WO2001063219A2

WO2001063219A2 - Time domain reflectometry

Info

Publication number: WO2001063219A2
Application number: PCT/EP2001/002087
Authority: WO
Inventors: Eric Nicol Clouston; Alistair Guy Morfey
Original assignee: Cambridge Consultants Ltd
Current assignee: Cambridge Consultants Ltd
Priority date: 2000-02-23
Filing date: 2001-02-23
Publication date: 2001-08-30
Anticipated expiration: 2002-08-23
Also published as: GB0004306D0; WO2001063219A3; AU5033701A

Abstract

A time domain reflectometry probe for fluid level measurement has a first portion which is vertical and a second portion which is horizontal. The horizontal portion has greater resolution than the vertical portion to give increased accuracy over a region of the measurement range. The first and second portions can also be at acute angles to the vertical and horizontal respectively.

Description

TIME DOMAIN REFLECTOMETRY

The present invention relates to time domain reflectometry (TDR) and m particular to the application of TDR to the sensing of fluid levels m containers.

Time domain reflectometry is the analysis of the reflection of electromagnetic waves propagating along a transmission line m order to derive information relating to discontinuities m the environment m which the transmission line is located.

A wave propagating along a transmission line will be reflected where there is an abrupt change m the impedance of the line, i.e. at an interface between two regions of different impedance. For a theoretically lossless transmission line, the (voltage) reflection factor for the amplitude of a reflected wave produced when an incident wave passes from a region of characteristic impedance Z_1# to a region of characteristic impedance Z₂ is given by

Z„ α = :D z_{1 +} z₂

Thus, where the pulse passes from a region of high impedance to a region of low impedance (Z_x > Z₂) , the reflection factor is negative indicating that the reflected pulse is m antiphase with the incident pulse. Similarly, when the pulse passes from a region of low impedance to a region of high impedance (Z₂ > Z_x) the reflected pulse is m phase with the incident pulse.

Reflections commonly occur at the end of transmission lines, where if there is an open circuit [ Z_Δ = ∞) the entire pulse is reflected m phase with the incident pulse, and if there is a short circuit (Z₂ = 0) the entire pulse is reflected m antiphase with the incident pulse. Reflections can be prevented at the end of a transmission line by terminating the transmission line with an impedance equal to the characteristic impedance of the line (Z_± = Z₂) .

In TDR, the difference m impedance between two regions is caused by a variation m the environment surrounding the transmission line. For example, if the transmission line is located m a vertical orientation m a partially full liquid container there will, m general, be an abrupt change m impedance at the level of the surface of the liquid due to the difference m the electromagnetic properties of the liquid compared to the atmosphere.

The impedance of the transmission line is influenced by the permittivity ε (and the permeability μ) of the material around the conductors which make up the transmission line. At the interface between two regions of different permittivity around the conductors there will be a change m impedance of the transmission line which will result m the reflection of a wave propagating along the transmission line.

The permittivity ε of the material around the conductors of the transmission line also determines the wave velocity v at which waves propagate along the transmission line. The effect of changes m the wave velocity v due to changes m the environment surrounding the transmission line on reflected waves can also be used to derive information about the environment surrounding the transmission line.

In general, a TDR system comprises a transmitter which generates a signal and applies the signal to a transmission line, also known as a probe. A receiver detects the signal close to the start of the probe and also any reflections due to changes m the environment around the probe. The signal applied to the probe by the transmitter is usually a pulse or a rapid transition from one level to another, such as a rising edge, as the time-limited signals makes the identification of reflections easier than if a continuous wave is used.

WO 90/15998 and US 5457990 disclose a system utilising TDR to detect an object m the vicinity of a transmission line or to detect the level of a fluid m a tank Further examples of fluid level sensors utilising TDR techniques are described m WO 98/03840, US 5651286, US 5609059, US 5610611 and US 4924700.

Although there have been many disclosures of theoretical TDR level sensing systems m the prior art, to date a low-cost commercial system suited to practical applications such as automotive fuel and lubricant level sensing has not become available. In general, TDR level sensing has represented an expensive technology m comparison to float-actuated potentiometers and the like which have traditionally been used m such applications. However, a TDR sensor with no moving parts is technologically attractive m, for example, automotive applications, if the cost is reasonable.

One reason for the expense of currently available TDR level sensors is the cost of the electronics required to process the signals generated by the transmission line. The electrical signals travel along the transmission line at approximately the speed of light (3xl0⁸ ms^"1) when the transmission line is m air and at around 4xl0⁷ ms ^x m water, which has a relative permittivity (dielectric constant) ε_r of 70. In hydrocarbons, which generally have a relative permittivity of around 4, this speed is around 1.5xl0⁸ ms^"1. Thus for a measurement accuracy of a few millimetres, the processing electronics must be able to resolve a time difference of much less than a nanosecond. One way to achieve th s is to use processing electronics that can operate at gigahertz frequencies, but these are expensive and therefore increase the price of the TDR device. It is also possible to use sampling techniques to reduce the frequency at which the analysis electronics must operate. However, the speed of the electronics required is still relatively high compared to the speed of commonly available and relatively inexpensive silicon CMOS integrated circuits. The inventors m the present case have been working to develop a TDR fluid level sensor which is sufficiently small and inexpensive for use m automotive level sensing applications.

According to the present invention there is provided a time domain reflectometry probe for a fluid level sensor comprising two elongate spaced conductors arranged to function as a transmission line, wherein the probe is configured such that, m the position of use, a first portion of the probe is at a first angle to the vertical direction and a second portion of the probe is at a second angle to the vertical direction, and the first angle is smaller than the second angle.

The inventors have realised that m the field of fluid level sensing it is not always necessary to measure the fluid level with the same degree of accuracy and confidence m all regions of the range of measurement of the sensor. For example, towards the bottom of a fuel tank it can be extremely important that the fuel level is accurately measured by the fluid level sensor, but it is less important to know exactly the fluid level of the tank around the half full level, for example. Likewise, at the top of a bath tub it is important to know accurately the fluid level m order to prevent overflow, but it is less important for the fluid level to be accurately detected at lower fluid levels. According to the invention, the TDR probe can be configured so that m regions where high accuracy and confidence are required, such as at the top of a bath or the bottom of a fuel tank, the angle between the probe and the vertical direction is greater than m regions where the accuracy of the measurement is less crucial . The closer a portion of the probe is to the horizontal, the greater the length of the transmission line that is provided for detecting the interface between a fluid and the atmosphere or another fluid at a given level. Thus, the ratio of distance along the probe to fluid depth is increased, and the effective accuracy of the probe is increased without having to increase the accuracy of the analysis electronics.

In the prior art, TDR probes are generally arranged vertically m a container or at best at a constant angle to the vertical. There is no suggestion m the prior art of a TDR probe with portions at different angles to the vertical for enhanced accuracy over a region of the measurement range .

The first angle may be any angle m the range 0° to 45° or 45° to 90°. The first angle may, m particular, be about 0°, i.e. the first portion of the probe may be substantially vertical m the position of use.

The second angle may be any angle within the ranges 0° to 45°, 45° to 90°, 90° to 135° or 135° to 180°. In particular, the second angle may be about 90°, i.e. the second portion may be substantially horizontal .

When the second angle is m the range 90° to 180°, the second portion runs m the opposite direction to the first portion. In this case, two reflections from a single fluid interface may be generated m the probe which may increase the accuracy or confidence level of the fluid level measurement.

Although the first and second portions of the probe have been described in terms of angles, the probe is not necessarily rectilinear in configuration. The probe may be curvilinear m configuration, and in this case the first and second angles may be considered to be the angles between the respective tangents to the first and second portions and the vertical direction. The first portion and/or the second portion need not be straight. In particular the first and/or second portion may be shaped, for example bent, to reduce the horizontal extent of the portion. In this way, the length of the first and/or second portion of the probe can be maintained to give a high ratio of probe length to depth while providing a horizontally compact probe. Thus, for example, the first and/or second portion may have a coiled, undulating, circular or spiral configuration .

The second portion may be adjacent the first portion or may be spaced therefrom by an intermediate portion. Indeed, there may be more than two portions, with each portion at a respective angle to the vertical. In one embodiment, the configuration of the probe is chosen with reference to the volume profile of the container into which the probe is to be fitted m order that the distance along the probe is proportional to the volume of fluid m the container.

The second portion may be an end portion of the probe, for example a lower portion and/or an upper portion of the probe. In such a case the accuracy of the level measurement at the top or bottom of a container may be enhanced. Furthermore, the provision of a, for example substantially horizontal, end portion of the probe may improve the distmguishableness of the reflections from the end of the probe from the reflections from an extreme (low or high) fluid level. The probe may be configured as a pair of parallel spaced conductors. In a preferred configuration, the conductors are arranged with an outer conductor surrounding an inner conductor, preferably substantially co-axially, so that the outer conductor provides electromagnetic screening to the inner conductor. This is desirable where the prevention of electromagnetic interference is important, such as m automotive applications . The conductors may be arranged such that one conductor only partially screens the other if it is desired to locate the transmission line close to a surface of a container. Thus, for example, an inner conductor may be positioned on one side against a surface of a container, for example a bottom surface, and an outer conductor may be provided to surround the inner conductor on the remaining sides thereof . This m itself is believed to be a novel configuration and thus viewed from a further aspect, the invention provides a time domain reflectometry probe comprising an inner conductor and an outer conductor, wherein the conductors are configured such that, m use, the inner conductor is positioned on one side against a surface of a container and the outer conductor surrounds the inner conductor on the remaining sides thereof. The advantage of this arrangement is that the probe is able to sense fluid levels very close to the surface of the container, for example fluid levels very close to the bottom of the container.

The transmission line may be terminated with an impedance substantially equal to the characteristic impedance of the line to prevent reflections at the end of the transmission line. However, it is preferred for the transmission line to be terminated by connecting the conductors with a short circuit. This is less expensive than providing a matched impedance and is generally unaffected by environmental change. Furthermore, the short circuit connection may be used to provide mechanical stability to the probe.

Thus, there may be provided a TDR probe comprising two conductors forming a transmission line wherein the transmission line is terminated m a short circuit.

The space between the conductors will m general be open to the fluid so that the fluid influences the impedance of the transmission line. However, it is possible for a dielectric material to be provided between the conductors m a region of the probe to provide a region which is electrically unaffected by the fluid. In particular, such a region may be provided at an upper or lower end of the probe in order to provide a delay between reflections from the end of the probe and reflections from the fluid level . Such an arrangement eases the processing of the reflected signals by ensuring a delay between reflections due to the end of the transmission line and reflections due to the fluid.

This m itself is believed to be new and thus from a further aspect the invention provides a TDR probe comprising two conductors forming a transmission line and having a first section which is open such that, in use, a fluid is able to enter a space between the conductors and affect the local impedance of the transmission line, and a second section into which a fluid is unable to pass, so that the impedance of this section is unaffected, in use, by changes in fluid level .

The measured fluid may be a liquid but could also be a powdered or particulate solid or any other fluid material . In general, the probe will have associated with it signal processing apparatus arranged to analyse reflections from the probe to determine a level of a fluid in which the probe is located. The signal processing apparatus generally comprises a transmitter for transmitting a signal, such as a pulse, along the transmission line and a receiver for detecting this signal and any reflections due to the fluid or the end(s) of the transmission line.

Some embodiments of the invention will now be described, by way of example only and with reference to the accompanying drawings, in which:

Figure 1 is a schematic sectional view of a coaxial TDR probe according to the invention;

Figure 2 is a schematic sectional view of a further TDR probe according to the invention;

Figure 3 is a schematic view of a section of a yet further TDR probe according to the invention; Figures 4a and 4b show two embodiments of a U- shaped TDR probe according to the invention;

Figure 5 is a schematic view of the general form of a TDR probe according to the invention; Figure 6 is a further schematic sectional view of a TDR probe according to the invention;

In the following, corresponding reference numerals are used for corresponding parts in the various embodiments . In traditional TDR probes, a two-wire probe has been used to give good characteristics as a broadband radio frequency (RF) transmission line. However, it has been found preferable to use a screened probe comprising one conductor enclosed by another conductor in order to reduce unwanted RF emissions and susceptibility to RF interference. An example of this is a coaxial L-shaped probe according to the invention shown in Figure 1, which comprises a first conductor 1 in the form of a rod, mounted in the middle of a second conductor 2 in the form of a tube. A space is provided between the conductors into which fluid can flow and thereby change the local impedance of the transmission line. The probe is terminated in a short circuit 3 with the first conductor 1 electrically connected directly to the second conductor 2. This provides both strength, simplicity of support and improved immunity to electrostatic discharge for the probe.

Alternatively, the probe can be terminated with a resistance (not shown) close to the characteristic impedance of the transmission line, to reduce the size of the reflection from the end of the probe. Compared with a short-circuit, this is more expensive to build, harder to make mechanically robust and more prone to problems over time, for example, if conductive particles build up in the bottom of the probe and reduce the effective resistance, but the processing of the reflected signal from the probe is simplified. The probe of Figure 1 is open at the top and the bottom, to allow the fluid to flow in and out between the conductors 1, 2. To measure multiple layers of different fluids, holes 4 are defined at regular intervals in the second conductor 2 along the probe, to allow the fluids to settle inside the probe at levels close to those outside.

A problem with prior TDR probes is that the reflected signal from the fluid surface can be difficult to analyse accurately when it is close enough to overlap the reflection from the lower end of the probe, or to overlap the initial signal, near the top of the probe. The simplest solution to this is to let the probe extend beyond the bottom of the container 10 in which the fluid is held, but this is often impractical to implement, because it may weaken the structure of the container. In car fuel tanks, for example, it is not possible for the probe to protrude without the risk of damage due to hitting the road surface in uneven driving conditions. The probe shown in Figure 1 connects to signal processing equipment (not shown) via electrical connections 5, 6 to the respective conductors 1, 2. At the top of the probe a dielectric 7, such as polythene or the like, is provided between the conductors 1, 2 so that this section of the probe is unaffected by changes in the fluid level and acts as a delay for the signal from the signal processing equipment to prevent confusion between the initial signal and the reflection from the highest fluid level . A dielectric section (not shown) may be provided at the bottom end of the probe in order to ensure a delay between reflections from the bottom end of the probe and those from a low fluid level . At the bottom of the probe, however, it is desirable for the open section, i.e. part of the probe without the dielectric material, to be as close as possible to the bottom of the fluid container 10. This is possible with a dielectric section, as described, because the dielectric section separates the reflection due to the fluid level from the reflection due to the end of the probe, (which may have a simple short-circuit or open-circuit termination) . The dielectric section of the probe is bent sideways or back on itself so that the open section of the probe is as close to the bottom of the container 10 as possible. The lowest fluid level which can be detected is therefore determined only by the probe width. If the L-shaped probe is open at its lower end and is 10 mm wide, say, the amplitude of the signal due to the fluid will decrease as the tank empties over the last 10 mm. The advantage of a probe which is open at its lower end is that it is likely to be easier to build than a probe which is closed with a section of dielectric. It also detects fluid all along the horizontal section, giving the possibility of greater sensitivity to the last few millimetres of fluid. There is also potentially useful information m the amplitude of the fluid reflection and m the delay of the end reflection and m the dynamics as the fluid sloshes.

To improve sensitivity to the last few millimetres of fluid m the bottom of the tank, an L-shaped probe as shown m Figure 1 may be used, but with the norizontal part effectively cut m half, with the screening removed from the bottom so that the inner conductor 1 can lie along the bottom of the container 10. This is shown m Figure 2 which also shows that the outer conductor 2 may be made of folded sheet metal to give a square-section construction. The dimensions should be tailored to give an acceptable characteristic impedance, minimising unwanted reflections from the bend m the probe.

In this way, rather than working normally down to the last 10mm, the probe can work normally down to the last couple of millimetres of fluid.

The horizontal section of the L-shaped probe may be bent round the bottom of the probe m an arc of a circle, up to one turn, and in a spiral if more than one turn, to save space. In situations where it is acceptable to use reduced screening, a simple arrangement, shown in Figure 3, has an outer conductor 2 in the form of a vertical tube with no horizontal section and a hole 8 defined in the side. The inner conductor 1 is arranged vertically down the centre of the outer conductor 2 and is bent through a right angle to emerge through the hole 8. The inner conductor 1 forms an arc around the outside of the outer conductor 2, before being welded to it to form a short-circuit termination. This may either be an open structure, or a closed structure, with dielectric filling the gap between the conductors 1, 2 around the outside. The above described L-shaped probes can be adapted to form J-shaped probes, by bending the horizontal section through an angle of more than 90°, for example 180° . Similarly, a U-shaped probe as shown in Figures 4a or 4b may be provided. Other possibilities include C-, S-, Z-, V-, N-; M- or W- shaped probes.

With a U-shaped probe, it is possible to separate the end-of-probe reflection completely from the reflections from the fluid near the minimum level and also achieve twice the resolution, because two reflections are generated by the fluid level. In this case, the transmission line extends to the lowest desired level, bends round, and returns in the opposite direction. Thus, the TDR probe has a first portion at 0° to the downward vertical and a second portion at 180° to the downward vertical. The signal processing equipment therefore receives a reflected signal where the signal enters the fluid (from air) and another reflected signal, of opposite polarity, where the signal leaves the fluid (and goes back into air) . The signal processing equipment can either: measure the delay between these two reflections and use an estimate of the speed of propagation in the fluid to derive the depth; or measure the distance from the top of the probe to the top of the fluid by reference to the wave velocity in air; or - measure the distance between the point where the signal exits the fluid and the far end of the probe .

In practice, all three measurements are made, and a weighted average is used to give the best estimate of the depth. The weighting may vary with depth, as the first method is more accurate for low fluid levels but the other methods are more accurate near higher levels.

The U-shaped construction may be extended to form a probe which passes up and down through the container a plurality of times, for example a W-shaped probe. The general form of such a probe is shown in Figure 5. This allows either more estimates to be made of the fluid depth, from the different reflections from different sections, or estimates to be made of the depth of fluid in different positions i.e. sensing in multiple locations with a single probe. In practice, multiple reflections between two instances of the fluid level should be minimised, for example using a probe with low sensitivity to the fluid. A proportion of the space between the conductors 1, 2 may be filled with solid dielectric, as described above, so that only a proportion is available to the fluid.

For ease of installation and for flexibility of installation, a flexible probe is advantageous. Such a flexible TDR probe with screening i.e. with one conductor larger than the other and at least partially protecting it from interference, may for example be in the form of a coaxial cable type arrangement, with some dielectric to support the inner conductor, but with the space between the conductors largely filled with air, and with the ability for fluid to flow freely along inside the probe. The embodiments of Figures 1 to 5 may be likewise constructed from flexible materials with an appropriate provision of dielectric spacers.

It will be apparent that a vertical probe gives a 1:1 correspondence between fluid depth and distance along the probe. A diagonal probe gives increased accuracy and resolution because distance measured along the probe is longer than the actual depth. According to the invention the probe is bent to a shape where it is close to vertical where normal accuracy is acceptable, but diagonal, or even close to horizontal where higher accuracy is required. This provides a significant improvement in resolution.

An example of such an arrangement is shown in Figure 6. This probe has a first section at an angle α to the vertical. Compared to a vertical section the depth resolution of this section is increased by a factor cos (α) . A second section is at a smaller angle β to the vertical and a third section is again at a larger angle γ . The configuration of this probe provides greater resolution at the upper and lower ends of the measurement range .

In general, the relationship between distance along the probe and fluid depth can be tailored as desired by choosing probe slope. In particular, the user often wants an indication of fluid volume rather than depth, and the relationship between them depends on the container profile. It is possible to tailor the slope of a probe to compensate for the container profile, so that the indicated distance along the probe is proportional to fluid volume.

If a TDR system is fitted with an open part of its probe horizontal, it can be used as a level switch. For example if the angle and/or the angle γ is increased to 90° in the configuration of Figure 6, the probe accurately indicates either full or empty, i.e. whether or not the fluid has reached the position where the horizontal portion of the probe is mounted. This gives less information than a fluid level measurement, but gives it with higher confidence, because the switching level is set when the probe is installed, and is independent of any calibration or errors due to ageing or changes m the electronics.

A probe can be roughly vertical over most of its length, but bent to be roughly horizontal over a section, and then roughly vertical again. This allows a combination of level measurement, as normal, with greatly increased sensitivity at a particular level . This approach allows the sensor to indicate with improved confidence whether the level is above or below that section of the probe, like a level switch, with the advantage that, if the position of the horizontal part of the probe is accurately known, there is no need to introduce errors by measuring relative to another feature, e.g. the top or bottom of the probe. Multiple roughly horizontal sections m the same probe can then give level measurement combined with multiple level switching.

In summary, a time domain reflectometry probe for fluid level measurement has a first portion which is vertical and a second portion which is horizontal. The horizontal portion has greater resolution than the vertical portion to give increased accuracy over a region of the measurement range. The first and second portions can also be at acute angles to the vertical and horizontal respectively. It will be appreciated that although distinct examples of probes according to the invention have been described, the individual features of these examples may be combined m any suitable manner to provide embodiments of the invention which are not specifically described herein.

Claims

1. A time domain reflectometry probe for a fluid level sensor comprising two elongate spaced conductors arranged to function as a transmission line, wherein the probe is configured such that, in the position of use, a first portion of the probe is at a first angle to the vertical direction and a second portion of the probe is at a second angle to the vertical direction, and the first angle is smaller than the second angle.

2. A probe as claimed in claim 1, wherein the first angle is about 0°, such that the first portion is substantially vertical.

3. A probe as claimed in claim 1 or 2 , wherein the second angle is about 90°, such that the second portion is substantially horizontal.

4. A probe as claimed in claim 1 or 2 , wherein the second angle is in the range 90° to 180°.

5. A probe as claimed in any preceding claim wherein the first and/or the second portion is shaped to reduce the horizontal extent of the portion.

6. A probe as claimed in any preceding claim, wherein the probe is configured such that the distance along the probe is proportional to the volume of fluid in a container in which the probe is to be installed, in use.

7. A probe as claimed in any preceding claim, wherein the second portion is an end portion of the probe.

8. A probe as claimed in any preceding claim, wherein the probe comprises an inner conductor and an outer conductor surrounding the inner conductor.

9. A probe as claimed in any preceding claim, wherein the transmission line is terminated in a short circuit .

10. A probe as claimed in any preceding claim, wherein the probe comprises an inner conductor arranged, in use, to be positioned on one side against a surface of a container, and an outer conductor arranged to surround the inner conductor on the remaining sides.

11. A time domain reflectometry probe comprising an inner conductor and an outer conductor, wherein the conductors are configured such that, in use, the inner conductor is positioned on one side against a surface of a container and the outer conductor surrounds the inner conductor on the remaining sides thereof.

12. A probe as claimed in any preceding claim comprising a first section which is open such that, in use, a fluid is able to enter a space between the conductors and affect the local impedance of the transmission line, and a second section into which a fluid is unable to pass, so that the impedance of this section is unaffected, in use, by changes in fluid level .

13. A time domain reflectometry probe comprising two conductors forming a transmission line and having a first section which is open such that, in use, a fluid is able to enter a space between the conductors and affect the local impedance of the transmission line, and a second section into which a fluid is unable to pass, so that the impedance of this section is unaffected, in use, by changes in fluid level.