GB2315866A

GB2315866A - Position detection

Info

Publication number: GB2315866A
Application number: GB9616151A
Authority: GB
Inventors: Christopher John Rainey; Richard William Fling
Original assignee: RADIODETECTION Ltd; Radiodetection Ltd
Current assignee: RADIODETECTION Ltd; Radiodetection Ltd
Priority date: 1996-08-01
Filing date: 1996-08-01
Publication date: 1998-02-11
Anticipated expiration: 2016-08-01
Also published as: GB9616151D0; GB2315866B

Abstract

A method of detecting the change in position of an underground object 2 comprises the steps of measuring the displacement of the object in its path of movement, measuring the angle the object makes with the horizontal or vertical, and calculating a change in position of the object from these measurements. The angle which the object makes with the horizontal or vertical may be detected by a tilt sensor (figs. 2-5) which uses a ferromagnetic fluid and which signals the tilt of the object to an outstation 8 via a modulated magnetic field. The outstation also determines any deviation in the heading of the object on the basis of the magnetic field generated by the object. The displacement of the object along its path is calculated by counting the number of rods 6 added behind the object, using magnetic sensors (figs. 6-9) to detect the passage of rod joints.

Description

POSITION DETECTION The present invention relates to a method for detecting the change in position of inaccessible objects, and apparatus useful in the method. Particularly, but not exclusively, it relates to detecting the change in position of underground boring tools, especially during horizontal boring.

When detecting the position of underground objects it is known to arrange for a magnetic field to be generated by the object or by a field source attached thereto. In the case of underground conductors such as cables or pipes, an alternating current can be applied to the conductor to induce a cylindrical magnetic field with a detectable field strength at ground level. By measuring the variation of the field strength at the surface, the path of the underground conductor can be determined.

The detecting of the position of objects such as sub-surface boring tools cannot normally be performed in this way because a cylindrical field cannot be generated.

There is also the additional complication that such objects necessarily change position. In a known method to detect the position of boring tools a magnetic field source is mounted on the boring tool and the field from that field source is detected. This field source can be a solenoid. When alternating current flows through the solenoid a bipolar magnetic field is generated which can be located at the surface by a person with a hand-held detector. The vertical component of the field at the surface will change direction when the field detector is directly above the solenoid. Therefore by noting the position in which that component of the field reverses, the position of the solenoid in a horizontal plane can be determined. If this is done continuously, the movement of the boring tool on which the solenoid is mounted can be tracked. The depth of the solenoid can also be gauged by measuring the attenuation of the field at the surface.

Of course, this requires the field strength at the solenoid to be known.

The method described above for detecting the position of a solenoid is a method of trial-and-error.

The field detector must be manoeuvred into the correct position, namely directly above the solenoid, for the solenoid to be accurately located. When the field detector is anywhere other than directly above the solenoid, the location of the solenoid is not known.

This method is often slow to perform, wasteful of manpower and can be impractical in many applications, for example if the solenoid is underneath a road or a waterway.

The present invention proposes a method of detecting the change in position of an underground object from a remote location, eliminating the need for detecting a field from directly above the object. Where the object is, for example, a sub-surface boring tool, its actual position in relation to a known starting point can be determined and it can then be controlled accordingly.

A first aspect of the present invention is concerned with determining the change in position of the object.

In this aspect, rather than measuring the horizontal distance between two points in the movement of the object, that distance can be calculated using the displacement of the object in its path of movement, and a measurement of the angle the object makes with the horizontal (or vertical). In a similar way, the depth of the object along its path of movement can be determined from the measured distance and angle.

In a first aspect, the present invention provides a method for detecting a change in position of an object, which object has a tilt sensor for measuring the angle of an axis of the object with respect to horizontal, the method comprising moving the object from a first position to a second position along its said axis, determining the distance moved by the object between the first and second positions, on the basis of the output of the tilt sensor determining the angle with respect to horizontal of a line joining the first and second positions, and calculating a change in the position of the object on the basis of the determined angle and distance.

If the first position is known, then the detected change in position can be used to calculate the actual second position. If the object is moved to third and subsequent positions, each change in position can be detected in the above manner and the current position of the object in relation to any previous position can be determined by accumulating the detected changes in position.

In practice, the change in position can be calculated by performing a finite integration of the change in position in the vertical and horizontal directions. This is a matter of simple trigonometry, represented by the following relationships: A vertical position = distance moved x sine 6 A horizontal position = distance moved x cosine e where e is the angle with respect to horizontal of the line joining the first and second positions.

In the case of boring machines, where rods are fed in from the surface behind the boring tool as it progresses through the ground, the distance moved by the boring tool can be determined by measuring the length of rod passing through the boring machine. Conveniently, since the rods are of known length, the distance can be determined simply by counting the number of rods and multiplying by the rod length. Where such a system having rods is used, the integration step for calculating the change in position of the drill head can conveniently be selected as a single rod length.

The angle with respect to horizontal of the line joining the first and second positions can be determined by assuming that the tilt angle of the object does not change between the two positions, in which case the output from the tilt sensor can be read at the first or second position only, or at any single position therebetween. More preferably, however, the angle of this line is determined on the basis of an average tilt angle of the object as it moves between the first and second positions. That is to say, the output of the tilt sensor is read and recorded at two or more points at or between the first and second positions and an average tilt angle is calculated on the basis of these readings. It is particularly preferred that readings from the tilt sensor are taken at least once per second when the object is moving.

The above method allows the detection of the-actual position of the object if it remains within a known vertical plane. However, in practice, the object may deviate left or right of the desired direction of travel and therefore out of the intended vertical plane. Such deviations may be brought about, for example, by changes in soil conditions. Therefore, in preferred embodiments of the first aspect of the present invention, deviations in the heading of the object are also detected so that the object may be controlled accordingly. For example, a magnetic field source such as a solenoid can be mounted on the object and changes in the magnitude and geometry of the magnetic field produced by this solenoid at a remote detection position can be used to detect a deviation of the object.

The output from the tilt sensor can be transmitted to a remote detection point using any suitable means.

Conveniently, in embodiments of the invention where the object has a solenoid mounted on it, the magnetic field produced by the solenoid can be modulated in a known manner to provide for transmission of the output from the tilt sensor.

Thus, a preferred practical embodiment of a system for performing the method of the present invention includes a boring tool having a tilt sensor mounted on it and also a solenoid for producing a magnetic field which can be used for electromagnetic communications as well as detecting changes in the heading of the drill head. A detector is used to extract the information from the field of the solenoid concerning the tilt angle of the head and any left/right deviation from the desired vertical plane, and the position of the drill head is detected on the basis of its calculated change in position for each rod fed into the ground. On the basis of this information, the drill head can be controlled using conventional control methods.

Considering the tilt sensor in more detail, since the axis of which the inclination is being measured can also be the axis of rotation of the boring tool, the output of the tilt sensor is preferably arranged to be insensitive to rotation about the axis of tilt. Although there are known sensors to perform this function, it has been found that their accuracy and stability over the typical operating temperature range of a boring tool can give rise to unsatisfactory results in a system operating in accordance with the method of the invention. A further aim of the present invention is therefore to provide a tilt sensor with high accuracy and stability which is suitable for use in the above method, as well as for other applications.

A second aspect of the invention is therefore concerned with the structure of the tilt sensor, and makes use of the fact that a fluid will tend to flow to the lowest part of a chamber partially filled with the fluid. The second aspect of the invention makes use of ferromagnetic fluids to form part of a magnetic circuit so that the alignment of that circuit is determined by the location of the fluid within the sensor.

Accordingly, in a second aspect, the present invention provides a tilt sensor comprising a pair of windings on respective ferromagnetic cores, the cores being spaced from one another with an air gap therebetween, wherein the air gap is part filled with a ferromagnetic fluid to form a magnetic circuit comprising the cores, air gap and ferromagnetic fluid.

The term "ferromagnetic fluid" as used in this specification is intended to denote a fluid having magnetic characteristics and a permeability substantially greater than 1. The ferromagnetic fluid can, for example, be a colloidal suspension of ultra-fine magnetic particles in a carrier liquid, to give it both liquid and magnetic properties.

If the sensor is rotated about its tilt axis, the magnetic circuit formed by the cores, ferromagnetic fluid and air gap, remains constant. However, when the sensor is tilted in that axis, the magnetic coupling across the ferromagnetic fluid/air gap alters as the fluid moves under gravity. That is, the coupling between the windings increases as more fluid fills the volume between pole pieces of the cores and decreases when there is less fluid.

Preferably, the ferromagnetic cores are cylindrical cores arranged coaxially and spaced apart in their axial direction with the windings thereon to form a gapped transformer. The two cores may be separated by a spacer, so that the gap between them is a cylindrical chamber formed by the faces of the windings and the spacer.

Preferably the complete arrangement is assembled into a sleeve, the material of which is not important, although metal may be used to provide EMC screening.

In a preferred embodiment of the sensor of the present invention, the air gap between the two cores is 50% filled with the ferromagnetic fluid. This arrangement can be used to provide a sensor with a response which is symmetrical to tilt about its axis and approximately linear about zero tilt. This is a direct volumetric relationship set by the dimensions of the cylindrical cavity containing the ferromagnetic fluid.

It is possible, however, to fill the air gap to a greater or lesser extent. In this way the sensor can be biased towards tilt in a particular direction.

In practice, the windings can be connected in series, either in phase or anti-phase, so as to form a balanced autotransformer. The sensor is excited by an alternating voltage and the output, which is taken at the point where the two coils are electrically connected, is preferably electrically processed by performing an AC ratiometric measurement between the input and output.

The sensor output can be characterised to determine the function relating the output and the tilt angle.

This characterised output can, for example, be stored in a numerical look-up table.

When the air gap of the sensor is 50% filled with the ferromagnetic fluid, it has been found that in practice a cubic polynomial fit provides a very close approximation to the function relating output and fill angle. When the sensor is balanced, corresponding to a horizontal orientation, the output is set at mid scale.

As the sensor is tilted the output increases or falls as the fluid couples one or the other of the axial poles of the cores.

Preferred practical embodiments of the sensor have a resolution of 0.1 or higher and an accuracy of 0.2 at least in a range minus 10 to plus 10 with respect to the horizontal. The accuracy outside this range is generally less critical, but it is still preferred that the accuracy is within 10 up to an angle of 450 either side of the horizontal.

As discussed above, where the method of the present invention is applied to a boring machine, the distance travelled by the boring tool can be determined by counting the number of rods fed through the boring machine. It is of course possible to count the number of rods added or removed manually, for instance by using a manually operated counter. However, this method of operation may be inappropriate in many applications and is quite likely to be unacceptable in practice.

Therefore, it is another aim of the present invention to provide an apparatus for automatically counting the number of rods added or removed from the ground. This is done in a third aspect of the present invention by recognising that the joint (typically in the form of a screw joint) between two metal rods, particularly if ferromagnetic, will alter a magnetic coupling between an a.c. excited coil, a further coil and a series of such jointed rods passing through the coils.

In a third aspect, the present invention therefore provides a rod counting method comprising (i) passing two or more jointed metal rods through a coil set comprising an A.C. excited coil and a detection coil magnetically coupled to the excited coil, a magnetic circuit being formed by the coils and the rods passing through them; (ii) detecting an output from the detection coil; and (iii) increasing or decreasing a stored value corresponding to a number of rods when it is determined from the output of the detection coil that a rod joint has passed through the coil set.

The question of whether the stored value is increased or decreased is dependent on the direction the rods are travelling. This can be determined by the operator of a system and the system set up accordingly.

Preferably, however, the coil set further comprises a second detection coil, the two detection coils being arranged symmetrically about the A.C. excited coil, preferably coaxially in the direction of rod movement.

With this arrangement, by detecting the order in which a joint passes through the two detection coils, the direction of travel of the rods can be determined, making it possible to automatically determine whether a rod has been added or removed. In addition to enabling the direction of travel to be determined, this arrangement is also beneficial to the sensitivity of the detection of the presence of a rod joint. This is because rather than detecting the absolute value of the output from a single detection coil, it becomes possible to detect a balance between the coil outputs. For instance, the outputs from the two detection coils can be arranged to be in balance when no rod joint is present in the coil set, and the coils moving out of balance can then be used to indicate the presence of a rod joint. This can help to eliminate the influence of external factors such as power surges to the excited coil.

The sensitivity of the rod travel direction detection with this arrangement is increased as the detection coils are moved further apart. However, as these coils are moved apart from one another they are necessarily spaced further from the central, excited coil, and this decreases the sensitivity of the detection of the presence of a rod joint. Therefore, it is even more preferred that two coil sets are used, these coil sets being spaced apart from one another in the direction of rod travel. In this way, each coil set can be optimised for the detection of the presence of a rod joint and the direction of travel can be accurately determined from the order in which the joint is detected to pass through each of the spaced apart coil sets. With this arrangement, each coil set need only comprise a single detection coil in addition to the excited coil.

However, it is preferred that two detection coils are used to reduce the sensitivity of the coil set to external factors as discussed above.

In practice, the various coils in the or each coil set can be wound on a cylindrical former, the bore of which is large enough to allow the rods to pass through it. Where two or more coil sets are used, they are preferably placed approximately 3cm apart along the former.

Preferably, a sensor comprising the or each coil set is arranged for mounting on the front end of a boring machine, such that the rods pass through it when going into (or coming out of) the ground. The output from the sensor can be fed to a processor to determine the order in which the coils or coil sets are triggered, this order being used to determine whether the rod is being added or removed.

The rod count information obtained in this way can be transmitted to a central controller and used in conjunction with a previously stored value of rod length to calculate the distance moved by the boring tool. This central controller can also receive information from the tilt sensor on the boring tool and thus calculate the incremental changes in position of the boring tool as discussed above.

Particularly preferred embodiments of the present invention will now be described, by way of example only, with reference to the accompanying figures, in which: Fig 1 schematically shows a system for carrying out the method of the first aspect of the present invention; Fig 2 is a schematic section of a sensor according to the second aspect of the present invention; Fig 3 is an end view of one of the cores of the sensor of Fig 2; Fig 4 is a circuit diagram for the sensor of Fig 2; Fig 5 is a plot of sensor output against tilt angle for the sensor of Fig 2; Fig 6 schematically shows a boring machine with a rod counter apparatus in accordance with the third aspect of the present invention; Fig. 7 schematically shows the rod counter of Fig 6 in more detail; Fig 8 illustrates the sensor output for the sensor of the rod counter apparatus of Fig 6; and Figs 9 (a) and (b) illustrate the rod counter output for a rod being added and removed respectively.

Fig 1 shows a system which can be used in accordance with the method of the first aspect of the present invention to detect a change in position of a boring tool 2 of a boring machine 4. As the boring tool 2 progresses through the ground, rods 6 are fed in behind the tool 2 by the boring machine 4. This can be carried out in a conventional manner, as can the control of the boring tool 2.

A solenoid for producing a magnetic field and a tilt sensor are mounted to the boring tool 2. The solenoid field is modulated in a known manner with the output from the tilt sensor. This field is detected at a remote location by an outstation 8, which has a demodulator to extract the tilt sensor output from the detected magnetic field and which is also able to determine any deviation in the heading of the boring tool 2 on the basis of the detected field. In this way, the outstation 8 can act as a steering target for the boring tool 2.

The boring machine 4 located at the entrance to the bore includes a rod counter 10 for determining the number of rods 6 added or removed from the bore. The output of the rod counter 10 is communicated to a basestation 12 which, on the basis of the communicated output, updates a stored value corresponding to the number of rods inserted into the ground. The base station 12 also stores a value corresponding to the length of an individual rod which can be used in combination with the output from the rod counter 10 to determine a distance moved by the boring tool 2.

A radio link 14 is provided between the outstation 8 and the basestation 12. Using this radio link 14, the output from the tilt sensor and also information with regard to the heading of the boring tool 2, is transmitted from the outstation 8 to the basestation 12.

A processor in the basestation 12 carries out a finite integration of the movement of the boring tool 2, on the basis of the detected distance moved and the output of the tilt sensor, to calculate the incremental change in the horizontal and vertical position of the boring tool 2. In practice, the integration step is chosen to be one rod length and the angle (6) of the axis along which the boring tool 2 has moved is taken to be the average of the output from the tilt sensor over the one rod length moved. The average tilt value is calculated on the basis of readings taken from the tilt sensor at least once every second. The incremental changes in vertical and horizontal position are calculated using simple trigonometry: A vertical position = rod length x sine 9 A horizontal position = rod length x cosine e The basestation 12 includes a display unit which is used to display the current position of the boring tool 2 in relation to its starting point (ie the point it entered the ground). The display can also be arranged to show information regarding the heading of the boring tool 2 and, for instance, to schematically show the path followed by the boring tool 2. A facility is also provided for entering topographic information concerning the ground through which the boring tool 2 is travelling.

Using this information, the basestation 12 can calculate the actual position and depth of the boring tool 2 in relation to the overlying ground and this information can be displayed.

Conveniently, the basestation 12 is mounted on the boring machine 4 so that an operator can observe the display of the basestation 12 and control the boring machine 4 accordingly in a conventional manner.

To perform its detecting function, the outstation 8 includes an aerial array formed by four aerials placed in different orientations and positions to create a mutually orthogonal aerial triplet with two axially displaced horizontal aerials. These horizontal aerials are arranged to be either in-line with an aerial axis of the solenoid on the boring tool or at 900 thereto. The other two aerials comprise a vertical aerial and a horizontal transverse aerial. Any deviation in the heading of the boring tool 2 can be determined, operating the outstation in a "remote steering mode", from the phase and amplitude of the output of the horizontal transverse and vertical aerials, and thus the necessary information for remotely steering the boring tool 2 can be obtained. In this remote steering mode, the outstation 8 is mounted in a stand at a fixed location, preferably along the intended line of the bore.

In some instances, it may be necessary to check the calculated position of the boring tool 2, for instance relative to buried services or a geographical feature.

The outstation 8 therefore has a second mode of operation, a "walkover mode", in which the output from the aerial array is used to provide guidance information for the operator to move the outstation 8 (eg. by carrying it) into a position directly over the boring tool 2. When that position is reached, the depth of the boring tool 2 below the ground can be determined on the basis of a detected field strength of the field generated by the solenoid.

In more detail, a signal processing system is used which enables the software to sample the following combinations of aerials: In the Remote Steering mode: Horizontal transverse and Vertical In the Walkover mode: Left/Right (far range): Horizontal transverse, Horizontal in-line SUM and Vertical Close to Sonde: Horizontal in-line SUM and DIFFERENCE, Vertical Depth: Horizontal in-line SUM, Horizontal Transverse, Vertical In both the remote steering and walkover over modes a signed output is used by the outstation to provide an indication of LEFT, ON-TRACK or RIGHT for the heading of the boring tool.

On this basis of this information, the basestation, to which the outstation transmits the steering information, displays a graduated indication of Left and Right in 4 levels for ease of steering control when the outstation is in remote steering mode. This steering information enables the operator to steer along a track, or "dead band", indications being displayed when the boring tool deviates either side of this. It is characteristic of the function that the width of the track increases with operating depth.

A calculation using the phase and amplitude of the vertical and transverse horizontal coils is made to determine the steering value This can be used by either the outstation or basestation to indicate the required steering correction, and is based on the following algorithm: If (square root (magnitude [vertical coil]/magnitude [transverse horizontal coil]) < threshold) and (Coils in phase) = > indicate direction 1 If (square root (magnitude [vertical coil]/magnitude [transverse horizontal coil])c threshold) and (Coils not in phase) = > indicate direction 2.

Where the "threshold^ is a preset value.

As can be seen, to determine which direction the correction to heading needs to be, the relative phase of the vertical and transverse horizontal coils are compared. If the coils are in phase "direction 1" is indicated, otherwise "direction 2" is indicated. To determine whether the coils are in phase the following algorithm is used: If (phase 2 > = (phase 1 - 90 degrees) and (phase 2 < (phase 1 + 90 degrees)) = > in phase.

The actual directions left and right are assigned to "direction 1" and "direction 2" depending on the phase conventions of the hardware employed.

It has been found that in practice this approach can give a "dead band" or steering corridor width of less than lm for bores up to 10m in depth, narrowing for shallower bores, and proportionally wide for deeper bores. It is also noted that a variable threshold could be applied as a correction for depth or for any other reason desired by the user.

When close to the position of the boring tool in walkover mode, an alternative method of determining Left/ Right deviation from the path of the boring tool is used.

This is because it is a characteristic of solenoid fields that the function sensitivity decreases asymptotically as the centre line of the field is approached, and steering information therefore deteriorates using the above approach. In practice this switch of method is made when the amplitude of the vertical aerial output falls below that of the horizontal in-line aerial.

With this alternative method the hardware is set so that the phase difference of the SUM in-line horizontal and DIFFERENCE in-line horizontal coils is used to determine Left/Right deviations. If the SUM and DIFFERENCE coil pairs are in phase "direction 1" is indicated, otherwise "direction 2" is indicated. LEFT and RIGHT are assigned to "direction 1" and "direction 2" depending on the conventions of the phase conventions of hardware used.

This is combined with a measurement of the amplitudes of the coil pairs, to determine whether the operator is on-track. Relative amplitudes of the SUM and DIFFERENCE coil pairs are compared, and If: Amplitude (DIFFERENCE coils) < Y x Amplitude (SUM coils) then the operator being guide towards the location of the boring tool is taken as being on-track. Here "Y" is selected to give the desired directional performance. A value of Y = 0.04 has been found to give adequate performance.

In the walk-over mode, as well as directing the operator left or right, once they near the line of the bore it is also necessary to direct them forwards or backwards along the line of the bore to the current position of the boring tool. This is done by comparing the relative phase of the vertical and horizontal in-line coils. If the coils are in phase "direction 3" is indicated, otherwise direction 4 is indicated.

As with the Left/Right function, FORWARDS and BACKWARDS are assigned to "direction 3" and "direction 4" depending on the phase conventions of the hardware employed.

To determine the alignment of the outstation aerial array to the axis of the solenoid on the boring tool the phase and ratio of the transverse and in-line horizontals are used.

The angle of rotation from a position of axial alignment is given by: Angle = Arctan ([magnitude of transverse coil]/[magnitude of in-line coil]) and the rotation direction (ie. clockwise or anti clockwise) is determined on the basis of the relative phase of the coils.

The outstation is arranged automatically to display a calculated depth of the boring tool when directly over it in walkover mode. This display is triggered automatically by the following condi equation relating to the absolute solenoid field strength (ie. vector sum of 3 dimensions) to determine the distance to the solenoid from the outstation. This equation includes terms for the unregulated solenoid output strength and hardware calibration. (Determined as part of the user calibration procedure.) The mathematical model of a solenoid predicts that the field strength vs. depth function follows a (one over depth cubed) law at distances greater than approximately 1 metre ie.

Field Strength = Where K is a constant dependent on the solenoid. Hence to calculate depth from field strength measurement it is necessary to establish the value of k from a calibration thus: - k = (calibration field strength) * (calibration distance) A3 The depth is then given by: Depth = cubed root(k/measured field strength) The calibration distance is preferably between 2 and 5 metres.

The solenoid output signal is unregulated and this will cause the calibration field to alter as the batteries powering the solenoid are depleted. A calibration procedure is carried out by the user to determine the solenoid signal strength at a known distance from the outstation and at a known battery level. The solenoid battery data can be subsequently used to compensate the depth calibration as the battery is depleted.

Referring now to Figs 2 - 5, a tilt sensor in accordance with the second aspect of the invention, which can be used in the above system, will be described. The sensor includes a pair of ferromagnetic cores 20 with central pole pieces 22. Two identical windings 24 are fitted on respective pole pieces 22 with the winding tails 26 exiting at the ends of the cores 20. The cores 20 face each other and are arranged coaxially and spaced apart from one another. A spacer 28 is provided between the cores 20 to hold them in this spaced apart relationship.

The coils 24 are held in an encapsulant 30 which fills the annular recess between the core body and the central pole piece 22, and the cores 20 and spacer 28 are assembled and held within a sleeve (not shown). With this construction, a cylindrical gap between the two cores 20 is defined by the inner surface of the sleeve 28 and the faces 32 of the coils/encapsulant. This cylindrical gap or chamber is 50% filled with a ferromagnetic fluid 34. A suitable ferromagnetic fluid is a proprietary fluid from Ferrofluidics Corporation, reference EMG905 as available, for instance, from Advanced Products and Technologies in the UK.

The operation of this sensor is as follows. The coils 20 are connected in series, in phase or antiphase, so as to form a balanced autotransformer. The sensor is excited by an alternating voltage (VAC) and an output (Vout) is electrically processed by performing an AC ratio metric measurement between the input and output.

This may be carried out, for example, using an ANALOG DEVICES AD2S93, which is a monolithic tracking LVDT to Digital Converter.

If the sensor is rotated about its tilt axis T, corresponding to rotation of a boring tool, a magnetic circuit formed by the cores 20, ferromagnetic fluid 34 and air gap is not changed. Whereas, if the axis T is tilted, as indicated by arrows A, corresponding to tilt of a boring tool, the ferrofluid magnetic coupling across the fluid/air gap alters as the fluid moves under gravity. As the sensor is tilted away from the horizontal, the coupling between the windings 24 increases as more fluid 34 fills the volume between the pole pieces 22 of the cores 20.

The sensor output is characterised to determine the function relating Vout and the tilt angle (in degrees).

In a practical embodiment of the sensor, it has been found that a cubic polynomial fit provides a very close approximation to this function (see Fig 5). Using this function, an output from the sensor can be translated into a tilt angle with respect to the horizontal. Thus, with the sensor mounted on a boring tool 2 with its tilt axis T coincident with the axis of the boring tool 2, the angle of the tool with respect to the horizontal can be detected.

Turning now to Figs 6 - 9, an embodiment of a rod counter according to the third aspect of the invention, which can be used as the rod counter 10 in the system of Fig 1, will be described. The rod counter 40, which is mounted to the front end of a boring machine 4 includes a sensor 42 and rod counter circuits 44 arranged to trigger outputs 46 when a rod joint 6a between two rods 6 is detected to pass through the sensor 42.

The sensor 42 comprises two coil sets 48a, 48b axially spaced from one another by about 3cm. The coil sets 48a, 48b are wound on a cylindrical former (not shown) having a bore large enough for the rods 6 to pass through, in the direction indicated by arrow B (Fig 7).

The rod counter operates by using an electromagnetic detection means, in which the magnetic anomaly of a joint in an otherwise continuous metallic rod is detected by a set of three coils, and the sequence of response created by the same joint passing through a second similar set of coils is used to determine the direction of traverse.

Fig. 7 shows the arrangement in principle. The two sets 48a, 48b of coils 50a, 50b, 50c and 52a, 52b, 52c are fixed in coaxial relationship with each other, with the rod 6 passing through them. The centre coil 50b, 52b of each set is A.C. energized from a suitable source, and the voltage developed in the outer coils 50a, 52a, 50c, 52c by electromagnetic coupling is sensed. When there is no rod joint 6a within the coil sets, the coil outputs will be in balance. As the joint 6a enters the first coil set 48a, its magnetic anomaly will affect the coupling from the energized coil 50b firstly to coil S0a and secondly to coil 50c (Fig. 8). The same effect will be repeated with passage through the second coil set 48b; a joint passing through in the opposite direction will of course reverse the sequence of output signals. From the signal sequence it is therefore possible to determine the presence and direction of passage of a rod joint, and give an 'add' or 'subtract' output signal accordingly.

While the arrangement is shown with six coils to illustrate the principle, it is in fact possible to combine the adjacent coils 50c, 52a of the two coil sets so that only five coils are needed.

The output from the coil set can be processed by the counter circuit 44 to produce the output 46 of the rod counter 40. The output may take the form of two quadrature signals (output 1 and output 2) respectively corresponding to a joint passing through the first coil set 48a and the second coil set 48b. As indicated in Fig 9, the order of the triggering of these two outputs 46 can be used to determine the direction the rods 6 are passing through the sensor 42, and hence whether a rod has been added or removed.

Alternative schemes for determining the direction of movement of the rods can be used, for instance schemes based on "state machine operation" could be used to process the outputs from the coils. That is to say, a predetermined detection sequence of the outputs from two sets of coils can be used to determine the presence and direction of travel of the rod joint passing through the counter. Typically this would be done by digitising the analog outputs from the coil sets using an ADC, and then using a microprocessor to implement the determination sequence.

The various aspects of the present invention have been described above by way of example only with reference to specific exemplary embodiments. Variations and modifications from that which has been specifically described and illustrated are possible within the scope of the invention, as will be apparent to persons skilled in the art.

Claims

CLAIMS:

1. A method of detecting a change in position of an underground object, comprising the steps of measuring the displacement of the object in its path of movement, measuring the angle the object makes with the horizontal or vertical and calculating a change in position of the object from these measurements.

2. A method of detecting a change in position of an underground object, which object has a tilt sensor for measuring the angle of an axis of the object with respect to horizontal, the method comprising moving the object from a first position to a second position along its said axis, determining the distance moved by the object between the first and second positions, on the basis of the output of the tilt sensor determining the angle with respect to horizontal of a line joining the first and second positions, and calculating a change in the position of the object on the basis of the determined angle and distance.

3. A method according to claim 2, wherein the object is a boring tool, boring rods being fed in from the surface behind the boring tool as it progresses through the ground, and the distance moved by the boring tool between said first and second positions is determined by measuring the length of rod fed.

4. A method according to claim 3, wherein the rods -are of known length, and the distance moved by the boring tool is determined by counting the number of rods fed and multiplying this number by the known rod length.

5. A method according to any claim 3 or claim 4, wherein the boring rods are of metal and are passed through a coil set comprising an A.C. excited coil and a detection coil magnetically coupled to the excited coil, a magnetic circuit being formed by the coils and the rod passing through them, the output from the detection coil is detected, the passing of a rod joint through the coil set is determined from the output of the detection coil, and a stored value is increased or decreased corresponding to the number of rods.

6. A method according to any one of claims 2 to 5, wherein the angle of the line joining said first and second positions is determined on the basis of an average tilt angle of the object as it moves between the first and second positions, the output of the tilt sensor being read and recorded at two or more points at or between the first and second positions to calculate the average angle.

7. A method according to any one of claims 2 to 6, wherein the tilt sensor output is transmitted to a remote detection point.

8. A method according to claim 7, wherein the object has a solenoid mounted on it, and the magnetic field produced by the solenoid is modulated to transmit the output from the tilt sensor.

9.- A method according to any one of claims 2 to 8, wherein the object is subsequently moved to one or more further positions, the change in position is detected between each pair of adjacent points, and the position of the object is calculated by accumulating the detected changes in position.

10. A method according to any one of the preceding claims wherein deviations in the heading of the object from a known vertical plane are detected.

11. A rod counting method comprising (i) passing two or more jointed metal rods through a coil set comprising an A.C. excited coil and a detection coil magnetically coupled to the excited coil, a magnetic circuit being formed by the coils and the rods passing through them; (ii) detecting an output from the detection coil; (iii) determining from the output of the detection coil that a rod joint has passed through the coil set and (iv) increasing or decreasing a stored value corresponding to a number of rods upon such determination.

12. A method according to claim 11, wherein the coil set further comprises a second detection coil, the two detection coils being arranged symmetrically about the AC excited coil, and the order in which a joint passes through the two detection coils is detected to determine the direction of travel of the rods.

13. A method according to claim 11, wherein two coil sets are used, the coil sets being spaced apart from one another in the direction of rod travel, and the direction of rod travel is determined from the order in which the joint is detected to pass through each of the spaced apart coil sets.

14. A method according to claim 13, wherein each coil set comprises at least two detection coils.

15. A rod counting apparatus comprising one or more coil sets through which a plurality of jointed rods can pass, the or each coil set having an A.C. excited coil and a detection coil magnetically coupled to the excited coil, and control means for detecting an output from the or each detection coil, determining from the output of the detected coil that a rod joint has passed through one or more coil set and increasing or decreasing a stored value corresponding to a number of rods or such determination.

16. A rod counting apparatus according to claim 15, wherein the or each coil set further comprises a second detection coil, the two detection coils being arranged symmetrically about the A.C. excited coil.

17. A rod counting apparatus according to-claim 15 or claim 15, further comprising means for mounting the apparatus on the front end of a boring machine, such that boring rods fed by the machine pass through the or each coil set when going into or coming out of the ground.