LOCATOR

BACKGROUND OF THE INVENTION

FIELD OF THE INVENTION

The present invention relates to a locator for

locating a conductive object. It is particularly, but not

exclusively, concerned with a locator for locating an

underground conductor such as a buried cable or pipe.

SUMMARY OF THE PRIOR ART

The proliferation of networks of buried cables and

pipes for many different utilities (electricity, gas, telecommunications, etc) has meant that any excavation of the ground is likely to be in the vicinity of a buried

cable or pipe, and such excavation involves a risk of

damage or interference to the buried cable or pipe,

unless the location of that buried cable or pipe is

precisely known.

In particular, the growth in the use of fibre-optic

communication systems for telephones, cable television,

etc has significantly increased the problems associated

with excavation. Such fibre-optic communication systems

have a much higher communication capacity than metallic conductors, but the costs consequent on damage or

interference to the fibre-optic communication system are

significant. Moreover, if damage occurs, it is more difficult to repair a fibre-optic connection than it is

to repair a metallic connection. For this reason, owners

and/or operators of fibre-optic communication systems

normally require that, before any excavation can occur in the vicinity thereof, the location of the fibre-optic

connection should be determined precisely, both by suitable location system and by visual inspection. In

practice, this means that an initial excavation needs to

be made to permit the official inspection of the cable, before any more extensive excavation can be carried out in the vicinity. Moreover, each preliminary location and excavation to enable the fibre-optic connection to be

inspected must be repeated along the length of the fibre- optic connection, and this requires a significant amount of time and effort.

One type of conventional locator detects alternating fields from signal currents in a conductor, by means of a

suitable antenna assembly incorporated in a hand-held

receiver. Such an arrangement is applicable to fibre-

optic connections because such connections normally have

a metal sheath for protection purposes, and a signal can

be applied to that metal sheath and detected.

In such conventional systems, a user carries the receiver and repeatedly makes measurements adjacent the

target conductor until the receiver indicates that the

conductor is present. Then, in order to obtain the

visual inspection referred to previously, an excavation

is made at the site determined by the locator, until the

pipe, cable or fibre-optic connection is exposed.

SUMMARY OF THE INVENTION

In a first aspect of the present invention, a locator is provided with at least two antennae of known separation, each antenna being able to measure electro¬

magnetic field components in the direction of the

separation of the two antennas, and perpendicular to that

direction to the object. Suitable processing means is then provided to derive co-ordinate information defining

the direction and separation, of the locator relative to

the conductor which generates the electro-magnetic

fields. In the following specification, the direction corresponding to the separation of the antennas will be referred to as the X direction, and the perpendicular

direction which intersects with the object will be

referred to as the Y direction. Thus, the present invention derives X and Y co-ordinate information of the

object relative to the locator.

Preferably, the locator is in the form of a ground

penetration probe. Then, as the probe penetrates the ground, the antennae detect electromagnetic signals from

a conductor of an underground object to be located, and

can determine the position of the probe relative to the

underground object. Thus, the probe can be driven into

the ground towards the underground object and the user will be provided with information which indicates the separation of the probe from the object, to enable the

probe to be brought into close proximity to the

underground object without the risk of the probe damaging the underground object due to forceful impact.

In order to determine the separation of the locator from the object, at least two antennas must be provided

at spaced-apart locations of the locator. The difference

in electromagnetic fields detected by the two antennas then enable the distance from the locator to the object

to the calculated, to generate suitable information to

the operator. Preferably, three or more antennas are

provided, as this then permits more accurate measurements

to be made as the locator approaches the underground

object by choice of sensing antennas utilized. This is

particularly useful when the locator is a ground penetration probe, so the accuracy of location is

improved as the probe approaches close proximity to the

underground object.

At least in theory, the position of the object

relative to the locator can be determined by assessment

of the electromagnetic fields at the two antennae, and by

calculations using simple trigonometry. However, in

practice, it is likely that there will be asymmetries in the electromagnetic field generated by the conductor, for example because of the presence of other adjacent

conductors, and therefore it is preferable that the

processing means is provided with suitable compensation for such errors. Moreover, although it is possible to have antennae with coils of common centres, it is often

more practical to have coils with displaced centers, in

which case suitable compensation must be provided for

this as well.

Although it is possible for each antenna to have

two-axis coils, it is preferable that three-axis coils

are provided since this enables further information to be

derived which enables the inclination of the direction of extension of the object relative to the Y direction to be

determined.

In a further development, the locator is provided

with tilt sensing means which enables the inclination of

the locator, and hence the X direction, to be determined

relative to the vertical. The processing means then

makes use of this tilt information, and it is then

possible to derive a determination of the location of the

object, in terms of its vertical and horizontal

separation from the locator, independent of the orientation of the locator. This is important since it is not easy for any operator to ensure that the locator is held absolutely vertically.

This aspect of the present invention is applicable

to the location of any conductive object from which an

electromagnetic signal can be transmitted. As has

previously been mentioned, the present invention is primarily concerned with the location of a buried fibre-

optic connection, by applying signals to a metal sheath

of such a connection and detecting the electromagnetic

fields generated therefrom, but the present invention is not limited to this field of application.

As has been mentioned above, a locator with three or

more antennae is preferred as it enables more accurate

measurements to be made as the locator approaches the object. Since the position of the object relative to the

locator can be determined by any pair of antennae, it is

possible for the choice of antenna pair to be changed in

dependence on the separation of the locator and the

object. This switching between antenna pairs, when

carried out automatically, represents a second

independent aspect of the present invention.

Also as mentioned above, the locator of the present invention may have three-axis antennae which enable the relative position of the object located to be determined,

but also the angle of inclination of the direction of

elongation of the object relative to the Y direction. As

a result, it then becomes possible to predict the

magnetic field which will be generated by the object at a

position close to, but separated from, the current

position of the locator. If the locator is then moved

to that position, and the field measured, the

correspondence between the measured field and the

predicted field gives a measure of the confidence level

of the measurement. This operation may be carried out by

moving the locator to a predetermined lateral

displacement. However, a similar effect can be achieved

by changing the inclination of the locator relative to the vertical. Since that change in inclination can be

measured by a tilt sensor, it is then not necessary to

move the locator by a predetermined amount, because the tilt sensor can then determine any change in inclination.

This simplifies the actions needed by the operator, since

the operator merely needs to change the angle of the

locator relative to the vertical in order to make a

measurement of the degree of confidence in the location

of the object. This way of obtaining a confidence measurement, by moving the locator, therefore represents

a third independent aspect of the present invention.

If the locator is a ground penetration probe, it

will be normal for the probe to be a drilling device which is driven into the ground at the approximate

location of the underground object by the operator.

Using the information from the antennae, the locator can

bring the probe hip into close proximity to the

underground object, since the operator is provided with information relating to the separation therebetween, and

can control the movement of the probe appropriately.

The probe may have an outer sheath from which the

rest of the probe can be removed. Then, the probe

together with, the sheath is inserted into the ground until the buried object is reached, and then the rest of

the probe removed from the sheath to permit access for

inspection or maintenance. For example, an endoscope may

be inserted into the sheath to give visual information

about the underground object. The sheath may be left in

place for subsequent access, or as a marker.

Alternatively, the probe leaves a hollow since this

permits visual inspection of the underground object by endoscope or other inspection means inserted into the

space left by the probe once the probe has been brought

into close proximity with the underground object and removed from the ground.

Where the probe is a drilling device, the

information from the antennae may be used to control the

drilling force. For example, when the probe is a long

way from the underground object, the drilling force can be large so that the probe moves rapidly towards the

underground object. As the probe approaches the

underground object, and to prevent forceful impact, the

drilling force may be reduced so that the drilling force

is minimal as the probe reaches the immediate vicinity of the underground object.

In the aspects of the present invention discussed

above, antennae detect signals from the object.

Normally, signals are applied to a conductor of that object from a separate transmitter. However, if such a

system is applied to the location of underground objects

using a ground penetration probe, there is a risk that

there may be other objects which could be hit by the

probe as the probe penetrates the ground. For non-

metallic objects, this problem can be resolved by providing other sensing means, e.g. radar or

accelerometer, to enable probe movement to be halted on

or just prior to impact with such an object. That other

sensing means may also be used for specific sensing

and/or locating tasks connected with solid objects. If the underground object has a conductor, however, it is possible for the probe to have a transmitter therein

which transmits signals which induce further signals in

the underground object, which further signals can then be

detected. This may be useful, for example, where there

is a dense network of different conductors at the site to be investigated.

Alternatively, the probe may detect and locate the

underground object on the basis of signals already

present on the object, for example 50Hz or 60Hz mains

power or radio signals.

In a further development, the object may have one or

more devices thereon which are able to transmit a .

predetermined signal. Such transmitters are known, in

themselves from e.g. from animal husbandry, in which they

are referred to as RFID systems. A locator is provided

with means for detecting such devices, so further

identification of the object can be achieved.

In the present invention, the relationship in space between the object and the locator is determined. Hence,

with the present invention, it becomes possible to

generate a visual display showing its spatial relationships thereof, rather than e.g. by a audible signal which varies its frequency. Such a generation of

a visual display is therefore a fourth aspect of the

present invention.

In this fourth aspect, the operator of the locator

e.g. a ground penetration probe may make use of the

visual display in order to control the movement of the

locator. Where the locator is a ground penetration probe

e.g. a drill, the user may alter the display and change

the direction of movement of the drill so that the drill

approaches the underground object. The user can

therefore ensure that the drill is at all times targeted

towards the underground object. Since the user is presented with a visual display of the separation (and it

may be possible for that display to change in magnification as the probe approaches the object), it would be possible for the user to stop the movement of

the probe very close to the object. However, it is

preferable, as previously described, that there is automatic control of the speed of movement of the probe

in accordance with the second aspect of the present invention, to reduce the risk of accidental damage to the object.

The visual display of this fourth aspect is

preferably a head-up display which presents the spatial relationship of the probe and the underground object in

the normal line of sight of the user. This has the

advantage that the user may simultaneously view that

display and at the same time see the movement of the

probe.

Indeed, such a head-up display may be applied to

other types of locators and is thus a fifth independent,

aspect of the present invention.

It is often the case that a site to be investigated has a plurality of adjacent objects. For example, the

need for cables of different utilities to follow similar

routes means that it is often necessary for the operator

of a ground penetration probe to be aware of all the

underground objects at the particular site. If all the

underground objects generated the same electromagnetic

signals, the antennae of the locator would simply record

the composite field generated. However, if the different signals are applied to the different underground objects, for example signals of different frequencies, then it is possible for the locator to distinguish between the

different objects by suitable modulation of the signals

received. Hence, by suitable analysis, it is possible to

determine the separation of the locator from each of the

underground objects, and for each underground object to be displayed on the display. Hence, the operator is then

presented with information showing the location of all

the underground objects at the site. This is important,

for example, in ensuring that the ground penetration

probe approaches only the underground object of particular interest, and avoids other underground

objects. Because the operator is presented with a visual

display showing the relevant positions of the objects,

because their positions relative to the locator are

known, it is possible for the operator to bring a ground

penetration probe into close proximity with one underground object, while avoiding contact with other

underground objects at the site.

As an alternative to applying different signals to

the underground objects, which is not always practical,

the underground objects at the site may be distinguished if they each carry active markers as described

previously, with each active marker generating a coded signal which identifies the particular underground object. Since the locator can then detect information which identifies the number of underground objects at the site, the locator can then resolve signals it receives

into different components corresponding to the different objects.

As has previously been mentioned, the ground

penetration probe is driven into the ground towards the

underground object. For very soft ground, this could be

done simply by the user applying force to the ground

penetration probe, but preferably a mechanical drive is provided.

According to a sixth aspect of the present invention, that drive is provided by contra-rotating masses. If two masses are driven at angular velocities,

about a pivot point, there will be a net force on the

pivot point which is determined by the phase between the

two masses, and by the masses themselves. By suitable

arrangement of the masses, and the phase therebetween, it

is possible to arrange for the variation in force to be

such that the downward force has a magnitude which is greater than any upward force, even if the time-average is zero. Then, bearing in mind that the penetration probe must overcome friction with the ground in order to

move, the forces can be such that the downward force is

sufficient to drive the ground penetration probe into the

ground but the upward forces do not overcome friction

sufficiently to drive the ground penetration upwards to the same extent, so that there is a net downward

movement. Similarly, by altering the phase of rotation of the masses, it is possible to re-arrange the system so

that the upward forces exceed the downward forces in

magnitude, so that the ground penetration probe will be

driven out of the ground. Preferably, two such pairs of masses are used to cancel-out lateral forces.

RRTEF DESCRIPTION OF THE DRAWINGS

Embodiments of the present invention will now be

described in detail, by way of example, with reference to

the accompanying drawings, in which:

Fig 1 is a schematic view from the side of the relationship between a ground penetration probe according to the present invention and an underground object;

Fig 2 corresponds to the ground penetration of Fig 1, but viewed from above;

Fig 3 is a schematic view showing measurement of the relationship between the ground penetration probe of Fig 1 and two underground objects;

Fig 4 shows the movement of the ground penetration

of Fig 1 relative to the underground objects;

Fig 5A shows practical embodiment of the ground

penetration probe of Fig 1;

Fig. 5B shows in more detail the handle of the

embodiment;

Fig. 5C shows in more detail the arrangement of

coils in an antenna in the embodiment of Fig. 5A;

Fig 6 shows a signal processing system for use with

the ground penetration probe of Fig 1;

Fig 7 shows a schematic view of a detailed embodiment of a locator embodying the present invention;

Figs 8 to 10 are views of a tool head in the

embodiment of Fig 7, Fig 8 being a transverse sectional

view, Fig 9 being a sectional view in plan, and Fig 10

being a side view, partially in section;

Figs 11a and lib illustrate the movement of masses

in the eccentrics in the tool head of Figs 8 to 10;

Fig. 12 is a graph showing the acceleration characteristics that are then achieved in the tool head of Figs 8 to 10; and

Fig. 13 is a schematic view of the display generated

by the head-up display of the location of Fig. 7;

Fig. 14 shows a portable locator being a further embodiment of the present invention;

Fig. 15 shows in more detail the locator and display

unit of the embodiment of Fig. 14; and

Fig. 16 shows a typical display generated by the

display unit of Fig. 15.

DETAILED DESCRIPTION

As has previously been described, the various

aspects of the present invention discussed above are

based on the detection of the separation of a locator and

an object. The present invention is particularly, but not exclusively concerned with the detection of the

separation of a ground penetration probe and an

underground object. The basic principles underlying such

detection will now be discussed with reference to Figs 1

to 4

Referring first to Fig 1, a ground penetration probe

24 has three antennae 21, 22, 23 thereon. The antennae

detect electromagnetic fields generated at an underground object 26, such as an underground conductor carrying

current. The structure of antennae capable of detecting

such radiation is, in itself, known and will not be discussed further.

One of the antennas 21 is located at, or a known distance from, the tip of the ground penetration probe

24, and the other two detectors 22, 23 are at known

separations SI and S2 respectively along the ground

penetration probe 24 from the first antennas 21. The

ground penetration probe 24 ensures that the antennas 21

to 23 have a known separation and fixed orientation so

that signals detected thereby can be processed.

When the ground penetration probe 24 is moved into

the ground proximate an underground object 26, which

underground object carries an AC signal, electromagnetic fields will be detected by the antennas 21 and 22

enabling position vectors V _x and V ₂ to be calculated, and

permitting the separation of the ground penetration probe

24 from the underground object 26 to be determined e.g.

in terms of X and Y position coordinates. Such

calculation may be carried out by a suitable processor 25

connected to the antennas 21 to 23 via the probe 24.

Thus, if:

B is the total field detected at antenna 21, T is the total field detected at antenna 22, Bh is the horizontal component at antenna 21,

Th is the horizontal component of the field at

antenna 22,

Bv is the vertical component of the field at antenna 22,

then the X and Y position co-ordinates are given by:

Y ^~~ -

B'

Bv_

X ^~~ Y. Bh

Since the X and Y co-ordinates can be calculated the

position of vectors V and V ₂ can be calculated. These equations hold true for the simple case of a single

conductor, and are given by way of example.

Furthermore, if the antennas 21 to 23 are each based

on a three-axis orthogonal aerial system, it is also

possible to obtain a third orientation vector V ₃ as shown in Fig 2, to derive the angle θ being the plan rotation

of the ground penetration probe 24 relative to the

underground conductor 26. Therefore, by knowing the X

and Y coordinates and the angle θ, a user can be presented with information indicating the separation of the ground penetration probe 24, and in particular the

tip thereof, from the underground object 26. The user

may then direct the movement of the ground penetration

probe 24 to reduce that separation, e.g. to bring the tip

of the ground penetration probe 24 immediately adjacent

the underground object 26.

From the above description, it can be seen that only

two antennas 21, 22 are needed in order to determine the

separation of the tip of the ground penetration probe 24

and the underground object 26. However, as the tip of

the ground penetration probe 24 approaches the

underground object, it is possible that the determination

of the separation makes use of the third antennas 23, as

shown in Fig 3. Measurements made on the first and third antennas 21, 23 then offer a more accurate measurement of

the separation of the tip of the probe 24 from the

underground object 26 when that separation is small.

Furthermore, Fig 3 also shows that, as the tip of

the probe 24 approaches the underground object 26, the effect of other generators of electromagnetic fields, e.g. object 27 is reduced since the antennae 23 is used only when the tip of the probe is close to the underground object 26. The change in ratio of antenna

separation distance to underground object distance will

reduce the relative effect of signals from object 27

enabling it to be identified as being other than the

desired underground object 26. Furthermore, the known proximity of the underground object 26 also enables

signal current direction and amplitude to be determined

with greater confidence than if only wider spaced

antennas are used.

As has previously been mentioned, the processor 25 determines the separation of the tip of the probe 24 from

the underground object 26. Therefore, if the angle of

orientation of the probe 24 is changed, this should not

affect separation of the tip of the probe 24 from the

underground object 26. Thus, if the ground penetration probe 24 is pivoted through an angle ø from position A to

B in Fig 4, the measurement of the position of the

underground object 26 should not change. Similarly, if

the probe 24 is moved laterally, e.g. to the position C

shown in Fig 4, then the lateral movement from position A to C corresponds to the change in position of the object 26 relative to the probe 24. In other words, the object

26 should maintain its absolute position.

One possible impediment to the accuracy of derived positional information is tilt of the locator axis from true vertical. This may be countered by incorporation

of a 2-axis tilt sensor of any suitable design giving

electrical data corresponding to angular deviation from

the vertical gravitational axis, from which the true

location of the target object can be computer, by

appropriate compensation of the locator data.

A practical limitation of the locator system as described with reference to Figs 1 and 4 is that it uses

antenna arrays having common centres and vertical axes.

This is difficult to achieve in practice, even with spherical cores common to orthogonal coils; with separate

solenoidal coils and cores, or corresponding magnetic

field sensor arrays of other types, achievement of common

centrality to a high degree of accuracy is impracticable.

It is therefore preferable to incorporate mathematical

compensation for the deviations from centrality of the

coils of the antennae.

Another limitation to the accuracy of positional data from multi-antenna locators is perturbation of the

magnetic field. It is therefore desirable to

incorporate mathematical compensation for field

perturbation of the sensed field, to increase the location accuracy, or establish a confidence level for the data derived.

The mathematical compensations needed to take into

account deviations from the centrality of the close of

the antennae and for field perturbation means that the equations referred to earlier to determine X and Y cannot

be used. Instead, more complex equations are needed, as

will now be described. Referring to Fig. 1, but assuming

that the ground penetration probe 24 is inclined by an

angle φ, then:

It is then possible to define the angle to the

cable from the antenna 21 and angle β to the cable from

the antenna 22. Then the field in the vertical coil of

antenna 22 is:

_j , _ /cos(π-β)

where I is the cable current.

Similarly the field in one of the horizontal coils

of coils of antenna 22 is:

/sin(π-β) _osø

and the field in the other horizontal coil of antenna 22

perpendicular to the first horizontal coil is:

_τ _ /sin(π-β)

.sinθ

Similarly, the field in the vertical coil of antenna 21 is:

_D _ /cos(π-α)

The field in one of the horizontal coils of antenna 21 is:

/sin(π-α) ^

'2

and the field in the other of the horizontal coils of antenna 21 is:

Tsin τ-α) ^

'3

Since the signals measured by the antennae 21,22 relate directly to the fields, those signals can be used

directly in the calculations of cable position.

Thus, the total field T at the antenna 22 is:

and the total field at the antenna 21 is:

V*L is then the solution to a quadratic equation, and hence there are two possible solutions:

The choice of which of solutions -^l) and V ₂ (2) is

correct depends on the sign of the horizontal field at

antenna 22:

Thus,

The above calculations then need to be corrected if

the axes of two horizontal coils do not intersect.

Define the sum H ₂ of the horizontal fields in one

direction at the two antennae 21,22

Ho = B + T _' .

Similarly, define the sum H ₃ of the horizontal fields in the perpendicular direction at the two antennae

21,22

H ₃ = B ₃ + T ₃

Then θ is the arctan of the ratio of H ₂ and H ₃ .

Then the correction to X needed if the axes of the

two coils are displaced by a distance d is:

These differences are then corrected for the probe

tilt to give the vertical distance D from the lower sensor 21 to the cable 26 and the horizontal distance H

from the lower sensor 21 to the cable 26

_D=Λ ^r cosφ+_Ksinφ

H-^sinφ + yicosφ

Figs. 5A to 5C show in more detail a practical

embodiment of the ground penetration probe being an

embodiment of the present invention.

In this embodiment, three antennae 150A, 150B and

150C are contained within the tubular housing 160, so

that the separation between the antennae 150A, 150B and

150C is fixed. The antennae thus correspond to the

antennae 21 to 23 in Figs. 1 to 4.

Each antenna 150A, 150B,150C, nominally horizontal

coils 151 and 152 at about 90° to each other, and a

nominally vertical coil 153. When designed for

incorporation into the tubular housing 160 of small diameter such as a ground penetrating probe, the coils

151,152,153 have inevitably a low ratio of length to

diameter, so that small dimensional variations result in

significant departures from true perpendicularity. In addition, the vertical separation of individual coils in each antenna 150A, 150B, 150C means that the three axes

of measurement are not on common centres, although the

vertical separation of the three antennae from each other

can be controlled quite accurately by the support

structure 154. This is typically made from rigid

plastic, incorporating appropriately positioned slots to

accommodate the coils, as in detail Fig. 5C. The support

structure 154 is firmly located within the outer tubular

housing 160. This tube may have a handle 56 when used as

a portable locator above the ground, or it may be the ground penetrating tube in drilling applications. The

tubular housing 160 provides for electrical interconnections 57 (see Fig. 5B) and can also

accommodate circuit boards and other sensing means as

required, e.g. at 158, and tilt sensor e.g. at 159.

One possible sensor that could be used as a sensing

means 158 in Fig. 5 is a sensor designed to interrogate

markers of the transponder type. Technology of such markers is known as such, and the sensor interrogates

the marker in a way which identifies the marker. The

marker incorporates a transponder tuned to a specific carrier frequency. The sensor 158 in the locator then transmits energy to the transponder which is converted by induction in the transponder, using a tuned pick-up coil,

to power a re-transmitting circuit of the marker. The

output of the re-transmitting circuit is at the carrier

frequency, but is modulated by appropriate means within

the marker to encode data identifying the marker, and

hence the object to which the marker is attached. The modulation is decoded by a receiving circuit of the

sensor 158. Once the marker has been identified, an

appropriate display may be generated as will be described

in more detail later.

It was mentioned above that, if the probe 24 in

Figs. 1 to 4 is moved laterally, or pivoted, the

measurement of the position of the underground object 26

should not change. This enables the degree of confidence

of the location of the underground object 26 to be

measured easily. Referring to Fig. 4, suppose that a

measurement is made of the position of the underground

object 26 at position A, and the ground penetration probe

24 is then moved by a known amount to position C. If the processing means of the locator is aware of the lateral separation of positions A and C, the measurement

at position A should enable a prediction to be made of

the result of the measurement at position C. Since the

location of the object 26 does not change, since the

separation of positions A and C is known, the fields

measured by the antennae 21,22 at position C is

predictable. Therefore, if the probe 24 is moved to

position C, and measurements made, a comparison can be

made between the actual measurements at position C and

the predicted measurements from the measurement at A. If

these coincide, there is a high degree of confidence that

the object 26 has been located accurately. If, however,

there is a substantial divergence between the predicted

measurements at C and the actual measurements at that

position, the accuracy of location of the object 26 is

questionable, so there is then a low degree of

confidence. This procedure requires the lateral separation of positions A and C to be known. The user of

the ground penetration probe 24 must therefore move that

probe 24 by that known amount. This may be inconvenient,

or difficult to achieve practically. However, a similar effect can be achieved by pivoting the probe 24 through the angle φ from position A to B in Fig. 4. Again, assuming that measurements are made in position A, it is possible to predict the results of the measurements made

in position B, assuming that the angle is known. It is

then possible to measure angles by using the tilt sensor

described with reference to Fig. 5, and that measurement

of the angle can be applied to the prediction. Asa a

result, it is not necessary for the user to move the

probe 24 by a known amount, since the angle of tilt

can be measured independently by the tilt sensor. As a

result, the user determines the position of the object 26

at the position A, tilts the probe 24 by any suitable

amount φ, and the processing means then calculates the

predicted measurements based on the measured angle of tilt φ, and at the same time determines the actual

measurements at that angle of tilt φ. This permitting

the actual and predicted measurements to be compared to

give a measure of the degree of confidence of the measurement.

Fig 6 shows the signal processing system in each

antennae 21 to 23. The signal IP from the corresponding

antennae 21 to 23 is passed via an amplifier 30 and a low

pass filter 31 to an analog-to-digital converter 32. The

low pass filter 31 eliminates unwanted frequencies in the signal, so that the signal generated by the analog-to- digital converter 32 may be passed to a digital signal

processor 33 to permit amplitude and phase signals to be

generated. Fig 6 also shows that the digital signal processor 33 can be used to control gain of the amplifier

30. Amplitude and phase signals thus generated from each

of the antennae 21 to 23 are passed to a microcomputer 34

which calculates the X, Y and ø measurements defining the

relationship between the probe 24 and the underground

object 26, and may be stored in a suitable recording system 35 and/or used to generate a display 36 (e.g. a head-up display).

Another embodiment of the present invention is illustrated in Fig 7. In this embodiment, the ground

penetration probe is a drilling bar 40 which is driven from a tool head 41. The drilling bar 40 contains the

antennas 21 to 23 described previously, but those

antennas are contained within the drilling bar 40 and are

therefore not visible in Fig 7. The tool head 41 receives power via a base station 42 and power line 43,

the base station 42 either containing its own power

source or being powered from a separate power supply 44.

That separate power supply 44 is vehicle-mounted in the embodiment of Fig 7, so that the whole system is transportable. The base station 42 also contains the

processor 25 described previously, with the processor

receiving signals from the antennas in the drilling bar

40 via a line 45 which is also connected to a pack 46 on the belt of the user. Thus, the processor may send

signals via the line 45 and the pack 46 to a head-up

display 47 to permit the user to obtain an immediate

visual indication of the separation of the tip of the

drilling bar 40 from the underground object.

The base unit 42 may be equipped with suitable

memories to store data derived from the antennae, to

provide a more permanent record of the movement of the drilling bar 40 relative to the underground object. If the system is also arranged to provide warning of other

sources of electromagnetic fields, or also a warning of

close proximity of the drilling bar 40 to the underground

object, these may be passed to the base unit 42 via the

line 45 and the pack 46 to earphones 48 for the user.

The powering of the drilling bar 40 by the tool head

41 will now be described with reference to Figs 8 to 10.

The sectional view of Fig 8 shows that the drilling bar 40 is clamped by a collet 50 to a hollow shaft 51 which is rigidly fixed to the casing 52 of the tool head 41. Handles 53 are resiliently attached to the casing 52

via torsion assemblies 54. In Fig 8, the handle 53 and

torsion assembly 54 on the left-hand side are shown in

section, whilst the handle 53 and torsion assembly 54 on

the right-hand side are shown from the exterior thereof.

The torsion assemblies 54 reduce vibration passed to the

user from the drilling head 41. A control switch 55 is

preferably provided adjacent one of the handles 53 to

permit the user to control the action of the drilling bar.

As can been seen from Fig 9, which shows the tool

head 41 in plan view, the switch 55 is connected via a wire 56 to a control unit 57 which controls a motor 58.

The motor 58 is also shown in the side view of Fig 10,

which illustrates the separate casing 59 for the motor

58, and also the right-angle gear box 60. That gear box

60 connects the motor 58 to a shaft 61 which is coupled

by a gear train 62 to a pair of contra-rotating shafts

63. Those shafts 63 are each rigidly coupled to an

eccentric 64 and by lost-motion to a phased eccentric 65.

The purpose of the use of lost-motion is to alter the net

effect of the rotation of the eccentrics to vibrate the casing 52 of the tool head 41 so as to drive the drilling

bar 40 downwardly or upwardly according to motor direction.

The principle of the drive system will now be

explained with reference to Figs 11a, lib and 12a and 12b.

Consider first the case show in Fig. 11a, in which

masses mj and m ₂ are rotated with the same annular

velocity ω, but in opposite directions, about respective

axes P* _L and P ₂ . Then, there is a net force f generated on

the support zone at axis i and P which varies

sinusoidally.

Next, consider the case where there are two further masses m ₃ and m ₄ rotating respectively about the axis P^

and P ₂ , with an angular velocity of ω and in opposite

directions. Then, as shown in Fig. lib, a force f ₂ is

generated which is again sinusoidal. The net force then

depends on the masses and angular velocities, which

determines the phase of the rotation of the masses.

Since forces f] _^ and F ₂ are both sinusoidal, it is

evident that the net force over a whole cycle will be zero. At first sight, therefore, the ground penetration probe will not move. Hover, this does not take into account the friction between the ground and the ground

penetration probe, which must be overcome before any

movement occurs. It is then possible to arrange for the maximum force to be either up or down, and since the

maximum force will then overcome friction by a greater

amount than other forces, this will impose a net movement

on the ground penetration probe.

In this embodiment of the present invention, axes _λ

and P correspond to the shafts 63, and masses m ₁ and m ₂

are equal, as are masses m ₃ and m ₄ . Moreover, assuming

masses m ₃ and m ₄ are smaller, ω ₂ is twice ω ₁ .

Suppose further that phases between the masses is

chosen so that there is a point in the cycle which masses

m and m ₃ are simultaneously directly above point _λ and

similarly masses m ₂ and m ₄ are directly above point P ₂ .

The resulting acceleration profile is then shown by the

solid line in Fig. 12. Figure 12 shows that the upward

acceleration has a maximum value which is greater than

the downward acceleration at any time of the cycle, although the time averaged acceleration over a whole

cycle will be zero. Then, account must be taken of the

friction that will exist between the ground penetration probe and the ground. Suppose a net acceleration of x in Fig. 12 is needed to overcome that friction. It can then be seen from Fig. 12 that the upward acceleration is

sufficient to overcome the frictional resistance, so that

the ground penetration probe moves upwardly. The

downward acceleration, however, is never sufficient to

overcome the friction and thus there is net movement of

the ground penetration probe. Similarly, by arranging

for the phases of the masses to be such that masses nr^

and m ₃ are aligned directly below the point P*L and the

masses m ₂ and m ₄ are directly aligned below the point P ₂ ,

the net acceleration then corresponds to the dotted line in Fig. 12. It can readily be appreciated that the

downward acceleration overcomes the frictional force, thereby imposing movement on the ground penetration

probe, whereas the upward force does not.

Hence, by suitably controlling the phases between

the masses, upward or downward movement of the ground penetration probe can be achieved by the effect of the

rotating masses and their interaction with frictional

forces.

Referring again to Fig. 7 the head-up display 47 gives a visual display of the relative positions of the ground penetration probe on the underground object'. The structure of that head-up display may correspond to that

display produced by Fraser-Nash Technology Limited, in

which an image is projected on to a semi-transparent

spherical mirror so that the light from the mirror is

collimated and is thus perceived by the viewer as being at a distance.

The display that may be generated with the present

invention is shown in Fig. 13. The display has three

windows 80, 81 and 82. Window 81 is a magnified view of

the image appearing in window 80, around the tip of the

ground penetration probe. Each of the windows 80 and 81 display an image 83 of the ground penetration probe, a

region 84 corresponding to the ground, and an image 85 of

the underground object. Thus, the user can see the approach of the image 83 of the ground penetration probe towards the image 85 of the underground object, and can

therefore control the movement of the ground penetration

probe 24 to ensure appropriate movement of the ground

penetration probe 24 into the ground towards the underground object 26. Hence, the user can see, in his direct line of sight, images corresponding to the

separation of the ground penetration probe 24 and the

underground object 26. The image for the window 80 may

be generated using antennae 21 and 22, whereas the image for the window 81 may be generated using the antennae 21 and 23.

The window 82 has a first region 86 which gives a

plan view of the image 85 of the underground object

relative to the longitudinal axis of the ground

penetration probe, and also gives numerical data relating

to the separation of the ground penetration probe 24 and the underground object 26.

In order for the region 84 in Fig. 13 corresponding

to the ground to have a surface which is horizontal, it

is then useful for the probe to contain a tilt sensor,

since the information from that tilt sensor may then be used in the generation of the display shown in Fig. 13. If the probe is not maintained in a horizontal position,

the information from the tilt sensor can then be used to

incline the image 83 relative to the region 84. This may

be useful not only to give the user a warning that the

probe is inclined, but also permits the user to steer the

probe if an inclined approach to the object is necessary,

because of difficulty of access to the surface directly above the object. In the absence of such a tilt sensor, the user must maintain the probe vertical, but this can

be achieved by e.g. a spirit level on the probe itself.

However, in that case, the display shown in Fig. 3 will not respond to the tilting of the probe and the boundary

of region 84 corresponding to the surface will then

always be perpendicular to the image 83, even if the

probe itself is not vertical. This will not affect the display of the approach of the probe to the object, corresponding to the approach of the image 83 towards the

image 85, but the display of Fig. 13 will then give a

less useful guide to the true position.

One possible problem which may arise when a ground

penetration probe is driven into the ground towards

underground cable is that there may be other underground objects, such as other utility cables, in the immediate

vincity. If all are generating electromagnetic signals,

the resulting signals detected by the antennae of the

ground penetration probe will correspond to the composite signal detected, thereby giving an inaccurate measurement

of the position of the underground object of interest.

The fact that the measurement is faulty will be

detectable by moving the ground penetration probe a predetermined distance, or by tilting it, and comparing the predicted and measured locations of the object, as previously described. However, although the operator

will then know that the underground object has not been located accurately, he would not be able to make an accurate location of it. However, if each underground

object at the site to be investigated carries alternating

currents of different frequencies, the electromagnetic

signals generated by each underground object will

similarly be at a different frequency, and therefore

resolvable by modulation at the locator. Therefore, if

the operator applies, by a suitable power source,

alternating currents of different frequencies to each

underground object at the site to be investigated, the

locator can then determine the separation, in terms of both X and Y coordinates, of the locator and each

underground object. Moreover, since the positions of all

the underground objects are known relative to the

locator, they are also known relative to each other.

Hence, the display shown in Fig. 13 may display more than

one underground object. The image of such a second

underground object is shown at 87. Hence, by applying

different signals to the underground objects at the site to be investigated, the operator of the ground penetration probe may be presented with a display showing

the position of all the objects at the site, so that the

ground penetration probe can be controlled so as to

approach the underground object of interest, and to avoid all the others.

As a further alternative, each underground object

may carry an active marker. Such active markers are

known in themselves, and have a transponder tuned to a specific carrier frequency. When the transponder

receives a signal at that carrier frequency, the energy

is converted by induction in a tuned pick-up coil to

power a re-transmitting circuit of the transponder to

generate an output, that output being frequency modulated

at the carrier, frequency, so that it carries encoding data identifying the active marker, and hence the object

to which it is attached. If each underground object at a

site carries such an active marker, and each active

marker is tuned to a different carrier frequency, then

the active marker of any one object can by triggered to

identify itself by the input of a signal at the carrier

frequency generated e.g. by the locator itself. Hence,

the locator can identify the underground objects at the site, which again permits a visual display similar to

Fig. 13 to be generated.

Although the embodiment of Figs 7 to 13 made use of

a drilling bar to form the ground penetration probe, the

present invention is not limited to use of such a drilling bar and other penetration probes such as blades

may be used. Furthermore, although the driving of the ground penetration probe by a motor-driven system has

been illustrated, other drive arrangements may be used to generate a hammering or vibrating action, such a

pneumatic, hydraulic, or electric arrangement. The

drilling bar may have a mechanical stop to limit the

penetration thereof into the ground to a predetermined

depth. It may also have means for attaching a sleeve

thereto, particularly when the ground is soft, to prevent collapse of soil and maintain visual access. Indeed,

suitable vacuum extraction means 60 (see Fig 7) may be

provided to remove material from the site of ground

penetration, to clear a finer layer of material

immediately adjacent the underground object.

Fig. 14 shows the use of a portable locator, being

an alternative to the drilling probe of Fig. 7. It

comprises a locator 141 corresponding generally to the

embodiment of Fig. 5, and a signal processing and display unit 142, typically supported by shoulder and/or waist straps; they are shown separately in Fig. 15.

Figs. 16 shows a typical form of display by display unit 142, showing a vertical cross-section of ground and locator, with positional data regarding separation of target line and locator.

While the foregoing has described the primary

application of the locator to subsurface objects, it may be applied equally to elongate conductors on or above the surface, e.g. to follow a guidance cable.