EP2917581A1

EP2917581A1 - Peristaltic pump and method of controlling a peristaltic pump

Info

Publication number: EP2917581A1
Application number: EP13771568.6A
Authority: EP
Inventors: Duncan Guthrie; Adrian CLARKSON
Original assignee: Vapourtec Ltd
Current assignee: Vapourtec Ltd
Priority date: 2012-09-14
Filing date: 2013-09-13
Publication date: 2015-09-16
Anticipated expiration: 2033-09-13
Also published as: WO2014041361A1; EP2917581B1; GB201216462D0; US20150240802A1

Abstract

Provided is a pump and a method of controlling a pump. The pump and method are particularly for use in dispensing reagents, for example in flow chemistry, and more particularly on a laboratory scale. The pump aims to provide a substantially constant output flow of fluid. The pump comprises: a motor; a peristaltic pump having a rotor driven by the motor; a pressure sensor monitoring the pressure of the pumped fluid downstream of the pump; and a control unit which controls the motor by adjusting the standard operating speed of the motor according to the pressure detected by the pressure sensor, such that the pump operates continuously at a rate set by an operator. Embodiments of the pump make use of lookup tables to determine a desired position of the pump rotor at each point in the cycle with the entry point into the lookup table being determined by the feedback from the pressure sensor. Long term changes in the performance of the pump can also be accounted for by changing the entry in the lookup table which is consulted.

Description

PERISTALTIC PUMP AND METHOD OF CONTROLLING A PERISTALTIC PUMP

Field of the Invention

The present invention relates to a pump and a method of controlling a pump. The invention is particularly, but not exclusively, concerned with a pump for use in flow chemistry or for dispensing reagents and a method of controlling such a pump.

Background of the Invention

Flow chemistry is a process whereby a chemical reaction is run in a continuously flowing stream rather than in one or more batches. Although long used in manufacturing processes on a large scale, flow chemistry has only recently been developed in the research/laboratory environment.

A key characteristic of flow chemistry is the need to pump liquids to and normally through reaction vessels. However, this pumping can be the source of problems, particularly on the laboratory scale, for example where corrosive reagents or reagents with a high concentration of particulates are used in the reaction.

Where reagents contain particulates existing pumps used in laboratory scale flow chemistry tend to be peristaltic pumps operating at around 1-2 bar of output pressure. These pumps have typically been unable to deal with corrosive reagents, have operated at relatively low pressure and have had irregular output flow rate. Recent advances in materials have led to the development of composite tubes containing perf!uoroe!astomers and fiuoroeiastomers. Examples of these materials include the style 400 and style 500 tubes supplied by GORE© Reinforced Elastomers W.L.Gore & Associates, Inc. 2401 Singerly Road Elkton, MD 21921. By selecting the correct material for the reagent being pumped the tubes are able to cope with acids, bases and organic solvents without corrosion or physical deterioration, but this does not address the problems of low pressure and irregular flows.

Various forms of pumps have been developed which address the above problems. Syringe pumps are used, but these are obviously limited in their capacity, even if paired, which requires change over in order to provide a continuous supply of reagent. Positive

displacement pumps have also been used which have small pistons (of the order of 50μΙ) which constantly draw in reagent from a supply and pump it out to the reaction chamber. However, the pistons in such pumps require non-return valves to prevent the reagent from being forced back into the supply rather than out to the reaction chamber and such valves do not cope well with precipitates, air bubbles in the reagent or immiscible liquids. Figure 6 shows the performance of a typical prior art peristaltic pump, which is controlled to achieve a constant speed of rotation (middle graph). The top and bottom graphs illustrate that, in this configuration, there is a wide variation in pressure and flow rate downstream of the pump.

Aspects of the present invention seek to address, overcome or ameliorate the above shortcomings of existing pumps used in flow chemistry. Preferably such pumps are able to cope with a wide range of reagents, including those with high levels of precipitates, and/or able to maintain a constant or substantially constant flow rate and/or able to provide reagents at a high pressure (e.g. up to 10 bar or more) to a reaction chamber.

Summary of the Invention At its broadest, a first aspect of the present invention provides a pump which uses feedback from a pressure sensor to control the operation of the pump. The pump is preferably suitable for use in dispensing reagents, and particularly, but not exclusively, in flow chemistry applications.

Accordingly, a first aspect of the present invention preferably provides a pump for use in dispensing reagents, the pump comprising: a motor; a peristaltic pump having a rotor driven by the motor; a pressure sensor monitoring the pressure of the pumped fluid downstream of the pump; and a control unit which controls the motor by adjusting the standard operating speed of the motor according to the pressure detected by the pressure sensor, such that the pump operates continuously at a rate set by an operator. Adjusting the operating speed according the detected pressure provides a more accurate and more rapid feedback than feedback based on flow rates and eliminates the need for mass flow sensing which can be highly problematic with corrosive solutions.

The output flow of the pump has been found to be very closely correlated to output pressure, even though it is pumping against a back pressure regulator which attempts to regulate pressure. Pressure regulators are slightly non-linear, so that an increase in flow produces a small increase in pressure. Other effects, such as system tubing losses, and inertia! mass of the fluid being pumped, combine to the effect that variations in outlet flow produce detectable variations in outlet pressure.

Therefore, precisely regulating outlet pressure, by means of adjusting rotor speed, will produce a steady outlet flow. It has been found that pumps according to embodiments of the present invention are able to maintain outlet flow downstream of the pump within a 5% variation of the desired flow rate.

Preferably the control unit compares the pressure detected by the pressure sensor with a target pressure and adjusts the speed of the motor accordingly. Preferably the sole user input to the controller is a speed set point. However, in other arrangements the user input may be a desired flow rate.

Preferably the control unit also receives a measurement of the current position of the rotor and compares that position to a desired position of the rotor, and adjusts the speed of the motor to compensate for any difference between the current position and the desired position.

In this configuration, there are two feedback parameters available the actual rotor position and the fluid pressure at the pump outlet.

Preferably the control unit determines a desired position of the rotor throughout its rotary cycle from a lookup table according to the rate set by the operator, The transfer function of a pump is typically highly non-linear, and varies over operating pressure, and with tubing type and condition. To compensate for this, the lookup table can provide curves describing the inverse of the transfer function which can be captured using a purely open-loop controller, running at constant pressure and flow.

Ideally several such transfer function maps are captured and provided at different pressures. In this arrangement the desired position of the rotor throughout the rotary cycle in the lookup table entries can be determined by the position of the rotor required to maintain a constant output pressure.

Preferably the control unit adjusts or selects the point of entry into the lookup table from which the desired position of the rotor is determined according to a target pressure. The appropriate transfer function map can thus be selected for use, with the target pressure largely dictating the map choice. This transfer function map then forms an idealised model of where the rotor should be at any point in time, to produce steady flow, and preferably dominates the behaviour of the controller to provide a system largely immune to outside disturbances.

Preferably, if the lookup table does not contain an entry for the output pressure in question, the control unit interpolates between adjacent entries in the lookup table to determine the desired position of the rotor throughout the rotary cycle.

The entries in the lookup table are typically discrete entries for the selected pressures.

These pressures may not exactly match the target pressure. Accordingly, the lookup preferably interpolates between pairs of adjacent transfer functions, so effectively a continuous range of curves are generated for a continuous range of pressure.

Preferably the control unit receives a measurement of the current position of the rotor and compares that position to a desired position of the rotor, and adjusts the speed of the motor to compensate for any difference between the current position and the desired position.

In a particularly preferred arrangement the control unit calculates the speed that will cause the difference between the current position and the desired position to zero by the time of the next measurement.

In one embodiment, the motor is a stepper motor. A stepper motor is particularly preferred as it provides constant information about the position of the motor drive shaft and therefore the position of the rotor. However, other motor types could be used, but preferably the motor chosen allows precise speed control and feedback information about the position of the rotor either from the drive of the motor itself or from a position encoder.

Preferably the control unit monitors the performance of the pump over time using the measured pressure and adjusts the entry in the lookup table which is consulted according to the observed performance of the pump.

This allows the control unit to take account of variations in the tubing properties in the pump, and over the life of the tube, and possibly between tubes of the same type due to

manufacturing differences. For example, the control unit may be arranged such that, if the measured pressure undershoots a target pressure after the high speed section, a higher pressure map is used with a longer fast section. Conversely, if the measured pressure overshoots after the fast section, a lower pressure map is selected. Preferably any change is blended in gradually over the course of a revolution of the pump.

Providing long term matching of the selected lookup table entry based on the performance of the pump allows lower short term control inputs to be made, which aids stability and immunity to fluctuations in flow caused by pressure transients from sources external to the pump, such as another pump operating upstream or downstream of the pump in question. Preferably a target pressure level is maintained to provide a set point for the control unit. In a preferred arrangement, the target pressure is a low-pass filtered version of the outlet pressure. The filter is preferably designed to maintain a constant level during normal operation, ignoring the pressure transients caused by the fast section of rotor travel. The filter may be gated so that it will track to the actual pressure level quickly during system start-up and pressure regulator adjustments, if the actual pressure maintains a level significantly different from the set point pressure for too long (several hundred milliseconds) the target pressure is quickly adjusted to match.

Preferably the control unit is arranged to adjust the value used to look up in the lookup table, not the output value of the lookup fable. Typically for a peristaltic pump, the transfer function has a period of slow movement (corresponding to the standard rotation of the rotor with the rolier(s) in contact with the tubing) and a short period of fast movement (corresponding to the point at which a roller disengages with the tubing and so the rotor has to accelerate to maintain the pressure in the tubing and avoid flow back into the section of tubing behind the disengaged roller. By adjusting the input to the lookup table based on the measured pressure, the effect of the control unit feedback can be limited to a small range of travel on the input axis, which produces an even smaller effect on the rotor movement during the slow portions of the transfer function. However, there is some variability in the timing of the release of the nip and the above described arrangement can allow the control unit to execute the full magnitude of the fast section of travel early, or delay it slightly, depending on the pressure sensor feedback. The fast section of the transfer function of a peristaltic pump can provide for about 50% of the total movement of the rotor in a single cycle at higher pressures, so the control unit would have to have limits of at least this range if it were adjusting map output values. Using the input side of the lookup table, adjustments of only around 5-10% of the input value may be needed.

Furthermore, the limited influence of the feedback from the pressure sensor in the slow portions of the lookup fable can provide significant immunity of the pump control to external pressure disturbances, such as those caused by another pump operating further

downstream. The pump of the first aspect may have some, ail or none of the above described optional and preferred features.

At its broadest, a second aspect of the present invention provides a method for controlling a pump which uses the pressure detected downstream of the pump to control the operation of the pump. The pump being controlled is preferably suitable for use in dispensing reagents, and particularly, but not exclusively, in flow chemistry applications.

Accordingly, a second aspect of the present invention preferably provides a method of controlling a peristaltic pump having a rotor driven by a motor, the method comprising the steps of: receiving a desired operating rate from an operator; detecting the pressure of the pumped fluid downstream of the pump: and adjusting the standard operating speed of the motor according to the detected pressure such that the pump operates continuously at the desired operating rate.

Adjusting the operating speed according the detected pressure provides a more accurate and more rapid feedback than feedback based on flow rates and eliminates the need for mass flow sensing which can be highly problematic with corrosive solutions. The output flow of the pump has been found to be very closely correlated to output pressure, even though it is pumping against a back pressure regulator which attempts to regulate pressure. Pressure regulators are slightly non-linear, so that an increase in flow produces a small increase in pressure. Other effects, such as system tubing losses, and inertial mass of the fluid being pumped, combine to the effect that variations in outlet flow produce detectable variations in outlet pressure. Therefore, precisely regulating outlet pressure, by means of adjusting rotor speed, will produce a steady outlet flow. It has been found that methods according to embodiments of the present invention are able to maintain outlet flow downstream of the pump within a 5% variation of the desired flow rate. Preferably the method further includes the steps of: comparing the pressure detected by the pressure sensor with a target pressure; and adjusting the speed of the motor accordingly.

Preferably the sole user input to the controller is a speed set point. However, in other arrangements the user input may be a desired flow rate.

The method preferably further includes the steps of: measuring the current position of the rotor; comparing that position to a desired position of the rotor, and adjusting the speed of the motor to compensate for any difference between the current position and the desired position.

In this configuration, there are two feedback parameters available the actual rotor position and the fluid pressure at the pump outlet. Preferably the method further includes the step of determining a desired position of the rotor throughout its rotary cycle from a lookup table according to the rate set by the operator.

Preferably the desired position of the rotor throughout the rotary cycle in the lookup table entries is determined by the position of the rotor required to maintain a constant output pressure, The transfer function of a pump is typically highly non-linear, and varies over operating pressure, and with tubing type and condition. To compensate for this, the lookup table can provide curves describing the inverse of the transfer function which can be captured using a purely open-loop controller, running at constant pressure and flow.

Ideally several such transfer function maps are captured and provided at different pressures. In this arrangement the desired position of the rotor throughout the rotary cycle in the lookup table entries can be determined by the position of the rotor required to maintain a constant output pressure. The method preferably further includes the step of adjusting the point of entry into the lookup table from which the desired position of the rotor is determined according to a target pressure.

The appropriate transfer function map can thus be selected for use, with the target pressure largely dictating the map choice. This transfer function map then forms an idealised model of where the rotor should be at any point in time, to produce steady flow, and preferably dominates the behaviour of the control method to provide a system largely immune to outside disturbances.

The entries in the lookup fable are typically discrete entries for the selected pressures.

Accordingly, the method preferably further includes the step of, if the lookup table does not contain an entry for the output pressure in question, interpolating between adjacent entries in the lookup table to determine the desired position of the rotor throughout the rotary cycle

In a particularly preferred arrangement the method calculates the speed that will cause the difference between the current position and the desired position to zero by the time of the next measurement.

In one embodiment, the motor is a stepper motor. A stepper motor is particularly preferred as it can provide constant information about the position of the motor drive shaft and therefore the position of the rotor. However, other motor types could be used, but preferably the motor chosen allows precise speed control and feedback information about the position of the rotor either from the drive of the motor itself or from a position encoder.

Preferably the method further includes the steps of: monitoring the performance of the pump over time using the measured pressure; and adjusting the entry into the lookup table which is consulted according to the observed performance of the pump.

This allows the method to take account of variations in the tubing properties in the pump, and over the life of the tube, and possibly between tubes of the same type due to manufacturing differences. For example, the method may operate such that, if the measured pressure undershoots a target pressure after the high speed section, a higher pressure map is used with a longer fast section. Conversely, if the measured pressure overshoots after the fast section, a lower pressure map is selected. Preferably any change is blended in gradually over the course of a revolution of the pump.

Providing long term matching of the selected lookup table entry based on the performance of the pump allows lower short term control inputs to be made, which aids stability and immunity to fluctuations in flow caused by pressure transients from sources external to the pump, such as another pump operating upstream or downstream of the pump in question. Preferably a target pressure level is maintained to provide a set point for the control method. In a preferred arrangement, the target pressure is a low-pass filtered version of the outlet pressure. The filter is preferably designed to maintain a constant level during normal operation, ignoring the pressure transients caused by the fast section of rotor travel. The filter may be gated so that it will track to the actual pressure level quickly during system start-up and pressure regulator adjustments. If the actual pressure maintains a level significantly different from the set point pressure for too long (several hundred milliseconds) the target pressure is quickly adjusted to match.

Preferably the method works by adjusting the value used to look up in the lookup table, not the output value of the lookup table. Typically for a peristaltic pump, the transfer function has a period of slow movement (corresponding to the standard rotation of the rotor with the rolier(s) in contact with the tubing) and a short period of fast movement (corresponding to the point at which a roller disengages with the tubing and so the rotor has to accelerate to maintain the pressure in the tubing and avoid flow back into the section of tubing behind the disengaged roller. By adjusting the input to the lookup table based on the measured pressure, the effect of the control feedback can be limited to a small range of travel on the input axis, which produces an even smaller effect on the rotor movement during the slow portions of the transfer function. However, there is some variability in the timing of the release of the nip and the above described arrangement can allow the control method to execute the full magnitude of the fast section of travel early, or delay it slightly, depending on the pressure sensor feedback. The fast section of the transfer function of a peristaltic pump can provide for about 50% of the total movement of the rotor in a single cycle at higher pressures, so the control unit would have to have limits of at least this range if it were adjusting map output values. Using the input side of the lookup table, adjustments of only around 5-10% of the input value may be needed.

downstream. The method of the second aspect may have some, ail or none of the above described optional and preferred features.

The method of this aspect is preferably used in conjunction with a pump according to the above first aspect, including some, all or none of the optional or preferred features of that aspect. However, the method may be used in conjunction with alternative pumps. Brief Description of the Drawings

Embodiments of the invention will now be described by way of example with reference to the accompanying drawings in which:

Figure 1 shows a perspective view of a pump according to an embodiment of the present invention; Figure 2 shows a detailed view of the pump of Figure 1 with the casing removed; Figure 3 shows a sectional view of the pump of Figure 1 ;

Figure 4 illustrates, in schematic form, a method of controlling a pump according to an embodiment of the present invention;

Figure 5 is a graph showing the desired rotor position against time for a range of different operating pressures of a pump;

Figure 8 is a graph showing the operation of a prior art peristaltic pump and has already been described; and Figure 7 is a graph showing the operation of a pump according to an embodiment of the present invention.

Detailed Description

Figure 1 shows a perspective view of a peristaltic pump 10 according to an embodiment of the present invention. The pump has a stepper motor 20, a pump unit 30 and circuit board 40. Further control circuitry and connections, discussed in more detail below, connect the circuit board 40 and the stepper motor 20 and control the operation of the pump 10.

The pump unit 30 is arranged to pump reagents from inlet 34 out through outlet 32. The inlet 34 is generally connected, in use, to a source of a reagent, such as a storage vat or bottle, or to the output of an earlier reaction system. The outlet 32 is generally connected, in use, to a reaction chamber for conducting flow chemistry. Suitable sources of reagents and reaction chambers are well known in the art and will not be described further here.

Figure 2 shows the detail of the pump unit 30 with the front cover removed. The pump unit consists of a standard, albeit high quality and rugged, peristaltic pump configuration in which a flexible eiastomeric tube 33 which provides fluid communication between the inlet 34 and the outlet 32 is sandwiched in a U-shaped configuration between the fixed block 37 of the pump unit and a rotor 35. Mounted on the rotor are three rollers (also commonly referred to as "shoes" or "wipers") 36a, b & c. When the pump unit 30 is driven by the stepper motor 20, the rotor 35 rotates (in a clockwise direction as viewed in Figure 2) causing the rollers to engage with the tube 33 to create a "nip" which is then moved around the tubing by the motion of the roller in the known manner, causing fluid to be transferred from the inlet 34 to the outlet 32 and pressure to be applied to the fluid in the subsequent tubing downstream from the outlet 32 in the known manner.

Where the pump is to be used with strong acids, a fluoroelastomer tubing 33 is preferred. Where the pump is to be used with organic solvents, a perfiuoroeiastomer tubing 33 is preferred.

Although the pump unit 30 shown in Figure 2 has three rollers, any number of rollers may be used, in particular, there may be six rollers mounted on the rotor. The selection of the number of rollers will depend on the intended use of the pump. As is well known, a greater number of rollers generally gives a smoother flow and can allow increased pumping pressure. However, a larger number of rollers generally necessitates a larger and therefore more expensive pump.

Figure 3 shows a cross-section of the pump 0 as viewed from the side. The drive shaft 21 of the stepper motor 20 is connected to the drive shaft 23 of the pump unit 30 by a rigid coupling 22, The provision of a rigid coupling between the motor and the shaft avoids oscillation of the pump drive shaft (and therefore the rotor 35 and rollers 36), particularly when the pump is operating at high pressures and therefore there is considerable change in the reactionary torque on the pump drive shaft 23 when a roller 36 engages or disengages the tube. An optocoupler 25 is mounted on the coupling 22 to provide a shaft position reading once every revolution to the control unit (not shown). The optocoupler comprises a optical sensor 25a through which a slotted disc 25b which is rigidly mounted to the coupling 22 passes. This provides an input to the motor controller as described in more detail below.

Downstream of the outlet 32, a pressure sensor 38 is mounted which detects the pressure in the tubing and feeds this back to the control unit so that the motor speed can be adjusted accordingly, as described below. The pressure sensor is a strain gauge pressure transducer with wetted parts manufactured from alumina ceramic as supplied by Roxspur Measurement and Control Ltd 1-4 Campbell Court, Bramiey, Tadley, Hampshire, RG26 5EG, UK.

The peristaltic pump unit 30, when pumping against a back pressure, has a highly nonlinear transfer function. With the motor running at constant speed, output flow is in the forwards direction until a roller 36 releases a nip in the tubing, which causes the pressurised fluid downstream of the pump to rush backwards into the now unpressurised section of the tubing, resulting in momentary reverse flow and, in most cases, a loss in output pressure (depending on the volume and compliance of the system downsteam of the pump.) Accordingly, the pump of an embodiment of the present invention has a control system which controls the stepper motor based on the detected pressure. The method of controlling the pump set out below constitutes a further embodiment of the present invention. Figure 4 shows a schematic overview of the control system and its operation.

Viewed as a whole, the objectives of the control system are twofold: - To increase the speed of the rotor as the nip releases, so that a steady flow is produced. - To govern the mean speed of the rotor so that a predetermined flow rate is maintained.

Therefore, precisely regulating outlet pressure, by means of adjusting rotor speed, will produce a steady outlet flow. If the outlet flow and pressure are steady, the effects of the system downstream of the pump (which may have variable volume and compliance) are negated.

The pump aims to deliver the correct flow rate in the presence of system pressure variations, including pressure fluctuations introduced by other pumps in the system. This is achieved by closely following a predetermined operating curve, and only allowing the pump position to deviate slightly from this curve to correct for pressure errors, the main source being some variability in the precise moment the roller releases the tubing nip.

The sole user input to the controller is the speed set point. There are two feedback parameters available (as discussed above in relation to the pump): the actual rotor position, and the fluid pressure at the pump outlet. The transfer function of the pump is highly non-linear, and varies over operating pressure, and with tubing type and condition. To compensate for this, curves describing the inverse of the transfer function are captured using a purely open-loop controller, running at constant pressure and flow. Ideally several transfer function maps 52 are captured at different pressures. The appropriate map is selected for use by the Performance Monitor 55, with the current

Target Pressure largely dictating the map choice. This forms an idealised model of where the rotor should be at any point in time, to produce steady flow, and dominates the behaviour of the controller to provide a system largely immune to outside disturbances. The map lookup interpolates between pairs of adjacent maps, so effectively a continuous range of curves are generated for a continuous range of pressure.

An actual set of eight maps derived at different operating pressures from a pump according to the embodiment described above is shown in Figure 5, with time on the X axis, and desired rotor position on the Y axis. The higher line on the left hand side of the point of inflexion of the graphs (and the lower line on the right hand side of the point of inflexion) is at zero pressure, whilst the lower line on the left hand side of the point of inflexion of the graphs (and the higher line on the right hand side of the point of inflexion) is at maximum pressure. As the pump has 3 rollers, only 1/3 of the rotation is recorded and this information is used 3 times for each complete revolution of the rotor.

The Stepper Motor Driver 53 controls the speed of the motor, and provides position feedback. In this embodiment, a stepper motor is used with a 1 : 1 drive gearing to the rotor. Other motor types could be used, providing they allow precise speed control and the position of the rotor is available for feedback, either as a model in the driver or via a position encoder, The actual motor position is subtracted from the demand position derived from the selected Transfer Function Map 52, to produce a position error. The speed controller calculates the speed that will bring the calculated error to zero by the time of the next controller cycle. This allows the stepper motor to be micro-stepped smoothly using a 20kHz signal generator, while using a much slower controller cycle, without the controller cycle frequency being audible on the stepper drive.

The Target Pressure signal largely dictates which transfer function map(s) are used. The Performance Monitor 55 adjusts the map selection slightly based on the observed pump performance. The elastomer tubing properties vary between tubing types, and over the life of the tube, and possibly between tubes of the same type due to manufacturing differences. If the pressure undershoots after the high speed section, a higher pressure map is used with a longer fast section. If the pressure overshoots after the fast section, a lower pressure map is selected.

Only small changes to the map selection are made, and the change is blended in gradually over the course of a revolution of the pump. Matching the map choice to the tubing properties allows lower proportional integral derivative (PID) controller gains to be used, which aids stability and immunity to fluctuations in flow caused by pressure transients from sources external to the pump, such as another pump operating upstream or downstream of the pump in question. A target pressure level is maintained to provide a set point for the PID Controller 57. The target pressure is a heavily low-pass filtered version of the outlet pressure. The filter is designed to maintain a constant level during normal operation, ignoring the pressure transients caused by the fast section of rotor travel. The filter is gated so that it will track to the actual pressure level quickly during system start-up and pressure regulator adjustments. If the actual pressure maintains a level significantly different from the set point pressure for too long (several hundred milliseconds) the target pressure is quickly adjusted to match.

The PID Controller 57 is arranged to match the outlet pressure to the target pressure. If the outlet pressure falls, pump speed is increased, and vice versa. The controller regularly experiences saturation conditions: the motor speed and acceleration are limited, as is the magnitude of the PID controller output. Under saturation, the integrator is disabled to prevent integral wind-up in the saturated direction.

The integrator also has a time-decay function that is dependent on the rotor speed, to prevent it slowly winding up to saturation over time.

The pump is constrained to closely follow the transfer function map. This is achieved by limiting the PID controller output level, which corresponds to X-axis displacement in Figure 5.

The index used to look up into the Transfer Function Map 52 is driven by the set-point speed. The speed value is integrated with respect to time in integrator 51 to calculate the open-loop index.

The output of the PID Controller 57 is used to modify the lookup index derived from the Integrator 51 slightly, to allow the PID Controller 57 a small influence over the rotor position. The PID level is added onto the open-loop index to produce the actual index used for lookup.

The PID Controller 57 adjusts the value used to look up into the map, not the map output value. This has the advantage that the controller can be limited to a small range of x-axis travel, which produces an even smaller y-axis displacement during the linear portions of the map. However, there is some variability in the timing of the release of the nip, so this arrangement allows the controller to execute the full magnitude of the fast section of travel early, or delay it slightly, depending on the P!D controller output.

The fast section (the near vertical chart area around the point of inflexion in Figure 5) occupies about 50% of the y~axis at higher pressures, so the PID controller would have to have limits of at least this range if it were working with map output values. Using the input side of the map, a range of only around 5-10% of the map is needed.

The very limited influence of the PID Controller 57 in the linear sections of the map provides a large amount of immunity to external pressure disturbances, such as those caused by another pump. The pump speed can preferably be adjusted over a range of at least 00: 1 without changing the transfer function maps. At lower speeds, the map curves are too slow for effective operation if the controller were purely open loop. The closed loop controller causes the lookup index to accelerate as the pressure drops, so the fast section of travel is executed in the same time as when the pump is running at full speed, and limited only to the motor acceleration and maximum speed settings.

The performance of the pump of the above embodiment, as operated by the method described above, is illustrated in Figure 7. it can be seen that the instantaneous speed of the rotor increases massively (middle graph) at the point at which each roller 38 disengages the tubing 33 in order to accommodate the resulting decrease in downstream pressure (top graph). The downstream pressure is therefore controlled so that it only transiently spikes downwards as each roller 36 disengages the tubing 33 with the control unit operating to immediately rectify this decrease and return the operation of the pump to the target pressure. As a result of this, the flow rate downstream of the pump (bottom graph) is maintained broadly constant over the operation of the pump. Using the pump or method of the above embodiments, it is possible to maintain the flow rate of the fluid being pumped downstream of the pump within 5% of the desired flow rate at all times during the steady state operation of the pump.

The method of the above embodiment may be implemented in a computer system (in particular in computer hardware or in computer software) in addition to the structural components and user interactions described. The term "computer system" includes the hardware, software and data storage devices for embodying a system or carrying out a method according to the above described

embodiments. For example, a computer system may comprise a centra! processing unit (CPU), input means, output means and data storage. Preferably the computer system has a monitor to provide a visual output display (for example of the operation of the pump, or of various real time outputs such as the speed, pressure or flow rate). The data storage may comprise RAM, disk drives or other computer readable media. The computer system may include a plurality of computing devices connected by a network and able to communicate with each other over that network. The method of the above embodiment may be provided as computer programs or as computer program products or computer readable media carrying a computer program which is arranged, when run on a computer, to perform the method(s) described above.

The term "computer readable media" includes, without limitation, any medium or media which can be read and accessed directly by a computer or computer system. The media can include, but are not limited to, magnetic storage media such as floppy discs, hard disc storage media and magnetic tape; optical storage media such as optical discs or CD-ROMs; electrical storage media such as memory, including RAM, ROM and flash memory; and hybrids and combinations of the above such as magnetic/optical storage media.

The method and pump described in the above embodiments are preferably combined and used in conjunction with each other, but this is not necessary and, in particular the method may be used to control a pump with an alternative configuration.

While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.

All references referred to above are hereby incorporated by reference.

Claims

1 . A pump for use in dispensing reagents , the pump comprising:

a motor;

a peristaltic pump having a rotor driven by the motor;

a pressure sensor monitoring the pressure of the pumped fluid downstream of the pump; and

a control unit which controls the motor by adjusting the standard operating speed of the motor according to the pressure detected by the pressure sensor, such that the pump operates continuously at a rate set by an operator,

2. A pump according to claim 1 wherein the control unit compares the pressure detected by the pressure sensor with a target pressure and adjusts the speed of the motor accordingly.

3. A pump according to claim 1 or claim 2 wherein the control unit receives a

measurement of the current position of the rotor and compares that position to a desired position of the rotor, and adjusts the speed of the motor to compensate for any difference between the current position and the desired position,

4. A pump according to any one of claims 1 to 3 wherein the control unit determines a desired position of the rotor throughout its rotary cycle from a lookup table according to the rate set by the operator.

5. A pump according to claim 4 wherein the desired position of the rotor throughout the rotary cycle in the lookup table entries is determined by the position of the rotor required to maintain a constant output pressure.

6. A pump according to claim 5 wherein the control unit adjusts the point of entry into the lookup table from which the desired position of the rotor is determined according to a target pressure.

7. A pump according to claim 6 wherein, during steady-state operation of the pump, the target pressure is the pressure detected by the pressure sensor and subjected to low-pass filtering.

8. A pump according to any one of claims 4 to 7 wherein, if the lookup table does not contain an entry for the output pressure in question, the control unit interpolates between adjacent entries in the lookup table to determine the desired position of the rotor throughout the rotary cycle.

9. A pump according to any one of claims 4 to 8 wherein the control unit monitors the performance of the pump over time using the measured pressure and adjusts the entry in the lookup table which is consulted according to the observed performance of the pump.

10. A pump according to any one of the preceding claims wherein the motor is a stepper motor.

1 1 . A pump according to any one of the preceding claims further including a sensor arranged to determine the position of the rotor.

12. A method of controlling a peristaltic pump having a rotor driven by a motor, the method comprising the steps of:

receiving a desired operating rate from an operator;

detecting the pressure of the pumped fluid downstream of the pump: and

adjusting the standard operating speed of the motor according to the detected pressure such that the pump operates continuously at the desired operating rate.

13. A method according to claim 12 further including the steps of:

comparing the pressure detected by the pressure sensor with a target pressure; and adjusting the speed of the motor accordingly.

14. A method according to claim 12 or claim 13 further including the steps of:

measuring the current position of the rotor;

comparing that position to a desired position of the rotor, and

adjusting the speed of the motor to compensate for any difference between the current position and the desired position.

15. A method according to any one of claims 12 to 14 further including the step of determining a desired position of the rotor throughout its rotary cycle from a lookup table according to the rate set by the operator.

16. A method according to claim 15 wherein the desired position of the rotor throughout the rotary cycle in the lookup table entries is determined by the position of the rotor required to maintain a constant output pressure,

17. A method according to claim 16 further including the step of adjusting the point of entry into the lookup table from which the desired position of the rotor is determined according to a target pressure.

18. A method according to claim 17 wherein, during steady-state operation of the pump, the target pressure is the detected pressure which is then low-pass filtered.

19. A method according to any one of claims 15 to 18 further including the step of, if the lookup table does not contain an entry for the output pressure in question, interpolating between adjacent entries in the lookup table to determine the desired position of the rotor throughout the rotary cycle.

20. A method according to any one of claims 15 to 19 further including the steps of:

monitoring the performance of the pump over time using the measured pressure; and adjusting the entry into the lookup table which is consulted according to the observed performance of the pump.