IMPROVEMENTS RELATING TO THE METERED DISPENSATION OF FLUID

The present invention relates to the metered dispensation of fluid. A preferred application of the invention relates to dispensing nutrients in the art of hydroponics to regulate the concentration of such nutrients in a solution that supports plant growth. The invention may also involve dispensing acid or alkali to control the acidity of a nutrient solution to regulate the pH level of such a solution.

Hydroponics, also known as hydroculture, relates to the growth of plants without the need for soil. Traditionally, soil provides support for the roots of a plant and acts as a reservoir for nutrients. When water is added to the soil, the nutrients dissolve; this enables the roots to absorb the nutrients. Hydroponics relies upon the fact that if the required nutrients are introduced into the water supplied to the roots, soil is no longer required.

In hydroponics, essential nutrients are provided to a plant in a solution in which the plant is grown. The nutrients are typically dispensed into the solution in a closely-monitored series of fluid dosages. The success of hydroponics relies greatly on accurately monitoring and finely adjusting the concentration of nutrient dissolved in the solution and/or the acidity of the solution.

Hydroponics is merely a preferred example of the need to monitor and adjust the concentration or acidity of a solution by periodic controlled dosing of the solution with a solute. Another example is the concentration of disinfectant such as chlorine in swimming pool water. Many other examples will be evident to those skilled in the art, such as drug administration. In each case, a fluid is dispensed in metered fashion from a reservoir and typically enters another reservoir or conduit such as a tank or a tube containing a solution or another fluid with which the first fluid is mixed.

Various different methods are available for the metered dispensation of a fluid. Typically such methods involve the use of one or more donor tanks containing a fluid or fluids to be dispensed. Where the fluid being dispensed is a solute, the fluid is typically dispensed into a receiving tank holding a solution. The concentration of the solute in the solution contained in the receiving tank is monitored. If the concentration of the solute falls below a threshold value, more solute is dispensed from the holding tank into the solution. This

process lends itself to automation, with a feedback loop maintaining the desired concentration of solute within the solution.

Many known dosing systems used in hydroponics include a gravity-fed receiving tank. As shown in Figure 1 of the drawings, a donor tank 4 containing a solute 14 is situated above the receiving tank 5. A feed tube 12 leads from the donor tank 4 to the receiving tank 5. Dispensation of the solute 14 through the feed tube 12 is regulated, by a solenoid valve 50. A sensor 8 immersed in the solution 15 in the receiving tank 5 enables the concentration of solute 14 in the solution 15 to be monitored by an electronic control unit (ECU) 6. The ECU 6 then controls the opening and closing of the solenoid valve 50, allowing solute 14 to flow down the feed tube 12 and into the receiving tank 5 under gravity. The dosage of solute 14 administered to the solution 15 in the receiving tank 5 is regulated by the period of opening of the solenoid valve 50.

A gravity-fed system may have more than one donor tank, each for a respective different solute. In that case, each donor tank has an associated feed tube leading to the receiving tank, and each feed tube is associated with a respective solenoid valve.

Several problems are inherent in the design of gravity-fed systems. The or each donor tank must be situated above the receiving tank. Typically the donor tank must be situated at a height of between 0.1 metres and 1.5 metres above the receiving tank. This can lead to problems of mounting the tanks and, in turn, issues of portability. The donor tank may need to be placed on an expensive and cumbersome floor stand or wall mounted.

Additionally the solenoid valves have parts that are in direct contact with the solute. When the solute is aggressive, for example a corrosive fluid, the parts of the valve in contact with the solute - especially the valve seats - must be resistant to it. This, too, can increase the cost of the system.

Figure 2 shows another known dosing system used in hydroponics. Here, the donor tank 4 is sealed so as to remain airtight and an air pump 20 is connected to the donor tank 4.

Solute 14 within the donor tank 4 passes to the receiving tank 5 via a feed tube 12. As air is pumped into the donor tank 4, the air pressure within the donor tank 4 increases, thus forcing the solute 14 out along the feed tube 12 and into the receiving tank 5. An air gap is maintained between the fluid in the receiving tank 5 and the feed tube 12 to avoid reverse flow of the contents of the receiving tank 5 due to siphoning.

Again, this system has significant disadvantages. The donor tank has to be located at a lower level than the receiving tank end of the feed tube. The air gap between the receiving tank and the feed tube is required to prevent siphoning but any air gap will cause the solute to disturb the surface of the solution when being dispensed. Splash back at the point of impact between solute and solution can be particularly dangerous when dispensing aggressive chemicals. Also, the air pump has to be powerful enough to displace the solute and may be expensive and unreliable. More significantly, the compressibility of air means the system has a sluggish and unpredictable response. Coupled with the slow switching time of an air pump, accurately metered dispensation of the solute is difficult to achieve.

In another dosing system known in hydroponics as shown in Figure 3, a peristaltic pump 51 is used to administer metered dosages of solute 14 to a receiving tank 5. A peristaltic pump 51 is a type of positive displacement pump. The solute 14 is contained within a flexible tube 52 fitted inside the pump casing. A rotor, with a number of rollers attached, is rotated by a motor such that the rollers perform a squeezing action along the length of the flexible tube 52. This causes the solute 14 within the flexible tube 52 to flow in the direction of the rotation.

A peristaltic pump 51 has the advantage that the pump mechanism is not exposed to the solute 14, which is retained within the feed tube 12 as the feed tube 12 passes through the pump 51. However, a peristaltic pump 51 cannot accurately dispense amounts in the order of a tenth of a millilitre unless it uses a specialised and expensive stepper motor drive system in conjunction with a non-return valve 53. This prevents the pressure in the feed tube 12 dropping, thus ensuring the same volume of solute is dosed during each operating cycle. These components increase the cost and complexity of the system - both in purchase and maintenance - and threaten its reliability. In particular, as a peristaltic pump 51 relies upon repeated squeezing of the flexible tube 52 carrying the solute, the pump 51 has to be regularly maintained. For example, a flexible tube 52 under continuous use in a peristaltic pump 51 may have a life as short as three to four months. The peristaltic pump 51 has to be regularly maintained. Grease needs to be regularly applied to the flexible tube 52. Insufficient grease will cause the motor to heat up beyond its operable level, leading to malfunction. The rollers on the rotor may also have to be replaced and the motor may have a lifetime of less than 10,000 hours.

The requirement for regular maintenance of the pump means it has to be installed in a readily-accessible area. Moreover, downtime of the pump and replacement of parts serve to increase running costs. The outlet of the feed tube must be above the solution in the receiving tank with an air gap maintained between the two. If the peristaltic pump stops with the drive axle in the horizontal position then the pump will not act as a nonreturn valve. The air gap prevents solution siphoning back into the donor tank.

In theory, it would be possible to use any other type of pump, such as a centrifugal pump or a diaphragm pump, to dispense fluid in a metered fashion. However, existing pumps all suffer from problems in terms of their inherent cost of purchase and maintenance, their reliability and/or their accuracy in micro-dispensing.

It is against this background that the present invention has been made.

From one aspect, the invention resides in a system for metered dispensation of fluid, comprising:

a conduit for receiving fluid from a reservoir, the conduit having a downstream section for dispensing fluid through a downstream end of the conduit;

an airflow inlet upstream of the downstream section of the conduit; and

an airflow controller for controlling airflow through the airflow inlet;

wherein the airflow controller is arranged to admit air to the conduit through the airflow inlet, the admitted air defining an upstream end of a slug of fluid created in use in the downstream section of the conduit.

The airflow controller may comprise: a valve, pump or actuator; and a control unit that controls the valve, pump or actuator in response to a fluid dispense signal indicating a requirement for fluid to be dispensed. The control unit suitably controls the valve, pump or actuator in response to a feedback sensor that determines a requirement for fluid to be dispensed. For example, the feedback sensor may be arranged to measure concentration of the dispensed fluid in a solution. In that case, in response to a fluid dispense signal from the feedback sensor, the control unit is arranged to determine

inadequate concentration of the dispensed fluid in the solution and, in response, to control the valve, pump or actuator to cause more fluid to be dispensed into the solution.

A fluid presence sensor may be positioned on the downstream section of the conduit downstream of the airflow inlet for controlling operation of the airflow controller in response to presence or absence of fluid in the downstream section of the conduit, wherein the slug of fluid is created in use between the airflow inlet and the fluid presence sensor. Advantageously, the fluid presence sensor can be positioned at different locations along the conduit. This varies the distance between the airflow inlet and the fluid presence sensor, and therefore allows a user to adjust the length of a slug of fluid. The fluid presence sensor is preferably held at a chosen location with respect to the conduit when the fluid presence sensor is released. For example, the fluid presence sensor may surround the conduit, and may be slidably mounted to the conduit whereby, when released, the fluid presence sensor is held at the chosen location by friction with the conduit. The conduit may have indicia along its length against which the position of the fluid presence sensor can be read.

In some embodiments of this invention there may be a speed-measuring arrangement for measuring the speed at which the fluid flows along the conduit in use. The speed- measuring arrangement could comprise a plurality of fluid presence sensors spaced along the conduit. Furthermore, a timer could act on the airflow controller to control the duration of airflow through the airflow inlet.

The conduit may have an upstream section between the reservoir and the airflow inlet. Fluid may be lifted against gravity from the reservoir in the upstream section of the conduit, and may be dispensed under gravity from a downwardly-facing leg of the downstream section of the conduit.

The airflow through the airflow inlet may be driven by a pump, which may run continuously or may be switched on and off each time fluid is dispensed. For example, the airflow through the airflow inlet may be driven by a blower pump upstream of the downstream section for applying elevated air pressure to an upstream end of the downstream section. In that case, a switch valve downstream of the blower pump may be switchable to direct elevated air pressure to the reservoir to drive fluid from the reservoir and into the conduit. There may also be an auxiliary blower pump for applying elevated air pressure to the reservoir to drive fluid from the reservoir and into the conduit.

To achieve this, the operation of the blower pump and the auxiliary blower pump may be toggled.

It is also possible for airflow through the airflow inlet to be driven by a suction pump downstream of the downstream section. This pump could, for example, be used to apply reduced air pressure to the downstream end of the conduit. Advantageously the suction pump is disposed in a vessel for receiving fluid dispensed from the downstream end of the conduit. For example, the suction pump could receive the fluid dispensed from the downstream end of the conduit and mix that fluid with a fluid already in the vessel. Also, the suction pump may draw air from the conduit and expel that air into a fluid within the vessel that receives fluid dispensed from the downstream end of the conduit. Preferably the suction pump is surrounded by, and more preferably submerged in, fluid already in the vessel.

Fluid may also be moved along the conduit from the reservoir by a pump. Indeed, the pump that moves the fluid along the conduit may also be used to drive airflow through the airflow inlet. In some arrangements, the movement of fluid along the conduit may be driven by a blower pump upstream of the reservoir for applying elevated air pressure to the reservoir to drive fluid from the reservoir and into the conduit. To achieve the desired airfiow and fluid flow, a non-return vaive may be disposed between the pump and the reservoir.

In other arrangements, movement of fluid along the conduit may be driven by a suction pump downstream of the downstream section that, in use, applies reduced pressure to the downstream end of the conduit. For example, the suction pump suitably applies reduced pressure to a column of air in the conduit to draw fluid from the reservoir and, once a slug of fluid is formed, to draw the slug along the downstream section of the conduit toward the downstream end of the conduit. As the slug moves along the downstream section of the conduit, the slug can draw air into the conduit through the airflow inlet, upstream of the slug.

In some embodiments of the invention, the airflow controller comprises an air valve operable to open and close the airflow inlet. The air valve may, for example, close as the suction pump draws fluid from the reservoir and open the airflow inlet to the atmosphere as the slug moves along the downstream section of the conduit. When, in use, the air valve is open, an upstream section of the conduit leading from the reservoir to the airflow

inlet is conveniently at atmospheric pressure. Advantageously, the air valve can be separated from fluid moving along the conduit; this avoids corrosion or other damage to the air valve, and allows the valve to operate more responsively to control signals.

The air valve may comprise a push-action solenoid acting on a valve head that is reciprocally movable within a chamber defined within a valve body. In that case, a valve head seal can seal the valve head to a valve seat to close the airflow inlet when the solenoid advances the valve head within the chamber.

In preferred embodiments, the downstream end of the conduit communicates with a throat of a venturi tube associated with the suction pump. This venturi tube may be associated with an inlet of the pump or could be associated with an outlet of the pump. The venturi throat can be spaced from the inlet or outlet of the pump by a tapering section of the venturi tube that widens from the throat to the inlet or outlet.

As an alternative to a venturi tube, the downstream end of the conduit may communicate with a shroud around an inlet of the suction pump.

In some embodiments, the airflow inlet is defined by an upstream end of the conduit and the airflow controller controls airflow through the airflow inlet by an actuator arranged to cause relative movement between the upstream end of the conduit and the surface of fluid in the reservoir. Preferably, a clearance is created between the upstream end of the conduit and the fluid surface in the reservoir to admit air into the conduit. Conversely, the upstream end of the conduit may be immersed in the fluid in the reservoir to admit fluid into the conduit, with the upstream end of the conduit suitably being movable relative to the fluid surface in the reservoir. The upstream end of the conduit may be movable relative to a support and the support may be movable relative to the reservoir in response to changes in the level of fluid within the reservoir. Elegantly, the support can simply float on the fluid in the reservoir.

A plurality of reservoirs, each having a respective conduit for receiving fluid from the associated reservoir may be used. Each conduit may have a respective airflow inlet, a respective airflow controller for controlling airflow through the airflow inlet and a respective fluid presence sensor. Advantageously, each reservoir could contain a different fluid, with the feedback sensor capable of detecting concentration of each fluid

in a solution and acting on the respective airflow controllers to control dispensation of the fluids into the solution independently.

From another aspect, the invention resides in a method for metered dispensation of fluid, comprising causing a body of fluid to move along a conduit from a reservoir and admitting air to the conduit to divide a slug from the body of fluid and to define an upstream end of the slug.

The body of fluid may be lifted from the reservoir against gravity. The slug of fluid is dispensed from a downstream end of the conduit and may be dispensed with the assistance of gravity. Elevated air pressure may be applied to the airflow inlet such that fluid is driven from the reservoir and into the conduit under the elevated air pressure applied to the reservoir. In some arrangements, the elevated air pressure is applied to the airflow inlet and the reservoir in alternation.

As in the system of the invention, airflow through the airflow inlet may be driven by reduced air pressure applied to a downstream end of the conduit. Advantageously, a suction pump produces reduced air pressure and agitates fluid within a vessel as aforesaid. Such agitation can advantageously mix the slug of fluid with the fluid in the vessel. The suction pump can also draw air from the conduit and expel that air into the fluid within the vessel, hence oxygenating and further agitating the fluid.

Reduced pressure may be applied to air in the conduit to draw the body of fluid from the reservoir into the conduit and, once a slug of fluid is formed, to draw the slug toward the downstream end of the conduit. As the slug of fluid moves along the conduit, the slug may draw air into the conduit upstream of the slug.

Preferably, as in the system of the invention, airflow into the conduit is controlled by an air valve operable to open and close an airflow inlet. For example, in use, the air valve can close the airflow inlet as the body of fluid is drawn from the reservoir into the conduit and the air valve can open the airflow inlet to atmosphere as the slug of fluid moves along the conduit. When the air valve is open, an upstream section of the conduit leading from the reservoir to the airflow inlet is suitably at atmospheric pressure.

A downstream end of the body of fluid may be sensed at a location along- the conduit, and when fluid is sensed at said location in the conduit, air may be admitted to the

conduit upstream of said location. Airflow into the conduit may be controlled by causing relative movement between an upstream end of the conduit and the surface of fluid in the reservoir. In that case, a clearance may be created between the upstream end of the conduit and the fluid surface in the reservoir to admit air into the conduit. The upstream end of the conduit may also be immersed in the fluid in the reservoir to admit fluid into the conduit. This may be achieved by the upstream end of the conduit being moved relative to the fluid surface in the reservoir or by the upstream end of the conduit being moved relative to a support, and the support moving relative to the reservoir in response to changes in the level of fluid within the reservoir.

Reference has already been made to Figures 1 to 3 of the accompanying drawings in which:

Figure 1 is a schematic sectional side view that illustrates the known prior art of a gravity-fed metered dispensation system;

Figure 2 is a schematic sectional side view that illustrates the known prior art of an air-displacement metered dispensation system; and

Figure 3 is a schematic sectional side view that illustrates the known prior art of a peristaltic pump utilised in a metered dispensation system.

In order that the invention may be more readily understood, reference will now be made, by way of example, to Figures 4 to 25, in which:

Figure 4 is a schematic sectional side view that illustrates a first embodiment of the invention with a venturi pump and single feed tube;

Figure 5 is a schematic sectional side view that shows the venturi pump of the first embodiment in use;

Figures 6 to 9 are schematic sectional side views that show how, in use of the first embodiment, solute flows along a donor tube and feed tube to be mixed into a receiving tank via the pump;

Figure 10 contains two flowcharts, one outlining a main sequence of operation of the first embodiment and the other outlining an interrupt sequence of operation of the first embodiment;

Figures 11(a), 11(b) and 11(c) are sectional views illustrating a venturi tube of the venturi pump of the first embodiment, Figure 11 (a) being a transverse section through an inlet end of the tube, Figure 11(b) being a longitudinal section along the tube, and Figure 11(c) being a transverse section through an outlet end of the tube;

Figure 12 is an enlarged detail perspective view showing a sensor for sensing the presence of fluid in a feed tube of the first embodiment, with the sensor mounted on the feed tube;

Figures 13(a) to 13(d) are views illustrating a low pressure air valve that may be used in some embodiments of the invention, Figure 13(a) being a transverse section through one end of the valve, Figure 13(b) being a longitudinal section along the valve, Figure 13(c) being a transverse section through another end of the valve and Figure 13(d) being a side view in a direction parallel to the plane of the section in Figure 13(b);

Figure 14 is a sectional exploded side view of the low pressure air valve of Figures 13(a) to 13(d);

Figure 15 is a schematic sectional side view that illustrates a second embodiment of the invention, having more than one donor tank;

Figure 16 is a schematic sectional side view that illustrates a third embodiment of the invention, having the feed tube connected to the outlet of the pump which runs in a reverse sense compared with the preceding embodiments;

Figures 17 and 18 are schematic sectional side views that illustrate fourth and fifth embodiments of the invention, each having a negative-pressure chamber associated with a mixing pump that omits a venturi tube;

Figure 19 is a schematic sectional side view that illustrates a sixth embodiment of the invention that works on the air-displacement principle and has a two-way air switch between an air pump and a donor tank;

Figure 20 is a schematic sectional side view that illustrates a seventh embodiment of the invention that again works on the air-displacement principle but has two air pumps;

Figure 21 is a schematic sectional side view that illustrates an eighth embodiment of the invention that again works on the air-displacement principle but has one air pump;

Figure 22 is a schematic sectional side view that illustrates a ninth embodiment of the invention, in which two tube fluid presence sensors are mutually spaced along the length of the feed tube;

Figure 23 is a schematic sectional side view that illustrates a tenth embodiment of the invention, in which the size of a solute slug is controlled by a timing arrangement so that a tube fluid presence sensor is not required; and

Figures 24 and 25 are schematic sectional side views of an eleventh embodiment of the invention, having a donor tube that reciprocates between a raised position shown in Figure 24, in which the donor tube is withdrawn from a solute, and a lowered position shown in Figure 25 in which the donor tube is dipped into the solute.

Dimensions given in the following specific description are by way of illustration only, and do not limit the scope of the invention. Additionally, the illustrated embodiments are not drawn to scale. Whilst the specific description puts the invention into the context of a hydroponics system, it is emphasised that the invention has broader application as aforesaid.

A first embodiment of the invention will now be described with reference to Figures 4 to

14. Referring firstly to Figure 4, a hydroponics system comprises a donor tank 4 containing a solute 14 such as concentrated fertilizer, a receiving tank 5 containing a solution 15 such as fertilizer dissolved in water, and a conduit 16 for transporting a

metered amount of solute from the donor tank 4 to be mixed into and dissolved in the solution 15 in the receiving tank 5. In use, a mixing pump 7 within the receiving tank 5 mixes the solute 14 into the solution 15. The donor tank may optionally have a lid (not shown).

In hydroponics practice, the receiving tank 5 can be used for two purposes. One is to grow plants directly with their roots immersed in the solution 15 within the receiving tank

5. The other is to use the receiving tank 5 as a mixing and holding tank from which solution 15 is drawn off to feed plants growing elsewhere. In both cases, a solvent (fresh water) has to be supplied to the receiving tank from time to time as evaporation or supply of solution 15 takes place or as plants absorb the solution 15. A fresh water inlet has been omitted from the drawings for brevity, as has an outlet for drawing off the solution 15.

It will be apparent that the concentration of solute 14 within the solution 15 has to be monitored and regular doses of solute 14 must be administered into the solution to maintain the concentration of solute 14 within the solution 15 within a desired range.

The concentration of solute 14 within the solution 15 is monitored by a sensor 8 which is suspended in the solution and connected to an electronic control unit (ECU) 6. The ECU

6, in turn, controls the opening of a low pressure air valve (LPAV) 1 that controls the flow of solute through the conduit 16.

The conduit 16 has two sections: a donor tube 11 being an upstream section and a feed tube 12 being a downstream section. The donor tube 11 rises vertically from the donor tank 4 with one end immersed in the solute 14 contained in the donor tank 4 and the other end connected to a first arm of a 'T' connector 10. The feed tube 12 is connected to a second arm of the 'T' connector 10. A portion of the feed tube 12 then extends horizontally from the 'T' connector 10 before a vertical leg of the feed tube 12 connects to an inlet end of the mixing pump 7 submerged in the solution 15 in the receiving tank 5.

The third available arm of the T connector 10 is connected to an air intake tube 17 that links both the donor tube 11 and feed tube 12 to the outlet of the LPAV 1.

It is not essential that the donor tube 11 extends precisely vertically or that the feed tube 12 extends precisely horizontally. For example, the donor tube 11 and the feed tube 12

may be formed of flexible tubing and can take any path from the donor tank 4 to the receiving tank 5.

A tube fluid presence sensor (TFPS) 3 connected to the ECU 6 monitors the presence of fluid along the horizontal portion of the feed tube 12. The TFPS 3 may, for example, comprise an infrared emitter on one side of the feed tube 12 and an infrared sensor on the other side of the feed tube 12 whereby the infrared sensor produces an output change when fluid is present in the feed tube 12 at the location of the TFPS 3.

The position of the TFPS 3 with respect to the length of the horizontal portion of the feed tube 12 is adjustable by sliding the TFPS 3 along the horizontal portion. The arrangement of the TFPS 3 around the feed tube 12 is best shown in the enlarged detail perspective view of Figure 12. The TFPS 3 is a close sliding fit around the feed tube 12 such that friction holds the TFPS 3 in a desired position with respect to the feed tube 12 when released.

The sensor 8 suspended in the solution in the receiving tank 5 is a control parameter probe. The sensor 8 communicates with the ECU 6 such that a predetermined parameter of the solution 15 in the receiving tank 5 - specifically the concentration of solute 14 in the solution 15 - can be monitored. When the control parameter is within a specified range, the system remains in a non-active state. A non-active state is defined by the LPAV 1 being in the open position as illustrated in Figure 4. The LPAV 1 is in the open position when it is de-energised and is in the closed position when energised.

When the control parameter determined by the sensor 8 differs from the acceptable predefined range, the ECU 6 triggers the system to dispense an amount of solute 14 defined by an algorithm programmed into the ECU 6. The amount of solute 14 and the rate of dispensation is governed by that algorithm. The rate of dispensation depends on how far the control parameter is from the acceptable range. Typically, the rate of dispensation is proportional to the difference between the control parameter and the acceptable range. The ECU 6 is generally set to dose at a higher rate of dispensation the further away the value of the control parameter is from the specified range. This prevents over dispensation taking the control parameter back out of range.

Operation of the first embodiment is as follows. The mixing pump 7 continually circulates the solution 15 within the receiving tank 5. Solution 15 is drawn into the inlet side of the

mixing pump 7 via a venturi tube 2 as indicated by 9 in Figure 4. A float (not shown) may be attached to the top of the mixing pump 7. The float sits on the surface of fluid in the receiving tank 5 to ensure that the mixing pump 7 is held close to the surface; this improves suction.

Figures 11 (a), (b) and (c) are sectional views of the venturi tube 2. The housing of the venturi tube 2 is outwardly cylindrical and contains an axially symmetric chamber running its length. This chamber is in four successive sections from left to right as shown: a first section 40, a second section 41 , a third section 42 and a fourth section 43.

The first section 40 comprises an inwardly tapering frusto-conical inlet, through which fluid enters the venturi tube 2. The second section 41 comprises a narrow parallel-sided duct where fluid passing through the venturi tube 2 accelerates, causing its pressure to drop. Thus, the action of drawing solution 15 through the venturi tube 2 causes a pressure drop at the second section 41. The feed tube 12 is connected to the intake of the mixing pump 7 by a spigot 44 at the second section 41. Consequently, the pressure drop in the second section 41 of the venturi tube 2 causes the air pressure within the feed tube 12 to drop. Connecting the venturi tube 2 to the intake of the mixing pump 7 causes a further pressure drop due to the pump's intake suction.

The chamber then opens out into the frusto-conical third section 42 which smoothly disperses and decelerates the fluid flow into the fourth section 43 which is a wide parallel-sided duct to match the larger-diameter cross section of the intake of the mixing pump 7. In an alternative configuration (not shown), the venturi tube 2 may comprise only the second section 41 , third section 42 and fourth section 43.

In its non-active position, shown in Figure 4, the LPAV 1 is open to the surrounding air at atmospheric pressure. As the air pressure within the feed tube 12 drops, the higher pressure air of the surrounding atmosphere is drawn into the LPAV 1 and along the feed tube 12. This air emerges from the feed tube 12 as a stream of bubbles that passes through to the outlet side of the mixing pump 7 as shown in Figure 5. A corresponding drop in air pressure also occurs in the donor tube 11. However, the flow of air through the open LPAV 1 is sufficient to prevent the pressure drop in the donor tube 11 lifting solute 14 up the donor tube 11 from the donor tank 4.

The air bubbling turbulently from the mixing pump 7 is an advantage in the field of hydroponics. Where the fluid in the receiving tank is a fertilizer solution, the resulting oxygenation increases root activity of the plants being grown, to the benefit of overall productivity. Hydroponicists often buy aquarium air pumps and air stones for this purpose, whereas use of the invention would save them this cost.

The system assumes an active state when the LPAV 1 moves to the closed position as shown in Figure 6. In the closed position, the feed tube 12 and the donor tube 11 are shut off from the surrounding atmosphere. Consequently, the pressure drop in the feed tube 12 caused by the venturi tube 2 of the mixing pump 7 results in a pressure drop in the donor tube 11 that is no longer relieved by air flowing into the LPAV 1. The atmospheric pressure acting on the solute 14 within the donor tank 4 therefore forces the solute 14 along the donor tube 11 , through the T connector 10 and along the feed tube 12.

When, as shown in Figure 7, the solute 14 reaches the TFPS 3, the output signal of the TFPS 3 changes. The ECU 6 detects this change and instructs the LPAV 1 to return to its open position, as shown in Figure 8. Again, air from the surrounding atmosphere is drawn into the LPAV 1 and along the feed tube 12. Any solute 14 that has been drawn along the donor tube 11 but that has not passed the T connector 10 and entered the feed tube 12 drops back into the donor tank 4 under gravity. The remaining slug of solute

14 travels along the feed tube 12, into the mixing pump 7 and is mixed with the solution

15 in the receiving tank 5, as shown in Figure 9.

The size of solute slug, and hence the amount of solute 14 dispensed in each cycle, is determined by the position of the TFPS 3 along the horizontal portion of the feed tube 12. The position of the TFPS 3 with respect to the horizontal portion of the feed tube 12 is adjustable as mentioned above.

The flowcharts in Figure 10 illustrate how the firmware of the ECU 6 may operate; other routines are possible. The firmware program is based around a timer interrupt function found on low-cost microcontroller microchips. The program consists of a main sequence and an interrupt sequence. The main sequence is run continuously when the ECU 6 is powered on. The interrupt sequence is then run periodically, interrupting the main sequence when it does so. After each interrupt sequence has run, the ECU 6 switches back to continue running the main sequence. The program is split into a continually-

running main sequence and an interrupt sequence for simplicity. The operations performed by the interrupt sequence do not need to be constantly implemented, whereas the operations performed by the main sequence do.

The main sequence allows the ECU 6 to monitor the output signal of the TFPS 3. If the TFPS 3 senses the presence of fluid, the ECU 6 de-energises the LPAV 1 and thus the LPAV 1 moves to the open position, shown in Figure 8. All other functions of the ECU 6 (e.g. control parameter calibration, control parameter range setting, display driving, dispensation rate calculation, etc) are implemented in the interrupt sequence and are referred to collectively as 'miscellaneous ECU functions' in the interrupt sequence flowchart of Figure 10. This is not indicative of a function hierarchy. Continual monitoring of the presence of fluid at the TFPS 3 is preferred as this optimises the response and accuracy of the dispensing mechanism.

Typically, the LPAV 1 comprises a known type of solenoid air valve. Alternatively, the LPAV 1 comprises a push-action solenoid 30 that controls the movement of a push rod 37 as shown in detail in Figures 13(a), (b), (c) and (d) and Figure 14. The solenoid 30 is disposed at one end of a valve body 31. The push rod 37 acts on a valve head 33 that is surrounded by the valve body 31 , being reciprocally movable within a cylindrical chamber 38 defined within the valve body 31. The valve head 33 is a cylinder that is axially mounted on the push rod 37.

A valve face end cap 35 closes the otherwise open end of the valve body 31 opposed to the solenoid 30. A grub screw 32 retains the end cap 35 on the valve body 31. An air tube connector 36 protrudes from the centre of the end cap 35, communicating with the chamber 38 within the valve body 31. The air tube 17 that leads to the conduit 16, as shown in Figure 4, is attached to the air tube connector 36.

A valve head seal 34 seals the valve head 33 to the valve face end cap 35 when the solenoid 30 extends the push rod 37.

The valve face end cap 35, valve head 33, valve head seal 34 and air tube connector 36 are all axially symmetric.

An air intake hole 39 extends radially through a side wall of the valve body 31 , near the end of the valve body 31 opposed to the solenoid 30. The hole communicates with the chamber 38 within the valve body 31.

Figure 13(b) shows the position of the valve head 33 in the LPAV 1 in its de-energised state. In that state, there is an air gap between the valve head seal 34 and the valve face end cap 35. Air can flow through the air intake hole 39 in the side wall of the valve body 31 , through this air gap and through the air tube connector 36. When the solenoid 30 is energised, the push rod 37 extends, pushing the valve head 33 toward the valve face end cap 35. The valve head seal 34 forms an airtight seal between the valve head 33 and the valve face end cap 35. This blocks airflow through the LPAV 1.

Figure 15 illustrates a second embodiment of the invention. In this embodiment two donor tanks 4a and 4b are each associated with a respective LPAV 1a and 1 b, such that more than one type of solute 14a, 14b can be dispensed. Specifically, the second embodiment employs an additional donor tank 4b, donor tube 11 b, feed tube 12b, LPAV 1b and TFPS 3b. Feed tubes 12a and 12b link to donor tanks 4a and 4b respectively via their associated T connectors 10a and 10b and donor tubes 11a and 11b. Each feed tube 12a, 12b is connected to the mixing pump 7 at the first section 40 of the venturi tube 2.

The second embodiment works to the same principle as described in the first embodiment of controlling the actuation of the LPAVs to control dispensation of the respective solutes 14a and 14b from the respective donor tanks 4a, 4b. To this end, the sensor 8 suitably has the ability to sense two different parameters of the solution 15 in the receiving tank 5, specifically the respective concentrations of solute 14a and solute 14b, and the ECU 6 has the ability to process data relating to both parameters and to act on that data independently. This embodiment can be extended to utilise any number of donor tanks containing many varieties of solute 14a, 14b.

The sensor 8 need not be required to sense more than one parameter of the solution 15. For example, in the case of controlling the acidity of the solution 15, acid and alkali could be held in the respective donor tanks 14a and 14b with sensor 8 measuring one parameter of the solution 15, namely its pH level. The sensor 8 acting on the ECU 6 may thereby cause the pH level to change by adding acid or alkali to the solution 15 as appropriate.

A third embodiment is shown in Figure 16. The third embodiment is structurally identical to the first embodiment but the mixing pump 7 runs in reverse. Thus, the venturi tube 2 is located at the outlet of the mixing pump 7. Again, fluid flowing through the narrow second section 41 of the venturi tube 2 produces a pressure drop in the feed tube 12 connected to the second section 41.

The third embodiment has an advantage in some applications as the air flowing from the feed tube 12 does not enter the mixing pump 7: such air entering the mixing pump could otherwise decrease the output flow rate of the mixing pump. This embodiment avoids that possibility. Multiple donor tanks 4 and feed tubes 12 for dispensing a plurality of solutes 14, as described in the previous two embodiments, can also be applied to this configuration of mixing pump 7.

A venturi tube 2 is not essential. Alternative methods can be used to create the required pressure difference and hence the air flow along the donor tube 11 and feed tube 12. Whilst other effective solutions are possible, Figures 17 and 18 show the use of a negative-pressure chamber 13 in conjunction with the mixing pump 7 to create the desired air flow. The intake of the mixing pump 7 causes a pressure drop within the chamber 13. This generates the necessary pressure difference to create an air flow along the donor tube 11 and feed tube 12. The negative-pressure chamber 13 can be attached directly to the intake of the mixing pump 7 as in the fourth embodiment shown in Figure 17. The negative-pressure chamber 13 could alternatively surround the mixing pump 7 as in the fifth embodiment shown in Figure 18, provided that the outlet of the mixing pump 7 is outside the negative-pressure chamber 13. As before, multiple donor tanks 4 and feed tubes 12 for dispensing a plurality of solutes 14, as described in the previous three embodiments, can also be applied to these two configurations of mixing pump 7.

A donor tube 11 , feed tube 12, T connector 10 and TFPS 3 can be applied to known air- displacement techniques to achieve metered fluid dispensation. Figures 19 and 20 show further embodiments of the invention in this regard.

A sixth embodiment is shown in Figure 19. This embodiment works on the principle of air displacement with an air switch 21 between an air pump 20 and a donor tank 4. The air switch 21 has two switchable outlets 23, 24, each having a respective non-return valve

22. The non-return valves 22 prevent fluid flow from the donor tank 4 towards the air switch 21 in use.

A first outlet 23 communicates with the feed tube 12 through the T connector 10. The second outlet 24 communicates with the donor tank 4 through an air inlet 25. The donor tank 4 is closed apart from the air inlet 25 and the donor tube 11. The inlet end of the donor tube 11 is immersed in the solute 14 in the donor tank 4 and has a non-return valve 22 and the outer end of the donor tube 11 protrudes from the donor tank 4 to connect to the T' connector.

Air flows from the pump 20 to the air switch 21. The pump may be run continuously or may be switched on and off each time fluid is dispensed to prolong pump life. If the pump 20 is running continuously, when the system is in a non-active state, the air switch 21 allows air to flow through the first outlet 23, along the feed tube 12 and out of the outlet of the feed tube 12 disposed in the solution 15 in the receiving tank 5. The non-return valve 22 in the donor tube 11 prevents air from flowing down the donor tube 11 and into the solute tank 4, thereby avoiding a build-up of pressure in the donor tank. That non-return valve 22 also prevents solution 15 from entering the solute tank 4 in the event of a reverse flow of solution 15 along the feed tube 12 and down the donor tube 11 into the solute tank 4 due to siphoning.

When the ECU 6 determines via the sensor 8 that a dose of solute 14 needs to be dispensed, the ECU 6 toggles the air switch 21 to allow air to flow through the second outlet 24 and the air inlet 25 into the donor tank 4. As air is pumped into the donor tank 4, the air pressure within the donor tank 4 increases, thus forcing the solute 14 out along the donor tube 11 , through the T connector 10 and into the feed tube 12.

As in previous embodiments, the signal output from the TFPS 3 changes when the solute 14 passes it. This change in signal is registered by the ECU 6 which in turn toggles the air switch 21 back to the non-active position. Air flows through the first outlet 23 of the air switch 21 along the feed tube 12 upstream of the solute 14 that has entered the feed tube 12 from the donor tube 11. This defines a slug of solute 14 between the T connector 10 and the TFPS 3, which is then forced by the airflow from the outlet of the feed tube 12 and into the solution 15 in the receiving tank 5.

Solute 14 that remains in the donor tube 11 is confined there by the non-return valve 22 and the pressure of the air flowing from the first outlet 23 of the air switch 21 along the feed tube 12. As described in the previous embodiments, multiple donor tanks 4 and feed tubes 12 for dispensing a plurality of solutes 14, can also be applied to this configuration of mixing pump 7.

Figure 20 illustrates a seventh embodiment of the invention. This embodiment is comparable with the sixth embodiment, but here two air pumps 20a and 20b replace the single air pump 20 and air switch 21 of the sixth embodiment. The same principles of air flow control the dispensation of solute 14. The timing of the operation of the two pumps 20a and 20b is controlled by the ECU 6. The outlet of pump 20a communicates with the interior of the donor tank 4 via a non-return valve 22 and an air inlet 25. This generates the air pressure required to force solute 14 out of the donor tank 4 along the donor tube 11 , into the feed tube 12 and past the TFPS 3. In response to the resulting change of output signal from the TFPS 3, the ECU 6 shuts off the pump 20a and turns on pump 20b which generates the air pressure required to force the slug of solute 14 into the receiving tank 5. As described in the previous embodiments, multiple donor tanks 4 and feed tubes 12 for dispensing a plurality of solutes 14, can also be applied to this embodiment.

Whilst Figure 20 illustrates an arrangement in which the pumps 20a, 20b are run intermittently, it would be possible to run them continuously and to vent air pressure via a vent valve under control of the ECU 6 when the pumps 20a, 20b are not required to exert air pressure upon the donor tank 4 or upon a slug of solute 14 in the feed tube 12.

An eighth embodiment is shown in Figure 21. This embodiment also works on the principle of air displacement and is comparable with the sixth embodiment, but here a two-way low pressure air valve (LPAV) 1 replaces the three-way air switch 21 of the sixth embodiment. When the system is in a non-active state but the pump 20 is running (again, the pump 20 can run continuously or intermittently), the LPAV 1 is in the open position and allows air to flow through the LPAV1 , along the feed tube 12 and out of the outlet of the feed tube 12 disposed in the solution 15 in the receiving tank 5. When the ECU 6 determines via the sensor 8 that a dose of solute 14 needs to be dispensed, the ECU 6 instructs the air valve to adopt the closed position, which forces air to flow into the donor tank 4. As air is pumped into the donor tank 4, the air pressure within the donor

tank 4 increases, thus forcing the solute 14 out along the donor tube 11 and into the feed tube 12.

As in previous embodiments, the signal output from the TFPS 3 changes when the solute 14 passes it. This change in signal is registered by the ECU 6 which in turn instructs the

LPAV 1 to open again. Air then flows through the LPAV 1 , along the feed tube 12 upstream of the solute 14 that has entered the feed tube 12 from the donor tube 11. This defines a slug of solute 14 between the T connector 10 and the TFPS 3, which slug is then forced by the airflow from the outlet of the feed tube 12 and into the solution 15 in the receiving tank 5.

Solute 14 that remains in the donor tube 11 may fall back into the donor tank 4 under gravity or may remain in the donor tube 11 , being confined there by the pressure of the air flowing from the LPAV 1 along the feed tube 12. As described in the previous embodiments, multiple donor tanks 4 and feed tubes 12 for dispensing a plurality of solutes 14, can also be applied to this embodiment. It is also possible to have a nonreturn valve in the donor tube 11 as in preceding embodiments.

A ninth embodiment is shown in Figure 22. This embodiment is structurally identical to the first embodiment but has two TFPS, numbered 3a and 3b, mutually spaced along the length of the feed tube 12. As with previous embodiments the LPAV 1 is closed to cause the solute 14 to be drawn along the donor tube 11 to the feed tube 12. By having more than one TFPS 3a, 3b positioned along the length of the feed tube 12, the ECU 6 can choose to dispense either a small slug or a large slug depending on how far from a pre- set value is the control parameter of the solution 15 as measured by the sensor 8. In other words, if the control parameter of the solution 15 is close to the pre-set value, the ECU 6 may dispense a small slug defined by the shorter length of feed tube 12 between the upstream end of the feed tube 12 and the upstream TFPS 3a. This .maintains fine adjustment of the control parameter. Conversely if the control parameter of the solution 15 is far from the pre-set value, the ECU 6 may dispense a large slug defined by the greater length of feed tube 12 between the upstream end of the feed tube 12 and the downstream TFPS 3b. This coarsely but swiftly adjusts the control parameter to bring it closer to the pre-set value as quickly as possible.

The two mutually-spaced TFPS 3a and 3b of the ninth embodiment can also allow the ECU 6 to measure the time taken for the solute 14 to pass from one TFPS to the next.

By knowing the separation between TFPS 3a and TFPS 3b and the cross-sectional area of the feed tube 12 which determine the volume of solute 14 moving between TFPS 3a and TFPS 3b, the flow rate of the solute 14 along the feed tube 12 may thereby be estimated. The ECU 6 can then calculate the timing of the opening and closing of the LPAV 1 necessary to dispense the required amount of solute, knowing that if the fluid velocity along the feed tube 12 is X m/s and the LPAV 1 is open for Y seconds, the length of the slug will be XY and the volume of the slug will be XYA where A is the cross- sectional area of the feed tube 12. As the cross-sectional area A can be assumed to be a constant programmed into the ECU 6 and as X can be measured by the time taken for solute 14 to pass between TFPS 3a and TFPS 3b, the required slug volume can be decided by a simple algorithm used by the ECU 6 to implement the necessary opening time Y of the LPAV l

A tenth embodiment is shown in Figure 23. This embodiment is structurally identical to the first embodiment except that the system in this embodiment operates without a TFPS 3. As before, the ECU 6 controls the actuation of the LPAV 1. The LPAV 1 is closed in order for solute 14 to be drawn along the donor tube 11 and, subsequently, the feed tube 12. In this case the size of the slug of solute 14 is determined not by the position of the TFPS 3, but by the LPAV 4 closing and then opening after a pre-set time period.

The method used by the tenth embodiment to create a slug of solute 14 is not as accurate as in previous embodiments. The advantages it holds over previous embodiments are simplicity, reliability and cost. The timing method of dispensation may be used wherever highly-accurate metering of solute 14 is not required.

In a variant of the tenth embodiment, not illustrated, a TFPS may also be used in conjunction with the timing method of dispensation. In this variant, accurately-metered dispensation of solute 14 may be controlled by the position of a TFPS along the length of the feed tube 12, as in preceding embodiments. Thus, when the control parameter of the solution 15 as measured by the sensor 8 is close to a pre-set value such that the control parameter is within a certain range, accurately metered dispensation of solute 14 into solution 15 would be performed using a TFPS in order to make fine adjustments to the control parameter of the solution 15. When the control parameter of the solution 15 is outside that range, larger, less accurate doses of solute 14 would be created using the timing method to bring the control parameter within the desired range as quickly as possible.

As with previous embodiments, the tenth embodiment could be used in conjunction with multiple donor tanks 4 and feed tubes 12 for dispensing a plurality of solutes 14. It could also be used with any configuration of mixing pump 7 described in previous embodiments

Turning finally to Figures 24 and 25, these show an eleventh embodiment of the invention in which the receiving tank 5, ECU 6 and mixing pump 7 are omitted for clarity, it being understood that the feed tube 12 extends downstream of the TFPS 3 to the receiving tank 5 as above. This embodiment has a movable donor tube 11 supported above a donor tank 4 by a dipping mechanism 26. By virtue of the dipping mechanism 18, the donor tube 11 reciprocates between a raised position shown in Figure 24, in which an inlet end of the donor tube 11 is withdrawn from the solute, and a lowered position shown in Figure 25 in which the inlet end of the donor tube 11 is dipped into the solute 14.

The reciprocation of the donor tube 11 is controlled by the ECU 6 acting on a solenoid actuator 23. In its non-active state, the solenoid actuator 23 is de-energised so that a push rod 27 of the actuator 23 remains retracted. This, in turn, maintains an air gap between the inlet end of the donor tube 11 and the solute 14 in the donor tank 4. The air gap allows the mixing pump 7 (not shown) connected to the feed tube 12 to draw air through the conduit 16 of which the donor tube 11 is a part, and into the receiving tank 5.

As before, when the control parameter determined by the sensor 8 in the solution 15 in the receiving tank 5 differs from an acceptable pre-defined range, the ECU 6 triggers the system to dispense an amount of solute 14 defined by an algorithm programmed into the ECU 6. The amount of solute 14 and the rate of dispensation is governed by that algorithm.

To dispense solute 14, the ECU 6 signals for the solenoid actuator 23 to be energised. This extends the push rod 27, hence lowering the inlet end of the donor tube 11 into the solute 14. Hence, solute 14 is now drawn along the conduit 16 instead of air. As with previous embodiments, the solute 14 is drawn along the donor tube 11 and into the feed tube 12. The solute 14 passes the TFPS 3 that is positioned at a desired location along the feed tube 12. Solute 14 passing the TFPS 3 causes an output signal change in the TFPS 3 that is detected by the ECU 6. This causes the ECU 6 to de-energise the

solenoid actuator 23, hence causing the push rod 27 to retract and pull the inlet end of the donor tube 11 clear of the solute 14. The air gap between donor tube 11 and solute 14 is thereby restored, allowing air to be drawn into the inlet end of the donor tube 11 once more. The incoming air defines the upstream end of a slug of solute 14 of known volume between the inlet end of the donor tube 11 and the TFPS 3. That slug is drawn into the receiving tank 5, completing a cycle of metered solute dispensation.

Figures 24 and 25 show that the dipping mechanism 26 is supported by a frame 18 upstanding from floats 28 that float on the surface of the solute 14 in the donor tank 4. Thus, the dipping mechanism is supported at a constant level with respect to the surface of the solute 14, even though that level changes in use.

As in the previous embodiments, the eleventh embodiment can be adapted such that more than one donor tank 4 could be used, each donor tank 4 thus having a respective donor tube dipping mechanism 26.

The invention as described in the above embodiments has many advantages over current metered dispensation apparatus. Through the use of the invention, small amounts of fluid can be dispensed, typically in the range of 0.1 millilitres to 10 millilitres. The fast switching time of the LPAV 1 coupled with the fast response time of the TFPS 3 and ECU 6 means that small, accurate pulses of solute can be dispensed. The ability accurately to dispense small amounts of solute means that the solute can be of a higher concentration. As a result, the donor tank 4 can be smaller; moreover, the cost of the solute that has to be used may be lower.

Another distinct advantage of the invention over the prior art is the comparatively low cost of components. For example, the venturi tube 2 and LPAV 1 (as shown in Figures 13(a), (b), (c) and (d) and Figure 14, as well as known solenoid air valves to be used with the invention) are made up of shapes that are simple to manufacture. In comparison, the components of prior art systems are relatively numerous and complex. Also the LPAV 1 only needs to form a low-pressure air seal that requires little force to produce, which means that its actuator solenoid can be a small low-cost unit. This is as opposed to the heavier-duty actuator needed in the gravity-fed solenoid or the higher-cost motors used on direct chemical pumps, such as peristaltic pumps, diaphragm pumps, centrifugal pumps and so on. This reduces the purchase cost and subsequent service costs of a system that uses the invention.

Another major advantage of the invention's simplicity is that it is easy to service and maintain. This is particularly important for the consumer market where the owner usually services the system. The LPAV 1 is the only part that needs to be serviced and its assembly/disassembly is relatively simple as per Figure 14. In the medium term, over say three to five years, the valve head and seal will need replacing: however, this is a very low-cost item. In the longer term, the solenoid will need to be replaced but this is a complete assembly that attaches to and detaches easily from the valve. This is in contrast to the more complex integrated assembly of a direct-contact solenoid valve as in the gravity-fed example or the motor/gearbox of a peristaltic pump. Also, none of the moving parts of the system of the invention come into contact with any aggressive fluids that might be dispensed, particularly acids and alkalis. Aside from minimising corrosion, this makes servicing easier and safer as the parts do not need to be thoroughly washed.

By virtue of the invention, the size of the fluid slugs or pulses can be simply and intuitively adjusted by moving the TFPS up and down the feed pipe. If the user knows the bore (B) of the tube and the length (L) of feed pipe between the TFPS and the T-piece connector, the user can calculate the pulse volume (PV) using the simple formula PV=BπL. Alternatively the manufacturer could place successively numbered stripes of predetermined width (W) along the feed tube. So if the user knows what number (n) stripe that the TFPS is positioned next to, the user can work out the length L by L = nW and consequently the pulse volume with PV=BπnW. Alternatively, a distributor of systems according to the invention could calculate the pulse volume for each numbered stripe and then provide that information on a simple look-up chart. This could, for example, read "if the TFPS is next to stripe number x, then the volume of each individual pulse will be y ml". It would, for example, be possible to define the number of stripes and their width to realise a metric graduation, for example 0.1ml to 2ml in 0.1ml graduations.

Many variations are possible within the inventive concept. For example, the use of an infrared TFPS is not essential. Any sensor that gives a change in output signal when solute is present in the feed tube could be used. An example is a capacitive sensor that measures the difference in the dielectric constant of the feed tube, thus being able to determine when fluid is present. It is also possible for the timing method of dispensation be used in conjunction with a TFPS or instead of a TFPS in any of the embodiment herein. Similarly, the design of LPAV described above is not essential to the invention.

The key attribute of the LPAV is the ability to control air flow through a conduit.