AUTOMATIC PRECISION NON-CONTACT OPEN-LOOP FLUID DISPENSING

Abstract

A rugged, all-electronic fluid dispensing system for use with pipettes or in other contexts indirectly measures fluid flow by using a non-linear system model to correlate vacuum existing at the top of a column of suspended fluid. Non-contact operation is provided to eliminate the need for contact-type closed-loop fluid flow sensing and associated potential cross-contamination risks. In one particular exemplary non-limiting illustrative implementation, an electronic controller within a gun-shaped, cordless self-contained pipetter housing dynamically calculates valve opening time based on a non-linear equation. Calibration is used to derive equation constants, and column vacuum pressure before the valve is opened is used as the independent variable to derive a valve opening time that will result in accurate dispensing of a desired programmed fluid quantity. Repetitive automatic dispensing with accuracies greater than 1% are possible within the context of a relatively inexpensive portable pipette or device without the need for mechanically-complex positive displacement arrangements.

Description

CROSS-REFERENCES TO RELATED APPLICATIONS

This application is a divisional of application Ser. No. 10/700,374 filed Nov. 4, 2003 entitled “Automatic Precision Non-Contact Open-Loop Fluid Dispensing” (attorney docket 4324-2) the entire contents of which is incorporated herein by reference.

FIELD

The technology herein relates to fluid dispensing, and more particularly to dispensing fluid from a narrow tube or pipette. Still more particularly, the technology herein relates to non-contact, open-loop automatic dispensing of precise quantities (aliquots) of fluid in the context of a relatively inexpensive, portable handheld gun or other shaped pipetter.

BACKGROUND AND SUMMARY

Life science research has developed into an extremely important area of modern scientific inquiry. Such research is used for example to discover new drugs, to investigate and sequence DNA and other genetic material, and to culture tissues for disease diagnosis to name a few of many areas of concentration.

Laboratory personnel in such laboratory environments are often required to accurately and safely handle and dispense relative small quantities of fluids. For example, a lab technician may prepare an aqueous (water-based) solution of cell nutrient that must be distributed in accurate quantities among a relatively large number of different containers (e.g., small culture tubes, test tubes, microcentrifuge tubes, etc.). Often, the technician is faced with a need to dispense precise amounts of such a prepared solution into a large number of containers in multiple trays. See FIG. 1. To provide accurate results, the liquid must be dispensed in relatively accurate and consistent amounts (e.g., better than 5% accuracy).

One common way of dispensing relatively small quantities of liquid is to use a narrow glass or plastic tube called a pipette. Most of us have, at one time or another, experimented with dispensing liquid using a drinking straw. Think of a drinking straw inserted into a glass of liquid so the liquid partially fills the straw. If you seal the uppermost open end of the drinking straw with your finger or thumb, you will be able to remove the drinking straw from the glass of liquid while still retaining the liquid within the straw. The liquid column remains in the straw because a vacuum is created at the top of the liquid column due to the force of gravity pulling the liquid down toward the bottom of the straw. The outside atmospheric pressure presses against the liquid at the open bottom end of the straw to maintain the liquid within the straw. When you release your finger or thumb to open up the drinking straw's top end, the vacuum is filled by atmospheric pressure rushing in to the top end of the straw and the liquid immediately runs out of the straw's bottom end.

Of course, laboratory researchers generally do not use drinking straws to handle and dispense liquids, but they use something quite similar in principle—a narrow disposable glass or plastic tube pipette. Such pipettes come in various standard sizes such as 5 ml, 10 ml, 20 ml, 50 ml, etc. Typically, the pipette has graduations so that the laboratory researcher can read the level of liquid in the tube as it is being dispensed.

Several decades ago, it was common for laboratory researchers to apply mouth suction to the top end of the pipette to suck or “aspirate” a column of liquid into the tube—thus allowing the level of liquid in the pipette to rise above the liquid level in container from which it was being drawn. However, this was relatively time consuming and could be dangerous if the fluids being dispensed were hazardous to health. In addition, mouth suction techniques were not conducive to a sterile environment or the exacting procedures required for genome sequencing and tissue culturing. Accordingly, there came a time several decades ago when various companies began developing “pipetter” handheld devices that accepted common disposable or non-disposable pipettes and which would supply powered suction and positive pressure to the open uppermost end of the pipette to draw up and release liquids. A leader in this development effort was Drummond Scientific Co. of Broomall Pa.—the owner of this patent. Drummond's vanguard development efforts resulted in a number of issued United States patents including for example U.S. Pat. No. 3,834,240; U.S. Pat. No. 3,963,061; U.S. Pat. No. 4,461,328; U.S. Pat. No. 4,624,147; U.S. Pat. No. 5,104,625; U.S. Pat. No. 5,214,968; U.S. Pat. No. 5,294,405; U.S. Pat. No. 5,616,871; and U.S. Pat. No. 5,090,255.

Drummond Scientific's associated pipetter products have been highly successful in the marketplace—making Drummond a leader in the pipetter field. For example, Drummond was one of the first if not the first to develop a practical, economical handheld gun-shaped portable pipetter device that allowed a laboratory technician or other user to depress variable-stroke push buttons to vary the amount of suction applied to the uppermost end of a pipette. To dispense liquid using this type of device, the user simply attaches a pipette to the gun-shaped handle and places the bottom end of the pipette into a liquid to be dispensed. Depressing the top button with a forefinger causes the pipetter to apply suction that draws liquid up into the pipette tube. This power suction allows the pipette to draw a liquid level higher than the level in the liquid reservoir sourcing the liquid being drawn. Upon attaining a desired column height, the user releases the top button to seal the top end of the pipette tube and thus maintain the liquid column level in the tube. The user may then lift the pipette out of the initial fluid reservoir and place it into or above the container into which the fluid is to be dispensed.

The user dispenses the fluid by depressing a down button while watching the descending column height relative to the graduations marked on the pipette tube. The user releases the down button when the desired quantity has been dispensed. The user may dispense additional quantities, or “aliquots,” into additional receptacles until most or all of the fluid within the pipette tube has been dispensed. The entire process may be repeated multiple times. Power dispensing reduces dispensing time and can also help to mix the fluid with contents already present in the container into which the fluid is being dispensed.

In this type of device marketed by Drummond in the past, the up and down buttons are coupled to needle or other valves having variable apertures. This allows the user to control the speed of aspiration or dispensing by varying the amount of pressure he or she applies to the up and down buttons respectively. A light touch on the button results in slower aspiration or dispensing, while a more firm depression increases the rate at which the fluid is drawn up or dispensed from the pipette tube. In come contexts, users may wish to dispense with some force so the dispensing agitates and mixes the resulting solution in the receptacle into which the liquid is being dispensed. In other cases, the user may be very concerned about dispensing nearly exact quantities and so will use a slower dispensing speed while more carefully watching the fluid column height relative to visual graduations on the pipette tube.

The Drummond products described above have worked extremely well over the years in a wide variety of laboratory contexts and have therefore been very successful. However, there are some instances when it would be desirable to reduce the amount of skill and potential tedium required to accurately dispense a large number of nearly identical quantities of fluid aliquots into a number of receptacles. In the industry, there has been a long felt but unsolved need for a relatively inexpensive, handheld or other dispensing apparatus that can be coupled to a standard laboratory pipette and which can be programmed to accurately and repetitively dispense a precise amount of liquid.

In the early 1990s, Drummond Scientific worked to solve this problem by developing an automatic pipetter based on a precision syringe and piston. See U.S. Pat. No. 5,090,255. A microcontroller operated a motor which in turn was mechanically coupled to the piston via a threaded shaft. Moving the piston out of the syringe by a precise displacement created suction which drew liquid into the pipette. Plunging the piston into the syringe a precise displacement caused a precise corresponding quantity of fluid to be dispensed from the pipette. The amount the piston was displaced precisely controlled the amount of fluid being “aspirated” or dispensed. This design was quite successful in automatically repetitively dispensing programmable amounts of liquid with a high degree of accuracy and precision. However, a disadvantage was the relatively high cost and complexity of the positive-displacement syringe-and-plunger arrangement. Positive-displacement-type devices are often handicapped by slower dispensing speeds and total column-height volumes that are limited to the plunger displacement volume. This means that handheld devices are generally limited due to the portability issue. It would therefore be highly desirable to provide automatic dispensing functionality in the context of a less expensive, more portable, all-electronic design not subject to these limitations.

One of the challenges to providing an improved automatic dispensing design relates to the number of variables that computer control needs to take into account in the context of a so-called “non-contacting” open-loop system to provide a requisite degree of accurate dispensing. One might initially think, for example, that it would be relatively straightforward to use a liquid flow sensor to accurately measure the amount of liquid being dispensed in the context of a conventional closed-loop control system. However, it must be remembered that many laboratory procedures require that no part of the dispensing apparatus other than the disposable or non-disposable, sterilized pipette come into contact with the fluid being dispensed. It is therefore undesirable or impossible in many contexts to use a flow sensor in contact with the fluid being dispensed to monitor fluid flow amount.

We have now discovered a way to control a relatively simple, inexpensive pipetter or other fluid dispenser to provide precision, repetitive, automatic dispensing of programmable fluid quantities. One exemplary, illustrative implementation of our technique mathematically models the pneumatic system of the dispensing apparatus—including the removable pipette tube—with a non-linear model. There are various methods by which the pipette and pipetter systems can be modeled. One exemplary illustrative non-limiting arrangement is aspirating to a specific and consistent column height, and dispensing in fixed time increments. Through such non-linear mathematical modeling, a computing element such as for example a relatively inexpensive microprocessor can be used to accurately control valve aperture and/or pump power to achieve relatively high precision of dispensing quantity in the context of an inexpensive handheld gun shaped or other pipetter or other dispensing system.

Non-limiting, exemplary illustrative advantages of our approach include for example:

• • avoids need for positive displacement type syringe-piston arrangements and/or expensive, complicated peristaltic or other pumps
  • mathematically and physically models non-linear system to provide a high degree of accuracy and precision
  • open-loop system—avoids need for closed-loop control
  • relatively light weight
  • inexpensive, rugged design
  • simple, reliable mechanics
  • easy to operate, intuitive operation
  • precise repetitive automatic dispensing of aliquots
  • relatively quiet operation
  • can correct for a variety of factors including, for example, dispensing at different angles, different pipette diameters, changing pump motor efficiency, different fluid viscosities, other
  • drip prevention or elimination
  • users not required to constantly pay attention to graduations on pipette tube during automatic dispensing
  • electronic controller/substantially all electronic design
  • programmable dispensing amount
  • electronic valve and pressure sensor
  • no contact/no fluid sensing orifice
  • pressure and vacuum operation
  • handheld (e.g., gun shaped)
  • controllable automatic dispensing rate and quantity
  • accommodates differently sized pipettes in some implementations
  • high accuracy (e.g., 1% or greater)
  • high repeatability
  • relativistic in operation
  • self-powered
  • intuitive graphical display and associated user interface
  • vane type electronic pump in some implementations
  • reversible orifice valve in some implementations
  • reversible pump motor in some implementations
  • plural pressure sensors in some implementations
  • look-up table in some implementations
  • no pressure sensors in some implementations
  • combined high-speed multiple decrements to achieve single larger volume aliquot dispensings
  • blocked filter detector
  • substantial increase in speed and efficiency of dispensing, reducing lab labor expenses and time
  • use of large volume pipettes to accurately dispense very small aliquots, reducing aspirations and decreasing the quantity of pipettes required and/or changed
  • automatic aspiration to preset column heights
  • automatic aspiration when pipette tip is inserted in liquid
  • look-up table or formula-based dispensing
  • single calibration for all pipette sizes
  • make use of any volume pipette, including 100 ml or greater, without being negatively impacted in dispensing accuracy or precision.

In one non-limiting, exemplary illustrative implementation, a pressure calibration technique is used to establish a base line. In one exemplary illustrative implementation, two column height pressure readings are taken: one for a given column height near the top of the pipette and another for a given column height near the bottom of the pipette. These pressure readings are used to calculate constants for a mathematical equation that outputs valve open time and/or pump power for dispensing a desired quantity of liquid. During active dispensing, column height pressure is continually monitored and used to calculate or look up the corresponding valve and/or pump control parameters. Accuracies better than 1% have been achieved.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other features and advantages will be better and more completely understood by referring to the following detailed description of exemplary, illustrative, non-limiting implementations in conjunction with the drawings, of which:

FIG. 1 shows an example laboratory environment in which a pipetter dispensing device might be useful;

FIG. 2 shows an example non-limiting illustrative schematic diagram of a pipetter dispensing device design;

FIG. 3 shows an example non-limiting illustrative gun-shaped implementation;

FIG. 4 shows an example more detailed electronic schematic block diagram of an exemplary non-limiting illustrative implementation;

FIG. 5 shows an example main software routine of an exemplary non-limiting illustrative exemplary implementation;

FIG. 6 is a flowchart of an exemplary non-limiting illustrative user-controlled mode software routine implementation;

FIG. 7 is a flowchart of an exemplary non-limiting illustrative software routine implementation for a calibrate mode;

FIG. 8 is a flowchart of an exemplary non-limiting illustrative automatic dispensing mode software routine;

FIGS. 9A and 9B together are a flowchart of a more detailed exemplary non-limiting illustrative implementation of an auto dispensing mode;

FIG. 10 is an exemplary illustrative non-limiting program mode software routine flowchart;

FIG. 11 is a flowchart of an exemplary illustrative non-limiting change pipette size software routine; and

FIGS. 12-14 are graphical illustrations of exemplary non-limiting illustrative system modeling parameters.

DETAILED DESCRIPTION OF EXEMPLARY NON-LIMITING ILLUSTRATIVE IMPLEMENTATIONS

FIG. 1 shows an illustrative exemplary non-limiting implementation of a hand-held electronic pipetter 50. Illustrative non-limiting pipetter 50 in this exemplary implementation is a self-contained, hand-held, lightweight, relatively inexpensive, easy-to-use device that allows a user U to easily and quickly automatically dispense repetitive quantities of liquids or other fluids with a high degree of accuracy and precision. As FIG. 1 illustrates, a user U such as a lab researcher or technician first removably attaches a conventional or non-conventional pipette P to the pipetter 50. In the exemplary illustrative implementation, pipette P may, for example, comprise a standard conventional off-the-shelf pipetter of the type commonly available from a wide variety of laboratory supply houses. Such a pipette P may be made for example of glass or plastic, and it may be disposable or non-disposable. It may have graduations imprinted on its outside surface, or it may not. Different standard laboratory pipettes come in different lengths and diameters for use with different quantities of liquid. Exemplary illustrative non-limiting implementation of pipetter 50 can accept various different standard-capacity pipettes P.

Once user U attaches the pipette P to pipetter 50, the user grasps the gun-shaped handle 102 and inserts the pipette lowermost end into a receptacle R of fluid F to be dispensed. The user U then depresses an uppermost button 106 with his or her “trigger” finger to command pipetter 50 to draw fluid F from receptacle R into the pipette P. The user U may now release the upper button 106 and lift the pipette P out of receptacle R. With both buttons 106, 108 released, the pipetter 50 seals the open valve-end of pipette P to retain fluid F within the pipette P.

The user U may now position the lowermost end of pipette P over a further receptacle R1 into which some desired quantity of fluid F is to be dispensed. In the preferred exemplary illustrative implementation, the user may program pipetter 50 with this desired quantity (e.g., 5.2 ml). When the user U depresses lowermost button 108 with his or her index finger, pipetter 50 automatically dispenses substantially the programmed quantity into the receptacle R1. The user U may repetitively depress lowermost button 108 to accurately dispense substantially the same programmed quantity into each of a number of additional receptacles R2, R3, etc. In the exemplary illustrative non-limiting implementation, when insufficient quantity of fluid F remains within pipette P, the pipetter 50 automatically tells the user U that the pipetter needs to be refilled. The user U may then return pipetter P to the initial receptacle R to draw an additional quantity of fluid F to be dispensed. This process may continue repetitively until all of the desired receptacles have received desired quantities of fluid.

Using the illustrative exemplary non-limiting implementation of pipetter 50 disclosed herein, the user during the dispensing process does not need to pay any attention to the graduations on pipette P (in fact, such graduations need not even be present in some implementations). Rather, the user U knows that each time he or she depresses the lowermost button 108, device 50 will automatically, reliably dispense the programmed desired quantity. This automatic operation not only speeds up dispensing dramatically, but also reduces the tedium and required skill level needed for accurate dispensing. Because of the consistency and precision at which exemplary illustrative non-limiting implementation of pipetter 50 dispenses programmed quantities, lab results may have a higher degree of reliability even when conducted by less skilled laboratory personnel. This can be especially important in the health care and basic research fields where a patient's diagnosis depends on accurate lab results or where basic research may be called into question because of less precise lab work.

Exemplary Illustrative Non-Limiting System Design

FIG. 2 shows a schematic diagram of an example illustrative non-limiting implementation of fluid dispensing system 50. System 50 includes a coupler 52 to which is removably attached a disposable or non-disposable conventional pipette 54 or other conventional dispensing fluid tube. Coupler 52 may be designed to be compatible with any number of different conventional or non-conventional pipette 54 designs including for example standard 25 ml, 50 ml or other capacity glass or plastic pipettes for serological or other laboratory use, micropipettes, or any other desired dispensing tube. In one exemplary illustrative implementation, a filter 56 that is in fluid contact with (and which may be part of) coupler 52 prevents fluid drawn into the pipette 54 from reaching the remainder of the components of system 50. If a user over-draws the fluid into pipette, the fluid will be blocked by filter 56 which may be removed and replaced to maintain sterile conditions and eliminate cross-contamination. The filter used in the pipetter can be, for example, a hydrophilic filter, which completely blocks the passage of air when it becomes damp (ensuring that the pneumatic system is not compromised or contaminated). The current system is able to detect when the filter has been compromised in the following manner: (1) valve closed; pump activated; pump-side sensor read. (2) valve opened; sensor read continuously for “x” milliseconds. (3) If there is no measurable pressure delta detected, the filter is determined to have been compromised by liquid contact, thus preventing air flow.

A pump 58 is pneumatically connected to coupler 52 through a valve 60. Plastic molded or other passages 62, 64 fluid-couple the pump 58 to valve 60 and valve 60 to coupler 52 respectively. An electronic pressure sensor 66 of conventional design monitors the pressure within passageway 62 between pump 58 and valve 60, and a second electronic pressure sensor 68 monitors the pressure within passageway 64 between valve 60 and coupler 52. Pressure sensor 66 is sometimes referred to in this specification as the “pump-side” pressure sensor because it monitors pump output pressure, and pressure sensor 68 is sometimes referred to as the “pipette-side” pressure sensor because it monitors the pressure at the top of the column of liquid within pipette P. In some implementations, only one pressure transducer may be used or required.

Pressure sensors 66, 68 generate outputs that are applied, in this exemplary illustrative non-limiting implementation, to a computer 70 such as for example a conventional microprocessor unit. Computer 70 also generates, by itself or with assistance of other associated driving circuitry, a pump drive signal to drive pump 58 and a valve drive signal VD to drive valve 60. Computer 70 receives user inputs and generates user outputs via block 72 shown in FIG. 2. In the exemplary implementation, computer 70 generates appropriate pulse-width modulated outputs to pump control 59 (FIG. 4) and to valve 60 under software control in response to depression of buttons 106, 108 and also in response to the pressures measured by pump-side sensor 66 and pipette-side sensor 68. Computer 70 stores software routines in firmware along with operating formulas with stored constants defining a model of system 50. Computer 70 in the exemplary illustrative non-limiting implementation includes an internal writable non-volatile memory used to store calibration data for different pipette configurations and diameters. We have designed the device to be field-reprogrammable by the user. The product will have a standard PC USB-type interface. User's will be able to go to Drummond's website and purchase and/or access upgrades, more specific troubleshooting and calibration routines, etc.

In the exemplary illustrative non-limiting implementation shown in FIG. 2, accurate precision dispensing is achieved without directly measuring the flow rate of fluid flowing into or out of pipette 54. In more detail, computer 70 indirectly ascertains fluid flow by measuring pressure. It reads the pressure (vacuum) P at the top of fluid column F at various points, and uses this measured pressure information along with empirical data and parameters stored within the computer to calculate a control output.

In one exemplary illustrative non-limiting implementation, the control output comprises a time value T used to control the opening of valve 60. In certain exemplary illustrative non-limiting implementations, computer 70 may, in addition or instead of controlling valve 60 opening time, control the valve's opening aperture and/or the power and/or direction of pump 58. In certain non-limiting exemplary implementations, pump 58 is a conventional continuous air pump with a fixed pumping rate while another exemplary illustrative non-limiting implementation's pump 58 is a vane or other type variable-speed pump the pumping rate of which can be controlled by computer 70. In still other implementations, pump 58 could be replaced or supplemented with a tank or other reservoir of pressurized gas (e.g., a pressurized CO₂ cartridge). In some non-limiting exemplary illustrative implementations, valve 60 is a digital on/off valve that is either open or closed, whereas in other exemplary illustrative non-limiting implementations the valve has a variable aperture opening that computer 70 may control by applying a variable signal to the valve. In some exemplary illustrative non-limiting implementations, the pump-side pressure sensor 66 may be eliminated and computer 70 may in such cases rely on only one (or no) pressure sensor.

FIG. 3 shows a cutaway view of an exemplary illustrative non-limiting handheld implementation of the FIG. 2 dispensing system 50. The handheld gun-shaped implementation shown in FIG. 3 is only one example—a variety of other implementations are possible including stationary, tabletop, embedded, shapes other than a gun, and many other implementations. In the non-limiting exemplary illustrative implementation shown in FIG. 3, a gun-shaped housing 100 includes a handle portion 102 and a main housing portion 104. Up pushbutton 106 and down pushbutton 108 are disposed partially within handle portion 102. The user depresses up pushbutton 106 to aspirate liquid into pipette 54, and depresses down pushbutton 108 to dispense liquid from the pipette. In the exemplary illustrative implementation shown, pushbuttons 106, 108 are variable-travel pushbuttons coupled to Hall effect magnetic or other sensors that provide the user with variable control over aspiration and dispensing rate.

At a distal end 110 of main housing portion 104, a conventional coupling arrangement 112 is disposed to accept and retain the open end of pipettes 54. In one conventional design, a user couples a disposable pipette to the coupler 52 to dispense a certain liquid, and then removes and throws away the disposable pipette once all of the liquid has been dispensed. The user uses a fresh sterilized disposable pipette to dispense a different liquid. In other arrangements, laboratories may use higher precision pipettes made of glass or other materials that are washed and sterilized after each use. In still other implementations, it could be desirable to semi-permanently or permanently couple a pipette or other dispensing tube to system 50 in which case coupler 52 might not be needed in the configuration shown. In one exemplary implementation, a switch 112 is used to detect whether a pipette 54 has been coupled to coupler 52 and to feed that information to computer 70.

In the example shown, computer 70 is mounted on a printed circuit board 114 within main housing portion 104. Other components such as capacitors, resistors and the like may also be disposed on a printed circuit board 114. Pump 58 is connected electrically to computer 70 via appropriate conventional motor drive circuitry also disposed on printed circuit board 114. Pressure sensors 66, 68 are also preferably provided on printed circuit board 114 in the exemplary illustrative non-limiting implementation shown. The FIG. 3 implementation may be entirely self-contained (e.g., including rechargeable or non-rechargeable batteries not shown that may be disposed within housing handle portion 102), or the system 50 may rely on external battery and/or other external components to operate. However, in one exemplary illustrative non-limiting implementation, lightweight self-contained, portable, cordless operation is achieved.

FIG. 4 shows an exemplary illustrative non-limiting more detailed schematic block diagram wherein main microprocessor 70 is coupled to a math coprocessor 70a and a graphical display system coprocessor 70b. Math coprocessor 70a efficiently performs mathematical computations in real time, whereas graphical display system coprocessor 70b provides graphic handling to generate images for viewing on a graphical display unit such as a liquid crystal display and associated depression switch or switches 72d. Main microprocessor 70 controls pump 58 via a pump control unit 59. Pump control unit 59 may for example comprise a pulse width modulated motor controller that receives a pulse switch modulated control signal of approximately 1 kHz from main microprocessor 70. Valve 60 may be controlled by main microprocessor 70 through a field effect transistor or other electronic switch not shown. In the exemplary embodiment, pump-side pressure sensor 66 and pipette-side pressure sensor 68 each comprise conventional differential pressure sensors generating analog outputs wherein the approximate center of the sensor output range indicates atmospheric pressure with, for example, an increased output indicating negative pressure and a decreased output indicating positive pressure. In one exemplary implementation, pressure sensors 66, 68 generate outputs between 0 and 5 volts DC with 2.5 volts DC output indicating atmospheric pressure. Main microprocessor 70 in the exemplary implementation includes conventional analog-to-digital converters that convert the pressure transducer output signals into digital bit values for storage and manipulation. In this exemplary illustrative implementation, an output from one of pressure transducers 66, 68 that is greater than 2.5 volts DC indicates vacuum, while an output that is less than 2.5 volts DC indicates positive pressure relative to atmospheric.

Exemplary Illustrative Non-Limiting Software Architecture

FIG. 5 is a flowchart of an exemplary illustrative non-limiting main routine performed by microprocessor 70. In this exemplary illustrative non-limiting implementation, upon unit start up (block 302) the main microprocessor initializes its registers, defines variables, reads stored calibration and default stored calibration and default variables, and performs other housekeeping functions (block 304). For example, in one exemplary illustrative non-limiting implementation, the main microprocessor 70 initializes its port A for analog input readings, port C for outputs, port D for other inputs and also sets a pulse-width modulation (PWM) register in addition to turning on a pulse-width modulator timer. The main microprocessor may also read data from EPROM or other non-volatile storage of stored calibrated values and variables.

In one exemplary illustrative non-limiting implementation, the main microprocessor 70 then controls the graphical display processor 70b to display a main menu of operating modes selectable by the user (block 306). Operating modes may include for example:

• • user-controlled mode,
  • program mode,
  • auto mode,
  • calibrate,
  • choose pipette size,
  • diagnostic mode,
  • other.

These operating modes are exemplary only. There may be other additional modes such as for example a “blast” mode that causes dispensing system 50 to operate in a squirt gun or “water pik” type pulsating rapid dispensing operation for use as an agitator. Other implementations may have fewer or different modes than the ones described above. For example, in one exemplary illustrative non-limiting implementation, the diagnostic mode 318 may simply show the firmware version of software. In other exemplary implementations, other more complicated functions may be invoked depending upon need.

In the exemplary illustrative implementation, mode selection is accomplished without requiring additional complex input controls through use of a simple, easy to understand graphical user interface that can be displayed on a compact display such as a liquid crystal display. For example, the display 72d can display a menu item at the top of the display with an arrow indicating that if the switch is pressed, that menu item will be performed. As the up and down buttons 106, 108 are pressed, the menu items scroll in that direction—and thus the same buttons used for dispensing can also be used for menu navigation (in one exemplary illustrative implementation menu option selection is accomplished by the user depressing the liquid crystal display 72d itself—which actuates a switch closure). The main microprocessor 70 reads the pushbutton states 106, 108 as well as an additional select switch that may be part of the liquid crystal display 72d. If the up or down pushbutton 106, 108 reaches a threshold, then the main microprocessor scrolls the menu up or down indicated by the buttons. If the switch is pressed, the main microprocessor jumps to the menu item's location and software. As the menu items scrolls up or down, the menu follows. If the last menu item is at the top, then the first menu item scrolls back to the top following the last menu item. This takes place until the switch is pressed. In the exemplary illustrative implementation, the menu will scroll once per up or down button depression. The system 50 waits until the button is released before it continues and performs the indicated task. The system scrolls once per button press in the exemplary illustrative implementation.

In one exemplary illustrative non-limiting implementation, a user-controlled mode 308 is provided. Under this mode of operation, the dispensing system 50 will aspirate or dispense directly in response to depression of pushbuttons 106, 108 by the user. Microprocessor 70 automatically performs software controlled functions in response to such button depressions. Although the “user controlled mode” is operated by the pushbuttons, there are distinguishing characteristics that demonstrate that the “user controlled mode” is not a computer-facilitated manual mode, but actually a significantly enhanced function that cannot be emulated in any current manually deployed system. A true manually operated pipetter in the aspirate mode does not permit the liquid level to decrease—it either remains stable or rises. Conversely, in the dispensing mode, the liquid only falls. When flow is valve aperture-dependent, emulating that mode electronically is very simple and straightforward. When the valve aperture is fixed, however, and flow is pump-dependent, the opportunity exists to use electronic control to greatly enhance the user's ability with respect to precision, particularly if pressure on either side of the valve can be ascertained. By constantly monitoring the liquid column height (pipette-side sensor) and comparing it to the pump-side sensor, the aspirate button can be used to both aspirate and dispense, whereby the valve is opened initially upon the sensors' outputs being equal, and remaining open until the aspirate button is fully released. For example, if the liquid column height in a 50 ml pipette were 40 ml, and additional liquid were to be aspirated from that starting point, the aspirate pushbutton would need to be depressed until the pump-side pressure sensor indicated an equivalent (or better) pressure relative to the pipette-side sensor. Once the valve was opened, the aspirate button could then be both increased and decreased (by displacement) to precisely allow the user to raise or lower the column height with respect to the desired pipette fill volume. Manual systems require both the aspirate and dispense buttons be used in the event of any “overshoot”.

FIG. 6 shows an exemplary illustrative non-limiting flowchart for a user-controlled operating mode 308. In this example, system 50 will continue to operate in the user-controlled mode until the graphical display switch is pressed. Using the up and down pushbutton switches, microprocessor 70 controls pump 58 to aspirate or expire liquid based on the pushbutton switch 106, 108 being depressed. When a pushbutton 106, 108 is pressed, a magnet on the end of the pushbutton comes closer to a Hall effect sensor. The distance between the magnetic and sensor is translated into a voltage which is converted to a digital bit count, which is compared to a threshold bit count set in software to see if the button is pressed far enough to turn on the pump (FIG. 6, blocks 352, 354, 356). Depressing the up button 106 will aspirate fluid—meaning that microprocessor 70 turns on pump 58 in an appropriate direction to generate vacuum (block 360). Depressing the down button 108 causes microprocessor 70 to turn on pump 58 in a direction to generate pressure (block 362). In other exemplary embodiments, a unidirectional pump could be used with different valve-controlled ports being opened to provide positive pressure and vacuum respectively. Thus, if the up button 106 is pressed, pump 58 will turn on in the direction to provide vacuum and the speed will change based on an algorithm using the bit count of the voltage reading from the Hall effect sensor. The speed of pump 58 will increase as the button is pressed further into the button assembly. Maximum speed will be attained when the magnet comes in closest contact with the sensor. As the button is retracted, the speed will decrease based on the same algorithm in the exemplary illustrative non-limiting implementation. The bottom button 106 in the exemplary illustrative non-limiting implementation acts the same way for speed control but reverses to produce pressure. In the exemplary illustrative non-limiting implementation, the pump speed is changed based on the pulse width modulated pulse output from microcontroller 70. The pulse-width modulation is the amount of time or duty cycle a pulse is on during one cycle of a timer. Generally, the duty cycle of a pulse-width modulated signal can range from 0% to 100%. However, in particular implementations, pump 58 may not be able to handle so large a dynamic range. In one exemplary illustrative implementation, the pump 58's pulse-width modulation range is constrained to be within the range of 33% duty cycle to 70% duty cycle but can change based on modeling of the system 50. At 33%, or low range, the pump is in its maximum speed operation. At a 70% duty cycle in the exemplary illustrative embodiment, the pump operates at its slowest speed.

FIG. 7 shows an exemplary illustrative non-limiting implementation for a calibrate mode 314. In the example shown, the calibrate mode is used preparatory to automatic dispensing in the automatic mode 312. In the calibrate mode shown, microprocessor 70 causes display 70d to display an instruction to the user to aspirate a predetermined amount of liquid which is preferably near the top of the maximum column height of the pipette being used. For example, in the case of a 50 ml pipette, system 50 may direct the user to aspirate 60 ml of fluid into the pipette (block 402 Many standard pipette volumes have approximately 20% additional column height capacity. For example, a standard laboratory 50 ml pipette is graduated to 60 ml. We can use the 60 ml graduation to achieve greater accuracy during calibration (correlation between different pipette sizes is a linear function). This level is just an example—other levels may be used. In general, this particular level should however be near the maximum quantity that the pipette 52 can draw to avoid extrapolation errors later.

In response to the displayed message, the user depresses the up pushbutton 106 just as in the manual mode until the predetermined requested level is reached by eye (FIG. 7 blocks 404, 406, 408, 410, 412). The down button 108 is also available if the user mistakenly over aspirates in order to correct the level to exactly the desired one (blocks 414, 416). Once the user has aspirated the desired level, the user presses the switch on the graphical display 70d. At this point, system 50 takes multiple (e.g., 5) readings of the column height pressure using the pipette-side sensor 68, averages the multiple readings and stores the resulting average as a variable (block 414).

In the exemplary illustrative non-limiting example, the user is then asked to aspirate to a predetermined level near the bottom of useful column heights accommodated by pipette 52. In one example, the user may be asked to take the liquid level down to for example 10 ml (FIG. 7, block 416). The user may of course do this by fully exhausting all of the liquid already in the pipette 52 and starting again, or he or she may simply dispense all but the desired liquid level that is already within pipette 52 from the initial calibration aspiration. As before, the user may depress up and down pushbuttons to achieve the desired aspiration level (FIG. 7, blocks 422-432). Once the desired level has been obtained, the user depresses the button on the graphical display 70d (decision block 422) which results in microprocessor 70 storing the corresponding pressure reading from the pipette-side pressure sensor 68 (FIG. 7, block 434). If desired, multiple pressure readings may be taken and averaged as above. System 50 then calculates the bits per ml based on the calculation of the high and low aspiration level pressures stored in blocks 418, 434 (block 436). System 50 may also at this time calculate the lowest possible level for dispensing from a particular pipette 52 (e.g., 3 ml). All calibration variables are stored in EPROM in the exemplary illustrative non-limiting implementation for later retrieval and calculations. The system 50 remains in this mode until the graphical display switch 70d is pressed at which time the system returns to the main menu.

FIG. 8 is a flowchart of an exemplary illustrative non-limiting implementation of a software-controlled automatic dispensing mode. In the exemplary embodiment, after calibration and before automatic dispensing, the user may wish to program in a variable quantity to be dispensed repetitively in the automatic mode. This program mode 310 is shown in FIG. 8 and is illustrated in more detail in FIG. 10, but it may be bypassed if a previously programmed or preset quantity is being used. In the exemplary illustrative implementation, system 50 retains previously-programmed quantities values and uses them at next power up until changed. To change the program quantity, the user will direct the display to change the amount to be dispensed. The display indicates the amount to be dispensed (“ATD”) by reading the EPROM data that has the number stored. A default setting of 1.0 ml or other desired amount may be provided. System 50 looks at the “ATD” value and if the pipette size has been changed to a smaller pipette and the “ATD” is larger than the maximum amount of the pipette, the system will automatically change the ATD to the maximum amount. For example, if the original pipette size was 50 ml, the maximum ATD would be 50.0 ml. If the user changes the pipette size to 25 ml, the system will automatically change the ATD to 25.0 ml. These may be default settings that can be overridden. To override the defaults, the user changes the ATD by pressing the push buttons (FIG. 10, blocks 472, 474). As the buttons 106, 108 are depressed, the system increments or decrements a displayed ATD until a desired ATD is reached. In the exemplary illustrative implementation, as the button is depressed passed a threshold set in software, the system will increase the speed at which the numbers change. As the pushbutton is released and passes a threshold, the numbers revert back to change at a slower speed. If the up button 106 is depressed, the ATD increases in 0.1 ml increments in one exemplary illustrative non-limiting implementation. If the down button 108 is depressed, the ATD decreases in 0.1 ml increments in one exemplary illustrative non-limiting implementation. Although one implementation may be to increase and decrease in increments as stated, one exemplary illustrative non-limiting arrangement (in that operation mode) will not be incremental in nature, but more flow rate specific. The minimum ATD may be set to some desired level e.g., 1.0 ml. A “learning” mode based on actual fluid dispensation under user-control may also be used to program amount to be dispensed. The maximum ATD may be based on pipette size. In the exemplary illustrative non-limiting implementation shown, system 50 remains in the ATD program mode until the graphical display switch 70d is depressed. At that time, the ATD is stored in EPROM so that if the unit is shut off, it will automatically be returned to the settings that are stored in EPROM. The system then returns to the main menu.

Referring once again to FIG. 8, once the desired dispensing quantity has been programmed into system 50 (or if a default or previously programmed amount is to be used), the system reads that ATD value and displays it on a graphical display 70d. The user may be asked to confirm that this is amount to be dispensed by depressing the switch of the graphical display 70d (block 502). In the automatic mode, each time the down button 108 is depressed passed a certain threshold set in software, the system will dispense the programmed amount of liquid. The system will repetitively dispense the desired quantity in response to successive depressions of down button 108 until the liquid level is below a minimum level (e.g., 3 ml of column height for a 50 ml pipette). In the exemplary illustrative non-limiting implementation, the vacuum pump speed is based on the depression of the buttons 106, 108. Thus, in the automatic programmable dispensing operation illustrated, users are able to control the rate at which system 50 aspirates and dispenses. This may be useful in order to, for example, provide a desired degree of agitation. If during this operation, the system reads that the up button 106 is depressed (i.e., the user wants to refill the pipette), the system will turn on pump 58 for vacuum and will stay on until either the user releases the button or until the amount of pressure indicated by pressure sensor 68 specifies that the column height is greater than the pipette allowable amount. Alternatively, the auto mode may employ several dispensing speed presets.

In more detail, when the system 50 detects that the down button is depressed (FIG. 8, block 504), it checks to see if the pipette 52 is empty by reading the pipette-side pressure transducer 68 output to determine column height and comparing that determined column height to the calculated minimum column height (e.g., 3 ml) (FIG. 8, block 506). If the microprocessor 70 determines that the unit is empty based on the comparison, it displays an error message (block 508) that requests the user to fill the pipette and does not attempt to dispense any additional liquid. Once the user fills the pipette above the minimum amount (FIG. 8, blocks 510-518), system 50 automatically recognizes that the unit may now dispense the requested amount.

The pump turns on first for a predetermined time interval, followed by the valve opening once the pressure developed is constant. The pump remains on until after the valve is closed. The pump may run for a set time prior to value opening—e.g., 250 ms. Assuming the unit is not empty, system 50 then calculates the amount of valve opening time needed to dispense the desired programmed amount based on column height as indicated by the pressure sensor 68 and based upon the programmed amount (FIG. 8, block 520). This calculation is performed (and/or results are looked up from memory) based on a formula derived from a model. Once the time has been calculated, system 50 turns on pump 58 for pressure for that amount of time at a pump speed also derived from modeling (FIG. 8, block 522). Once the time has expired, microprocessor 70 turns off pump 58 (FIG. 8, block 524) and returns to await further depression of one of buttons 106, 108.

FIGS. 9A and 9B are a flowchart of a more detailed exemplary illustrative non-limiting implementation of automatic dispensing operation. In this example, system 50 displays a prompt to the user asking the user to specify what quantity of liquid to aspirate or may use a default (FIG. 9A, block 602). Upon ascertaining the quantity to dispense (FIG. 9A, block 604), the user then aspirates to any desired column height (FIG. 90A, block 606). At this point, microprocessor 70 reads the pipette-side pressure sensor 68 (block 608) and detects whether the pressure corresponds to an expected range (decision block 610). Assuming the read pressure value is as expected (“yes” exit to decision block 610), microprocessor 70 calculates or otherwise determines column height of the aspirated amount (e.g., based on the baseline values previously determined during calibration) (FIG. 9A, block 614). Microprocessor 70 then uses system modeling results to determine valve opening time (block 616) and waits for the user to depress down button 108 (block 618).

Although we could implement the use of absolute pressure sensing to establish to pump PWM, one exemplary non-limiting arrangement will likely not contain that feature. When the user depresses the down button, system 50 powers on pump 58 at a desired flow rate (block 620) and optionally reads the pump-side pressure sensor 66 to determine whether it is within an expected range (block 622, decision block 624). Pump output can change over time based on heating, wear, etc.; the test performed by decision block 624 gives system 50 a chance to correct pump output to compensate (block 626).

Microprocessor 70 then opens valve 60 for a time T that is determined based on system modeling (block 628). This valve opening at the desired pressure generated by pump 58 results in dispensing the programmed quantity of liquid. Upon expiration of the calculated valve opening time, microprocessor 70 closes valve 60 (block 630) to cease liquid dispensing. Note that in the exemplary illustrative non-limiting implementation, pump motor 58 remains active during the entire time that valve 60 is open—the pump starting before the valve opens (e.g., 250 ms before) and turning off after valve closure.

In one exemplary illustrative non-limiting arrangement, once microprocessor 70 closes valve 60 at the termination of the calculated valve opening time, the microprocessor may control pump 58 to reverse its direction in order to generate suction rather than positive pressure (block 632). Microprocessor 70 may then open valve 60 for a very short time (a few milliseconds) to prevent dripping and to decrease system settling time (block 634). Microprocessor 70 may then close valve 60 (block 636) and shut off pump 58 (block 638).

Example Illustrative Non-Limiting Non-Linear System Modeling

As discussed above, a non-linear system model is used in the exemplary illustrative non-limiting device 50 to provide accurate automatic dispensing of fluid quantities from pipette P. As the height of the fluid column in pipette P falls during dispensing, the amount of time the valve needs to open to dispense the same amount of liquid changes. In the exemplary illustrative implementation, column height is indirectly measured by measuring the vacuum at the top of the column and the top of the column is sealed. This vacuum pressure is used to determine how long the valve must be opened to dispense a given desired amount of liquid. The model used in the exemplary illustrative non-limiting implementation takes into account and models the non-linearity in the relationship between column height vacuum pressure and valve opening time.

Before reaching the model, some background discussion about “accuracy” and “precision” in pipetting are in order.

Accuracy

A pipette is accurate to the degree that the volume delivered is equal to the specified volume. Accuracy is expressed as the mean and standard deviation for replicate measurement:

$\begin{array}{c}\stackrel{\_}{E}\%=\frac{\stackrel{\_}{V}-\mathrm{Vn}}{\mathrm{Vn}}\times 100,\mathrm{where}\\ \stackrel{\_}{E}\%=\mathrm{Accuracy},\\ \stackrel{\_}{V}=\mathrm{Mean}\mathrm{Volume},\mathrm{and}\\ \mathrm{Vn}=\mathrm{Nominal}\mathrm{Volume}.\end{array}$

Precision

Precision generally refers to the repeatability of the pipette sampling. Precision is expressed as the coefficient of variation (CV). System 50 modeling will greatly influence pipetting precision because of reduced dependency on laboratory practices (that are dependent upon human intervention, manual dexterity, and eye-hand coordination):

$\begin{array}{c}S=\sqrt{\frac{\sum _{n=1}^n{\left(w_s-\stackrel{\_}{w}\right)}^2}{n-1}},\\ \mathrm{where}S=\mathrm{Standard}\mathrm{Deviation},\\ W_s=\mathrm{Individual}\mathrm{Weighting},\\ \stackrel{\_}{W}=\mathrm{Mean}\mathrm{Weighting},\mathrm{and}\\ n=\mathrm{Number}\mathrm{of}\mathrm{measurements}.\end{array}$

This equation can be expressed as a coefficient of variation:

$\mathrm{CV}\%=\frac{S}{\stackrel{\_}{w}+\stackrel{\_}{e}}\times 100$

Empirical Data Collection & System Testing

An, expected physical response of the system was that the result of a given pressure applied to a standing column of liquid would be a direct function of both column height and time. For example, the greater the column height of liquid in the pipette, the less time (in milliseconds) it would take to dispense a specific volume if the delivered pump pressure and valve aperture remained constant. In order to determine the non-linear nature of the system response, we empirically modeled several characteristics:

• • 1) Dispensed Volume as a function of liquid Column Height
  • 2) Dispensed Volume as a function of Pump Pressure
  • 3) Dispensed Volume as a function of Valve Aperture
  • 4) Dispensed Volume as a function of Time
  • 5) Dispensed Volume as a function of Fluid Viscosity

By constraining the dispensing pressure developed by the pump to a constant mid-range value during empirical data collection of the system (e.g., the pump PWM constrained to 175 bits; 255 bits being OFF and 1 bit being full ON), the variability of the pump speed could be used to slightly compensate for system variations during actual deployment, as well as compensate for its own wear (or thermal pressure delta over operating time). The nominal pump PWM value determined for the 50.0 ml pipette was 175 in one exemplary non-limiting illustrative implementation. Empirical modeling can use this pump setting as the “standard” programmed value, although any pump setting can be effectively modeled.

Once an effective method for closing the valve after a dispensing cycle is in place, the 50.0 ml pipette and system can be empirically modeled as follows:

• • 1) Liquid is aspirated to predetermined column heights and dispensed at varying time intervals. The dispensing cycle begins with the valve closed and the system pressure stable. The pump is then energized to a predetermined power level (e.g., always 175) and monitored for consistent pressure (e.g., delivery for a predetermined time period such as 250 milliseconds), at which time the valve is completely opened (e.g., bit count 255), with a predictable reaction (lag) time approximately 15 milliseconds according to manufacturer's specifications. The valve then remains open for a preset period of time (measured in milliseconds) while the pressure displaced a given volume of liquid into a beaker. The valve is then abruptly turned OFF, and the reversing algorithm discussed above and shown in FIG. 9B is activated (or not).
  • 2) The dispensed liquid is measured with a precision scale and tared after each measurement. An equivalency of 1.00 gram per 1.00 milliliter may, for example, be established for the purpose of scaling the quantities as closely as possible (Note: this is a directly scaleable quantity, and can be offset operationally in production or during calibration without difficulty).
  • 3) For each preprogrammed volume to be dispensed, presets such as 50.0, 40.0, 30.0, 20.0, and 10.0 milliliter column heights can be used for benchmarking. Using algebraic equivalencies, the time (in milliseconds) can be determined that precisely dispenses the preprogrammed amount. In one example illustrative non-limiting implementation, the empirical measurement margin for error acceptable (as a function of weight displacement after measurement) during this benchmarking is 1.0% for 1.0 and 2.0 ml amounts, and decreases as the volume dispensed increased (10.0 ml displacements can for example be constrained to 0.04% of the preprogrammed amount for the empirical data to be considered valid and recorded). The amounts preprogrammed and benchmarked can for example be at intervals such as 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, and 10.0 milliliters.
  • 4) The empirical data can be gathered in for example 1.0 ml column height increments from 50.0 down to 6.0 ml (45 measurements). All other measurements may be sampled at the intervals cited in the section above. See the following table:

Empirical Data Analysis

Exemplary actual 1.0 ml empirical data is as follows:

XY *	X Value	Y Value
1	6.0000000	291.00000
2	7.0000000	286.00000
3	8.0000000	281.00000
4	9.0000000	276.00000
5	10.000000	271.00000
6	11.000000	265.00000
7	12.000000	260.00000
8	13.000000	256.00000
9	14.000000	252.00000
10	15.000000	247.00000
11	16.000000	243.00000
12	17.000000	240.00000
13	18.000000	236.00000
14	19.000000	233.00000
15	20.000000	229.00000
16	21.000000	224.00000
17	22.000000	221.00000
18	23.000000	218.00000
19	24.000000	215.00000
20	25.000000	212.00000
21	26.000000	209.00000
22	27.000000	207.00000
23	28.000000	204.00000
24	29.000000	201.00000
25	30.000000	198.00000
26	31.000000	195.00000
27	32.000000	193.00000
28	33.000000	192.00000
29	34.000000	189.00000
30	35.000000	186.00000
31	36.000000	184.00000
32	37.000000	182.00000
33	38.000000	180.00000
34	39.000000	177.00000
35	40.000000	175.00000
36	41.000000	173.00000
37	42.000000	171.00000
38	43.000000	170.00000
39	44.000000	167.00000
40	45.000000	164.00000
41	46.000000	163.00000
42	47.000000	161.00000
43	48.000000	160.00000
44	49.000000	159.00000
45	50.000000	158.88888

• • where
  • XY=Displacement Cycle,
  • X=Column Height value prior to the displacement cycle, and
  • Y =Dispense Time (ms) required to displace 1.0 ml from the given Column Height (X reference).

It should be noted that whole number dispensing volumes from pre-selected pipette graduations were only used for the purpose of data clarity in this example. In general, the empirical data set can be derived from any manner of volume displacement and pipette column height, understanding that it is only a matter of mathematical presentation (variation in formulae) that would change.

FIG. 12 shows one exemplary illustrative non-limiting empirically derived dispensing times derived by equation based on such modeling procedure.

The FIG. 12 exemplary plot also contains the curve-fit equation of:

y⁻¹=a+bx

• • where
  • y=Time (in milliseconds) to dispense the preprogrammed amount (each equation is different as a function of amount to be dispensed),
  • x=Column Height (the starting Column Height of the dispensing cycle)
  • a=a constant determined by the empirical data, and
  • b=a constant determined by the empirical data.

Note that it is desirable to take the inverse function of “y” for the Time to be ascertained. The equation can then be rewritten as:

y=1/(a+bx)

The following data is exemplary “A” and “B” constants from the curve-fit analysis of the above equation for the 1.0 through 10.0 ml data:

Volume (ml)	A Constants	B Constants
1	0.003031292	6.70367000E−05
2	0.001500689	3.07417000E−05
3	0.000970459	2.03847000E−05
4	0.000706635	1.53658000E−05
5	0.000577874	1.16246000E−05
6	0.000462246	1.00846000E−05
7	0.000393395	8.55889000E−06
8	0.000338824	7.49410000E−06
9	0.000295163	6.68268000E−06
10	0.000263529	5.96896000E−06

Using only the equations which result from plugging in the “A” and “B” constants might possibly limit the pipetter 50 to whole ml volume dispensing, with a decreased accuracy and precision due to liquid column heights that are not absolute whole number increments. Therefore, in an attempt to further mathematically map the system as a single equation, the “A” and “B” constants are both analyzed in the same way the XY empirical data was evaluated, as a function of dispensing volume.

The exemplary illustrative graphs of FIGS. 13 and 14 are curve-fit plots of the “A” and “B” constants, respectively. The exemplary A constants curve-fit takes the form of:

y=a+b/x

• • where
  • a=−4.7998E−05,
  • b=0.003079612, and
  • x=Amount To be Dispensed.

The exemplary B constant takes the familiar curve-fit form of:

y=1/(a+bx)

• • where
  • a=−2251.50489,
  • b=17181.58587, and
  • x=Amount To be Dispensed.

Rewriting both of the above equations for clarity yields:

A=−4.7998E−05+(0.003079612/AMT)

• • and

B=1/(−2251.50489+(17181.58587*AMT))

• • Where AMT=Amount To be Dispensed (both equations).

Recalling the curve-fit equation where the result was the Time (ms) required to dispense a given volume from a known Column Height:

y=1/(a+bx)

• • and rewritten for clarity as follows:

Time=1/((−4.7998E−05+(0.003079612/AMT))+((1/(−2251.50489+(17181.58587*AMT)))*CH))

• • where
  • Time=Time in milliseconds required to dispense the AMT preprogrammed
  • CH=liquid Column Height
  • AMT=Amount To be Dispensed

Perhaps the simplest equation to mathematically develop for inexpensive microcontrollers that will provide the greatest accuracy for system 50 with a 50.0 milliliter pipette, and is similar in format to that which would be deployed for the 10.0 and 25.0 milliliter pipettes, while further being flexible enough to have scaleable capabilities with which to account for varying liquid viscosities, is as follows:

Time=1/((−4.7998E−05+(0.003079612/AMT))+((1/(−2251.50489+(17181.58587*AMT)))*CH))

• • Where
  • Time=Time in milliseconds required to dispense the AMT preprogrammed,
  • CH=liquid Column Height, and
  • AMT=Amount To be Dispensed.

After significant testing, the above equation has been effectively deployed such that from virtually any column height, any incremental volume of liquid can be precisely and accurately displaced (within the framework of the parameters earlier presented).

As will be understood, part of the modeling described above is based upon microprocessor 70 using prestored values and constants associated with a particular size pipette. System 50 in one exemplary illustrative non-limiting implementation may accommodate a variety of differently sized conventional or unconventional pipettes. In such exemplary illustrative non-limiting implementation, an operating mode 316 is provided to allow the user to program the pipette size. See FIG. 11. When the user presses this menu, system 50 allows the user to pick the pipette size he or she will be using. Exemplary size choices are 1 ml, 10 ml, 25 ml and 50 ml. The user picks the pipette size by using the up or down pushbuttons 106, 108. Pushing the up pushbutton increments the pipette size choice while pushing the down pushbutton decrements the pipette size choice. When the maximum or minimum pipette size is shown on the display screen 70d and the user presses the pushbutton, the system rolls over the choice. For example, the screen shows 1 ml, 10 ml, 25 ml, 50 ml, 1 ml, etc. if the up pushbutton is pressed, or 50 ml, 25 ml, 10 ml, 1 ml, 50 ml . . . if the down button is pressed. The system will remain in this mode until the graphical display switch button 70d is pressed. At that point, the pipette size is stored in the system EPROM and the system returns to the main menu.

Because the system is capable of being so accurate, in vertical dispensing applications it wouldn't be necessary to use pressure sensors as long as the starting column height was known or could be input by the user. If the starting column height is known, and because the dispensing is so accurate, subsequent column height measurements could be derived mathematically as opposed to an absolute pressure measurement. Summarily, all dispensing would be by equation alone for the entire column height.

Because the empirical modeling data points decrease as the valve open time increases, the ability to accurately model higher dispensing volumes is not as good as the multi-point data for smaller volumes. In order to improve the accuracy and precision of higher volume aliquots, an alternative method to a single valve open time would be as follows: (1) energize pump and allow constant pressure to stabilize; (2) open the valve for a time consistent for a smaller volume aliquot (i.e., 5 ml); (3) close the valve; continue to run the pump; recalculate the next dispensing quantity mathematically (or by use of the pressure sensors); (4) and open the valve for the ATD required; (5) repeat until the desired total volume aliquot has been dispensed.

There is a mathematical correlation between the various pipette sizes and column height. The relationship is linear, and is a function of cross-sectional volume. This permits calibration to be done on only one volume pipette while allowing any different subsequent volume pipette to used without further calibration. The only other consideration with respect to varying sizes of pipettes is the tapered dispensing tip of the pipettes, but this can be accommodated in firmware and is not an issue.

When the microcontroller has determined that the pipette is substantially empty (the pipette-side sensor has detected the low threshold for the given volume pipette), if the sensor is continuously monitored the microcontroller can determine when the tip of the pipette has been inserted into liquid (it can actually determine how far below the surface the tip has been inserted), and automatically aspirate to a predetermined column height. One obvious advantage would be the reduction of hand stress (a plus for those with carpal tunnel syndrome).

A math coprocessor may not be necessary, depending upon the microcontroller and/or programming language used.

While the technology herein has been described in connection with exemplary illustrative non-limiting implementations, the invention is not to be limited by the disclosure. For example, the technology herein can be applied to a wide variety of applications including fluid handling systems, foot control operation, tabletop designs, media bag reservoirs, etc. The invention is intended to be defined by the claims and to cover all corresponding and equivalent arrangements whether or not specifically disclosed herein.

Claims

1. A method of automatically dispensing liquid from a pipette comprising:

restricting gas flow into said pipette to thereby substantially retain a column of liquid there within;

computing a timing parameter based at least in part on (i) height of said liquid column within said pipette, and (ii) a desired quantity of said liquid to dispense; and

automatically controlling gas flow into said pipette in accordance with said timing parameter to thereby dispense substantially said desired quantity of liquid from said pipette.

2. The method of claim 1 wherein said restricting includes maintaining a valve closed, and said controlling includes opening said valve for a time period based at least in part on said timing parameter.

3. The method of claim 1 wherein said computing includes accessing a stored mathematical model.

4. The method of claim 3 wherein said computing further includes calculating said timing parameter.

5. The method of claim 3 wherein said computing further includes looking up said timing parameter from a stored table of empirical values.

6. The method of claim 1 wherein said controlling includes actuating a pump.

7. The method of claim 1 further including measuring negative pressure at the top of said column, and ascertaining liquid column height based at least in part on said measured negative pressure.

8. The method of claim 1 further including contacting said liquid only with said pipette during said aforementioned steps.

9. The method of claim 1 wherein said controlling includes using an open-loop control mechanism.

10. The method of claim 1 wherein said controlling controls said dispensing without using feedback.

11. The method of claim 1 further including decoupling said pipette from a handheld fixture and disposing of said pipette after use.

12. The method of claim 1 further including repeating said computing and controlling to repetitively dispense said desired quantity.

13. The method of claim 12 wherein said repeating is performed in response to user actuation.

14. A non-contacting, open loop automatic fluid dispensing method comprising:

programming an amount of fluid to be dispensed from a reservoir;

using a mathematical model to generate a time parameter; and

controlling dispensing of said fluid based at least in part on said time parameter,

wherein said method accurately dispenses said desired quantities over different column heights within said reservoir.

15. The method of automatically dispensing fluid from a fluid reservoir comprising:

programming an electronic controller with a desired quantity of fluid to be dispensed;

measuring a pressure associated with the fluid reservoir;

deriving at least one open-loop control output parameter from said measured pressure and said programmed desired quantity; and

applying a signal corresponding to said derived open-loop control parameter to at least one fluid flow control element coupled to said fluid reservoir,

wherein said applying causes substantially said desired fluid quantity to be automatically dispensed.

16. The method of claim 15 wherein said reservoir comprises a pipette.

17. The method of claim 15 wherein said flow control element includes an electric valve and said control output parameter comprises opening time of said valve.

18. The method of claim 15 further including powering an electric pump.

19. The method of claim 15 wherein said reservoir provides a suspended fluid column and said measuring includes measuring the vacuum at the top of the suspended fluid column.

20. The method of claim 15 further including issuing a control output causing said reservoir to be at least partially refilled with fluid.

21. The method of claim 15 wherein said deriving includes calculating at least one non-linear equation.

22. The method of claim 15 further including a look-up table.

23. A system for automatically dispensing quantities of fluid from a reservoir, comprising:

a storage medium that stores a parameter relating to a desired quantity of fluid to be dispensed;

a pressure transducer that measures a pressure associated with the reservoir;

at least one fluid flow control element coupled to said reservoir; and

a processor coupled to said storage medium, said processor deriving at least one open-loop control output parameter from said measured pressure and said stored quantity parameter; and applying an open-loop control signal corresponding to said derived open-loop control parameter to said fluid flow control element to control substantially said desired fluid quantity to be dispensed.

24. The system of claim 23 wherein said reservoir comprises a pipette.

25. The system of claim 23 wherein said flow control element includes an electric valve.

26. The system of claim 23 wherein said flow control element includes an electric pump.

27. The method of claim 23 wherein said transducer measures the vacuum at the top of a suspended fluid column.

28. The method of claim 23 further including means for issuing a control output causing said reservoir to be at least partially refilled with fluid.