Pumping Systems

Abstract

A system and method for pumping fluid. The system includes a sequence of two or more positive-displacement sub-systems each having a respective one-way inlet. A respective one-way flow path links each adjacent two of the sub-systems. A one-way outlet from a last of the sub-systems is provided. The system is capable of a mode of operation in which at least some of the sub-systems are substantially in phase with respect to each other to cause the system to draw fluid from more than one of the one-way inlets; and another other mode of operation in which at least some of the sub-systems are substantially in antiphase with respect to each other to increment a pressure of the fluid as the fluid moves along the sequence.

Claims

1. A method comprising: pumping fluid with a pumping arrangement, which comprises: a first system for pumping fluid, the first system comprising: a sequence of two or more sub-systems each being a respective positive-displacement sub-system and having a respective one-way inlet; a respective one-way flow path linking each adjacent two of the sub-systems; and a one-way outlet from a last of the sub-systems; wherein the pumping comprises: operating the pumping arrangement in a first mode of operation in which at least some of the sub-systems are substantially in phase with respect to each other to cause the first system to draw the fluid from more than one of the one-way inlets; and operating the pumping arrangement in a second mode of operation in which at least some of the sub-systems are substantially in antiphase with respect to each other to increment a pressure of the fluid as the fluid moves along the sequence to cause the pumping arrangement to deliver higher pressure and lower flow than the pumping arrangement delivers in the first mode; and wherein at least one of the sub-systems has a variable stroke length; and wherein a ratio of a stroke length of at least one of the sub-systems to a stroke length of another of the sub-systems downstream of the at least one of the sub-systems is lower, for the first mode than the ratio is for the second mode, to compensate for compression of the fluid as the fluid moves along the sequence.

2. The method of claim 1 wherein in the first mode of operation all of the sub-systems are substantially in phase with respect to each other.

3. The method of claim 1 wherein in the second mode of operation the sub-systems of each adjacent two of the sub-systems are substantially in antiphase with respect to each other to increment the pressure through each of the sub-systems.

4. The method of claim 1 wherein the sequence comprises more than two of the sub-systems.

5. The method of claim 1 wherein the sequence comprises four of the sub-systems.

6. The method of claim 5 wherein: in the first mode of operation the four sub-systems are in phase with respect to each other; the method comprises operating the pumping arrangement in an intermediate mode of operation in which: a first adjacent two of the sub-systems are substantially in phase with respect to each other; and a last adjacent two of the sub-systems are substantially in phase with respect to each other and substantially in antiphase with respect the first adjacent two of the sub-systems to deliver higher pressure and lower flow than the pumping arrangement delivers in the first mode; and in the second mode of operation, the sub-systems of each adjacent two of the four sub-systems are substantially in antiphase with respect to each other to deliver higher pressure and lower flow than the pumping arrangement delivers in the intermediate mode.

7. The method of claim 1 wherein two or more of the sub-systems are substantially identical to each other.

8. The method of claim 1 wherein all of the sub-systems are substantially identical to each other.

9. The method of claim 1 comprising the pumping arrangement automatically responding to feedback to deliver a target pressure when the first system is in at least one of the first mode of operation or the second mode of operation.

10. The method of claim 1 comprising the pumping arrangement automatically responding to feedback to deliver a target flow rate when the first system is in at least one of the first mode of operation or the second mode of operation.

11. The method of claim 1 comprising the pumping arrangement automatically responding to feedback, to maximize a delivery of the first system while maintaining one or more parameters of the first system within one or more respective limits, when the first system is in at least one of the first mode of operation or the second mode of operation.

12. The method of claim 1 comprising the pumping arrangement automatically responding to feedback to transition from the first mode of operation to the second mode of operation.

13. The method of claim 1 comprising the pumping arrangement automatically responding to feedback to equalize at least one parameter of at least one of the sub-systems with a corresponding parameter of at least one other of the sub-systems.

14. The method of claim 13 wherein the parameter is drive unit temperature.

15. The method of claim 1 wherein: the pumping arrangement comprises a second system, comprising: a sequence of two or more second-system sub-systems each being a positive-displacement sub-system and having a respective second-system one-way inlet; a respective second-system one-way flow path linking each adjacent two of the second-system sub-systems; and a second-system one-way outlet from a last of the second-system sub-systems; the pumping arrangement comprises double-acting pumps, each of which comprises: a respective sub-system of the first system; a respective second-system sub-system; and a respective drive arrangement for driving the respective sub-systems in antiphase to each other; the first mode of operation is a mode in which at least some of the second-system sub-systems are substantially in phase with respect to each other to cause the second system to draw the fluid from more than one of the second-system one-way inlets of the second system; and the second mode of operation is a mode in which at least some of the second-system sub-systems are substantially in antiphase with respect to each other to increment a pressure of the fluid as the fluid moves along the sequence of second-system sub-systems.

16. The method of claim 15 wherein: each respective pump, of the double-acting pumps, includes: a screw; a nut engaged with the screw; and a drive for rotationally driving the nut; and pumping the fluid comprises rotationally driving the nut.

17. The method of claim 16 wherein the drive comprises a stator coaxial to the screw.

18. The method of claim 16 wherein the nut and screw are together a ball screw.

19. The method of claim 1 wherein the pumping arrangement is capable of delivering about 80 kilopounds per square inch (ksi).

20. The method of claim 1 wherein the pumping arrangement is capable of delivering about 0.28 cubic feet per minute (cfm) (8 liters per minute (L/min)) at about 80 ksi.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0073] FIG. 1 illustrates a pumping arrangement in a first mode of operation;

[0074] FIG. 2 is an enlargement of detail 2 from FIG. 1;

[0075] FIG. 3 illustrates the arrangement of FIG. 1 in a second mode of operation;

[0076] FIG. 4 illustrates the arrangement of FIG. 1 in a third mode of operation;

[0077] FIG. 5 illustrates another pumping arrangement in a first mode of operation;

[0078] FIG. 6A illustrates plumbing by which a pump A is connected to a pump B;

[0079] FIG. 6B illustrates a detail of a variant in which a pair of conrod driven pistons are mounted in opposition in a common bore to define a pressure space;

[0080] FIG. 7 illustrates the arrangement of FIG. 5 in a second mode of operation;

[0081] FIG. 8 illustrates the arrangement of FIG. 5 in a third mode of operation; and

[0082] FIG. 9 is an axial cross-section view of an exemplary pump.

DESCRIPTION OF EMBODIMENTS

[0083] The pumping arrangement 1 includes four double-acting positive-displacement pumps A, B, C, D shared between the two pumping systems 3, 3′. The system 3, 3′ are plumbed in parallel between an inlet path 5 and an outlet path 7.

[0084] The arrangement 1 further includes another pump 9, in this case a centrifugal pump, for drawing fluid (in this case water) from an inlet path 11 and pressurising the inlet path 5. In this example the pump 9 provides 32 litres per minute at 100 psi to the inlet path 5. T-junction 13 divides the flow from the inlet path 5 between the systems 3, 3′.

[0085] The system 3 and pump A each include a movable element 15.sub.A for positively displacing fluid. In this example, the movable element 15.sub.A is a plunger carried within a cylinder 17.sub.A. In another example the element movable within the cylinder may be a piston. In yet another example the movable element might take the form of a diaphragm.

[0086] The pump A is a substantially symmetrical double-acting pump having the plunger cylinder arrangement 15.sub.A, 17.sub.A of the system 3 at one end and a plunger cylinder arrangement 15.sub.A′, 17.sub.A′ of the system 3′ at its other end. The plungers 15.sub.A, 15.sub.A′ are mutually connected by a screw portion (not shown) so that the plungers 15.sub.A, 15.sub.A′ and screw portion together form a unit (i.e. an arrangement that is movable as a unitary body which may or may not be fully rigid).

[0087] The screw portion is part of a drive 19.sub.A by which the unit 15.sub.A, 15.sub.A′ is reciprocally driven, that is alternately stroked left and right as illustrated in FIG. 1. The drive 19.sub.A includes a nut arrangement engaged with the unit 15.sub.A, 15.sub.A′ and encircled by a stator coaxial to the screw portion. The stator co-operates with the nut arrangement to form an electric motor by which the nut arrangement is rotated. The nut arrangement co-operates with the screw portion to convert this rotation into linear movement of the unit 15.sub.A, 15.sub.A′. A suitable encoder is provided to provide feedback to a control arrangement of the motor whereby the motor is a servo motor.

[0088] Preferably the nut arrangement incorporates rolling elements. In this particular example, the nut arrangement and screw portion are together a ball screw. Other examples of the pump A may take different forms, e.g. the form of a double-acting hydraulic intensifier.

[0089] The piston cylinder arrangement 15.sub.A, 17.sub.A constitutes a positive-displacement sub-system of the system 3. The sub-system has an inlet 21.sub.A by which fluid is supplied from the T-junction 13 to the sub-system 15.sub.A, 17.sub.A. The inlet 21.sub.A is equipped with a check valve 23.sub.A and thereby constitutes a one-way inlet. The pumps A, B, C, D are substantially identical to each other although in other examples they may be mutually different. Each pump’s respective inlet is arranged to in at least one mode of operation draw fluid from the T-junction 13, e.g. inlet 21.sub.B of the pump B is arranged to draw fluid from the junction 13.

[0090] The pump B includes a plunger 15.sub.B and cylinder 17.sub.B together forming a sub-system 15.sub.B, 17.sub.B. The sub-system 15.sub.A, 17.sub.A is connected to the sub-system 15.sub.B, 17.sub.B by a flow path 25.sub.AB defined by a conduit. A check valve 27.sub.AB is mounted along the flow path 25.sub.AB whereby the flow path 25.sub.AB is a one-way flow path. Similar one-way flow paths 25.sub.BC, 25.sub.CD mutually connect the pumps B, C and pumps C, D respectively whereby the portions shared by the pumps A, B, C, D and the system 3 form a sequence of sub-systems along which the fluid is pumped from the T-junction 13 to the outlet 7.

[0091] The pump D defines a last sub-system 15.sub.D, 17.sub.D of the sequence. An outlet path 25.sub.D connects the sub-system 15.sub.D, 17.sub.D to a T-junction 29. The T-junction 29 also opens to the outlet path 25.sub.D′ to receive fluid from the system 3′ and to the outlet. The flow path 25.sub.D is equipped with a check valve 27.sub.D and is thereby a one-way flow path.

[0092] FIG. 2 illustrates a hatched region 31.sub.B corresponding to an internal pressure volume of the sub-system 15.sub.B, 17.sub.B including an elongate portion extending through a tubular cavity from the plunger-receiving bore of the cylinder 17.sub.B. The internal pressure volume 31.sub.B includes a portion connecting the flow path 25.sub.AB to the flow 25.sub.BC. In this example the flow path 25.sub.BC is equipped with a check valve 27.sub.BCi: at its start in addition to the check valve 27.sub.BC at its end.

[0093] The check valve 23.sub.A is an inlet check valve to the sub-system 15.sub.A, 17.sub.A whilst the check valve 27.sub.AB functions as an outlet check valve from that system whereby reciprocal movement of the plunger 15.sub.A within the cylinder 17.sub.A pumps fluid from the inlet 21.sub.A to the outlet 27.sub.AB. In this example the plunger 15.sub.A is movable whilst the cylinder 17.sub.A is stationary although relative movement could be achieved in other ways, e.g. the plunger 15.sub.A could be held stationary whilst the cylinder 17.sub.A is moved.

[0094] FIG. 1 illustrates a mode of operation in which each of the four sub-systems are operated in phase with respect to each other. At the moment illustrated in FIG. 1 the plungers 15 are simultaneously advancing towards the right-hand side on their respective compression strokes whereby each of the sub-systems of the system 3 simultaneously delivers about the same volume of water (8 litres per minute in this example) to about the same pressure (20 ksi in this example). During this phase of operation, the delivery of the sub-system 15.sub.A, 17.sub.A passes through the internal pressure space 31.sub.B of the second sub-system 15.sub.B, 17.sub.B.

[0095] In this example the plungers 15.sub.A, 15.sub.A′ are part of a common unit and the cylinders 17.sub.A, 17.sub.A′ are fixed relative to each other whereby the sub-system 15.sub.A, 17.sub.A always operates in antiphase to the sub-system 15.sub.A′, 17.sub.A′. As such in the moment illustrated in FIG. 1 each of the four sub-systems of the system 3′ draws in about the same volume (in this case about 8 litres per minute) of fluid at about the same pressure (about 100 psi in this example).

[0096] The pump 9 supplies the systems 3, 3′ with about 32 litres per minute at about 100 psi. At the moment illustrated in FIG. 1, the system 3 passes on that about 32 litres per minute at about 20 ksi to the outlet 7. When the pumps reach the end of the illustrated stroke and reverse their direction the arrangement 1 switches to the sub-system 3′ pumping that volumetric rate at that pressure.

[0097] The mode of FIG. 1 is a high-flow low-pressure pumping mode in which the pumps A, B, C, D are effectively in parallel to each other. FIG. 3 illustrates an intermediate pressure, intermediate flow mode. Pumps A, B are operated in phase with respect to each other whilst the pumps C, D are operated in phase with respect to each other but in antiphase to the pumps A, B.

[0098] In the moment of FIG. 3 the sub-systems 15.sub.A, 17.sub.A and 15.sub.B, 17.sub.B are on their compression stroke to supply to the fluid path 25.sub.BC about 16 litres per minute at about 20 ksi. The sub-systems 15.sub.C, 17.sub.C and 15.sub.D, 17.sub.D are on their intake strokes to receive the fluid from the path 25.sub.BC.

[0099] A stroke length of the sub-systems 15.sub.C, 17.sub.C and 15.sub.D, 17.sub.D is reduced relative to a stroke length of the sub-systems 15.sub.A, 17.sub.A and 15.sub.B, 17.sub.B by an amount commensurate with the compressibility of the fluid between a pressure at the inlet 21.sub.A (100 psi in this example) and the desired line pressure along the conduit 25.sub.BC (20 ksi in this example). Accordingly, the two upstream sub-systems hydraulically power the two downstream sub-systems; the 20 ksi pressure does work on those sub-systems by pushing the plungers 15.sub.C, 15.sub.D to the left as drawn in FIG. 3. The units 15.sub.C, 15.sub.C′ and 15.sub.D, 15.sub.D′ each constitutes a mechanism by which the power from the preceding sub-systems is utilised to pump fluid. In this example the 20 ksi pressure acting on the plunger 15.sub.C helps to move the unit 15.sub.C, 15.sub.C′ to the left as drawn whereby power from the sub-systems 15.sub.A, 17.sub.A and 15.sub.B, 17.sub.B is utilised to pump fluid from the sub-system 15.sub.C′, 17.sub.C′.

[0100] As such the pumps C, D are able to pump fluid to 40 ksi whilst the load borne by each of the drives 19.sub.C, 19.sub.D corresponds to only 20 ksi corresponding to the 20 ksi pressure difference between the cylinders 17.sub.C, 17.sub.D on the one hand and the cylinders 17.sub.C′, 17.sub.D′ on the other hand.

[0101] Accordingly, in the moment of FIG. 3 the first two sub-systems of the system 3′ draws about 16 litres per minute at about 100 psi from the inlet 5 whilst the last two sub-systems of the system 3′ delivers about 16 litres per minute at about 40 ksi to the outlet path 25.sub.D′ with the assistance of 20 ksi of supporting pressure from the system 3. At the end of the illustrated stroke, the systems 3, 3′ switch roles whereby the outlet 7 provides a delivery of about 16 litres per minute at about 40 ksi. The delivery is substantially continuous but for a pressure pulse associated with the stroke reversal. An accumulator may be provided to attenuate the pressure pulse.

[0102] FIG. 4 illustrates a high-pressure low-flow mode in which the arrangement 1 delivers about 8 litres per minute at about 80 ksi. The pumps are effectively in series to each other. In this mode each adjacent two of the pumps A, B, C, D (and the sub-systems defined thereby) are operated in antiphase with respect to each other. From sub-system to sub-system the stroke length is incrementally reduced by an amount corresponding to the incremental compression of the fluid between the sub-systems. As such within the system 3, the pressure is incremented through the first sub-system to 20 ksi, through the second sub-system to 40 ksi, through the third sub-system to 60 ksi and through the fourth sub-system to 80 ksi in the outlet path 25.sub.D. Each compression stroke is partly powered by fluid pressure from the corresponding intake stroke of the counterpart sub-system of the other system whereby again, each of the drive mechanisms sees resistance corresponding to the 20 ksi pressure difference across the two plungers with which it is associated.

[0103] Many variations of the illustrated principles are possible. By way of example, for applications for which 40 ksi is sufficient pressure the plungers 15.sub.A, 15.sub.B and cylinders 17.sub.A, 17.sub.B may together constitute a single sub-system and share a single one-way inlet between them.

[0104] In another variation one of the pumps A, B, C, D might be replaced by an alternate, potentially lower cost, pump that does not have a variable stroke length in which case the stroke length of the other pumps might be varied in relation to the stroke of the fixed-stroke pump. Indeed, the illustrated principles can be applied to a pumping system having no variable-stroke sub-systems, e.g. a bore diameter of the pumps A, B, C, D might decrement along the sequence by amounts corresponding to the compressibility of the fluid in the high-pressure/low-flow mode.

[0105] Systems having at least one variable-stroke pumping sub-system are preferred in that they are suited to more advanced control strategies to better share the pumping burden between the sub-systems; e.g. whilst a simple implementation may entail each sub-system having a respective fixed stroke length for each of the operating modes, other control strategies may take account of feedback, e.g. feedback from pressure and/or temperature sensor(s), to dynamically vary the stroke length(s) of one or more of the sub-systems by shifting one or both end points of the stroke. For example, in one example where it is desirable to maximise the output of the pumping system, the stroke length of each sub-system could be controlled in response (e.g. in negative relation to) a temperature of its respective drive unit, whereby each drive unit could be worked to its maximum sustainable temperature. The temperature of the drive unit could be measured via a sensor and/or inferred from data related to the impedance of the drive unit’s windings.

[0106] In other examples, e.g. wherein the life of the ball screw is a limiting factor by which service intervals are determined, it may be desirable to control each of the stages to deliver a constant force. In yet other examples, it may be desirable to cause each sub-system to deliver a common amount of fluid power.

[0107] In the illustrated example the mechanism in which power from a preceding chamber is utilised to pump fluid takes the form of a unit for transmitting power from one chamber to another. In another example a sub-system that is powered during its intake stroke may utilise the power from that intake stroke during its compression stroke e.g. a sub-system could take the form of a simple single piston compressor driven by an internal combustion motor and having a flywheel mass which flywheel mass is accelerated by the fluid power during the intake stroke then returns that power to the fluid during the compression stroke. With such a pump, the preceding parallel-series principles disclosed herein could be implemented in a simple two-cylinder system.

[0108] According to preferred variants of the described systems, the described phase shifting is sufficient to change between the described modes and thereby change the pressure flow characteristics of the pumping system. Advantageously the cost associated with dedicated control valves (etc) can be avoided by making better use of the drive-control arrangements; e.g. there are no electromechanical control valves along any flow path connecting any adjacent two sub-systems. Indeed, preferred forms of the system do not include any electronically switchable valves. This advantageously eliminates a number of potential failure points and thereby improves reliability and reduces maintenance costs relative to switching between parallel and series operation by switching a multitude of electromechanical valves.

[0109] The pumping arrangement 1 is well adapted to pumping against variable resistances such as pressuring a vessel. The present inventors have considered one application that entails pressuring a vessel to 87 ksi to process food. On the basis that the food contained in the vessel has similar properties (compressibility) to water, an additional 12% of water must be pumped into the vessel. The pressure against which the pump must work ranges from zero (atmospheric) pressure at the outset to 87 ksi at completion. Using a variant of the arrangement 1 in which the plungers each have a diameter of about 32 mm acceptable filling times are achievable.

[0110] Plunger diameters of (approximately) 14 mm, 15 mm, 16 mm, 17.5 mm and 19 mm are also contemplated as are stroke lengths in the range of 120 mm to 170 mm. Maximum plunger speeds in the vicinity of 350 mm per second corresponding to a stroke rate of about 55 cycles per minute are also contemplated.

[0111] FIGS. 5 to 8 illustrate yet another alternate implementation of the principles disclosed herein. The system 100 includes the pumps A, B, C, D connected by an alternate plumbing arrangement. The pumps A, B, C, D are fluidly connected end to end. Of course, they need not be physically end to end. The connecting conduits could be curved, etc. FIG. 6A illustrates plumbing by which the pump A is connected to the pump B. Each adjacent two of the pumps B, C, D are mutually connected by a similar plumbing arrangement.

[0112] A conduit 133.sub.AB mutually connects cylinders 17.sub.A, 17.sub.B so that those two cylinders are at substantially the same pressure throughout all modes of operation. An inlet 135.sub.AB connects the conduit 133 with an inlet rail 139. The inlet 135.sub.AB is equipped with a check valve 137.sub.AB and is thereby a one-way inlet. Likewise, an outlet 141.sub.AB connects the conduit 133.sub.AB to an outlet rail 145 and is equipped with a check valve 143.sub.AB so as to be a one-way outlet. The plungers 15.sub.A, 15.sub.B′, cylinders 17.sub.A, 17.sub.B and check valves 137.sub.AB, 143.sub.AB together define an internal pressure space 147.sub.AB as suggested by hatching in FIG. 6A.

[0113] The first and last systems 15.sub.A′, 17.sub.A′ and 15.sub.D, 17.sub.D are likewise connected between the inlet and outlet rails 139, 145 via check valves.

[0114] The system 15.sub.A′, 17.sub.A′ defines an internal pressure space 147.sub.A and the system 15.sub.D, 17.sub.D defines an internal pressure space 147.sub.D each of which is akin to the pressure space 147.sub.AB.

[0115] FIG. 5 illustrates the system 100 in a first mode of operation being a relatively high-flow low-pressure mode. Fluid is supplied to the intake rail 139 at about 32 litres per minute and about 100 psi via the inlet 5. Each adjacent two of the plungers is operated in phase with respect to each other, e.g. the plunger 15.sub.A and plunger 15.sub.B′ are each on their intake strokes so as to increase the volume of the pressure space 147.sub.B and thereby draw about 16 litres per minute of fluid into the pressure space 147.sub.AB via the inlet 135.sub.AB. Mutually adjacent plungers 15.sub.B, 15.sub.C′ are each on their compression stroke whereby about 16 litres per minute of fluid at about 20 ksi is driven from pressure space 147.sub.BC. The plungers 15.sub.A′, 15.sub.D are each on their compression stroke to each drive about 8 litres per minute of fluid towards the outlet rail 145.

[0116] As such, at the moment illustrated in FIG. 5 [0117] Fluid is being drawn into the pressure spacers 147.sub.AB, 147.sub.CD at a combined rate of about 32 litres per minute; and [0118] Fluid is being expelled from the pressure spacers 147.sub.A, 147.sub.BC, 147.sub.D at about the same rate to supply about 32 litres per minute at about 20 ksi to the outlet 7.

[0119] FIG. 7 illustrates the system 100 in an intermediate flow/intermediate pressure mode in which the plungers 15.sub.A, 15.sub.B′ are operated in antiphase to each other. At the moment illustrated in FIG. 7 the plunger 15.sub.B′ is on its compression stroke whilst the plunger 15.sub.A is on its retraction stroke the length of which is adjusted to hold the pressure space 147.sub.AB at about 20 ksi to power the pump A to deliver to the rail 145 fluid from the pressure space 147.sub.A at about 8 litres per minute at about 40 ksi whilst the drive 19.sub.A only has to work against a pressure difference of about 20 ksi. The adjacent plungers of the pumps C, D are likewise operated in antiphase to each other whereby the pressure space 147.sub.CD does not itself act as a pumping chamber mutually connecting the rails 139, 147 but instead serves as a fluid-power transmission by which the pump C powers the pump D. The adjacent plungers of the pumps B, C are operated in phase with each other whereby at the moment illustrated in FIG. 7 about 16 litres per minute of fluid is drawn into the pressure space 147.sub.BC and upon reversal of the stroke direction fluid is pumped from that space at about the same rate to the rail 145.

[0120] FIG. 8 illustrates the system 100 in its high-pressure/low-flow mode in which each adjacent two of the plungers are operated in antiphase to each other so that the intermediate pressure spaces 147.sub.AB, 147.sub.BC, 147.sub.CD form fluid-power transmissions akin to the spaces 147.sub.AB, 147.sub.CD in the intermediate mode.

[0121] In the high-pressure/low-flow mode only the spaces 147.sub.A, 147.sub.D convey fluid from the rail 139 to the fluid 145.sub.B. In the moment illustrated in FIG. 8 the pump D receives power from the pumps A, B, C, D to enable it to deliver about 8 litres per minute at about 80 ksi. At this moment the pressure in the intermediate pressure spaces 147.sub.AD, 147.sub.BC, 147.sub.CD are held at about 20 ksi, 40 ksi and 60 ksi respectively, whereby each of the drive mechanisms works against only an about 20 ksi pressure difference.

[0122] FIG. 6B illustrates a detail of a variant of the system 100 in which a pair of conrod driven pistons 115.sub.A, 115.sub.B are mounted in opposition in a common bore 117.sub.AB to define the pressure space 147.sub.AB.

[0123] The pump 201 of FIG. 9 is a two-stage pump including a first stage 203 serially connected to a second stage 205. By way of example, each of the stages 203, 205 may be defined by a respective gas head. A drive arrangement 207 in the form of a linear actuator 207a drives both stages 203, 205.

[0124] The stages are driven in antiphase to each other, i.e. when the first stage is on its intake stroke, the second stage is on its outlet stroke and vice versa. The first stage 203 includes a cylinder 203a in which a plunger 207b is received to define a pumping chamber 203b. The plunger 207b is stroked (i.e. moved back and forth) by the drive arrangement 207 to pump fluid through the chamber 203b.

[0125] An inlet 209 opens into the chamber 203b. The inlet 209 is equipped with a check valve 209a and as such is a one-way flow inlet.

[0126] The plunger 207b is one end of a unit. The other end of the unit is formed by a plunger 207c. The plunger 207c is received within a cylinder 205a to define a chamber 205b. Advantageously, a cross-sectional area of the plunger 207b is larger than the cross-sectional area of the plunger 207c. In this example, the first stage has a capacity (i.e. swept volume) of about 2 L whilst the second stage has a capacity of about 1 L.

[0127] A screw (not shown) connects the plungers 207b, 207c to each other and is embraced by a nut arrangement (not shown). Preferably the nut and screw are together a ball screw arrangement. Most preferably the nut is embraced by a stator coaxial to the screw to form a drive akin to the 19.sub.A.

[0128] A sensor is provided to provide an indication of the position of the unit 207b, 207c. In this case, the sensor takes the form of a Renishaw LM10 incremental encoder arranged to measure the rotation of the nut arrangement. That rotation is an indication of the position of the unit 207b, 207c in that it is relatable to the position of the unit via the pitch of the ball screw. Preferably the ball screw has a 25 mm pitch. Of course, other forms of sensor are possible – a linear encoder may be used to directly measure the position of the unit 207b, 207c, or some arrangement of proximity switches may be effective. Indeed, feedback from the motor itself may be useful. Potentially the unit 207b, 207c might be periodically homed against one end or the other to calibrate the position-monitoring system.

[0129] The drive arrangement 207 preferably includes a control arrangement responsive to this position feedback whereby the motor, sensor and control arrangement together form a servo motor.

[0130] An outlet 211 from the chamber 203b carries the pumped fluid to a heat exchanger 213 and onwards to the chamber 205b. The conduit 211 is equipped with a check valve 211a and as such constitutes a one-way flow path. Reciprocal operation of the second stage 205 pumps fluid from the conduit 211 to an outlet conduit 215.

[0131] In operation, the unit 207b, 207c is stroked along the stroke S. FIG. 9 illustrates the unit 207b, 207c being moved to the right, corresponding to the intake stroke of the stage 203 as suggested by the arrow A and the compression stroke of the stage 205 as suggested by the arrow B.

[0132] When compressing fluids, the heat of compression can be problematic. If the pump is not properly controlled, the stages 203, 205 can overheat, potentially resulting in damage such as damage to a seal between the cylinder 203a and the plunger 207b and/or damage to the check valves. Cooling water is supplied to the heat exchanger 213 to cool the fluid en route from the first stage 203 to the second stage 205. Nonetheless, the present inventors have recognised that the temperatures resulting from the heat of compression can be a limiting factor, and that when existing control strategies are implemented to limit the hottest end of the pump to an acceptable temperature, inevitably the other end of the pump will be cooler and therefore more than likely used to less than its full potential.

[0133] The present inventors have recognised that the two ends of the pump can be balanced and thereby more efficiently utilised by varying at least one end point of the stroke S. The shift may be to further balance other aspects of the system. In one example, the average load on each end may be balanced to achieve equal wear and loading on either end to maximise product life.

[0134] In one convenient implementation, a temperature of the second stage 205 may be measured at any convenient location (e.g. along the outlet conduit 215). This feedback can be provided to the control arrangement of the drive arrangement 207 to enable the control arrangement 207 to vary a stroke rate to hold the second stage 205 at a desirable temperature.

[0135] At the same time, a temperature of the first stage may be measured (e.g. at any convenient location upstream of the heat exchanger 213) and an end point of the stroke S adjusted to regulate that temperature.

[0136] Preferably it is the end point of the compression stroke of the plunger 207b that is limited so as to control the end clearance between the plunger 207b and the end of the chamber 203b. Since, in this example, the plungers 207b, 207c are part of a common unit, controlling the end point of the compression stroke of the first stage inherently also entails varying the start point of the compression stroke of the second stage 205. Of course, a change of a few millimetres at the end of a compression stroke has a far more significant impact on the compression ratio of a particular stage than the same variation of a few millimetres at the start of a stage’s compression stroke.

[0137] In this way, the temperatures at the ends of the pump may be relatively controlled (i.e. controlled relative to each other). This relative control may entail controlling one, or the other, or both of the ends of the pump. The relative control may be to equalise the temperatures at the ends of the pump. Alternatively, a fixed relativity may be maintained. By way of example, in some applications the seals at the lower-pressure end of the pump may be capable of withstanding higher temperatures than the seals at the higher-pressure end in which case it may be desirable to hold the temperature of the lower-pressure end a fixed proportion or amount above the temperature of the higher-pressure end so that both pumping chambers are used to their full potential.

[0138] Other implementations of the concept are possible. By way of example, instead of the two temperature sensors mentioned above, the end clearance(s) might be controlled in response to: [0139] pressure transducers indicating the output pressure of each cylinder, e.g. a trio of pressure sensors may be associated with the conduits 209, 211 and 215; [0140] feedback from the drive arrangement 207, e.g. to equalise the amount of mechanical work performed on each stroke; [0141] other variables; or [0142] some combination of two or more of the previously mentioned potential inputs.

[0143] In operation of the pump 201 gas may be [0144] supplied via the inlet 209 at 100 psi and about ambient temperature; [0145] pumped through the chamber 203b to about 600 psi and about 200° C.; [0146] cooled through the heat exchanger 213 to about ambient temperature; and then [0147] pumped through the chamber 205b to about 700 psi and 200° C.

[0148] A further heat exchanger (not shown) may be mounted along the outlet path to cool the fluid, for example en route to a vessel that is being filled.

[0149] The present inventors have recognised that balancing the output pressures of the cylinders of a multi-cylinder pump can be problematic. If the pump is not properly controlled, the output pressure of the first stage may reach its maximum pressure limit before the desired output pressure from the final stage is achieved.

[0150] The balance of pressures is complicated. It depends on the incoming fluid pressure and temperature, as well as the developed compression ratio in each stage, as well as the cooling of the intermediate stage.

[0151] In practice, the incoming pressure (e.g. the pressure at inlet 209) can be highly variable, depending on the equipment supplying the gas to the pump, and can have a large effect on the output pressure from the first stage.

[0152] For example, the system may be optimised for 100 psi incoming pressure, such that the inter-stage pressure just remains below its maximum pressure of 700 psi. If the incoming pressure is however increased to 150 psi, with all other factors remaining fixed, then the inter-stage pressure will exceed the maximum of 700 psi.

[0153] The present inventors have recognised that when existing control strategies are implemented to limit either end of the pump to be within their maximum pressure envelope the other end may not be utilised to its full capacity.

[0154] Preferred forms of the various disclosed apparatus are suited to pumping gases such as nitrogen, butane, hydrogen, carbon dioxide and oxygen. For example, a variant may be used to pump butane gas used in the extraction of oil from marijuana leaves, or to compress the gas coming off a hydrogen reformer. Other embodiments could be used for compressing gases for feedstock for an ethylene plant. Indeed, some variants may be configured to pump liquid, e.g. water for liquid-cutting such as waterjet cutting.

[0155] In FIG. 9 the stages 203, 205 are mutually connected in series. In another example the two pumping chambers may be plumbed in parallel to each other. Indeed, other embodiments of the principles disclosed herein may take a form entirely different to the pump 201. By way of example, in the context of a single piston compressor driven by an internal combustion engine and a crank and conrod arrangement, the end points of the stroke of the cylinder relative to its cylinder may be varied by axially moving the cylinder and/or varying an effective length of the conrod. Such an arrangement may advantageously allow the internal combustion engine to continue operating at peak efficiency whilst the pressure faced by the pump varies.

[0156] The principles discussed in relation to FIG. 9 may also be applied to variants of the pumping systems disclosed in the preceding figures. By way of example, the temperatures of the check valves 27.sub.AB, 27.sub.BC, 27.sub.CD, 27.sub.D may be monitored by suitable sensors and the stroke end points of the pumps A, B, C, D controlled, e.g. to limit the sensed temperatures to protect the seals of the pumping chambers. In a preferred form, a respective heat exchanger is positioned between each adjacent two of the sub-systems of the system 3.

[0157] Various control strategies may be implemented to shift the end point of a stroke. Various of the illustrated examples incorporate an encoder for providing feedback indicative of a position of the stroked element and in response to which the corresponding drive unit can be controlled to move the stroked element to a predetermined end point. Other implementations of the disclosed principles are possible without such an encoder or any similar positional feedback. By way of example: [0158] the pump A may have position switches in response to which the stroked element 15.sub.A, 15.sub.A′ is reversed; and [0159] the drive units of the pumps B, C, D may be controlled to deliver a constant force and to simply reverse direction whenever the pump A reverses direction.

[0160] Through this logic, the end points of the strokes of the pumps B, C, D may be shifted. Of course, other control strategies that result in the end point being shifted are possible.

[0161] The invention is not limited to the illustrated examples. Rather, the invention is defined by the claims.