Turbidity Meter With Parameter Selection and Weighting

A portion of the disclosure of this patent document contains material which is the subject of copyright protection. The copyright owner has no objection to the facsimile production by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.

1. Field of the Invention

This invention relates to turbidity meters for the measurement of solids suspended or colour bodies dissolved in a fluid.

2. Review of the Art

Whilst the term turbidity strictly speaking refers to the light transmission and scattering properties of a liquid containing suspended solids, similar techniques to those used for measuring turbidity may commonly be used for determining the degree of opacity of coloured liquids and of gases containing suspended solid or liquid particles. For convenience, the term turbidity will be utilized hereinafter in a broad sense to refer collectively to all of these phenomena unless otherwise indicated.

Whilst direct measurements of turbidity of a fluid can be made by separating the turbidifying phase from the fluid and measuring the weight of both phases separately, this is time consuming and difficult. For most purposes, measurement of the optical properties of the turbid fluid provides a more convenient although indirect way of quantifying turbidity. Commonly measured properties for this purpose are the attenuation and scattering of light passed through the fluid. Measurements of scattering (nephelometry) are usually most suitable when the degree of turbidity and thus attenuation is low, whilst measurements of attenuation (turbidimetry) are most useful when the degree of turbidity and thus attenuation is high. The measurement range of most existing instruments is limited, so that no single instrument can provide a capability extending over the full range of turbidities likely to be encountered in the field.

A further problem arises in instruments having measuring heads which must be submerged in the fluid under test, since material from the fluid or biological growth engendered by the fluid will tend to deposit on windows or lenses associated with light sources and detectors incorporated in the head, thus introducing measurement errors. In instruments that make measurements within the fluid, rather than withdrawing samples for analysis, further errors arise through spurious responses of the light detectors as a result of ambient light.

Instruments currently available are calibrated in turbidity units (Jackson T.V. or N.T.V.) which are based on a standard suspension with reproducible optical properties which are not simply related to the content of suspended solids. The presence of colour due to dissolved solids is interpreted by most current instruments as turbidity. Such instruments are adjusted manually, and measurements are also taken manually or from chart records. Manual adjustments and readings at a measurement location are inconvenient at best and possibly hazardous or impossible: remote operation would be desirable.

We have sought to address these problems and provide an instrument which can be constructed to provide measurements over a very wide range of turbidities, and which can prevent or reduce to deposition of material on its light sensors and sources exposed to a fluid under test. The instrument can be calibrated at a measuring location, can allow for fluid colour and stray light, and can be controlled and read remotely.

According to the invention a turbidity meter comprises:

(a) a light source, and a lens for focusing light from the source into a beam;

(b) a first light detector producing an output signal responsive to the intensity of light incident thereon;

(c) at least one second light detector producing an output signal responsive to the intensity of light incident thereon;

(d) means supporting said first and second light detectors in defined positions relative to the light source in a fluid under test such that said beam is directed through said fluid towards said first light detector, and each said second light detector is positioned to detect light scattered by said fluid at a predetermined angle to said beam;

(e) means to determine when the amplitude of at least one of the output signals obtained lies within a given range;

(f) means responsive to said determining means to change at least one parameter selected from the intensity of the light source, the effective sensitivity of at least one of the detectors, and the output signal chosen so that the output of at least one of said detectors falls within said range; and

(g) means responsive to the altered value of said at least one parameter and the identity of said at least one detector to provide an output signal weighted in accordance with the altered value of said at least one parameter.

The invention also extends to a turbidity meter comprising:

(a) a light source, and a lens for focusing light from the source into a beam;

(b) at least one light detector having a lens and producing an output signal responsive to the intensity of light incident thereon through the lens;

(c) means supporting the lens of each said light detector in a defined position relative to the lens of the light source in a fluid under test such that said beam is directed through said fluid, and each said light detector is positioned to detect light from said beam after one of transmission and scattering by said fluid;

(d) means processing output signals of said at least one detector to provide a signal indicative of the turbidity of the fluid; and

(e) means to prevent deposition of material upon the lenses of said source and each said detector;

wherein said means to prevent deposition upon the lenses comprise a source of pressurised fluid compatible with that under test, a flow forming chamber adjacent to each lens and conduits from said source to each said flow forming chambers, each flow forming chamber having an exit orifice partially surrounding the periphery of a surface of its associated lens nearest the fluid under test, whereby a flow of said compatible fluid is maintained over that surface of the lens.

Further features of the invention will become apparent from the following description of a preferred embodiment thereof.

FIG. 1 is an elevational view of an instrument embodying the invention;

FIG. 2 is a longitudinal cross section of a light source or sensor assembly utilized in the instrument of FIG. 1;

FIG. 3 is an end view of the assembly shown in FIG. 2; and

FIG. 4 is a block schematic diagram of electronic components of the instrument.

Referring to FIG. 1, the instrument consists of a control unit 2, a first junction box 4, a second junction box 5, a rigid tubular corrosion resistant support conduit 6, and a sensor head 8. The sensor head is conveniently a continuation of the support conduit 6, and supports a light source assembly 10 and two sensor assemblies 12, 14 for immersion in the fluid under test, which will usually but not necessarily be a liquid, typically water, containing suspended matter.

The junction box 4 establishes electrical connections through a cable 16 from the control unit 2 to the assemblies 10, 12, 14 by means of a connection block 17. It also passes a group of three tubes 22 extending from the junction box 5 to the assemblies 10, 12 and 14. The junction box 5 has a pressurized fluid supply pipe 18, typically connected to a piped water supply. A regulator valve 20 meters fluid from the pressurized supply through a manifold 21, which may include a pressure sensor 86 (see FIG. 4), to the tubes 22 which pass down the conduit 6 to the assemblies 10, 12, 14.

The assemblies 10, 12 and 14 are physically identical, an exemplary assembly being shown in FIGS. 2 and 3. Each has a body 24 secured over a slot 36 in the conduit 6 by screws 26, a watertight seal between the assembly and the body being assured by use of a silicone sealant. The body has a first bore 28 housing an optoelectronic device 30 having connections through a cable 32 to the cable 16 via the terminal block 17. A further parallel bore 34 is fitted with a spigot 38 connected to one of the pipes 22. A washer or gasket 40 has a slot extending over most of its diameter from one side of the assembly, over a lens 42 of the device 30, with the distal portion of the slot forming a flow forming chamber 44 at the outlet of the bore 34. A cover piece 46 is secured over the washer 40 by screws 48, and is also formed with a radial slot 50 overlying the slot in the washer 40 except in the region of the flow forming chamber 44.

The assembly 10 is supported directly opposite and in axial alignment with the assembly 12 (see FIG. 1) so that a beam of light emitted by the device 30 of the assembly 10 and focused by its lens 42 is directed onto the lens 42 and device 30 of the assembly 12.

In preferred arrangement, the device 30 of the assembly 10 is a light emitting diode (LED) of suitable spectral characteristics, whilst the device 30 in each assembly 12 and 14 is a phototransistor which conducts a current proportional to the intensity of light incident upon it. The phototransistors are matched to the light emitting diode of assembly 10 such that they respond to light of similar spectral characteristics to that emitted by the diode. By suitable choice of the diode and transistors, a fairly high degree of immunity to ambient light conditions can be obtained. Typically gallium arsenide diodes are used which emit monochromatic light in the infra red region at 940 nm. It should be understood however that other light sources such as incandescent lamps or solid state lasers, and other sensor devices such as photodiodes and bolometers can be utilized, possibly in conjunction with colour filters or other devices to restrict spectral response in a manner such as to reduce the influence of ambient light. A phototransistor operates over a very wide range of currents, and as such its effective sensitivity can be adjusted by modifying the gain of preamplifiers 54 and 56 (see FIG. 4).

The assembly 14 is disposed so that the optical axis of the system formed by its LED 30 and associated lens 42 intersects the common axis of the assemblies 10, and 12 at an angle of 45.degree. so as to receive light scattered rearwardly at that angle from the beam generated by the assembly 10. Whilst the exemplary embodiment shown is arranged so that the assembly 14 receives light scattered rearwardly at an angle of 45.degree., the position of the assembly can be changed so that it is disposed to receive light scattered at different angles, or so that it receives forwardly scattered light. Alternatively, additional assemblies 14 may be provided having different angular relationships to the source. The assembly 14 may be mounted at the bottom of the head rather than in the position shown.

The LED in the assembly 10 is connected by the cables 32 and 16 to a current source 52 (see FIG. 4) in the control box 2, whilst the phototransistors in the assemblies 12 and 14 are similarly connected to preamplifiers 54 and 56 respectively. Whilst these preamplifiers will usually be located in the control unit 2, they may be located in the connection box 4 or at the assemblies 12 and 14, particularly if a long cable 16 is required. The preamplifiers act in fact as current to voltage converters. At least the preamplifier 54 has a plurality of selectable levels of gain, the value of resistance in a negative feedback path of the amplifier being selected by means of a multiplexer 78 which decodes a digital input signal to switch different resistances or combinations of resistances into the feedback path.

The internal circuitry of the control unit 2 is described further with reference to FIG. 4. The instrument is under control of a microcontroller 58, which is conveniently of type 8052AH-BASIC from Intel Corporation. This controller includes an eight bit microprocessor, various peripherals and a dialect of the Basic language resident internally in read only memory so that the microcontroller can be programmed directly in BASIC; the BASIC program in this case is stored in a read only memory 60.

The controller is also provided with random access memory, including both volatile memory 62 and non-volatile memory 64, the former being utilized as working memory and the latter for the storage of instrument readings and parameters required on a longer term basis.

Parallel ports 66 are provided both by the built-in ports of the controller 58 and an external parallel port unit 66, conveniently of type 8255 from Intel Corporation. The ports 66 provide an interface between the controller 58 and a keypad 68 through which data and commands may be entered into the system. They also provide, through drivers 70, an interface with a display 72, typically consisting of a sixteen digit LCD display and several individual indicator LEDs. A control line is also provided to a piezoelectric buzzer 84 to provide audible feedback to a user. Lines from the parallel port unit are connected to the multiplexer 78 to vary the gain of preamplifier 54, and/or to a decoder 74 to provide one of four selection of four different currents which can be generated by the current source 52, which would in that case consist of four current sources with their outputs connected in parallel and inputs switched by field effect transistors connected via suitable drivers to the decoder outputs. An input line may also be provided from the pressure sensor 86, which would be monitored by the controller to verify the existence of adequate fluid pressure in the tubes 22, failing which an alarm output to the buzzer 84 and other outputs of the instrument would be provided. In the description that follows, a simplified arrangement is described in which only the gain of the preamplifier 54 is controlled, the current from current source 52 remaining constant and merely being turned on or off by the port unit 66.

A digital-to-analog converter 82 provides output data from the controller in analog form for use by an external process controller or alarm device, and preferably includes an output circuit providing a standard 4-20 ma output.

The controller is also connected to a serial interface 76 providing a serial port conforming with an accepted standard such as RS232C so that it may interchange data with a remote computer. Depending on requirements, this may be implemented externally, but a presently preferred embodiment uses internal serial ports of the controller with suitable external drivers.

The preamplifiers 54 and 56, and corresponding preamplifiers for any additional sensors which may be provided, are connected to the inputs of multiple channel analog to digital converter 80 whose digital outputs are available to the data bus of the controller 58 on demand.

The controller 58 incorporates a clock generator conventionally frequency stabilized by a quartz crystal, and a built-in real time clock and real time interrupt system. Further features of the controller 58 can be found in the relevant product literature published by Intel Corporation.

The ports incorporated in the interface 66, the serial interface 76, the LCD display, the multiplexer 78 and the analog to digital converter 80 are all mapped into the memory space of the controller 88.

It will of course be understood that the control unit 2 is provided with a suitable electrical power supply, either line or battery operated.

In use of the apparatus, the sensor head 8 is supported so that it is submerged in a fluid under test. It is assumed for the purposes of description that this is an aqueous liquid. The pipe 18 is connected to a piped water supply which in general will provide sufficient pressure to provide a desired flow through the apparatus. The regulator valve 20 is set to provide a flow through the pipes 22 such that water will leave the flow forming chambers 44 and pass across the lenses 42 at a flow rate preferably in the range of 0.5 to 2.0 cm/s. The pressure sensor 86, if provided, verifies that adequate pressure is available. Typically the height of each chamber 44, as determined by the thickness of the washer or gasket 40, is about 0.8 mm. The extent of the chamber, between the passage 34 and the exit from the chamber around one half of the periphery of the lens 42, is sufficient to produce a substantially uniform generally laminar flow from the flow forming chamber across the surface of the lens and out through the slot 50. Other arrangements of flow forming chamber may be employed that will achieve the same result, i.e. a curtain-like flow over the lens that will prevent deposition of material from the fluid under test. The small quantity of clean water introduced into the fluid under test will be insufficient to have any significant effect upon the turbidity readings obtained.

A known approach to preventing unwanted deposition on sensor or light source lenses has been to provide a ring of pressurized liquid jets surrounding the lens and parallel to its axis. The axis of the jets may be yawed from the axis of the lens to impart a slight swirl to the liquid. Such an arrangement however provides a greater interference with fluid under test in the path of light entering or leaving the lens than does the relatively thin laminar flow of the present arrangement, and is likely to require the use of greater quantities of a pressurized fluid.

In the example described, the pressurized fluid is water. Where the fluid under test is not aqueous, the pressurized fluid should be compatible with the fluid under test. If a pressure pump is provided, a part of the fluid under test may be pressurized and utilized, although in this case a filter should be provided to remove suspended matter from the fluid so utilized.

Whilst in the arrangement described, the lenses 42 would be directly exposed to the fluid under test were it not for the flow of pressurized fluid discussed above, in other arrangements the light source or sensors might be located behind windows. The term lens, as used here and in the appended claims should be read as including such windows or any other transparent layer through which light passes between the source or sensors and the fluid under test. The fluid flow from the flow forming chamber should be at least wide enough to cover the effective area of the lens, and preferably somewhat wider.

Once the sensor head 8 is installed and the regulator valve 20 is adjusted, the control unit 2 is powered up and the program stored in ROM 60 commences to execute. The program first initializes various storage arrays in RAM 62, sets up default values of variables used during operation, including variables indicating the addresses for the display, the analog to digital converter, the serial port and the ports provided by unit 66. It then sets up a record buffer in NOVRAM 64 for storage of data captured by the unit during use, and initializes the real time clock.

The system is next tuned for each possible source-detector combination. This involves activating the source assembly and selecting in turn each value of gain provided by the preamplifier 54, by means of the multiplexer 78, in order to determine which gain level is best suited for maintaining the output of preamplifier 54 within a preferred range. For each level of gain the outputs of preamplifiers 54 and 56 are sampled first with the LED source in assembly 10 turned on, and secondly with it turned off, the second samples being subtracted from the first so as to reduce the effect of ambient light. The preamplifier outputs are sampled by the A/D converter 80. In practice a set of 1-200 sets samples taken at intervals is obtained for each level of gain, and for each selection of sensor assembly, and each set of readings obtained is averaged. The averaged readings are reviewed to determine the most appropriate source-sensor pair to select for use, and the most appropriate gain to select for the amplifier 54 associated with assembly 12. If the current applied to the LED in assembly 10 is selectable, the effect of different current levels can be similarly tested.

Assuming that the maximum digitized signal level which can be handled by the converter 80 is 255, this signal level is subtracted from a value representing approximately the midpoint of the signal range, for example 128, and the result is squared, thus producing an output which increases according to the divergence of the reading from the mid point of the range. The most appropriate value gain to select is that which minimizes the output, i.e. which produces a signal level nearest the middle of the range. In fluids of low turbidity, each of the gain values will result in a signal from the sensor assembly 12 which saturates the converter 80 and thus this signal will have no influence upon the selection process, which will then select the sensor assembly 14.

Once the source detector pair and the gain value for the preamplifier 54 have been initially selected, the program enters a loop in which direct or scattered light readings from the sensors are taken at intervals determined by the real time clock and averaged, the averaged readings being compared with previous readings. If the new averaged reading is outside a defined range, then the tuning process described above is repeated, and a new set of readings is then taken and averaged. Otherwise the averaged readings are subjected to further processing by a data smoothing algorithm and undergo appropriate logarithmic transformation prior to display to allow for the gain value selected, and to allow for whether the signals selected are those from the sensor assembly 12 or the sensor assembly 14 (in some cases both).

The output may be fitted to a calibration curve so as to convert it to desired units for the purpose of display, or the data can be displayed in raw form. The calibration curve may for example translate the readings into J.T.V. or N.T.V. units, grains/ft.sup.3 or mg/L. In the example program provided in the appendix, this is achieved by applying a linear equation to the transformed signals to generate data in the desired units. Constants for use in the equation are supplied to the control unit either via the serial port or the keyboard. The data may be displayed directly by the control unit 2, and/or transmitted via the serial interface to a printer for printing or to another computer for further processing. Successive readings are stored in NOVRAM 64 so that they may be dumped or reviewed as required, and will be retained even when the instrument is powered down.

Successive raw readings are taken in sets which are averaged as described above before further processing. Provision can be made for readings which differ widely from immediately preceding readings to be rejected unless repeated several times; this enables erratic readings due to such phenomena as bubbles in the fluid to be rejected, and avoids output jitter.

It will be understood that various modifications to the arrangement described are possible, some of which have already been mentioned above. Thus whilst in the embodiment described, automatic range switching or `tuning` of the apparatus is performed by switching the gain of a preamplifier in a sensor, additional or alternative range switching can be implemented by changing the current applied to the LED in assembly 10. Mechanical means of range switching may also be employed, for example by physically moving the source and sensors towards or away from each other, or by applying iris arrangements to change the effective aperture of the lenses of the source and/or sensor assemblies.

In the arrangement described above, the apparatus by default preferentially uses the turbidometric mode of operation, the source-sensor pair provided by the assemblies 10 and 12 being selected if a suitable signal level can be obtained, with fall back to the nephelometric mode using assemblies 10 and 14 when turbidity is too low to provide usable signal levels in the turbidometric mode. Other forms of operation are possible. Thus in some cases it may be desired to disable either the nephelometric or the turbidometric mode, or to utilize data from both in combination. With suitable programming of the unit, the operating format can be selected either from the keyboard or by an external computer or control unit communicating with the control unit 2 through its serial port.

The NOVRAM 64 permits the instrument to serve as a monitor over an extended time period. In this mode, turbidity data readings are stored, together with a time from the real time clock, only when a change occurs, and sampling intervals are selected according the period over which monitoring is required and according to the memory available. In a typical configuration, about 1000 readings may be stored before early readings are overwritten; more extended periods of operation can be achieved by periodic dumping of data to an external printer or a host computer. The instrument can be utilized for process control or alarm purposes by use either of the analog output from the digital-to-analog converter 82, which may be provided with a standard 4-20 ma interface, or of digital data from the serial interface 76, each of which can provide the running average output signal discussed above. The use of the averaged signal provides some protection against erroneous readings produced by gas bubbles and the like, thus reducing the risk of false alarms.

Since the preferred form of the invention utilizes a light source producing monochromatic light, typically in the infra-red region, a correction is necessary to allow for changes in the apparent turbidity of the fluid under test as compared to observations made using visible light of conventional spectral distribution. Such correction, if required, can be incorporated in calibration data provided to the control unit.

Provision may also be made to compensate for ambient light by periodically determining the output of the sensor assemblies when the source assembly is turned off, so as to provide a correction to be applied to the readings. This technique has already been described above in connection with the tuning procedure. Any compensation necessary for stray light from the source, for example reflected from the walls of a vessel within which measurements are made, can be applied as part of the calibration data. Similarly, allowance for colour background can be made utilizing data from both the sensors 12 and 14 in conjunction with a mathematical model of the scattering phenomenon. Allowance for background can also be made during calibration.

The following Appendix represents the presently preferred version of the program stored in ROM 60, to which reference is made above. Copyright in this program is owned by Claritek Instruments Inc. ##SPC1##