GB2083225A - Fluid level sensor - Google Patents

Abstract

A fluid level sensor for detecting the presence or absence of a fluid in a predetermined region has two electrodes 9, 10 in this region. A voltage is applied to a first electrode 9 via an impedance 16. Means 19 are provided responsive to the voltage across this impedance and means 14 are provided responsive to the voltage between the second electrode 10 and the body of fluid. Fouling of the first electrode 9 will show on the means 19 as a fluid "absent" signal and fouling of the second electrode 10 shows on means 14 as a fluid "absent" signal. This sensor has a failure state (e.g. on power failure) corresponding to absence of fluid. The sensor is for use in situations where it is normally immersed in a conductive fluid such as water and enables fouling to be distinguished from the presence of the fluid. <IMAGE>

Description

SPECIFICATION Fluid level sensors and alarm and/or control apparatus employing such sensors This invention relates to fluid level sensors and to alarm and/or control apparatus employing such sensors.

In some. process plant, for example in large boilers, an exceptionally high degree of reliability is often required in the sensing of conditions of operation of the plant. A failure of the instrumentation may hazard the plant and may endanger life. Whilst it is essential that the particular parameter condition being sensed is properly detected and the alarm or control operation effected, it is important that the alarm or safeguard control should not normally be operated through any malfunction or failure in the alarm and/or control instrumentation.

In conventional sensing systems, an instrument malfunction might not only prohibit a wanted operation in the event of genuine condition pertaining but it might also give rise to false operation which may cause a costly and unnecessary shutdown of a large plant. In conventional instruments, high reliability is achieved by using a sound physical principle and by ensuring that the system is well engineered with high quality components. The confidence of the user in the ability of the instrument to perform correctly has to be supported by frequent functional checks. In an instrument used to detect an infrequent condition, checks must be made at relatively short intervals to reveal any inability to operate correctly. This is costly and disadvantageous because of the consequent hazards to the controlled plant.Even frequent checking however gives no guarantee of continuous instrument availability.

In U.K. Patent No. 1438271 and in U.K. Patent Applications Nos. 40600/77 and 23865/78 there are described means for enhancing the reliability of overall systems. In applying such techniques howeverto probe type level sensing systems for sensing the level of an interface between two fluids by determining which of two states, namely the presence or absence of one of the fluids (which is commonly a liquid above which is a gas), exists at a particular region within the plant, it is necessary for the sensor to have a "failure mode" such that in the event of the sensor failing, the output of the sensor reverts to that output state which would correspond to the abnormal plant state.In other words if the sensor is normally above the level of the lower fluid, a failure of the sensor should indicate the presence of the said fluid; if the sensor is below the normal level of the fluid, a failure of the sensor should then indicate the absence of the fluid. It will be appreciated that a sensor giving such output indication would normally be used in a system, such as is described in the aforementioned applications, in which checking is made between separate sensors to obtain an indication of sensor failure as well as an indication of the presence or absence of one of the two fluids which may typically comprise a liquid and a gas.

In a fluid level sensing system making use of an electrode in a vessel, the presence or absence of a conductive fluid at the level of the electrode may be determined by applying a voltage, between a counter electrode, which may comprise the body of the vessel, and the electrode. When the electrode is not immersed in the conductive fluid, there is a high impedance between the electrode and the vessel or immersed counter electrode and this can be detected by means responsive to the voltage between the electrode and the vessel or counter electrode. If a resistance is included in the circuit between the voltage source and the electrode, if the electrode is immersed in the conductive fluid, usually a liquid, current flows through this resistance and a voltage drop across it can be measured.If the electrode is immersed in non-conductive fluid, usually a gas, no current flows through this resistance and no voltage would be sensed across the resistance. It will be seen that a failure of the supply voltage results in neither of the responsive means giving any output. Hence such a failure can readily be detected. In a conductive fluid or liquid level sensor however one possible source or failure is fouling of the electrode by material which would normally be electrically conductive and which may bridge between the electrode and the body of the vessel, which is commonly metal. Such fouling may therefore be indistinguishable, by the sensing system, from the presence of a conductive fluid in which the sensing electrode is normally immersed.Failure of this nature therefore can only be detected if the electrode is not normally immersed in a conductive liquid since only then does the failure lead to an abnormal state of output of the sensor. Such a failure however cannot be detected, by consideration of the sensor outputs, if the electrode is normally immersed in the conductive fluid.

According to one aspect of the present invention, a sensor for sensing the presence or absence of a fluid in a predetermined region in a vessel containing two fluids, one being or higher electrical conductivity, permittivity or dielectric loss than and located below the other comprises at least two electrodes spaced apart in said region, means for applying a voltage (with respect to the body of said one fluid in the vessel) via an impedance to a first one of the electrodes, means responsive to the voltage across said impedance and means responsive to the voltage beween the second electrode and the body of said one fluid.

Very conveniently the two fluids will have differing electrical conductivity but, as will be apparent from the following description, the electrodes may be capacitive electrodes and the fluids might have differing dielectric constants such that the presence of one of the fluids between the electrodes decreases the impedance therebetween.

The term "fluid" is used to refer to anyflowable material, e.g. a powder. Most commonly the sensor would be used for sensing the presence or absence of a liquid, above which there may be a different liquid, a vapour or gas.

It will be seen that with this arrangement, the means responsive to the voltage across the impedance will give an indication of voltage if current flows into the first electrode as is the case when it is immersed in the conductive fluid. If the electrodes are immersed in the fluid so that the fluid bridges the region between the two electrodes, the potential drop between the two electrodes through the conductive fluid in the vessel can be detected by the means responsive to the voltage between the second electrode and the body of the said fluid. This voltage responsive means preferably therefore has a very high impedance compared with the impedance between the two electrodes. This may readily be achieved in practice since the impedance between the two electrodes can be kept small by ensuring close proximity and/or large size of adjacent surfaces of the electrodes.

The applied voltage may be a direct voltage or an alternating voltage. The impedance between the electrodes may be resistive or capacitive.

It will be seen that, with this construction, in the event of fouling of the first electrode occurring so that there is a low impedance path between the electrode and the main body of said one fluid (this low impedance path is produced by fouling between the electrode and the body of the vessel if the vessel is electrically conductive), the potential between the second electrode and the body of the conductive fluid (said one fluid) will tend towards a low or zero value even if the two electrodes are immersed. In other words such fouling of the first electrode could give rise to an output, from the means responsive to the voltage between the second electrode and the body of the said one fluid, indicative of the absence of the said one fluid.Similarly fouling of the second electrode will also reduce the voltage on that electrode to a low or zero value so thereby also giving rise to a false "fluid absent" signal. Any failure of the power supply of the voltage detector which is responsive to the voltage between the second electrode and the body of said one fluid will also result in an output indicative of the absence of said one fluid (the conductive fluid) between the electrodes.

It will be seen therefore that this fluid level sensor has a failure state corresponding to the absence of fluid.

Such a sensor therefore find particular application in conditions where the sensor is at a level in a vessel in which the electrodes would normally be immersed in a conductive fluid. The sensor may be used in an alarm or control system having two or more sensors with separate sensor paths responsive to the level of the interface between the conductive and non-conductive fluids and means for comparing the outputs of the sensor paths for providing a validated alarm or control signal when the sensors sense an abnormal level of the conductive fluid, as described in U.K. Patent No. 1438271.As described in the aforementioned Application No. 40600/77 indicator means may be provided arranged to indicate when the outputs of the sensor paths differ and therein each sensor with its sensor path is arranged so that a fault in the sensor or sensor path or the power supply thereto causes the output of the sensor path to revert to that corresponding to an abnormal level of the conductive fluid; thus the above-described fluid level sensor would be used where the electrodes are normally immersed in a conductive fluid and a sensor which fails to a "conductive fluid present" mode would be used in regions normally above the conductive fluid level, which can be utilised in conditions either normally immersed in the conductive fluid or normally above the conductive fluid and hence it is convenient, in the above-described conductive fluid level sensor, to provide a third means responsive to the voltage between the first electrode and the body of the fluid in the vessel. Use of the first and third voltage responsive means provides output suitable for use to give a "conductive fluid present" failure state.

In the following description, reference will be made to the accompanying drawings in which: Figure lisa simplified diagram illustrating in part a known form of liquid level sensor to which has been added a diagrammatic representation of a validation concept; Figure 2 illustrates a liquid level sensor constituting one embodiment of the present invention; and Figure 3 is a graphical diagram illustrating the variation of output voltage with liquid conductivity in one embodiment of the apparatus of Figure 2.

Figure lisa diagram to explain known techniques. In a steam raising boiler, it is desirable that a liquid level detector which is at a level normally in the steam region should have a failure mode in which a "water" indication is given. A level detector normally immersed in water on the other hand should have a failure mode giving a "steam" indication. For ease of explanation, it is convenient to consider these two forms of level detector separately and therefore, in Figure 1, there is shown in the upper part a level detector having a failure mode to water and, in the lower part, a level detector having a failure mode to steam. Thus a single electrode, situated anywhere in the boiler, may have an associated electronic circuit with one or other (or both) of two different failure modes.The two different circuits are illustrated respectively in the upper and lower parts of Figure 1.

Figure 1 illustrates in the upper part and in simplified diagrammatic form a known circuit which is widely used in liquid level sensing systems. The reference to failure modes on Figure 1 applies only to the electronic system, not to the sensing method itself. A vessel 1 contains an electrode 2 connected by a supply line to a source of electrical potential, illustrated as a transformer 4. The line 3 includes an impedance 5, in this case a resistive impedance. A high impedance voltage detector 7 is electrically connected between the electrode 2 and the body of liquid in the vessel. In this case, the vessel is an electrically conductive metal vessel which is electrically earthed. In the absence of a liquid in vessel 1 the potential of electrode 2 relative to the vessel itself approaches that of the supply line 3 which is fed from the source 4. It will be understood that although an A.C. source is illustrated the source may not necessarily be A.C. The indicator 7 will give zero volts output (OV) if the electrode 2 is immersed in the water but will give an output V, approaching the voltage of line 3, if the electrode 2 is above the water level.

In the lower part of Figure 1, there is shown a level indicator for use if the electrode 2 is at a level normally in the water. For convenience, the vessel 1, electrode 2 and other components have been shown again. In the lower part the electrode is energised from line 3 via impedance 5 as before. The voltage across this impedance 5 is measured by a voltage detector 6.Measurement of the potential across impedance 5 by the second voltage detector 6 and between the electrode 2 and earth by voltage detector 7 results in indications which can be summarised as follows in Table 1: TABLE 1 State of electrode Indication of V1 (6) Indication of V2 (7) Normal Supply or Electrode Normal Supply or Electrode Detector Fouled Detector Fouled Failure Failure Immersed in liquid V OV V OV OV OV Not Immersed OV OV V V OV OV V = voltage approaching voltage of line 3 of Figure 1 OV = voltage approaching zero.

These indications are illustrated diamgrammatically in Figure 1 and are processed to deliver an indication suitable for liquid presence detection applications. This is illustrated diagrammatically at the right hand side of Figure 1. Presence of OV from V2 and V from V1 simultaneously indicates water. Presence of OV from V2 and OV from V1 indicates a fault in the electronic equipment. Presence of V from V2 and OV from V1 indicates steam.

The system of Figure 1 can detect faults in the electronic equipment but cannot detect faults arising from fouling of the electrode system. When the electrode is immersed, conductive fouling of electrode 2 can result in an undetected fault state. Similarly fouling of the non-immersed electrode can result in a false "liquid present" indication. For these reasons an electrode sensing arrangement such as that illustrated in Figure 1 is unsuitable for use in applications in which the sensing electrode is normally immersed, because electrode fouling cannot be detected, the failure mode of electrode is normally immersed, because electrode fouling cannot be detected, the failure mode of electrode 2 of Figure 1 being towards "liquid present".

In a typical liquid level sensing system for a large steam-raising boiler as illustrated in Figure 1 the electrode 2 potential V2 measured by device 7 assuming no electrode fouling is 0.382 volts in water and 12.4 volts in steam. The current in the former case is 10.7 micro amperes corresponding to an impedance at 8 of 35.7 kilohms. The impedance 8 in steam is typically 18.6 megohms.

By way of an example Figure 2 illustrates in simplified form an arrangement, constituting one embodiment of the present invention, in which a second electrode 10 is introduced. The first electrode 9 corresponds to electrode 2 of Figure 1 and has a voltage supply source 18, series resistor 16, voltage detector 19 across resistor 16 and voltage detector 20 between electrode 9 and earth. These correspond to the circuit components and voltage detectors of Figure 1 which have already been considered. The second electrode 10 may be arranged at the same level as electrode 9 as illustrated or may be located above or below or co-axial with 9 as convenient.Electrodes 9 and 10 may comprise independent electrical conductivity or capacitance sensing electrodes using separate mountings, or they may be combined into a common assembly having two electrically independent terminals. In the conductivity case the coupling through the liquid between the two is arranged to be sufficient to allow the potential drop between the immersed electrode 9 through the liquid to the vessel 11 across notional resistor 12 (see inset diagram) to be measured via notional resistor 13 and electrode 10 with a voltage detector 14, having a high impedance 15. This is achieved by minimising impedance 13 which may be achieved by ensuring close proximity and/or large size of adjacent surface of electrodes 9 and 10.

The action of the voltage detectors 19 (measuring a potential V4) and 20 (measuring a potential V5) has been described with reference to Figure 1.

Considering first the non-immersed state, electrodes 9 and 10 are not bridged by liquid since one of the other or both of them are above the liquid level. Resistor 13 therefore assumes an inifinite or very high value so that detector 14 detects a substantially zero voltage V3 which is indicative of "liquid-not-present".

Considering secondly the immersed state, electrodes 9 and 10 are bridged by liquid since both are immersed below the liquid level. Resistor 13 falls to a relatively low ohmic value, so that a small voltage V4 is indicated by detector 19, corresponding to a "liquid-present" signal. The voltage on electrode 10 is lower than that on electrode 9 because of the potentiometric action of divider 12, 13, and the impedance of the liquid 17 which results from 10 being immersed.

The potential V5 on electrode 9 (detected by 21) results from the potentiometric action of impedances 16 and 12, the latter being provided by the liquid connecting the immersed electrode 9 to the vessel 11 by notional impedance 12 which may typically be of the order of 35.7 kilohms.

In the event of fouling of electrode 9 the potential V5 at 9 and hence at 20 falls to a low value, depending upon the choice of impedance 16, such that the potential V3 indicated at 14 tends towards a low or zero value which gives rise at detector 14to an, albeit false, "liquid-absent" signal. The low voltage detected by device 14 is indicative of absence of conductive liquid or an instrument fault.

Similarly fouling of electrode 10 reduces the indicated voltage V3 at detector 14 to a low or zero value giving rise to a similar false "liquid-absent" signal.

Any failures of the power supply or of the electronic voltage detectors 20 measuring V5 and 14 measuring V3 results in a "liquid-absent" signal.

Using two standard electrodes each having a 13 mm diameter cylindrical tip 28.5 mm long mounted with their axes parallel and centres 5 mm apart, the form of the variation of output voltage V3 for given cell constants and water conductivities is illustrated in Figure 3. The cell constants assumed are 0.05 from electrodes to vessel and 0.025 from one electrode to the other. The source voltage is assumed to be 10 volts from an impedance of 1 K ohm at 16 of Figure 2.

It will thus be seen that the circuit of Figure 2 permits the complete system to have a "failure mode" towards water absent. By using in an alarm or control system, two such liquid level sensors with separate sensor paths and by comparing the outputs, it is possible to obtain a validated alarm and/or control signal.

Essentially such a comparison or validation enables a determination to be made as to whether or not the indication given is false or true.

The sensing electrode system described above gives a failure mode towards "no-liquid-present" and this has marked advantage in certain forms of liquid level sensor particularly those which are in the normal state permanently immersed. Float switches are, for example, widely used in steam turbine and nuclear plant and whether they are permanently immersed or not in a given application there is usually no way of ensuring that the float is still free and is able to respond at all times to changes in liquid level. In such plant water level sensors having both failure modes are highly desirable so that full reliance can be placed on the devices to declare any incipient defect.

Because of this and because the "normal" state in some plant cannot be defined because changes in operating level may occur, it is sometimes also desirable to provide a sensor system having both failure modes. Referring again to Figure 2 the known circuit employing measurements V4 and V5 on electrode 9 can be combined with the arrangements defined herein where electrodes 9 and 10 are used together.

In one possible arrangement the known concepts of Figure 1 and the system outlined are combined as in Figure 2 in order to embrace both failure modes. By suitable processing of the output indications, it is possible to distinguish definitively by the indications of V3, V4 and V5 between true liquid level conditions within vessel 11 of Figure 2 on the one hand and any defects, such as a failure of the supply, the voltage detectors or fouling of the electrodes on the other hand.

The potential V3 sensed on electrode 10 by detector 14 of Figure 2 may be derived from the same source as it used for the known arrangements for the measurement of V4 and V5 of Figure 2 as illustrated.

Alternatively, it may be derived from a separate source applied through a suitable impedance to electrode 9.

The two systems may use separate frequencies, waveforms or polarities and/or may be time division multiplexed to operate alternately.

It will be appreciated that, in a system such as has been described above in which measurements are made dependent on the conductivity of the water, the system has to be designed in accordance with the actual conductivity magnitude. It will be seen from Figure 3 that there is a limited range of water (or liquid) conductivities over which any one system can be used and this depends on the electrode design and circuit component values.

The only type of fouling which cannot be sensed by the arrangement so far described is that in which a conducting deposit bridges the gap between electrodes 9 and 10 of Figure 3 reducing notional resistor 13 to a low value.

It will be noted that changes in liquid conductivity affect resistors 12, 13 and 17 in the same ratio so that under the normal "liquid-present" conditions the voltage V3 sensed at 14 can never normally exceed some maximum value which is less than the supply voltage. However, if conductive fouling bridges the gap between electrodes 9 and 10, reducing notional resistor 13 to a low ohmic value the potential V3 measured at 14 rises to an abnormally high value exceeding the normal value. If desired this over-voltage sensed at 14 may be arranged to be indicative of the presence of a conductive bridge between electrodes 9 and 10 and to give rise to an "instrument faulty alarm".

The resulting system has a failure mode which is towards the "non-liquid present" state and which is thus suitable for use in detection system in which the sensors are normally immersed.

This differs from the known electrode sensing system of Figure 1 in which fouling of the electrode 2 will result in a false liquid present signal, or in the immersed case be undetectable as an instrument fault.

Claims (12)

1. A sensor for sensing the presence or absence of a fluid in a predetermined region in a vessel containing two fluids, one being of higher electrical conductivity, permittivity or dielectric loss than and located below the other, comprises at least two electrodes spaced apart in said region, means for applying a voltage (with respect to the body of said one fluid in the vessel) via an impedance to a first one of the electrodes, means responsive to the voltage across said impedance means responsive to the voltage between the second electrode and the body of said one fluid.

2. A sensor as claimed in claim 1 wherein said impedance through which the voltage is applied to said first electrode is a resistive or capacitive impedance.

3. A sensor as claimed in claimed 2 wherein the means responsive to the voltage across said impedance is a high impedance voltage measuring device having an electrical impedance high compared with that of said impedance across which the voltage is measured.

4. A sensor as claimed in claim 1 wherein the means responsive to the voltage between the second electrode and the body of the conductive high permittivity or high loss fluid (said one fluid) has an impedance high compared with that of the fluid between the electrodes when the electrodes are immersed in said one fluid.

5. A sensor as claimed in any of the preceding claims wherein said means for applying a voltage to said first electrode is arranged to apply a direct voltage.

6. A sensor as claimed in any of claims 1 to 4 wherein the means for applying a voltage to said first electrode is arranged to apply an alternating voltage.

7. A sensor as claimed in any of the preceding claims and having the electrodes arranged in a vessel containing an electrically conductive, high permittivity or high dielectric loss liquid constituting said one fluid with a gas or vapor having lower conductivity, permittivity or dielectric loss above said liquid.

8. A sensor as claimed in claim 7 wherein the vessel is connected to or constitutes a boiler containing water and steam.

9. A sensor as claimed in any of the preceding claims and having also means responsive to the voltage between the first electrode and the body of said one fluid.

10. An alarm or control system for a vessel containing two or more fluid level sensors, each as claimed in any of the preceding claims with the two sensors being at a level such that they are normally immersed in said one fluid, separate sensor paths responsive to the fluid level, and means for comparing the outputs of the sensor paths for providing validated alarm or control signal when the sensors sense an abnormal level of said one fluid.

11. An alarm or control system as claimed in claim 10 and having indicator means arranged to indicate when the outputs of the sensor paths differ and wherein each sensor with its sensor path is arranged so that a fault in the sensor or sensor path or the power supply thereto causes the output of the sensor path to revert to that corresponding to the absence of said one fluid at the level of the sensor.

12. A sensor substantially as hereinbefore described with reference to Figure 2 of the accompanying drawings.