A reject valve of a reverse osmosis system maintaining the system pressure at a predetermined level despite fluctuation in the volume of the reject flow.

It is known that by throttling the flow exiting as reject from a reverse osmosis module, the pressure in the module can be adjusted while water is simultaneously being fed into the module with a high-pressure pump. The module refers to a conventional, standardized tubular pressure vessel and a reverse osmosis membrane located inside it. As a saline water stream passes through the module, the membrane separates fresh water from it. The remaining concentrate, from which fresh water has been separated, is called the reject.

In a reverse osmosis system, the number and type of membranes determine the limit values of the feed flow volume.

The higher the concentration of solids dissolved into water, mainly the

concentration of salts (TDS = total dissolved solids), the higher the osmotic pressure of the water. In order for the membrane to separate fresh water from saline water, the pressure in the module must be at least as high as the osmotic pressure of the concentrate flowing through the module. Fresh water separated by the membrane is referred to as its yield. The sum of the yield and the reject is the same as the feed flow volume.

If a constant throttle is used on the reject side, such as a choke throttling the flow, the problem is that the pressure in the system changes as the salinity of the water changes. In addition, the temperature of the water to be treated has a significant effect on the membrane yield, whereby the reject volume changes correspondingly. Another problem with a constant throttle is that the membrane yield is reduced as the membranes age, whereby the reject volume increases in relation to the feed flow volume. Constant throttling is only suitable for situations where the feed flow volume is constant and where the membrane yield remains constant.

In small reverse osmosis devices intended for brackish water and whose yield is a few dozen liters per hour, a manually adjustable needle valve is typically used as the reject flow throttle valve, which can be adjusted to the desired throttle level once the pressure pump of the system has been started. A practical problem with these devices is that the membrane yield stabilizes only after the system has been used for a while. Salt deposit may also form at the throttle valve, changing its flow area.

Particularly if the reverse osmosis system is to be used with energy obtained from solar panels, without batteries in between, the need for manual adjustment of the reject valve would be constant. This is due to the fact that power obtained from solar panels changes as the solar irradiance kW/m ² varies. Thus, also the rotating speed of the inverter-controlled feed pump varies and as a result, the feed volume and reject volume of the system fluctuate.

The reject valve of the invention solves the problem related to the adjustment of the reject valve of the reverse osmosis system. It maintains constant pressure in the system despite fluctuation of the flow volume. The valve is spring-actuated.

Structurally, all known spring-actuated valves, such as check valves, pressure relief valves, safety valves and bypass valves, resemble each other, but their operating principle and purpose of use are different.

The structure of the valve of the invention closely resembles a check valve, but its function and purpose of use are not that of a check valve or of a pressure relief valve, nor of any of the above-mentioned valves. It is a valve whose flow area adjusts dynamically according to the fluctuation of the inflow volume. The other above-mentioned valves do not do this.

The publication JP 3079258U describes a check valve, the structure of which closely resembles that of the valve of the invention. Therein, a spring presses the cone closing the valve against the wall of a conical valve seat. The cone sits deep in the valve seat and a seal ring has been mounted around the cone. The intended use of the valve is that while closed, the valve does not allow any flow in from the direction opposite to the inflow. As the cone sits mainly in the valve seat, the pressure attempting to open the inflow valve is applied mainly to the end of its arm and very little to the cone itself. When the force exerted by the inflow pressure on the cone and the arm thereof exceeds the cracking force exerted on the cone by the spring, the cone jumps up, opening the duct. In this case, the flow pressure between the cone and its conical valve seat immediately drops to close to zero. This is a result of the flow velocity increasing between the walls positioned close to each other. Fig.5 and Fig.6 illustrate the matter. Fig.5 describes the change of the flow pressure and velocity through the choke. In a valve of the type described in the publication, the inflow pressure is applied minimally to the cone itself Fig.6, thus, once the valve has opened, the spring force will pull the cone back into the valve seat. To prevent this from happening, the inflow volume must be high enough and the spring force low, otherwise the cone will start to beat against the valve seat.

The higher the spring force exerted on the cone, the worse this beating

phenomenon becomes. Therefore, the spring is intended to be as loose as possible and to only return the cone to the valve seat once the flow ceases. In practice, this type of check valve is always maximally open or completely closed. The structure of this type of check valve is not suitable for maintaining the inflow pressure constant when the flow volume fluctuates.

The publication WO2014168768 describes a valve structurally resembling the valve of the invention. It is intended for rapid pressure relief of the internal pressure of the system in a deep sea, high-pressure environment in case the pressure in the system increases for some reason. It is not intended for a fluid flow nor for maintaining inflow pressure. In a fluid flow, the same problems would persist as with the check valve described in the above-mentioned publication.

The operating principle of the valve of the invention is that the size of its flow duct adjusts automatically according to the inflow volume, i.e. as the volume increases, the flow area of the duct increases and as the volume decreases, it decreases correspondingly. Thereby the valve maintains the system pressure on the inflow side at a predetermined level on a predetermined inflow volume area.

Owing to the above-mentioned characteristic, the valve allows the electric motor of the high-pressure pump of the reverse osmosis unit to be used directly with power obtained from solar panels through an inverter, whereby the rotating speed and correspondingly, the feed flow volume vary according to the energy obtained from solar panels. As a result of the variation of the rotating speed of the motor, the volume of fresh water and correspondingly, the reject, produced by the reverse osmosis system within a unit of time varies.

Next, the structure and function of the valve according to the invention is described in more detail with reference to figures 1 - 7.

Figure 1 shows a check valve of prior art wherein the valve cone is supported by a conical valve seat. Figure 2 shows the structure of the valve of the invention.

Figure 3 shows a flow passing by the cone of the valve of the invention when the valve is closed. Figure 4 shows the cone of the valve of the invention when opened to its

maximum position at maximum reject flow volume.

Figure 5 shows a graphic representation of the change in flow pressure and velocity as the flow passes through the choke.

Figure 6 shows a graphic representation of the change in flow pressure and velocity as the flow passes through the check valve shown in figure 1.

Figure 7 shows a graphic representation of the change in flow pressure and velocity as the flow passes through the valve of the invention.

Figure 2 shows the cross-section of the reject valve of the invention. The basic structure of the valve is axially connected to the input pipe 11 of the reject flow.

The length of the pipe 11 is such that the arm 4 of the cone 5 closing the valve duct 2 and the support plate 9 extending close to the wall of the pipe 11 can move freely inside the pipe 11. A tensional compression spring 6 is supported between the sliding member 7 controlling the arm 4 supported on the frame 1 and the support plate 9, exerting force, when the valve is closed, that presses the cone 5 against the end 3 of the flow duct 2 on the outflow side. The compression spring 6 is located around the arm 4 and the outer diameter of the spring 6 is selected to be such that it extends close to the wall of the pipe 11, so that the spring 6 is prevented from twisting inside the pipe. The gliding member 7 is fitted with ducts 8 allowing flow through and also the support plate 9 allows flow-through.

When leaning on the outflow end 3 of the duct 2 the cone 5 never fully closes the outflow duct 2, but always allows partial flow-through arranged by means of flow grooves 10 on the circumference of the outflow end 3 of the outflow duct 2, corresponding to a constant choke when the cone 5 leans on the circumference of the outflow end 3, through which choke, just before the movement of the cone 5 starts to open the outflow duct 2, i.e. when the desired system pressure has been reached, a flow is allowed to pass through the reject valve, the volume of which amounts to 20 to 40 % of the maximum volume of the reject flow. Fig.3. It can be envisioned that there would be a separate constant choke next to the duct 2, but in addition to being hard to implement from a machining point of view, the saline reject concentrate would block it quickly due to poor rinsing. Particularly in apparatuses powered by solar energy, standing idle at night time, salt deposit would form at the choke. It is also preferred for the flow control function of the cone 5 that the entire volume of the inflow goes around it.

The cylindrical inflow duct 2 and the arm 4 of the cone 5 have been sized so that the diameter of the duct 2 is at least 2.2 times the diameter of the arm 4, whereby the cone 5 sits deep enough in the cylindrical duct 2 in all flow situations for the flow pressure on the input side to exert force on the cone 5 that is sufficient for the pressure control function of the valve also in the case where the valve has opened, but the flow volume is still under 45 % of the maximum flow volume. Fig.3, Fig.4 and Fig.7.

When the flow volume created by the pump of the system increases to 20 to 40 percent of its maximum volume, in a preferred case to 25 to 35 percent, the flow pressure on the input side increases to a predetermined system pressure, meaning that the reject flow also increases to 20 to 40 percent of its maximum volume. This also means that just before the cone 5 starts to move, 20 to 40 % of the maximum volume of the reject flows through the duct 10. The preferred geometric flank angle of the cone 5 is from 13 to 20 degrees, in a preferred case from 14 to 18 degrees. Thereby the cone 5 will not, due to the spring force exerted thereon, get stuck at the end of the flow duct 3 while leaning on it and because d/D < 4.5, the cone 5 sits at a depth in the flow duct 2 that is adequate for the function of the valve in each flow situation. It is essential for the structure of the valve that the cone 5 leans on the end of the cylindrical duct 2. A valve with a conical valve seat does not work for the intended use of the invention, which is clearly indicated by the graphic figure 6.

When the cone 5 opens the valve, the flow area of the end of the duct 2 increases and the spring 6 begins to compress. The longer the spring 6 used, the smaller the change in force caused by its compression. In order for the spring 6 to control the movement of the cone 5 according to the change in the flow volume, the spring 6 should be a straight and sufficiently long compression spring. Because the valve forms a preliminary assembly figure 2, a pipe 11 of any length can be connected to the frame 1 thereof with a tread connection.

In an exemplary embodiment, the diameter D of the duct 2 is 20 mm and the flank angle of the cone 5 is 17 degrees and the free length of the compression spring is 200 mm. For the sake of clarity, this example does not account for the cross- sectional area of the duct 10.

The spring has been prestressed to a length of 170 mm. This means that when the cone 5 has moved a distance of 1 mm opening the duct 2, the circular flow cross- sectional area has increased from zero to 19 mm ² and the force exerted by the spring 6 on the cone 5 has increased by 0.6 %. When the flow volume has increased enough for the cone 5 to have moved a distance of 5 mm opening the duct 2, the circular flow cross-sectional area has increased to 87 mm ² . This represents a 460 % increase in the flow cross-sectional area compared to the previous position, which also means a corresponding increase in the flow volume. Meanwhile, the force exerted by the spring on the cone has increased by only 2.4 %. It is concluded that an 87 mm ² flow duct corresponds to a pipe with a diameter of 10.5 mm. This exemplary embodiment proves that the force exerted by the spring 6 on the cone 5 remains essentially constant as the reject flow volume increases. The spring force of the spring 6 is determined so that when the flow pressure on the input side of the reject valve increases to 95 +/- 5 percent of a predetermined system pressure, the cone 5 starts to move, opening the outflow duct 2.

The arm 4 of the cone axially penetrates the flow duct 2 of the valve. The diameter of the arm 4 is d and the diameter of the duct 2 is D. It is essential for the valve that d/D < 0.45, because otherwise the valve is not able to control the flow adequately, as the surface area of the cone 5 that the inflow pressure is applied to would be too small. Fig.3, Fig.4 and Fig.7.

When the valve is open, the flow pressure at the end 3 of the outflow duct 2 is converted primarily to velocity. Fig.7. The flow velocity vl at the end of the duct 2 is obtained from the equation v = Co(2gFI) ^1/2 . Flerein, the pressure height FI corresponds to the pressure difference Dr = (pi - p2) on different sides of the end of the outflow duct.

The amount of flow-through of the flow duct 2 of the valve with a round cross- section is obtained approximately from the equation Q = C _D -A (2g H) ^1/2 , where Q [m ³ /s]; CD is the constant depending on the shape of the duct; A [m ² ] is the cross- sectional area of the duct; g is 9.81 m/s ² and FI is the pressure height in meters.

The qualities in the calculations must, however, be converted to cm ³ and mm ² . It is apparent from the equation that at constant throttle, when the flow volume increases by, for example, forty percent, the input pressure increases by one hundred percent. On the other hand, if the flow volume increases by forty percent and the cross-sectional area of the flow duct also increases by forty percent, the inflow pressure remains constant. The valve of the invention controls the flow cross- sectional area according to the flow volume so as to maintain the pressure constant.

As a result of the constant choke characteristic of the valve, the pressure in the reverse osmosis system does not suddenly increase as the high-pressure pump thereof starts and the rotating speed increases, because the volume of the flow through going through the duct 10 also increases with the system pressure. On the other hand, when the system is stopped, no residual pressure remains in the modules, which facilitates a gradual pressure increase when the system is restarted. If high pressure was to remain in the modules once the system is stopped, it would also cause the membranes to be clogged with salts, as no rinsing would take place. A continuous flow through the duct 10 also presents the particular advantage of the cone 5 not opening the duct 2 in one jump-like movement once the system pressure has been reached, but rather with continuous movement according to the increase in flow volume. Thus, the duct 10 presents many benefits.

The reject valve, whose outflow duct 2, cone 5 and arm 4 have been sized with relation to each other in such a way that when the valve is open, the force exerted on the cone 5 and the arm 4 thereof by the inflow is always equal to the reverse force exerted on the cone 5 by the spring 6, resulting in the cone 5 controlling the cross-sectional area of the outflow end 3 of the outflow duct 2 dynamically according to the fluctuation of the flow volume and the pressure in the system being maintained at a predetermined level.