heated, or not sufficiently above it for the Rankine-type heat transfer cycle to be used and provide heat at the desired rate. This cycle is similar to a conventional vapor-compression, heat pump cycle, but is controlled by novel techniques.

The air-assisted vapor-compression heating cycle is a conventional vapor-compression, heat pump cycle, but is controlled differently.

The Rankine power cycle is a conventional Rankine power cycle with negligible superheat. It is used by a thermal engine powered by the solar heat absorbed by the refrigerant in the collector. This engine is employed exclusively, or with the assistance of an electric motor, to drive the compressor(s) of the vapor-compression cooling cycle.

Finally, the vapor-compression cooling cycle is a conventional vapor-compression, refrigeration cycle, but is controlled differently in all cases where the medium to be cooled is the storage medium of a latent-heat reservoir.

The present disclosure further relates to novel systems in which apparatus used to instrument the five cycles listed earlier are combined in a number of different ways.

A typical configuration of a combination for heating two different media, and cooling one medium, inside a building is described under the heading "An Embodiment of the Invention, Overall System."

The solar collector and evaporator employed may use any type of tracking focused or planar collector, or any type of fixed planar collector. However, the foregoing novel methods for heating and cooling were devised primarily for solar collectors consisting of fixed planar modules whose surfaces have no more than two or three orientations. In this case, the preferred type of evaporator used is functionally similar to the type of evaporator described in the art as a flooded evaporator. However, its physical configuration is novel. Further details of the present invention are described and discussed in the context of a solar collector and evaporator consisting of planar modules lying in a single plane and using a flooded evaporator.

With the foregoing in mind, it is a primary object of the present invention to provide a new heating and cooling system which can extract and utilize the heat primarily from the sun's radiation, and secondarily from the surrounding air.

A further principal object of the present invention is to provide a new heating system which can extract heat from the sun's radiation and transfer it to the medium to be heated by using the same working fluid and the same means for distributing this fluid in the solar collector both in the case when the temperature of this fluid inside the collector is higher and lower than the medium to be heated.

Still another important object of this invention is to use the same working fluid and the same means of distributing this working fluid inside the collector to extract heat from the sun's radiation to power a heat engine used to drive the compressor(s) of a vapor-compression cooling cycle.

Yet another object of the present invention is to supplement solar heat with atmospheric heat at the cost of only an insignificant amount of additional equipment.

A still further object of the present invention is to provide a novel planar collector configuration whose absorbing surface is maintained at a uniform temperature, employs a working fluid which can absorb a given amount of heat while it is in the collector, and release it to the medium to be heating, with the lowest possible mass flow rate.

Still further primary objects of the present invention reside in the reduced operating costs involved in heating and cooling in accordance with the systems embodied in this invention.

These and other objects of the present invention will become apparent from the following description, specification, claims and accompanying drawings.

DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic illustration showing a typical assembly of four solar collector module panels of the present invention.

FIG. 2a is a perspective view of the evaporator in which the working fluid is evaporated in the solar collector in accordance with the present invention;

FIGS. 2b, 2c, and 2d are sections along lines 2b--2b, 2c--2c, and 2d-2d of FIG. 2a and show three typical configurations of the evaporator passageways;

FIGS. 2e and 2f are side elevations of the novel evaporator along section lines 2e--2e and 2f--2f of FIG. 2a showing two different vapor header shapes;

FIGS. 3a and 3b are side illustrations of two forms of solar collectors containing the novel evaporator of FIG. 2a in which only solar energy is collected in the FIG. 3a form and both solar and thermal energy are collected in the FIG. 3b form;

FIG. 4 is a schematic flow diagram of the Rankine-type heat transfer cycle of the present invention;

FIG. 5 is a typical pressure-enthalpy diagram for a refrigerant following the Rankine-type heat transfer cycle in accordance with the present invention;

FIGS. 6a and 6b are typical pressure-enthalpy diagrams for a refrigerant following the vapor compression, heat-pump cycle in accordance with present invention under ideal conditions and under actual conditions, respectively;

FIGS. 7a and 7b are a composite schematic flow diagram of the solar assisted, vapor compression heat pump cycle of the present invention with a typical control system therefor;

FIGS. 8a and 8b are a composite schematic flow diagram of the combined Rankine-power cycle and air-source, vapor-compression cooling cycle of the present invention with a typical control system therefor;

FIGS. 9a and 9b are a composite schematic flow diagram of the heating and cooling system of the present invention showing a typical application for a building; and

FIGS. 10a, 10b and 10c are schematic control diagrams which illustrate methods for choosing how Q_e, Q_c and Q_h /H_cp can be chosen as the independent variable in the solar-assisted, vapor-compression, heat-pump cycle.

DESCRIPTION OF PREFERRED EMBODIMENTS FOR TYPICAL APPLICATION

PRELIMINARY REMARKS

The preferred embodiments described hereinafter relate to a typical application of the invention. It will be readily appreciated by those skilled in the art that the heating and cooling system of the present invention has numerous applications, such as office buildings, hospitals, residential dwellings, industrial process heating operations, and any other application requiring heating or cooling within the temperature range of the equipment and working fluid used.

THE SOLAR COLLECTOR AND EVAPORATOR

In general, any solar collector configuration and, in particular, any of the conventional planar solar collector configurations used heretofore can be used in the instant invention. However, the advantages offered by the present invention including the novel controls described hereinafter are exploited to a much greater extent by using, in the case of planar solar collectors, the novel working fluid distribution configuration illustrated in FIG. 2a.

A partial assembly of a typical solar collector configuration containing only four modules is shown in FIG. 1. It will be appreciated that many modules are normally required for even the smallest solar collector and the number obviously depends on the size of the collector desired. The modules can conveniently be assembled at the building site.

The evaporator 100A of each solar collector module 100 consists of a liquid distribution header 101, a vapor collection header 102, and a set of passageways 103 connecting the liquid header to its vapor header. The function of this set of passageways is to bring the liquid refrigerant to intimate thermal contact with its sheet metal panel 104. The sheet metal panel 104, which extends over the entire surface of the solar collector module between the two headers 101 and 102 may be an integral part of the set of passageways 103, or may be a separate sheet of absorbing material in intimate thermal contact with the set of passageways, and joined to them by a process such as brazing or welding.

In the former case, the heat absorbing (metal) panel 104 itself may either consist of raised and depressed areas joined to another metal panel 105 with matching raised and depressed areas in such a way as to form channels, joining its liquid and vapor headers, through which the liquid refrigerants could flow, or it may be flat and joined to another sheet which has raised and depressed areas (see FIG. 2).

Such channels 103 could be typically joined by two fluted panels, joined together by seams 106 (which may be brazed, bonded or welded seams), whose flutes may have axes perpendicular to the axes of the two header. It is important to note, from the viewpoint of cost, that these seams need not be leak proof. In fact, from a functional standpoint, the two sheet metal panels could be two parallel plane sheets spaced, say, one eighth to one quarter of an inch apart. These two parallel panels are joined together only by leakproof seams 107 along its two panel edges perpendicular to the headers. The purpose of the intermediate seams is solely to increase the structural strength of the evaporator to allow a pressurized fluid at a pressure of up to, say, 250 to 300 lb/in² to be contained between two thin metal sheets, say, one eighth of an inch thick or less.

In the case where the sheet of absorbing material is not an integral part of passageway 103, the passageways may be a set of approximately parallel tubes 108 (See FIG. 2d) joined to the panel 104.

The absorbing panel 104 is joined to the headers 101 and 102 by brazed or welded seams 109 and 110, and so is the second panel or the tubes, as applicable.

The liquid header distributes liquid refrigerant to the set of passageways. These headers are connected together and the liquid manifold 111 (FIG. 2a) as shown in FIG. 1. The vapor header collects the refrigerant vapor that has been formed in the passageways of the evaporator by solar heat and passes it to vapor manifold 112. The vapor headers are connected together and to the vapor manifold as shown in FIG. 1. The ends 130 and 131 of the liquid and vapor headers that are not connected to another header are capped.

The liquid and vapor headers have orifices located so as to communicate with the passageways.

The purpose of the passageways is to distribute liquid refrigerant over the heat absorbing panel 104 so that the liquid can be evaporated (boiled) efficiently by the solar radiation impinging on and absorbed by panel 104, and the resulting vapor bubbles can escape to the vapor header.

Depending on the layout of a particular installation, it may be desirable to provide additional oil return paths to that provided by the manifold by tapping small tubes (not shown) into the line formed by the serially connected liquid headers at appropriate intervals. The slope of panel 104 must be such that the equilibrium level 115 of the liquid in the vapor header is high enough to keep the passageways 103 filled with liquid to ensure efficient evaporation, but not so high as to restrict significantly the flow of vapor in vapor header 102. Also the passageways 103 must be connected near the "bottom" of the vapor header 102 as shown in FIGS. 2e and 2f, where x--x is a horizontal reference line. The vapor header need not have a circular cross-section and elliptical shapes such as shown in FIG. 2f are also acceptable.

There are a large number of different ways the evaporator 100A shown in FIG. 2a could be incorporated into a complete solar collector module. No novel attachments or configuration are claimed for the overall module. However, for illustration, two typical configurations of an overall module are shown in FIGS. 3a and 3b.

In the FIG. 3a configuration, the refrigerant in the evaporator is used to absorb only solar radiation. The evaporator consists of liquid header 101, vapor header 102, absorbing panel 104, lower panel 105 which, together with panel 104, forms the passageways 103. The evaporator is contained within the external structure 116 and cover plate 117, which is transparent to solar radiation frequencies contained in the frequency spectrum corresponding to the temperature of the absorbing panel 104.

The space between the evaporator 100A and the structure 116 is filled with insulation 118. The space between the cover plate and the evaporator is airtight. Liquid and non-permeable insulation material is provided at 120, 121, 122 and 123, where the structure 116 comes in contact with the evaporator panels 104 and 105.

In the FIG. 3b design, the evaporator 100A is also used to absorb heat from the air surrounding the collector module. The space 119 is airtight when solar heat is used as the source of free energy, and open to the surrounding air when the thermal energy of the atmosphere is used as the source of free energy. Suitable dampers are provided at each end of a row of modules for this purpose. In most cases, the collector module itself would consist only of the portion shown in FIG. 3b above the line y--y. The channel 111 below the y--y line being formed by the supporting structure such as the roof, wall, etc. on which the collector module is mounted.

The preferred material for the evaporator 100A is aluminum. The absorbing panel 104 should be coated by a frequency-selective material.

A typical coating is copper oxide, which is deposited on panel 104 after its surface has been polished. Frequency-selective coatings that can be formed by an electro-plating process are also acceptable and inexpensive.

Since the maximum efficiency of the solar collector modules requires that the temperature of the liquid refrigerant entering the collector differ by only a few degrees from the temperature of the saturated vapor in the vapor header, the Rankine-type heat transfer cycle will be discussed next and the solar-assisted, vapor-compression cycle and the Rankine power cycle discussed later are controlled in such a manner that this condition is satisfied.

Some of the advantages resulting from the novel evaporator configurations and operation described heretofore are as follows:

(1) all of the solar collector heat absorbing panels are maintained at approximately the same uniform temperature say within 1°-2° F.;

(2) the speed of the liquid refrigerant in the passageways is low, typically one or two feet per minute, and is independent of the number of modules connected in a row; thus friction is negligible,

(b 3) the optimal sizes of the passageways is unaffected by the number of modules connected in series and hence also by the overall size of the solar collector. Thus, only the optimal size of the headers is affected by the number of modules connected in series, and oversizing only these headers, as opposed to oversizing all the passageways (tubing), for small installation so that they can also be employed for large installation entails only a small penalty in cost. Hence, for example, a single solar module design could be used for solar collection from, say, 600 to 12,000 square feet, and perhaps only one additional design for sizes between 8,000 and 160,000 square feet.

THE RADIATION EQUILIBRIUM TEMPERATURE

In the present invention, the term "radiation equilibrium temperature" of the heat absorbing panel 104 refers to that radiation equilibrium temperature attained by the panel when no heat is removed by the refrigerant. This differs markedly from earlier efforts, Marchant et al in U.S. Pat. No. 2,693,939. Where the patentees measure the radiation equilibrium temperature of the airspace between the collector transparent cover plate (5) and the energy absorbing panel.

In the present invention the "radiation equilibrium temperature" is measured continuously even when the working fluid (refrigerant) is removing heat from the solar collector. This is accomplished for example by using a temperature transducer where thermal bulb is in intimate contact with the heat absorbing panel of a small dummy collector module which uses exactly the same cover plate, panel spacing and panel material (including any frequency-selective coating), as the operating solar collector modules. However, the dummy collector contains no liquid and vapor header, and no refrigerant. The heat absorbing plate is however insulated from the surrounding air and supporting structure in a manner which replicates the rate at which the heat absorbing panel of a regular collector module loses heat to its surroundings when no heat is being removed from it by its working fluid. The dimensions of the dummy collector module must be large enough for so-called "end effects" to be negligible. A typical panel size is six inches by one foot. The thermal bulb of the temperature transducer is placed in the center of the panel, and the panel orientation must be the same as that of the solar collector itself. If the collector consists of two or three planes with different orientations, a different dummy collector module must be provided, and the radiation equilibrium temperature must be measured, for each such plane.

RANKINE-TYPE HEAT TRANSFER CYCLE

Known systems, developed or proposed, that use the sun's thermal energy to heat buildings employ a liquid or a gas to transfer heat from the solar collector to a medium in the building. This medium may be the air of a forced-air heating system, the water of a hot-water heating system, the storage medium of a heat reservoir, or the refrigerant of a heat pump. A gas, usually air, is employed only for small buildings. In most cases, a liquid is employed which normally consists of a mixture of water and an "anti-freeze" liquid such as ethylene glycol. In all these cases, the temperature of the fluid leaving the solar collector is higher than that of the medium to be heated.

The present invention contemplates instead that heat be transferred from the solar collector by a volatile fluid whose phase alternates between the liquid and vapor states so as to absorb latent heat while it is being evaporated by the sun's radiation and so as to release latent heat while it is being condensed by the heat receiving (cooler) medium. Volatile fluids used to transfer heat in this manner are called refrigerants. Such fluids are usually employed to transfer heat from a body at a given temperature to a body at a higher temperature, and not to a body at a lower temperature as proposed in the present invention.

I now discuss the foregoing binary-phase heat transfer cycle in greater detail.

In this cycle, referring to FIG. 4, the refrigerant, in its liquid phase, enters solar collector (and evaporator) 202 at point 201, absorbs heat in this collector, primarily while it is being evaporated (by the sun's heat), and exits the solar collector at point 203 after the liquid refrigerant has been transformed into vapor. The refrigerant passes through the servo-controlled throttling valve 204, and then enters at point 205 condensing coil 206 where it is condensed by releasing heat, to the medium 207 by conduction, free or forced convection, or radiation. The condensing coil may be bare or finned, and can have any known configuration compatible with the medium. The refrigerant exits this condensing coil at point 208 after the vapor has been condensed. From point 208, the refrigerant enters at point 209 the receiver (liquid reservoir) 210 and, after exiting at point 211, enters at point 212 the (liquid) feed pump 213. After the refrigerant is compressed, it exits this pump at point 214, enters at point 215 flow transducer 216 and, after exiting from the flow transducer at point 217, is returned to point 201. Safety valve 218 opens only if the refrigerant pressure at exit 203 exceeds the maximum design pressure.

The salient features of the cycle traversed by the refrigerant may be summarized by saying that it resembles a Rankine power cycle in which the engine has been omitted. This is the reason why this cycle was referred to earlier as the Rankin-type heat transfer cycle.

Solar collector (and evaporator) 202 can be any collector (and evaporator) whose heat absorbing panel remains at a temperature below the critical temperature of the refrigerant used. However, the details of the control techniques described hereinafter are tailored to solar collectors which possess the following properties:

(1) all collector (and evaporator) modules are planar and their cover plates lie in one or parallel planes;

(2) the heat-absorbing panels of these modules lose heat to their surroundings at a rate which depends essentially for a given radiation equilibrium temperature t_w of the heat-absorbing panels and the temperature t_a of the ambient air; and

(3) the temperatures of the heat absorbing panels are all essentially equal and uniform over the entire surface of each plate. The first condition simplifies the discussion. The second condition holds for well designed planar collectors. The third condition can be fulfilled easily with the evaporator described under the heading "Solar Collector and Evaporator".

The rate Q_av at which solar heat becomes "available" for absorption by the refrigerant in the solar collector reaches a maximum when the temperature t_w of the absorbing plate is equal to the ambient air temperature t_a. In practice, this uniform temperature t_w will differ insignificantly from the saturated-vapor temperature t_s at which the refrigerant is being evaporated in the solar collector.

The rate Q_av reaches a maximum, for a given value of the equilibrium radiation temperature t_eq, when t_s falls to the point where it becomes equal to t_a. This maximum value of Q_av is given by

Qav =ke (teq -ta), ts ≦ta(1)

In this expression, k_e is a factor which can be determined by calibration tests, and which remains approximately constant over a wide range of temperatures. The equilibrium radiation temperature t_eq as defined previously, is the steady-state temperature of the solar-radiation absorbing panel when no heat is removed from this panel by the refrigerant.

The values of Q_av for values of t_s greater than t_a, are given by

Qav =Ke (teq -ta)-ke (ts -ta)=ke (teq -ts)                                       (2)

The rate Q_abs at which the refrigerant flowing through the solar collector can absorb heat while it is being evaporated is given by

Qabs =hfg mf                                (3)

where h_fg is the latent heat of evaporation of the refrigerant per unit mass and m_f is the refrigerant mass flow rate.

Except at start-up, servo-controlled throttling valve 204 is always set to the "full open" position. The thermodynamic cycle traversed by the state of the refrigerant when this throttling valve is open is shown on the pressure enthalpy diagram of FIG. 5.

When the refrigerant enters (at point 201) solar collector and evaporator 202 in FIG. 4, it is in state A in FIG. 5. Upon entering solar collector and evaporator 202, the refrigerant temperature is raised a couple of degrees Fahrenheit and the refrigerant's thermodynamic state crosses the liquid boundary line at B. At exit 203 in FIG. 4, the refrigerant is approximately in state C. Friction in the piping between exit 203 and entrance 205 to condensing coil 206 causes the temperature and pressure to drop slightly at constant enthalpy h_c (see FIG. 5) so that the refrigerant reaches state D. The refrigerant leaves the condensing coil 206 and enters receiver (liquid reservoir) 210 in approximately state E from which it is raised in pressure, at approximately constant temperature, to state A by pump 213. In the case when the medium to be heated is a latent heat reservoir at temperature t_m, its temperature will typically be five to ten degrees Fahrenheit below the saturated-vapor temperature of the refrigerant in the condensing coil 207.

The electric motor 250, driving pump 213, controls the refrigerant mass flow rate m_f so that ##EQU1## and hence ensures, to a high degree of approximation, that

Qabs =Qav 

and also that essentially all the heat absorbed by the refrigerant while it traverses solar collector and evaporator 202, is absorbed in the form of latent heat.

Servo-controlled throttling valve 204 assists in getting the refrigerant, after start up, to traverse a thermodynamic cycle from which it will automatically, after a transient, change over to the final regime whose steady state is represented by the cycle described earlier. To this end, a fixed-logic central control unit, which I shall refer to as the CCU, is used to select the appropriate initial partially open position of the servo-controlled throttling valve 204. When only the Rankine heat transfer cycle is used in a given system, the complexity of the CCU would not exceed that of an inexpensive, hand-held computer. The CCU also selects the initial temperature to be exceeded by the saturated-vapor temperature t_s in solar collector and evaporator 202 before the throttling valve 204 is opened fully.

The electric motor 250, driving pump 213, after start-up and before the servo-controlled throttling valve 204 is fully open, controls the refrigerant mass flow rate m_f so that ##EQU2## The CCU also selects the temperature t_i. A typical value for this temperature is ##EQU3## When t_s exceeds the value (t_i -δ), where δ is a small positive quantity, the said servo starts to control the electric motor 250 so as to approximate the mass flow rate given by the right-hand side of equation (4) instead of that of equation (5).

A typical thermodynamic cycle at the time just before this change over is illustrated by the polygon A'B'C'D'E' in FIG. 5.

A reliable change over to the final regime, and a shorter transient following it, are ensured (i) if the initial regime reaches a steady state condition before changing over to the final regime, and (ii) if the saturated-vapor temperature of the refrigerant in the condensing coil exceeds that of t_m. To this end, the change over to the final regime is delayed long enough, after t_s has exceeded (t_i -δ), to ensure that the initial regime has reached a quasi steady-state condition, and the initial opening of the servo-controlled throttling valve 204 commanded by the CCU is, together with the chosen value of t_i, sufficient to ensure that the saturated-vapor temperature of the refrigerant in the condensing coil is higher than the temperature t_m of the medium to be heated.

Referring to the servo-controlled electric motor 250, the desired value m_f.sup.(d) of m_f after start up and before switchover is obtained by instrumenting equation (5) and, after switchover, by instrumenting equation (4) using multiplier 251. In either case the actual value m_f.sup.(a) of m_f is obtained by using multiplier 252 to instrument equation ##EQU4## where F_f is the refrigerant volumetric flow rate, as given by flow transducer 216, and v_f is the specific volume of the refrigerant. The expression k_e /h_fg and l/v_f are only slowly varying functions of the refrigerant pressure p. The pressure p, as measured by pressure transducer 253, provides in general, a close enough approximation to the refrigerant's pressure at transducer entrance 215 as well as of the refrigerant pressure near exit 203, for computing the expressions k_e /h_fg and 1/v_f supplied to multipliers 251 and 252, respectively. The pressure p is supplied to the CCU as indicated by symbol 254. The CCU computes the expressions k_e /h_fg and l/v_f as a function of p and supplies them to multipliers 251 and 252, respectively, as indicated by symbols 255 and 256. The servo controls the speed of motor 250 so as to tend to annul the servo error signal

ε=mf.sup.(d) -mf.sup.(a).

To this end, the function KG , respresenting the transfer function of servo-amplifier and frequency-dependent network 257 must include an integrator. A typical expression of KG is given by ##EQU5## where K is the amplifier gain, τ a smoothing constant and the Laplace transform of the error signal ε(t) supplied to the servo amplified and frequency-dependent network 257.

The quantities (t_eq -t_i) and (t_eq -t_s), used as appropriate, in deriving the servo reference signal representing the quantity m_f.sup.(d), are derived as follows. The solar collector radiation equilibrium temperature t_eq is obtained by using temperature transducer 258 as explained in conjunction with the description of the solar collector. The temperature t_m of the medium to be heated is obtained by using temperature transducer 259. ##EQU6## is instrumented by using adder 260 and divider 261. The saturated vapor temperature t_s of the regrigerant corresponding to a given pressure is obtained from the pressure p (as measured by pressure transducer 253) by instrumenting the single-valued function

ts =ts (p)                                       (9)

by means of function generator 262. This function can be derived from published tables or graphs for any one of the commonly used refrigerants. The differences (t_eq -t_i) and t_eq -t_s, respectively, by using adders 263 and 264, respectively.

The system employed to determine which of the foregoing two differences should be used consists of four devices: latching relay 265, changeover control relay 266, diode rectified 267, and latch-control relay 268. After start-up, and before changing over to using (t_eq -t_s), the relay contracts are in the position shown in FIG. 4. When the signal representing the quantity t_s -(t_i -δ), which is formed by using adders 269 and 270, becomes positive, diode rectifier 267 allows this signal to pass causing changeover control relay to allow this signal to be supplied to latching relay 265. This causes the positive supply voltage+V_s to be connected to the coil of latching relay 265, which, in turn, causes this relay to changeover to the latched position. In this position, latching relay 265 stops the signal representing the difference (t_eq -t_s) from being fed to multiplier 251, and substitutes the signal representing the quantity (t_eq -t_s). The purpose of the supply voltage supplied to latching relay 265 is to hold this relay in the latched position when the value of t_s falls below (t_i -δ) after changeover and relay 266 returns to its initial position (which is the one shown in FIG. 4). This purpose is achieved as long as latch control relay 268 is in the position that provides a closed path between this supply voltage and earth through the coil of latching relay 265. This closed path exists when latch control relay 268 is in the position shown. When the Rankine-type heat transfer cycle is stopped or used to heat a medium at a higher temperature to the one being heated, the CCU provides as indicated by symbol 271, a control signal +v_c which opens latch control relay 268 and causes latching relay 265 to return to its initial position.

The low-limit float switch 272 interrupts the power supply _s to motor 250 whenever the level of the liquid refrigerant is receiver 210 falls below a minimum preselected level. This prevents pump 213 from running dry.

If the solar collector consisted of modules lying in more than one plane, nearly all the components shown in FIG. 4 would have to be duplicated for each different plane.

In FIG. 4 none of the components which might be desirable for calibration or checking and servicing are shown. For example, most actual installations would include a temperature transducer to measure the working fluid temperature near exit 203. This temperature can be compared with t_s to determine the existence of undesirable superheat. This temperature transducer can also be used to help estimate the quantity of the refrigerant vapor at exit 203 when the refrigerant leaving this exit is not superheated. It will be obvious to those skilled in the art how a small signal can be added to the signal controlling motor 250, and how the amount of this bias required to cause a detectable amount of superheat to appear at exit 203 can be used to estimate the quality of the refrigerant at this exit. Also, by allowing a measurable amount of superheat to exist at exit 207 during normal operations after steady state conditions have been reached in the final regime, the servo controlled-motor 250 can, during this regime, be driven by the error signal

ε=Δsp.sup.(d) t-Δsp.sup.(a) t, (9a)

where Δ_sp.sup.(d) t and Δ_sp t.sup.(a) are the preselected small amounts of superheat used as the reference signal and the measured amount of superheat, respectively. This method of controlling motor 250 would cause a small degradation in heat-transfer efficiency but would allow less accurate values of m.sup.(d) and m.sup.(a) to be used.

Finally, in inexpensive systems, and where a positive-displacement type pump is used for pump 213, the signal from a suitably calibrated tachometer, driven by motor 250 may be used to provide a measure of the volumetric flow rate F_f instead of flow transducer 216.

If the centerlines of all the refrigerant piping are in the same horizontal plane, the cycle, once started, is self-sustaining in the absence of friction, and needs no pump. Under these ideal conditions the cycle also operates optimally without the help of a pump. As differences in elevation of the system components increase, or friction is increased by using smaller diameter piping, etc., the amount of heat transferred from the solar collector decreases. Further, if the elevation difference is excessive, the cycle is no longer self-sustaining. Although a feed pump can be used to offset the degradation in performance arising from differences in elevation and from friction, some residual degradation remains. Differences in elevation and the amount of friction should therefore be kept as small as possible.

In the absence of friction and differences in elevation ##EQU7## Consequently, t_s approaches t_m as the rate k_e /k_c tends to infinity. Since

Qav =ke (teq -ts),                     (2)

it also follows that, for a given heat transfer capacity k_e of the solar collector, the rate at which heat can be transferred from the solar collector, for given values of t_eq and t_m, increases with the heat transfer capacity k_c of the condensing unit. This conclusion holds in the presence of differences in elevation and friction. However, the value of Q_av will in this case be smaller than the one given by equation (10).

SOLAR-ASSISTED, VAPOR-COMPRESSION HEATING CYCLE

The thermodynamic cycles commonly referred to as a vapor-compression, heat-pump cycle and as a vapor-compression refrigeration cycle both belong to the same class of thermodynamic cycles: they both use fluids whose state alternates once between the liquid and vapor phase during a cycle, they both compress the fluid once during a cycle when the fluid is in its vapor phase, and they are both employed to absorb heat from one medium and to release heat to another medium at a higher temperature. The foregoing two cycles, in essence, only differ in their purpose. Namely, the purpose of the heat pump cycle is to supply heat to the medium at the higher temperature, and the purpose of the refrigeration cycle is to remove heat from the medium at the lower temperature. To help clarify the subsequent description and discussion of the novel control methods of the present invention, I shall use the term vapor-compression heating cycle to refer to the former, the term vapor-compression cooling cycle to refer to the latter, and the term binary-phase vapor compression cycle to refer to both. I shall also refer to the equipment used to instrument the first cycle as a heat pump, and the equipment used to instrument the second cycle alone or both cycles as thermal pumps.

The conventional control systems of heat pumps are designed to match the heating load by cycling compressors on-and-off either simultaneously or sequentially in steps. In either case, the "start" and "stop" signals originate either in one or more thermostats in the conditioned space, or in devices which sense the temperature of the fluid used to heat the conditioned space.

In large custom-engineered systems, the capacity of individual compressors, if variable, is also varied in response to changes in the heating load detected indirectly by temperature measurements of the conditioned space or of the fluid used to heat that space. The saturated-vapor temperature (or corresponding pressure) of the working fluid in the evaporator is not controlled independently by such control systems. Instead this temperature takes on--with varying heating-load requirements--values consistent with these requirements and the temperature of the medium from which free thermal energy is extracted.

The foregoing indirect measures of heating load do not provide means for controlling optimally the state of the working fluid during the heating cycle under large changes in heating load, and this especially true in the case when these changes in load occur rapidly. Furthermore, these indirect measures cannot be employed when the medium to be heated is a latent heat reservoir. Nevertheless, the conventional control systems discussed hitherto--although not optimal--are, in most cases, acceptable where the maximum rate at which heat can be extracted from the free source of thermal energy is not the factor governing maximum system heating capacity, or when no large-capacity heat reservoir is used. Both of these conditions hold in most conventional vapor-compression heating systems. Namely, these systems emply no large-capacity heat reservoir, and the factor limiting maximum heating capacity is, in general, always compressor capacity.

Solar-assisted heat pumps must use large-capacity heat reservoirs to provide large reductions in non-free (purchased) energy and most such reservoirs store latent heat so as to save space. Furthermore, for such heat pumps, the factor governing maximum heating capacity is often the rate Q_av at which solar heat is available and can be removed from the solar collector, and is not compressor capacity. Under these conditions, a control system designed to attempt to match the heating load would nearly always be pursuing the wrong objective. However, even when this objective is the right one, a different control system from the one used hitherto in conventional heat pumps or solar assisted heat pumps is desirable. This is also true for heat pumps using atmospheric heat as the source of free energy whenever they employ a heat reservoir.

The novel control system for the vapor-compression heating cycle of this invention is based on a cycle using "optimal subcooling." The term "optimal subcooling" is used to refer to the case where the temperature of the refrigerant leaving a subcooler following the condenser is essentially the same as the temperature of the refrigerant entering the evaporator of the solar collector. Such subcooling is practicable if the temperature of the solar collector working fluid is, say, at least 20° F. above the temperature of the ambient air. A slightly modified control system applicable to cases when the foregoing condition does not hold is also contemplated and is discussed later. The novel control system for the vapor-compression cycle using optimal subcooling ensures automatically that the refrigerant vapor always leaves (the evaporator in) the solar collector dry, and enters it in the fully liquid state. The former condition is also satisfied by conventional control systems at the expense of accepting whatever evaporator pressure p_e (and hence also saturated-vapor temperature t_es) is necessary to satisfy this condition. In the control system described here, the value of p_e (and hence the corresponding value of t_es) can be chosed independently. This cannot be done with conventional control systems.

The desuperheating and condensation takes place in a separate unit from subcooling. I shall refer to the former as the condenser and the latter as the subcooler. The control system can adjust the heat released per pound of refrigerant by the condensing coil precisely by adjusting the condenser temperature differential The control system performs this adjustment by selecting precisely the condenser pressure p_c. The subcooler is a forced-air condenser whose coil is cooled by entering ventilation air. The subcooler capacity is adjusted to provide the precise degree of subcooling needed by varying the amount of ventilation air bypassing the coil.

An examination of the "ideal" pressure-enthalpy diagram shown in FIG. 6A reveals the following facts:

(a) the complete cycle consists of four lines;

(b) the length, as well as the location, of the line

p=pe 

is determined by choosing the value of p_e since the control system controls the state of the refrigerant so that the line (segment) starts on the liquid boundary line and ends on the vapor boundary line.

(c) the choice of the value of p_c not only determines the location of the line

p=pc,

but also determines automatically the length of the segment, CF, and the lengths of the remaining two segments BC and FA, where BC is the isentropic compression line and FA is the constant-enthalpy throttlin line. It follows that the ideal cycle is determined unambiguously by the selection of the pressure p_e and p_c.

The foregoing statements remain true in essence in a "real" cycle, namely, one which the refrigerant would undergo in an actual (physical) installation. Such a cycle is shown in FIG. 6b.

The values of the pressures p_e and p_c are selected by the control system so that the desired value Q_e.sup.(d) of Q_e and the desired value Q_c.sup.(d) of Q_c are achieved, where Q_e is the rate at which the refrigerant in the solar collector and evaporator absorbs solar heat, and where Q_c is the rate at which the refrigerant in the condensing coil releases heat to the medium to be heated.

When the rate Q_c at which the refrigerant in the condensing coil releases heat to this medium can be controlled independently of the condenser temperature differential, the pressures p_e and p_c can vary independently. However, when the heat Q_c released to the medium can be controlled only by varying the condenser temperature differential, as for example, in most cases where the condensing coil is used to heat the storage medium of a latent heat reservoir, these two pressures cannot vary independently. The novel controls described next apply to both cases, except where otherwise explicitly stated.

These novel controls will, in general, receive certain inputs from a unit we shall refer to as the Central Control Unit (CCU) which may be a simple fixed-logic or small (pocket-size) programmable special-purpose computer.

FIGS. 7A and 7B show a method of implementing the novel control system of this invention in the case where a latent heat reservoir is used.

The refrigerant enters at 301 the solar collector (and evaporator) 302 in state A (see FIGS. 6A and 6B) and exits at 303 in state B. In this state, the refrigerant enters at 304 compressor 305 and exits at 306 in state C, passes through flow transducer 307, and enters in this same state, at point 308 condensing coil 309 embedded in storage medium 310 of latent heat reservoir 311. It exits condensing coil 309, at point 312, in state E, and enters, in this same state, subcooler coil 314 over which flows ventilation air 315. The refrigerant exits subcooler coil 314 at point 316 in state F and, after passing through throttling valve 317, exits at point 318 in approx. state A. After passing through receiver (liquid reservoir) 319, it enters pump 321 at point 320 and exits from this pump at point 322 again in approx. state A. From this point, the liquid refrigerant is returned to inlet 201 of the solar collector (and evaporator).

The control system for the vapor compression heating cycle of this invention consists essentially of four servos.

The purpose of the servo controlling the electric motor 323 which drives pump 322 through shaft 324, is to ensure that the evaporator pressure p_e assumes the value corresponding to the desired (chosen) value of Q_e and also to ensure that the refrigerant entering the evaporator at point 301 is in state A by causing the mass flow rate at which the liquid refrigerant enters at point 301 to match (namely to be equal to) the mass flow rate at which the refrigerant is evaporated in collector 302 by solar heat.

The purpose of the servo controlling electric motor 325 which drives compressor 305 through shaft 326, is to adjust the capacity of the compressor (only speed control shown) so that the mass rate at which refrigerant flows through the compressor matches the rate at which the refrigerant is being evaporated, and to ensure that dry vapor enters the compressor at point 304. To this end, the servo is controlled so as to provide a few degrees (say 5° F.) of superheat Δ_sp t_e. (This superheat is not shown in the pressure enthalpy diagrams of FIGS. 6A and 6B because it is insignificant compared to the amount of superheat of the refrigerant at the compressor exit 306, and would distort the diagram.)

The purpose of the servo controlling the throttling valve 317 is to ensure that the condenser pressure p_c assumes the value corresponding to the desired (chosen) value of Q_c. The purpose of flow transducer 307 is to measure the mass flow rate of m_g of the working fluid in its vapor phase. The value of m_g is used by the servo controlling the throttling valve 317 in one of the embodiments of this servo.

The purpose of the servo controlling damper 327, whose position determines the amount of ventilation air 315 passing over coil 314, is to ensure that the liquid at exit point 316 is subcooled by the desired amount.

Referring to the servo motor 323, the actual value Q_e.sup.(a) of Q_e corresponding to a given value of the evaporator pressure p_e is derived from the evaporator capacity coefficient k_e, the refrigerant saturated vapor t_es in the evaporator, and the radiation equilibrium temperature defined earlier. Multiplier 350 and adder 351 are used to instrument the equation

Qe.sup.(a) =ke (teq -tes).             (11)

The value of the temperature t_es is derived by using the equation

ts =ts (p)                                       (12)

which expresses the functional relationship of the saturated vapor temperature t_s that corresponds to the refrigerant pressure p. This relation can be deduced, for any gven refrigerant, from published graphs and tables. The value of the radiation equilibrium temperature t_eq is measured by temperature transducer 353 located in dummy collector 354. The pressure p_e.sup.(a) is measured by pressure transducer 355 and the corresponding saturated-vapor temperature t_es is obtained from it by using function generator 356 to instrument the functional relationship t_s t_s (p), which, as mentioned earlier, can be derived from published tables. Adder 351 and multiplier 350 are used to instrument the right hand side of equation (11) and thus obtain Q_e.sup.(a). The value k_e is provided by the CCU, as indicated by symbol 357. The value of Q_e.sup.(d) is provided either directly by the CCU, or indirectly by it in the manner described later. This fact is indicated by symbol 358. The value of k_e is computed by the CCU as a function of t_eq and t_s. This function, which is only a slowly varying function of t_eq and t_s can be determined by calibration tests and preprogrammed. The fact that the values of t_eq and t_es are supplied to the CCU is indicated by symbols 359 and 360, respectively.

The value Q_e.sup.(a) obtained, as already described is subtracted from Q_e.sup.(d) by using adder 361. The resulting servo error

ε1 =Qe.sup.(d) -Qe.sup.(a)

controls the speed of motor 323, via servo amplifier and frequency-dependent network 362, so as to cause ε₁ to tend to zero. To this end, the function K₁ G₁ (s) must include an integration. A simple typical expression for this function is ##EQU8## where k₁ is the amplifier gain τ₁ is a smoothing constant and s is the laplace transform of the error ε₁ (t).

The low limit float switch 363 is used to interrupt the power supply _s to motor 323 whenever the level of the liquid receiver 319 falls below a minimum preselected level. This prevents pump 321 from running dry.

Referring to the servo controlling electric motor 325, the desired value Δ_sp.sup.(d) t_e of refrigerant superheat, at a point in the vicinity of exit 303, is used as the reference signal of this servo and is preselected. If different preselected values of Δ_sp.sup.(d) t_e are desired for different operating conditions, they are provided by the CCU, as indicated by the symbol 368. The actual amount of superheat Δ_sp.sup.(a) t_e is derived from the refrigerant pressure p_e.sup.(a) as measured by pressure transducer 355 and from refrigerant temperature t_e.sup.(a) as measured by temperature transducer 364. The saturated vapor temperature t_es corresponding to the pressure p_e.sup.(a) is then subtracted from t_e.sup.(a) by using adder 365. The value of the saturated vapor temperature t_es, corresponds to the measured pressure p_e.sup.(a), is obtained by using function generator 356 to instrument the functional relation t_s =t_s (p) between the value of the saturated vapor t_s corresponding to a given pressure p. This functional relation can (as mentioned earlier) be deduced, for any given refrigerant, from published graphs and tables. Pressure transducer 355 and temperature transdcuer 364 are both located in the vicinity of exit 303. The difference ##EQU9## obtained by using adder 366, is used as the servo error signal. This signal, by means of servo-action, controls the speed of electric motor 325, via servo amplifier and frequency dependent network 367. The capacity of the compressor 305 can also be varied directly by such mechanisms as cylinder unloading, in conjunction with or in addition to motor speed control.

To the end of controlling the motor speed, the transfer function K₂ G₂ (s) must also include an integration. A typical simple expression for K₂ G₂ (s) is given by ##EQU10## where the symbols have a similar meaning to those given in equation (13).

Referring to the servo controlling the throttling valve 317, the actual value Q_c.sup.(a) of Q_c corresponding to a given value of the condenser pressure p_c.sup.(a) is derived from the condenser capacity coefficient k_c, the refrigerant saturated vapor temperature t_cs in the condenser, the amounts of refrigerant superheat Δ_sp t_c in the vicinity of the entrance 308 to condensing coil 309, and the temperature t_m of the medium to be heated. Multiplier 369 and adders 370 and 371 are used to instrument the equation relating to these quantities; namely,

Qc.sup.(a) =0.95kc [(tcs -Δsp tc)-tm ]. (16)

The value of t_m is measured by temperature transducer 372. The value of Δ_sp t_c is derived from the measured value t_c of the refrigerant temperature in the vicinity of entrance 308, the refrigerant constant pressure specific heat c_pg, the condenser capacity coefficient k_c, the saturated vapor temperature t_cs, and the mass flow rate m_g of the refrigerant at some point between the compressor exit 306 and the condensing coil entrance 308. Adder 373 and multipliers 374 and 375 are used to instrument the equation relating Δ_sp t_c, t_c, c_pg, k_c, m_g and t_cs ; namely, the equation ##EQU11## where in turn c_pg can be found in published tables for a given refrigerant and temperature range, t_c is measured by temperature transducer 376, k_c is known from design or calibration data, and m_g is, as explained later, obtained from flow transducer 307. As indicated by symbol 377, the quantity c_pg /0.95k_c is supplied by the CCU and the value of t_cs is obtained from the value of p_c.sup.(a), as measured by pressure transducer 378, by means of equation

ts =ts (p)                                       (12)

Equation (12) is instrumented using function generator 379, which is identical to functional generator 356.

The value of Δ_sp t_c can, alternatively, be obtained by instrumenting equation ##EQU12## where p_c and p_e are measured by the same means used in instrumenting equation (17); where c_pg is found as mentioned earlier; and where ##EQU13## where, in turn α₂ and α_cp are the angles defined in FIG. 6A. The expressions (19) and (20) are computed by the CCU. The functions α₂ (p) and α_cp (p) can be deduced from published pressure-enthalpy diagrams for a given refrigerant.

The value of Q_c.sup.(a), obtained as described, is subtracted from the desired value Q_c.sup.(d) of Q_c, which is, as indicated by the symbol 379, supplied by the Central Control Unit (CCU) (either directly or indirectly) by using adder 380. The resulting servo error signal ε₃, drives throttling valve 317, via servo amplifier and frequency-dependent network 381, represented by transfer function K₃ G₃ (s), so as to make Q_c.sup.(a) tend toward Q_c.sup.(d). In this case, the transfer function does not include an integrator. A typical simple expression for this transfer function is

K3 G3 (s)=K3 (1+τ3 s),             (21)

where the symbols have a similar meaning to those given in connection with equation (13).

Referring to the servo controlling damper 327, the desired value t_sb.sup.(d) for the reference servo signal is, for the case of full subcooling, equal to the saturated vapor temperature t_es of the evaporator plus, if desired, some small margin Δt_es (not shown in FIG. 7). Δt_es is a positive quantity. The servo error signal ε₄ is obtained by substracting the actual value t_sb.sup.(a) of t_sb, as measured by temperature transducer 381, from t_sb.sup.(d) using adder 382. The error signal drives, via servo amplifier and frequency-dependent network 383, represented by transfer function K₄ G₄ (s), damper 327 (which may contain its own small electric motor) in such a way as to make t_sb.sup.(a) tend toward t_sb.sup.(d). Again, no integration is contained in the transfer function, and a simple typical expression for this transfer function is

K4 G4 (s)=K4 (1+τ4 s),             (22)

where the symbols have a similar meaning to those given in connection with equation (13). All four of the transfer functions mentioned may contain higher order terms. However, the first two must give a constant output whenever the servo error signal is zero, whereas the last two must give a zero output whenever the servo error signal is zero.

The control system discussed so far allows the CCU to select Q_e and Q_c while maintaining the desired degree of superheat at exit 303 and "full subcooling", as defined earlier. The values of Q_e and Q_c cannot, howver, be selected independently. The CCU must therefore select compatible values of Q_e and Q_c. More specifically the foregoing two quantities are related by

Qc =Qe -Qsb +Hcp +Qex,            (23)

wheren Q_sb is the rate at which heat is released by the subcooler, H_cp is the rate at which enthalpy of the refrigerant is increased in the compressor, and Q_ex is the net amount of heat supplied to the refrigerant by the surroundings, excluding Q_c, Q_e and Q_sb, but including the heat from the electric motor(s) driving the compressor(s) where hermetically sealed compressors are employed. One major component of H_ex in poorly insulated installations is the heat entering or leaving the refrigerant piping. The value of H_ex, whenever it is significant, can be obtained from calibration tests and programmed in the CCU. However, because the servo controlling the subcooling rate Q_sb is controlled independently of the other three servos, any small error made in determining the values of Q_c, Q_e, H_cp and Q_ex, are compensated automatically by the servo controlling the subcooling rate.

It follows from the foregoing discussion that, in addition to obtaining Q_ex by calibration tests in cases where the value of this quantity is significant, the values Q and H_cp must also be computed. To this end, I use the expressions

Qsb =m(pc -pe).tan α1            (24)

and

Hcp =m(pc -pe)tanαcp             (25)

In theory, the value of m in equation (24) should be obtained from a flow transducer located between the condensing coil and the subcooler, and the value of m in equation (25) should be obtained from the transducer located between the compressor and the condenser. The system can be designed, however, with small enough transients for the value of the mass flow rate m_g, obtained from flow transducer 307, to be used to instrument both equations. This is the case for the system shown in FIGS. 7A and 7B.

Equation (24) is instrumented by using adder 384 and multipliers 385 and 386. Equation (25) is instrumented by using adder 384 and multipliers 387 and 388. The quantities tan α₁, tan α_cp are supplied by the CCU as indicated by symbols 389 and 390, respectively. The quantity ##EQU14## and the quantity tan α_cp is given by equation (20). The quantity m_g is obtained from the volumetric flow rate F_g, supplied by flow transducer 307, by multiplying F_g by 1/v_g, where v_g is the specific volume of the refrigerant.

The quantity 1/v_g is computed by the CCU as a function of t_c and p_c.This function can be deduced from published graphs and tables and preprogrammed. The symbols 391 and 392 indicate that t_c and p_c are supplied to the CCU, and symbol 393 indicates that 1/v_g is supplied by the CCU to multiplier 394.

Before proceeding to discuss the instrumentation of equation (23), it is important to recognize that there exists a large number of different ways of implementing the control concepts for the vapor compression heating cycle illustrated in FIGS. 7A and 7B. Two alternative ways of expressing and instrumenting the quantity Δ_sp t_c have already been mentioned. An example of an alternative way of instrumenting another quantity, namely the refrigerant mass flow rate m, can be mentioned. This rate could have been obtained without flow transducer 307 and the associated computation, by instrumenting the equation ##EQU15## where h_fg (p_e) is the latent heat absorbed by the refrigerant in the evaporator when "full subcooling" is used. Hence, it should be apparent to those skilled in the art that there are many techniques for instrumenting the control featurs of the instant invention without departing from my inventive concepts.

Turning to equation (23). Except for the quantity Q_ex, which is fixed by the design, and cannot be changed during operation, the CCU could, in principle, choose according to programmed rates the value of any one of the remaining quantities in this equation. In practice, Q_sb would rarely be chosen as the independent variable. Briefly, if it is desired to change the amount of heat released to the entering ventilation air, one would normally do this by supplementing Q_sb by other means. Also since H_cp represents approximately the work done by the compressor(s), constraints may be imposed upon the value of H_cp, which can be accomplished quite simply with the type of control system presently discussed, but one would not normally wish to choose H_cp as an independent variable whenever compressor capacity can be changed continuously--as assumed heretofore. Often, however, compressor capacity can only be changed in steps. In this case, the CCU can be programmed to select the capacity step that best meets a desired objective expressed in terms of another variable, such as Q_e ; and, after this has been accomplished, the particular value of H_cp corresponding to that step can be used as the independent variable, and the quantities Q_e, Q_c, Q_sb, controlled so as to assume values consistent with that particular value of H_cp. One may wish, depending on the application to choose either Q_e or Q_c as the independent variable, or one might also wish to choose the rate Q_h defined by

Qn =Qc +Qsb,

or the theoretical coefficients of performance ##EQU16## or the actual coefficients of performance ##EQU17## as the independent variable, where P_i may be either the electrical power consumed by the compressor(s), or by the compressor(s) and the pump(s) and possibly the fans (blowers) associated with a system.

The diagrams in FIGS. 10A-C show how Q_e, Q_c and Q_h /H_cp can be chosen as the independent variable. The method for choosing the remaining three quantities, or any other quantity from which either Q_e.sup.(d) or Q_c.sup.(d) can be derived, should be obvious to those skilled in the art.

The foregoing novel controls assume that the solar collector uses the type of evaporator described under solar collector and referred to therein as "the evaporator". This evaporator resembles functionally the type of evaporator referred to by those skilled in the art as a "flooded evaporator". The foregoing control apply, in essence, to any flooded evaporator whose source of heat is radiant energy.

In some installations employing multiple flooded evaporators it may be necessary to add individual low-pressure float controls fed by a single pump, or float switches controlling separate liquid pumps supplied from the same receiver (liquid refrigerant reservoir). In particular, such float controls will probably be required in large solar collector installations for each row or group of rows of collector modules.

AIR-ASSISTED, VAPOR-COMPRESSION HEATING CYCLE

The foregoing discussion under the heading SOLAR-ASSISTED, VAPOR-COMPRESSION CYCLE applies, in essence, also to the air-assisted, vapor-compression cycle with the following exceptions: (1) the rate Q_e at which heat can be absorbed by the device "collecting" atmospheric heat does not have a precise upper limit; instead it can be increased far beyond the system design capacity which is limited primarily by compressor capacity; and (2) "full subcooling" as defined cannot be achieved.

The description of the novel controls for the solar-assisted, vapor-compression cycle also applied in essence, to the air-assisted, vapor-compression heating cycle with the following exceptions: (1) the value of Q_e.sup.(a) is obtained by instrumenting the equation

Qe.sup.(a) =ke (ta -tes)

instead of equation (11), where t_a is the temperature of the air surrounding the device collecting atmospheric heat and the other symbols have the same meaning as those given in connection with equation (11); (2) the value of k_e usually remains constant within the normal operating range of the equipment (as governed primarily by compressor capacity); and (3) the reference signal for the servo controlling damper 327 in FIG. 7B is no longer t_es but some other attainable higher temperature, say 15° F., above the temperature t_a, whose precise value depends on system design and operating conditions and is supplied by the CCU.

The foregoing applies equally to any vapor-compression heating cycle in accordance with the present invention which obtains its thermal energy from a medium at a uniform temperature in which the "heat-collecting" device is embedded or immersed, such as in a lake or artesian well, provided the temperature of this medium is substituted for the temperature t_a of the air.

COMBINED RANKINE POWER CYCLE AND VAPOR-COMPRESSION COOLING CYCLE

A heat engine powered by solar radiation and employing a refrigerant as a working fluid is old, and so is the use of such an engine to drive a vapor-compression cooling cycle. However, no one prior to the instant invention has conceived an arrangement for using a solar radiation powered heat engine to drive a vapor-compression cooling cycle in which (1) a single (binary-phase) working fluid and in particular, a refrigerant is shared by both the engine's Rankine Power cycle and the cooling cycle; (2) a single condensing unit is shared by both cycles; (3) the radiation equilibrium temperature of the heat absorbing panel is used to predict the available engine power for a given installation, (4) the control system ensures, in the particular manner described later, that the working fluid and, in particular, refrigerant passing through the heat engine is preheated before it is returned to the solar collector by the amount required to cause the temperature of this refrigerant to be almost equal to that of the saturated-vapor temperature in the evaporator before the refrigerant is returned to the evaporator; (5) the control system ensures, in essence, that only the heat absorbed by the working fluid and, in particular, the refrigerant at the uniform temperature of the solar collector is supplied to the heat engine, and that the remaining working fluid bypasses the engine thus requiring a smaller engine to obtain the same performance for a given amount of solar heat absorbed, and (6) the control system allows the saturated-vapor temperature t'_es of the refrigerant in the solar collector, and the refrigerant mass flow rate m_p through it, to be chosen independently of the degree of insolation so as to optimize selected performance objectives (criteria). After reading the discussion of the vapor-compression heating cycle given earlier it will be obvious to those skilled in the art how with the type of controls described in the present invention, the combined power and cooling cycle can be controlled in other ways to achieve a variety of objectives similar to those requiring for example, a given value of cooling rate, or the maximization of the coefficient of performance of the combined cycle. The number of possible alternative objectives however is greater here because of the presence of two cycles instead of one which may, within limits, be controlled independently. In effect, to this end, one need only derive similar expressions and impose applicable constraints--in the manner done earlier--by examining the applicable pressure-enthalpy diagram.

The system shown in FIG. 8 illustrates the particular case where the objective chosen is the maximization of the thermodynamic enthalpy rate H_Eng of the heat engine, whenever its magnitude exceeds a minimum pre-assigned value, in the simple case where the vapor-compression cooling cycle is not controlled independently, and the refrigerant flow rate through the cooling cycle evaporator is controlled by a thermostatic expansion valve in a conventional manner.

Each of the foregoing six features are accomplished by the present invention.

Referring now to FIG. 8, the refrigerant enters solar collector 402 at point 401. After the refrigerant exits at point 403 it is apportioned at point 404 in part by pumps 405 and 406, and in part by flow control means described later, between heat engine 407 and heat exchanger 408. The fraction xm_p of the refrigerant flow rate m_p through the solar collector apportioned to the heat engine 407 flows through the outdoor condenser 430, the receiver 451 and is returned by pump 405 to solar collector 402 after passing through flow transducer 409 and through heat exchanger 408, entering at point 452 and existing at point 453. The fraction (1-x)m_p of the refrigerant apportioned to heat exchanger 408, after traversing it, flows through receiver 454 and is returned to solar collector 402 by pump 406 after passing through flow transducer 415. The receiver 451 supplies, in addition to pump 405, expansion valve 455, evaporating coil 456, and compressor 425, of the vapor-compression cooling cycle. As will be appreciated, a conventional vapor compression cooling cycle is employed in this illustrated embodiment of the invention and includes expansion valve 455, evaporating coil 456 of evaporator 461, compressor 425, outdoor condenser 430 and receiver 451. Thus condenser 430 serves both the Rankine power cycle and the cooling cycle simultaneously. The refrigerant, after passing through the compressor 425, merges with the refrigerant traversing engine 407 at point 457 and is returned to receiver 451 after passing through condenser 430.

The fraction (xm_p) of the total refrigerant flow rate m_p through the solar collector which is apportioned to the heat engine is determined by instrumenting the equation

[(1-x)mp ].sup.(a) =(xmp).sup.(a) ·cpf ·(tes '.sup.(a) -tcs '.sup.(a))        (28)

solved for x, to obtain the desired value of x.sup.(d) of x, namely ##EQU18## where the quantity (xm).sup.(a) is the actual steady-state mass flow rate apportioned to the heat engine 407, as deduced from flow transducer 409 and the refrigerant liquid density 1/v_f using multiplier 410; c _pf is the average specific heat of the refrigerant flowing through heat exchanger 408, as supplied by the CCU, in terms of temperatures t_es '.sup.(a) and t_cs '.sup.(a), which are in turn obtained from pressure transducers 411 and 412, and function generators 413 and 414; where is the actual steady-state mass flow rate apportioned to the heat exchanger 408, as deduced from flow transducer 415 and the refrigerant liquid-density or density 1/v_f using multiplier 416; and where h_fg is the latent heat per unit of refrigerant mass at the pressure p_e. As in earlier figures, the symbol . indicates throughout that the associated quantity is supplied by, or supplied to the CCU. The applicable alternative is indicated by the direction of the arrow. The left-hand side of equation (28) is denoted hereafter by the symbol Q_ph.sup.(a) and the right hand side by Q_ph.sup.(d). The quantity Q_ph.sup.(a) is obtained from [(1-x)m_p ].sup.(a) by multiplying it by h_fg using multiplier 417. Q_ph.sup.(d) is obtained from t_es '.sup.(a) and t_cs '.sup.(a) using adder 418 to obtain (t_es '.sup.(a) -t_cs '.sup.(a)). This last expression is multiplied first by c_pf, using multiplier 419, and next by (xm_p).sup.(a), using multiplier 420, to obtain Q_ph.sup.(d). The error signal

ε5 =Qph.sup.(d) -Qph.sup.(a)        (30)

is formed using adder 421 and is used to control electric motor 422, via an appropriate servo-amplifier and frequency-dependent network 423, represented by transfer function K₅ G₅ (s). The motor 422 in turn, drives pump 406. The desired value x.sup.(d) of x, as determined by equation (29), is derived by instrumenting this equation, as shown in FIG. 8, using multipliers 425 and 426 and adder 427.

The error signal

ε6 =(xmp).sup.(d) -(xmp).sup.(a)    (31)

is used to control motor 429 via servo-amplifier and frequency-dependent network 460, represented by transfer function K₆ G₆ (s), which in turn drives pump 405. The quantity (xm).sup.(d) is obtained, as shown in FIG. 8 using pressure transducer 411, function generator 413, multipliers 431, 432, 433, and 455, and adder 434 and 456 and divider 457. The error signal ε₆ is obtained by using adder 428.

The compressor 425 is driven by heat engine 407 or electric motor 435 or by both. In FIG. 8, the last of these three cases is illustrated. The rate at which the heat engine can deliver the thermodynamic work H_Eng is given by

HEng =(Pe '.sup.(a) -Pc '.sup.(a))·tan αEng ·(xmp).sup.(a).             (32)

This expression is computed in this particular form of the embodiment in terms of P_e '.sup.(a), P_c '.sup.(a), (xm_p).sup.(a) obtained as described earlier in connection with the vapor-compression heating cycle, and the quantity tan α_Eng computed and supplied by the CCU in the same manner as the quantities tan α₁, tan α₂, tan α_cp, discussed earlier in connection with the vapor-compression heating cycle. The subtraction in equation (32) is made using adder 436 and multiplier 437 and 438.

The value of H_Eng.sup.(a) is zero when

tes '.sup.(a) =teq or tes '.sup.(a) =ta (33)

For a value of t_es '.sup.(a) between these two extreme values, H_Eng.sup.(a) attains its maximum. This value may, for given values of t_eq and the temperature t_a of the ambient air, be a function, of t_es ' alone, or of both t_es ' and t_cs ' varying independently. If the vapor-compression cooling cycle is not controlled independently of the power cycle and if the coefficient of capacity k_c ' of the condenser 430 shared by both cycles cannot be varied, the first alternative applies because t_cs ' cannot be varied independently of t_es '. This is the case illustrated in FIG. 8. In this case, the maximum value of H_Eng is obtained when the derivative of H_Eng with respect to t_es ' is zero. Consequently, to maximize H_Eng, the derivative is obtained by using the differentiator 439 and the resulting quantity is used as the error signal driving electric motor 435 via servo amplifier and frequency-dependent network 440 represented by the transfer function K₇ G₇ (s). This motor regulates, in turn, the speed of compressor 425 and heat engine 407, and thus ensures that this heat engine operates at a speed which results in the maximum value of H_Eng. The temperature t_es '.sup.(a) of the refrigerant in the solar collector is, for given values of t_eq and t_a, controlled by this speed. Whenever the value of H_Eng supplied to the CCU falls below a minimum preselected value, the CCU provides a signal that disconnects, by using a clutch (not shown) or other means, the compressor from the engine drive. Under these conditions, whenever cooling is required, the compressor 425 is driven by motor 435 alone.

It should be apparent from the foregoing discussion of the combined power and cooling cycles how the first and second of the six features listed earlier are accomplished. I now elaborate on the way the remaining three features are accomplished.

The prediction of the available engine power corresponding to a given value of t_eq and t_a depends on the method chosen to control the engine speed and the cooling cycle. In general, the value H_Eng can be predicted by computing in the CCU the value of m_p, corresponding to given values of t_eq and t_a, from ##EQU19## the value h_Eng of H_Eng per unit mass from

hEng =(Pe '-Pc ')·tan αEng (Pe ', Pc ')                                                (35)

and by multiplying m_p by h_Eng to obtain

HEng =mp hEng,                              (36)

However, the value of p_c ' in equation (35) will depend on the particular method chosen to control the engine and the cooling cycle. For the particular method illustrated in FIG. 8, p_c ' can be determined for a given installation by calibration tests as a function of t_eq, p_e ' and t_a. Consequently, in this case, for given values of t_eq and t_a, H_Eng is a single-valued function of p_e ' or equividently of t_es ' and the available engine power can be predicted from H_Eng for any assumed value of t_es ' and, in particular for the value of t_es ' for which ##EQU20## The value of H_Eng thus determined gives the theoretical power available. The corresponding available shaft or brake horsepower can in turn be derived by multiplying this value by the overall engine efficiency.

How the fourth and fifth features are accomplished under steady state conditions and with perfect calibrations should be clear from the earlier discussion. However, to ensure that these two features are accomplished in the absence of perfect calibrations and steady state conditions, additional controls must be used. These controls consist of potentiometers 437 and 438 which add a bias to the voltage controlling the speed of motors 429 and 422, respectively, whenever floats 439 and 440 respectively, deviate from a preassigned reference level. These bias voltages, by slowing down or speeding up the motor they control so as to maintain the level in receivers 451 and 454 approximately constant ensure that the rates of evaporation in condenser 430 and in heat exchanger 408 adjust themselves automatically. Thus, the flow rates (xm_p).sup.(a) and [(1-x)m_p ].sup.(a) satisfy equation (28) at points 441 and 442 before the refrigerant enters receivers 451 and 454, respectively, as well as exit point 443 and 444 after the refrigerant leaves pumps 405 and 406 respectively. Steady state biases in system calibration are compensated by differences in the steady state levels of the liquid refrigerants in receivers 451 and 454. The bias voltage generated by potentiometer 437 is applied to motor 429 using adder 443 and that generated by potentiometer 438 is applied to motor 422.

TYPICAL HEATING AND COOLING SYSTEM FOR BUILDINGS

FIG. 9 illustrates a typical embodiment in accordance with the present invention of a heating and cooling system using all five of the cycles described in the preceding sections. In this particular embodiment, a hot water supply reservoir 509 is heated directly by an immersed condensing coil (not shown), and a high temperature latent heat reservoir 525 and a low temperature latent heat reservoir 543, are respectively, heated and cooled. The reservoirs 501 and 502 may then in turn be used to heat or cool a chosen fluid which distributes heating or cooling, for example, within a building, by techniques presently well known to those skilled in the art.

In the system illustrated in FIG. 9, the hot water supply 509 and the high temperature storage medium 525 can be heated by using either the Rankine type heat transfer cycle, the solar-assisted vapor-compression heating cycle, or the air-assisted vapor compression heating cycle. I shall refer to these cycles as Modes H1a, H2a, H3a and Modes H1b, H2b, H3b, where the letter "a" refers to cycles heating the hot water supply reservoir and the letter "b" refers to cycles heating the high temperature reservoir. I shall refer to the combined Rankine power cycle and air-source, vapor-compression cooling cycle as Mode P & C, and to the latter cycle alone as Mode C.

In the system illustrated in FIG. 9, the evaporator in the solar collector 503 is used to absorb solar heat and, as appropriate, atmospheric heat, as previously described.

In FIG. 9, only the refrigerant circuits (loops) and flow devices are shown. Controls are omitted. The flow transducers are used only in the appropriate modes. The refrigerant circuits used in each mode are indicated by the symbols defined earlier. The pumps are designed by P, the motors driving the pumps by M, the compressor by C, the heat engine by E, the transfer valve by TV and the start-stop valves by SV.

In mode H1a, the refrigerant enters solar collector 502 at point 501 and exits a point 503. Next, the refrigerant traverses (solenoid-actuated) transfer valve 504, exits at point 505, traverses (servo-controller) throttling valve 506 enters transfer valve 507 and exits at point 508, traverses the coil (not shown) of hot-water supply reservoir 509, exits transfer valve 511 at point 513, and enters receiver 514 at point 515.Finally, the refrigerant exits receiver 514 at point 516, traverses valve 517, and is returned by pump 518 to solar collector 502 after traversing flow transducer 519, and transfer valve 520 from which it exits at point 521. Valve 561 is a safety valve.

In mode H1b the refrigerant follows the same path as in mode H1a, except that instad of entering transfer valve 507, it enters transfervalve 522 at point 523, and traverses successively flow transducer 524 the coil (not shown) of indoor condenser and high temperature reservoir 525 and transfer valve 527 from which it exits at point 528, and finally enters receiver 514 at point 529. Thereafter the refrigerant exits this receiver at point 516 and follows the same path back to the solar collector 502 as in Mode H1a.

In mode H2a, the refrigerant exits the solar collector 502 at point 503, traverses transfer valve 504 exiting at point 530, traverses transfer valve 531 existing at point 532, traverses compressor 533, transfer valve 534, 535 and 507, the coil in the hot-water supply reservoir 509, transfer valve 511, throttling valve 530 and enters receier 514 at point 515. Thereafter the refrigerant exits the receiver at point 516 and follows the same path back to the solar collector 502 as in modes H1a and H1b.

In mode H2b, the refrigerant follows the same path as in mode H2a, except that instead of entering transfer valve 507, it passes from transfer valve 535 to transfer valve 522, flow transducer 524, traverses the condensing coil (not shown) of condenser and high-temperature reservoir 525, transfer valve 527 from which it exits at point 537, traverses the coil 538 of subcooler 539, throttling valve 540 and enters receiver 514 at point 529. Thereafter the refrigerant exits the receiver 514 at point 516 and follows the same path as in all the earlier modes.

In modes H3a and H3b, the refrigerant follows the same path as in modes H2a and H2b, respectively. However, whereas in the first three of these six modes the space behind the evaporator is airtight, in the last three modes outdoor air is forced through this space so that the evaporator (not shown) in the solar collector can absorb heat from this air. Air ducts (not shown) are used to channel this air into and out of the solar collector, and dampers (not shown) are used to isolate from or open to the outdoor air the air channel (air chase) 111 shown in FIG. 3b.

In mode C, the refrigerant leaves receiver 514 at exit 541, passes through thermostatic expansion valve 542, condensing coil (not shown) of the condenser and low-temperature heat reservoir 543, transfer valve 531, compressor 533, transfer valve 534, outdoor condenser 544 and returns to receiver 514 at point 545.

In mode C & P, the refrigerant traverses the same path as in mode C, but in addition, the refrigerant also circulates in the Rankine power cycle circuit. Namely, the refrigerant enters solar collector 502 at point 501 and exits at point 503. Next the refrigerant splits at point 550 into two parallel flow paths. One of these paths traverses transfer valve 504, exits at point 551, next traverses heat engine 552, outdoor condenser 544 and enters receiver 514 at point 515. Finally this flow path exits at point 516, traverses valve 517, pump 518, flow transducer 519, transfer valve 521 exiting at point 553, and returns to solar condenser 502 after passing through heat exchanger 554 and exiting at point 555. The second of the two aforementioned parallel flow paths traverses valve 556, heat exchanger 554, receiver 557, pump 558, flow transducer 559, check valve 560 and returns to solar collector 502 entering at point 501. This seconf flow path is only used in the C & P mode, and the refrigerant is prevented from flowing through it in other modes by check valve 560, and by valve 556 which is open only when the C & P mode is being used.

Although not shown in FIG. 9, (a) the heat released by the outdoor condenser can be used to preheat the cold water entering the hot-water supply system, (b) and/or be used to reheat the cool air supplied during mode C & P to obtain additional dehumidification in cases where the latent-to-sensible load ratio is high. It will be obvious to those skilled in the art how either of these two functions can be accomplished by conventional techniques or by using the Rankine-type heat transfer cycle described earlier.

One of the three modes, H1a, H2a or H3a, is selected (automatically by the CCU) whenever the temperature of the water at the bottom of the hot-water reservoir 509 falls below a preselected valve, say 130° F., and whenever the high temperature reservoir does not "require" heat to be supplied to it based upon preselected criteria, such as, the amount of heat then presently stored. Among these modes, Mode H1a is selected whenever the difference between the radiation equilibrium temperature and the temperature at the bottom of the hot-water reservoir 509 exceeds a preselected valve, say, 20° F. Mode H2a is selected whenever this difference falls below this preselected value and the radiation equilibrium temperature exceeds the ambient air temperature. Mode H3a is used when the radiation equilibrium temperature falls below the ambient air temperature and the later temperature is above a preselected temperature, say 40° F.

One of the three modes, H1b, H2b or H3b, is selected whenever the high temperature reservoir "required" heat to be supplied to it based on the aforesaid preselected criteria, such as, the amount of heat then presently stored. Among these modes, Mode H1b is selected whenever the difference between the radiation equilibrium temperature and the temperature of the high temperature reservoir exceeds a preselected value, say 30° F. Mode H2b is selected whenever this difference falls below the preselected value and the radiation equilibrium temperature exceeds the ambient air temperature. Mode H3b is selected whenever the radiation equilibrium temperature falls below that of the ambient air.

One of the two modes, mode C & P or mode C is selected whenever the low temperature reservoir "requires" heat to be removed from it, based on preselected criteria. Mode C & P is selected whenever the difference in temperature between the radiation equilibrium temperature and the ambient air temperature exceeds a preselected number, say, 50° F., for which the heat engine is known to produce useful output. Mode C is selected whenever the foregoing difference is less than the preselected value, say, 50° F.

Many of the novel features of this invention do not depend on the fact that the working fluid is a "refrigerant." The only essential requirement is that the working fluid alternate between the liquid phase and the vapor phase during one complete cycle. In particular, the novel configuration of the evaporator of the solar collector, the novel Rankine-type heat transfer cycle, and the novel technique used in the Rankine-power cycle to apportion the working fluid between a path that transverses the heat engine and a path that by-passes the heat engine, in the manner described herein, can be any fluid whose phase changes during a cycle between the liquid and the vapor phases such as, for example, water.

I wish to again emphasize that the typical embodiments of my invention described herein are provided for illustrative purpose only since many other applications and modifications and control techniques will occur readily to one skilled in the art. Accordingly, it is intended that my present invention not be limited in any way to the specific embodiments described, but rather be limited only as defined in the appended claims.