## Relief Well - Homing-In

1 Management summary

This document provides background information on the subjects of relief well planning, homing-in logging method selection, logging optimization and analysis of the obtained information.

Knowledge of the capabilities and limitations of homing-in logging should allow to develop a blowout contingency/abandonment plan at an early stage of well planning. In particular for slimhole and high pressure/high temperature wells, which are the most critical wells with regard to relief well drilling.

2 Specialist summary

The document addresses the topics: detection range, relief well planning, modelling of the expected homing-in logging signal and, based on a comparison with measured data, the estimation of the actual relief-to-target well distance and the direction. Both detection techniques, electromagnetic and magnetostatic, require conducting material, casing or drillpipe, in the target hole.

Homing-in can be used to guide the relief well planning and execution process, to allow an interception of a blowout well or a re-entry of a well with a severed casing, e.g. for abandonment.

Practical values for the detection range for both electromagnetic (typ. range < 60 m) and magnetostatic homing-in (typ. < 15 m for remanent fields on casings, 25 - 40 m for magnetized casing joints) tools are given. These values are based on field experience and quick-look approximations. Combined with the borehole position uncertainty of both the target and the relief well, they serve to make a first choice of the homing-in logging method.

A second selection criterion of homing-in tool type is given by the formation. If magnetic anomalies in the formation are likely to be encountered, magnetostatic homing-in should be disregarded. Electromagnetic homing-in should be disregarded, when the formation is anisotropic and dipping with respect to the relief well, when the approach angle between the relief and the target well is large and when the formation resistivity parameters fluctuate significantly over the logging range.

A third criterion may be cost and mobilization time. Both will in general be smaller for magnetostatic tools than for electromagnetic tools.

Based on models for the field intensity and direction, given a choice of logging method, a prediction of the tool response as a function of along hole depth in the relief well can be given. The response will depend strongly on the relief well design. The predicted response is used for two purposes. In the planning phase, it is used to estimate the detection range based on the minimum sensitivity of the tool and the magnetic fields associated with the steel in the target well. If this is not satisfactory, the detection method or relief well design should be changed. In the logging phase, it is used for for processing of the logging tool results to obtain position information on the target well relative to the relief well.

The modelling of the tool response and the calculation of the most likely position of the target well is done by software packages for electromagnetic tools, and magnetostatic tools.

3 Guide to the document

Homing-in logging is a rarely applied technique. Luckily because blowouts are seldom occurrences and most abandonments can be completed through the original wellhead. Over the last years, most homing-in jobs have been performed on troublesome abandonments.

Due to the fact that homing-in is not a core business, it is difficult to stay abreast of the development of homing-in technology and evaluation techniques. The minimum required of a homing-in focal point is to be familiar with the general theory, as presented in Section 5 to Section 9, Section 11 and Section 12.

The steps in the process of designing a homing-in job are

1.Evaluate the configuration and condition of the iron parts in the target well (sizes, along hole depth position and integrity)

2.Generate a set of potential relief well configurations (Section 11)

3.Collect directional survey data of the target well and petrophysical data from formations in the potential homing-in area

-petrophysical: i.e. resistivity, dip, resistivity, anisotropy

-mud types (OBM/WBM),

-magnetic anomalies (from raw MWD/EMS or dipmeter)

-directional survey data from nearby wells and side-tracks to evaluate a) collision risk for the relief well and b) risk of interference on the homing-in signal (Section 8)

4.Evaluate position uncertainty of the target well (Section 6)

5.Evaluate combined position uncertainty of target and relief well (Section 6) for each of the relief well options (make sure that target and relief well data are given relative to a common reference of depth and azimuth (Section 5))

6.Evaluate petrophysical and well status data to estimate the chance of success with a particular detection method (Section 8, Section 9, Section 12). (magnetostatic: expected pole strength (casing/DP size) and position (casing/BHA tally), interpretation; electromagnetic: expected background signals, current injection options)

7.Simulate the configuration of target and relief well with software(s) to determine the expected signal levels (Section 10, Section 10.4.3) and preferred AHD intervals for logging

8.Contact a homing-in contractor or surveying contractor (Section 8.1.5, Section 8.2.3) and ensure that the data will be made available in the required format for the software packages.

This list of steps will usually not be executed sequentially, but should be regarded as an iterative process.

The minimum dataset required to perform homing-in is:

·a target well survey (AHD, TVD, East, North)

·a relief well survey (AHD, TVD, East, North)

·homing-in logging data at regular AHD intervals

·homing-in logging-run toolface data (see Section 3)

·info on casing sizes and formation resistivity (electromagnetic homing-in)

·info on the expected magnetic pole pattern and strength (magnetostatic homing-in)

It should be made sure that all survey data is given with respect to a common depth and azimuth reference.

The toolface data can, depending on the tool type and borehole inclination, be obtained from the logging tool itself or from a tool run in conjunction with the logging tool. For hole inclinations (I) larger than 5 deg., gravity toolface can be used, which is available from both types of homing-in logging tools. For I<5 deg. and an electromagnetic tool, magnetic toolface can be used. (Beware for magnetic interference from the target casing at short distances to the target.) For I<5 deg. and magnetostatic homing-in, a gyro tool has to be run in conjunction with the magnetic tool to obtain toolface information.

4 Introduction

When the blowout of a target well cannot be controlled by surface capping operations, subsurface intervention by means of one or more relief wells is often required. Similarly if a well cannot be entered below a certain depth, a relief well may be required to reenter the part of the well below the obstruction or parted casing.

Because the combined downhole position uncertainty of the target and the relief well is too large for direct interception, homing-in logging is used to determine the direction and distance between the two more accurately than possible from their surveys.

Homing-in logging should be considered at an early stage of relief well operations due to its impact on relief well design.

This manual addresses the major issues relating to the planning and analysis of homing-in logs during the course of relief well drilling operations. It provides checklists, guidelines and computation methods, and it summarises currently available logging tools and logging problems.

1.The operations petrophysicist nominated as co-ordinator for homing-in logging. He should be confident with basic electromagnetic theory and should have a working knowledge of the various homing-in techniques and the software for data interpretation.

2.The deviation focal point should be to assess borehole position uncertainty for relief and target wells in order to plan directional survey programs and homing-in operations.

The operations petrophysicist, the deviation focal point and the drilling engineer in charge of the relief well should work in close cooperation so as to optimise the efficiency of the planning and execution of the relief well drilling process.

5 Borehole survey references

Almost all survey tools measure inclination and azimuth at particular along hole depths. These measurements must be tied to fixed reference systems so that the borehole course may be calculated and recorded. The recommended calculation method is the Minimum Curvature Method. The reference systems used are:

·inclination references (Section 5.1);

·depth references (Section 5.2);

·azimuth references (Section 5.3).

The exceptions are the inertial navigation tools that measure the accelerations and rotations of the tool with respect to the three coordinate axes.

5.1 Inclination references

Inclination - The inclination of a well at a given depth is the angle (in degrees) between the local vertical and the tangent to the wellbore axis at that depth. The convention is that 0° is vertical and 90° is horizontal. It is recommended to work in decimal degrees.

Local vertical - The local vertical reference coincides with the direction of the local gravity vector. The magnitude of gravity varies with latitude. Local deviations in both magnitude and direction can occur due to peculiarities in the Earth's crust (mountains, accumulations of dense materials). These local deviations, however small, may affect modern electronic survey tools and may limit the accuracy of the inclination measuring instrument.

5.2 Depth references

There are two kinds of depth: Along Hole Depth (AHD) and True Vertical Depth (TVD). Depth is measured or calculated with reference to a fixed point or datum.

Along hole depth (AHD) The distance measured along the actual course of the borehole from the surface reference point to the point of interest (e.g. as measured by wireline or drillpipe length).

True vertical depth (TVD) The vertical distance from the depth reference level to a point on the borehole course.

Depth reference systems During relief well drilling operations two depth references may be used: a common depth reference or working depth reference.

Common depth reference To compare individual wells within the same field, a common reference must be defined and always be referred to (i.e. a depth datum). The datum level is used when drilling a relief well into a blowing well, such that the difference in elevation between the wellheads will be accurately known. Offshore practice is to use the mean sea level (MSL).

Working depth reference In most drilling operations the Rotary Table (RT) elevation is used as the working depth reference. This is also referred to as derrick floor elevation. Depths measured from these references are often called below rotary table (BRT) or below derrick floor (BDF). For floating drilling rigs the rotary table elevation is not fixed and hence a mean rotary table elevation has to be used. A heave/tide gauge on the rig floor is often useful.

Prior to spudding-in a well, the rotary table elevation relative to the common depth reference must be established. For land wells normally the elevations of the top cellar to the chart datum is surveyed by the Topography Department prior to starting drilling. When the derrick is rigged up, the distance rotary table-top cellar is measured and thereafter the elevation of the rotary table to the chart datum can then be calculated. Offshore the distance from the observed sea level to the rotary table is measured and thereafter corrected for the variation with respect to the Mean Sea Level (MSL).

5.3 Azimuth references

Definition of azimuth - Azimuth is the angle between the horizontal component of the direction of the wellbore at a particular point measured in a clockwise direction from Magnetic, Grid or True North. Azimuth must be expressed as a reading on a 0°-360° system.

Azimuth reference systems - For directional surveying, there are three azimuth reference systems: Magnetic North (MN), True (Geographic) North (TN) and Grid North (GN).

Magnetic North (MN) - This is the direction of the horizontal component of the Earth's magnetic field lines at a particular point on the Earth's surface. A magnetic compass will align itself to these lines with the positive pole of the compass indicating North.

True (geographic) north (TN) - This is the direction of the geographic North Pole. This lies on the axis of rotation of the Earth. The direction is shown on maps by the meridians of longitude.

Grid north (GN) - The meridians of longitude converge towards the North Pole, and therefore do not produce a rectangular grid system. The grid lines on a map form a rectangular grid system, the Northerly direction of which is determined by one specified meridian of longitude (the central meridian). This direction is called Grid North. It is identical to True North only for the central meridian. Details about the coordinate system to be used and about applicable conversion factors and correction angles should be available from local Topography Department.

Azimuth reference system conversions - All azimuths must be quoted in the same reference system. This is usually the Grid North system. In practice, azimuths are often measured in systems other than the Grid North system. Two conversions normally have to be applied to the measured azimuths: Grid correction (to convert azimuth values between the True North and the specified Grid North systems). Magnetic declination cirrection (to convert azimuth values between Magnetic North and True North systems).

Grid correction angle (G) - A grid correction converts azimuth values between the True North system and the specified Grid North system. The grid correction angle is the angle between the meridians of longitude (TN) and the Northing of the particular grid system (GN) at the point. By definition, the grid correction is positive when moving clockwise from Grid North to True North, and is negative when moving anticlockwise from Grid North to True North. The value of grid correction angle depends upon location. Close to the Equator the correction is small; increases with increasing latitude.

Magnetic declination correction (D) Magnetic declination correction converts azimuth values between the Magnetic North and True North systems. The magnetic declination correction is the angle between the horizontal component of the Earth's magnetic field lines and the lines of longitude. When Magnetic North lies to the West of True North, the magnetic declination is said to be Westerly, and its value is defined as negative.

6 Borehole position uncertainty in relief well planning

Homing-in aspects of relief well planning initially centre on the design of the relief well and its effect on the combined bottom hole position uncertainty of the blowout and relief well and the range of the homing-in tool. The range of the tool determines the maximum bottom hole position uncertainty at which the tool can be used successfully. Range is defined as the distance at which the tool can determine the position of the blowout well relative to the relief well more accurate than is possible using survey data.

Relief well design affects both the range of the homing-in tool and the combined bottom hole position uncertainty. It is essential in the design of a relief well, especially in the initial planning stages, that there is an understanding of the relationships between bottom hole uncertainty, homing-in tool range and well design.

6.1 Borehole position uncertainty

Borehole position uncertainty is the range of the actual possible positions of a survey station from its calculated position. The magnitude of position uncertainty describes a three dimensional ellipsoid. The orientation of the orthogonal triad of the three major axes may vary along the borehole with respect to North East and Vertical. The axes only coincide with the local, casing fixed, coordinate axes for a straight well.

A study by Wolff and De Wardt on survey errors, revealed that the major source of position uncertainty is caused by systematic errors (i.e. errors which have a constant sign and therefore, do not average to zero over all or part of a borehole survey). The errors for gyros and magnetic surveys have been quantified, and are given for 'good' and 'poor' quality surveys.

The sources of systematic survey errors - For the Homing-in co-ordinator it is not necessary to enter this in detail, since it is a matter for the survey specialist. It is more important for the Homing-in co-ordinator to be aware of aspects of a relief well path that affects the bottom hole position uncertainty. Therefore, the method for estimating the position uncertainty as given below is only an approximation. The reader should refer to uncertainties calculated by well planning programs for final results.

6.2 Relative position uncertainty

Computation using the Wolff and De Wardt systematic error model can be used to produce curves of borehole position uncertainty, in the lateral, along hole and upward directions for quality magnetic and gyro surveys. For most practical cases lateral position uncertainty is the largest uncertainty and consequently can be used to estimate borehole position uncertainty. Pposition uncertainty is expressed in m/1000 m (ft/1000 ft) AHD.

It should be noted that these curves indicate the position uncertainty of a single well only (not the relative position uncertainty of two wells.

6.3 Estimating well position uncertainty

The curve of the lateral position uncertainty for a specified tool can be used to give a rough estimate of the position uncertainty of a well. As only the largest radius of the ellipsoid is used, a circle of position uncertainty is produced. The estimate is made by dividing the well into sections (e.g. vertical and tangent) and using the curves to estimate the uncertainty for each section. The results are then summed to obtain the total position uncertainty. For complicated well shapes a better estimate can be made by using smaller sections. However, the results obtained for such wells have very limited value.

6.3.1 Worked example

To explain the procedure in more detail, the position uncertainty of the target at 973 m AHD and total depth of the example well will be estimated for a good magnetic survey. The well can be simplified and divided into three sections:

·a vertical section from 0 to 550 m AHD

·a tangent section with an average inclination of 18.5° from 550 to TD at 973 m AHD.

·a tangent section with an average inclination of 18.5° from 973 to TD at 1142 m AHD.

The position uncertainty of each section can then be estimated.

6.3.2 The vertical section

From the lateral position uncertainty curve for good magnetic surveys at 0° inclination, the uncertainty radius is 1.8 m per 1000 m AHD. The along hole depth of this section is 550 m so the uncertainty radius is (550/1000) ´ 1.8 = 0.99 m @ 1 m.

6.3.3 The tangent section from 550 to 973 m AHD

From the position uncertainty curve for good magnetic surveys at 18.5° inclination the uncertainty radius is 8.25 m per 1000 m AHD. For this section the along hole depth is 973 - 550 = 423 m. Hence the uncertainty radius of this section is (423/1000) ´ 8.25 = 3.49 @ 3.5 m.

6.3.4 The tangent section from 973 to 1142 m AHD

The uncertainty of this section is the same as the last section and is 8.25 m per 1000 m AHD. The along hole depth of this section is 1142 - 973 = 169 m. Hence for this section the uncertainty radius is (169/1000) ´ 8.5 = 1.39 m @ 1.4 m.

6.3.5 Summation of errors from individual sections

From the above sections the position uncertainty at the target will be the uncertainty of section 1 added to section 2 which is 1 + 3.5 = 4.5 m. Hence the estimated position uncertainty at the target is a circle with radius 4.5 m. A total depth uncertainty will be the uncertainty of the target added to the uncertainty of section 3 which is 4.5 + 1.4 = 5.9 m. Hence at TD the position uncertainty is a circle of radius 5.9 m.

1.This method of estimating position uncertainty produces a circle of uncertainty, and should only be used as a rough guide to possible error.

2.Always consult the surveying focal point for the latest tool error values.

6.4 Combined bottom hole position uncertainty

The combined bottom hole position uncertainty of blowout and relief well will dictate, when it is possible to perform the first homing-in log in view of the detection limits of the homing-in tool.

Once a well is drilled and a homing-in operation is required, its bottom hole position uncertainty is fixed. Minimisation of the combined bottom hole position uncertainty can be achieved in a number of ways while planning a relief well:

1.Minimise the separation between the blowout and relief well surface location within limits of safety. This has the effect of reducing the along hole distance and inclination needed to reach the target depth.

2.Reduce the along hole distance, to the target depth.

3.Kick-off shallow and reduce the inclination of the relief well.

4.Re-survey cased hole section of the relief well with higher accuracy tools, compared with the magnetic tools used in the open hole.

5.Aim the relief well for a direct intersection with the surveyed (or assumed) position of the target well.

6.5 When to make the first homing-in run

Below is a checklist to assist in planning the depth at which the first homing-in run should take place:

Estimate the maximum distance at which the homing-in tool will detect the target well (covered in Section 8).

Determine (in cooperation with the surveying specialist) the bottom hole position uncertainty of the blowout and relief wells.

Find the shallowest horizon at which the maximum combined bottom hole position uncertainty is less than or equal to the detection limit of the tool. The well should be drilled to this horizon and logging can commence.

Ideally the circles of borehole position uncertainty should not intersect at the proposed first logging depth, unless it can be tolerated by drilling operations, i.e. kill facilities on standby in case the target well is intersected before logging.

The first logging run should be as early as possible in drilling the relief well, therefore giving the directional driller maximum along hole length to make steering corrections prior to reaching the proposed kill depth, without having to plug-back.

7 The earth's magnetic field

The Earth can be considered to be a magnet surrounded by its magnetic field. By definition the field strength vector is into the Earth at the magnetic North Pole and out of the Earth at the magnetic South Pole. Close to the Equator the field lines are parallel to the Earth's surface and at the magnetic poles they are near vertical.

For practical purposes, the Earth's magnetic field is resolved into a horizontal and a vertical component. A magnetic compass will align itself to the horizontal component of the Earth's magnetic field.

As the horizontal component of the Earth's magnetic field becomes smaller (i.e. towards the North and South Poles), a compass will become more sensitive to any extraneous magnetic field. Such a field may be due to remanent magnetism on steel components in a nearby target well).

7.1 Dip

Dip is defined as the angle between the Earth's magnetic field lines and the horizontal (see Fig. 1285). The dip ranges from -90° at the South Pole, through zero at the Equator to +90° at the North Pole.

7.2 Magnetic field strength

The total strength of the Earth's magnetic field Be is expressed in micro Tesla (mT) and varies from 40 mT at the magnetic equator to 60 mT at the magnetic poles. The horizontal component Bh of the magnetic field vector ranges from 40 mT at the Equator (where Bh = Be) to 0 mT at the magnetic poles. Data on magnetic field strength can be obtained from the local Topography Department or the Directional Drilling Focal Point.

8 Homing-in techniques

This chapter outlines briefly the operating principles of the two field proven techniques for homing-in. This chapter is intended to provide a list of advantages and limitations of the techniques, condensed into a few pages. This will enable the operations petrophysicist to have a 'quick look' set of guidelines to be aware of, prior to contacting homing-in contractors and attending the first relief well planning meeting.

Modern, field proven homing-in techniques require the presence of steel (casing or drillpipe) in the target well, upon which to home-in. Two homing-in techniques are possible:

1.Electromagnetic (or active), current injection method, which is the most widely used and most expensive method. A target well can be detected at a distance of 45-60 m (150-200 ft) (Section 8.1).

2.Magnetostatic (or passive) homing-in which is a cheaper technique. It relies on the remanent magnetism of steel in the target well. A target well can be detected at a distance of 15 m (50 ft), depending on the strength of the remanent magnetic field. In magnetostatic homing-in on artificially magnetized casing joints a detection range of 25-40 m can be obtained Section 8.2).

8.1 Principle of electromagnetic homing-in

Electromagnetic homing-in tools have two main components: a downhole or surface current injection electrode and a downhole magnetic field sensor sonde. The magnetic field sensor lies below the downhole electrode and is separated by an insulated bridle usually 91 m (300 ft) long. The insulating bridle then continues for another 30 m (100 ft) above the electrode so that it is isolated from the wireline which runs to surface.

8.1.1 Wireline current injection

The electromagnetic homing-in tool is run on wireline to the logging depth in the relief well and a low frequency (1-3 Hz) AC current of about 5 Ampere is injected via the downhole electrode in the relief well. This current flows into the formation and returns to be collected by remote surface electrodes. In formations of uniform conductivity and no target metal (i.e. casing or drillpipe in the target well), the current injector produces a spherically symmetric current flow which fluctuates in time. This current flow produces no net cross-axial magnetic field at the sensor position, if the sensor is also on the symmetry axis.

If there is a target well casing nearby, it offers a low resistance path for the current to flow, and thus short circuits the current flow along the casing hence disturbing the uniform flow. This current along the casing produces a fluctuating magnetic field in the plane perpendicular to the axis of the target well. Two orthogonal AC magnetometers in the sensor sonde will detect the intensity and direction of the induced magnetic field (H). The intensity of the signal recorded by the sensors, is proportional to the amount of current on the casing (it) and the inverse distance (1/dSS) of the sensor from the casing.

Measurements of the Earth's magnetic field and/or gravity field by the sensor sonde is used to indicate the orientation of the homing-in tool. When these orientation measurements are inaccurate, due to the verticality of the well (< 5° inclination) and the proximity of the target well (disturbance of the magnetic field), the tool orientation can be determined by running a gyroscope in conjunction with the sensor. Measurements of the low frequency AC magnetic field are analysed to estimate the distance and direction to the target. When the sensor proximity to the casing is less than 15 m (50 ft) the Earth's magnetic field may be distorted due to remanent magnetisation of the target casing, hence the direction to the target casing is reported relative to the Earth's gravity field. The distortion to the Earth's magnetic field measured by the sensor can also be analysed to determine the distance and direction to the magnetic pole, as is normally done for a magnetostatic homing-in tool.

8.1.2 Surface excitation system

The electromagnetic homing-in tool consisting of a sensor sonde and current injection electrode are run in the relief well on a wireline in the same configuration. However, the downhole electrode is switched off, and the current injection electrode is attached at surface directly onto the target well casing head. However, in case the target well is not accessible, a low frequency AC is injected into the ground by use of surface electrodes near the wellhead of the target well.

The casing or drillpipe will be the preferred path for conducting current down the target casing and into the formation. The return current is again collected by remote surface electrodes. The distribution of current on the casing however will differ from the downhole electrode case.

As with the downhole electrode case, the current along the casing sets up a fluctuating magnetic field which can be detected by the sensor sonde and interpreted to give a distance and direction.

8.1.3 Capabilities of electromagnetic homing-in tools

Electromagnetic homing-in tools require that the target well has some electrically conductive material such as a drillstring or casing in the region to be detected. The maximum range of the tools is subject to many factors including:

1.Integrity of the casing or drillstring in the target well.

2.Conductivity of the drilling mud used in the relief well.

3.Resistivity and homogeneity of the surrounding rock formation.

4.Inclination of the relief well relative to the target well at the ranging depth.

Rule-of-thumb: For relief wells drilled with water based mud in homogeneous rock with a resistivity of 1 ohm-m, good distance and direction to the target well can be obtained up to 60 m (200 ft) away.

When oil-based mud is used to drill the relief well, the detection range is reduced due to the insulating properties of the mud. Oil-based mud inhibits the current transmitted from the wireline electrode: therefore less current is available to give a signal on the target well. Much of the loss of signal strength can be made up by averaging the signal over a longer period of time.

Rule-of-thumb: Field tests by Kuckes have shown that oil based mud in the well reduces the detection distance to about one half of the original range.

If the target well was drilled with oil-based mud, the critical factor determining the strength of the signal measured by the homing-in tool is the contact resistance between the drill string or well casing and the surrounding rock. This can vary widely depending on the individual circumstances. Again, much of the signal loss can be made up by extended signal averaging.

Guideline: in special cases, implementing the Surface Excitation System provides sufficient signal strength when the downhole electrode system does not.

Heterogeneous surrounding rock can produce its own signal and bias the electromagnetic tool response. However, this effect is only important at low signal levels; within 30 m (100 ft) signal levels are sufficient that accurate ranging can be done. Further, if the relief well is passing by the target well, triangulation can be used for distance determination and the range is not reduced.

The magnitude of the AC magnetic field recorded by the sensor sonde decreases with increasing inclination of the relief well relative to the target well. This is due to the AC magnetic field sensors, which only record the component of the magnetic field in the plane perpendicular to the relief well axis. Hence 100% of the induced AC magnetic field will be measured when the relief well is parallel to the target. When using the wireline current injection electrode the separation between source and sensors (of 90 m) is such that the sensor sonde is in the horizon corresponding to the point at which maximum current builds up on the target casing.

Guideline: an angle of intersection between target and relief wells up to 30° will ensure good signal levels are recorded. Angles exceeding 40° will result in strongly reduced signal strength and may compromise interpretation of tool response.

When EM-homing-in tools with three AC magnetometers are developed, there will no longer be a limit on the range of intercept angles regarding adequate signal levels (see also Section 9.4.3).

The amount of current on the target well is reduced greatly near the bottom of the well, known as the 'pipe-end effect'. It is best to plan the relief well such that the target depth is well above the bottom of the well, even if the relief well is later drilled to the bottom for the kill. Similarly the current on the target well is reduced greatly near where the casing has been severed. This can lead to unexpected low signal strengths.

EM ranging tools can also be run inside drillpipe. If the openhole proves difficult to access, the tool can be run through an open ended drillpipe which spans the difficult section of hole. Using a side-entry sub in the drillstring, this type of operation will permit logging of large sections of openhole. The wireline current injection electrode can still be used. It should be noted that inside or near the drillstring the source will no longer act as a point source of current, but as a line source.

At present, the software does not provide routines to model this line source.

8.1.4 Strategies for well location

It is important at this stage to be aware of the various strategies for locating a target well using electromagnetic homing-in. Each method is covered in more detail later in the manual. Distance and direction to a blowout can be determined using four methods:

1.Triangulation by observing the change in direction of the target well depth.

2.Variation of the intensity calculations based on formation and well conductivity models.

3.Absolute intensity calculations based on formation and well conductivity models.

4.Analysis of perturbations of the Earth's magnetic field due to magnetization of the target casing.

When the wells are within 15 m (50 ft) of each other and casing magnetization would normally be a problem, gravity sensors in the homing-in tool allow accurate determination of inclination even in these circumstances, provided I > 5° (Section 8.1.1). Also the perturbation of the Earth's magnetic field from the target casing can be used to make an independent determination of distance and direction. This would be normally done in magnetostatic homing-in.

The best accuracy is possible in a pass-by situation. All four methods can be used to determine the distance and cross-checks between them can be made. When planning a pass-by it is prudent to make up to 4 initial homing-in runs before passing the blowout to be sure of not inadvertently intersecting the well at an unwanted position. As a guideline the runs can be made at separations of 50, 25, 10 and 3 m from the target well. If the relief well passes the target well, an additional run can be made at target depth.

The presence of another cased hole near the target well or of a sidetracked fish in the target well can cause deviations in the homing-in tool signal. It is best to plan the relief well so that homing-in logs are run in regions where these effects will be small. In general, if other casings or a fish are expected to be at least twice as far away as the target, they will have a minimal effect on the electromagnetic tool response.

When drilling a relief well with no bass-by, or when close to the target well, the EM-ranging tool can be used inside a non-magnetic drill collar. When landed into a directional drilling shoe, the angle between the target well and the shoe pin can be monitored. This permits the driller to home-in to the target well with a bent sub and motor drill without tripping out of the hole.

8.1.5 Commercial information on EM homing-in tools

WELLSPOT: Vector Magnetics Inc

MAGRANGE: Tensor Inc

The group has gained some experience with the WELLSPOT tool during an operation of BSP Seria [1748], in relief well Rasau 19. In this well the tool was run on a Schlumberger wireline unit. It was observed that all runs with the tool in this well had depth control problems. These problems were attributed to the fact that WELLSPOT is a lightweight tool (as opposed to a heavy Schlumberger tool for which the units are normally used) and the lack of cooperation between the Vector Magnetics engineer and the Schlumberger engineer.

The following recommendations are made to improve communication at the wellsite:

1.The wireline engineer has the responsibility for the running operation and depth control. Vector Magnetics should provide information from the motion sensors.

2.Prior to running in hole, Vector Magnetics should provide the Schlumberger engineer with the distance between the torpedo and the sensor point, and the distance between the sensor point and the bottom of the tool.

3.Zeroing of the tool, prior to running in hole, should be accomplished with the torpedo fastened and the wireline cable tensioned to remove slack.

4.Running in hole, the first 100 ft should be slow, to avoid oedometer wheel slipping. An operator should be assigned to watch the wheels.

5.A cable cleaner should always be used to avoid slipping.

6.Depth tie-in should be made whenever possible.

7.Upon pulling-out, the depth zero should be checked and the closure should be given in the report on the measurements.

8.A sketch and short report on depth control should be given by the Schlumberger engineer.

It was also found that excessive use of the motion sensors causes tool heating due to energy dissipation in the electronics. The maximum rating for the electronics is 60°C and the tool's thermal shield is a Dewar (Thermosflask).

8.2 Principle of magnetostatic homing-in

Magnetostatic or passive homing-in tools use any remanent magnetisation of the target well casing or drillpipe to find its relative position. The magnetic field measured with orthogonal (triaxial) DC magnetometers consists of a vector sum of the Earth's magnetic field and the target field. The influence of the Earth's field is subtracted such that only a 'signature' of the magnetic anomaly remains to be interpreted. The total magnetic field is recorded by a DC magnetometer triad. The measured field, Bm, is decomposed in a component Bmz along the relief well axis and a component Bmxy in the plane perpendicular to the relief well axis. Although the method is not limited by formation effects, pipe-end effects and intercept angle, the degree and distribution of magnetisation on the target is unpredictable since it depends on the magnetic and mechanical history of the steel in the blowout well. The detection range for magnetostatic homing-in is usually limited to less than 15 m (50 ft).

The detection distance of wells can be improved (before they blow-out) by including artificially magnetised steel components in the bottom hole assembly of a drillstring or including a magnetised joint at the shoe in a casing design. It has been found ([1736], [1737]) that detection ditance can be doubled, i.e. to a distance of 25-40 m (100 ft) with this relatively simple method.

8.2.1 Magnetic field modelling

The hardware required for magnetic homing-in logging, consists in principle of calibrated three-axis magnetometers and accelerometers. Traditionally magnetic homing-in has been performed by Tensor Inc. with their MAGRANGE tool. Now, in principle any electronic magnetic surveying tool, EMS and MWD, can be used to measure the downhole magnetic anomaly. Tools with a continuous readout at surface are best suited for observing the anomaly while logging. However, this is not a necessary requirement.

8.2.2 Interpretation of magnetostatic homing-in logs

Target direction is obtained immediately when the magnetic anomaly is a monopole field or when the relief and target wells are parallel; the total field is then, respectively, spherically or rotationally symmetric, pointing towards or away from the target depending on polarity. The polarity in turn can be determined from the axial field component Baz.

Direction determination is more cumbersome with multiple and interfering poles on a target which is not coplanar with the relief well. In this case modelling of the total magnetic field and curve fitting techniques must be applied.

There are two options for determining the target distance from the magnetic log. Each option has its preferred configuration:

1.When monopole or dipole signatures are recorded, the shape of the curves Bar(z) and Baz(z) is fully determined by the target distance d and is independent of the pole strength M. Hence d can be derived from the width of the curves Bar and/or Baz. A quick-look approach can be used at the wellsite, and computerised curve fitting by automated forward modelling can be applied in the office.

2.The 'overlay technique' for multiple and interfering poles is limited to parallel target and relief well paths. This method, introduced by BSP was used successfully on two relief well operations, where the GPIT (General Purpose Inclinometry Tool) part of the SHDT resistivity tool was used to detect the magnetic anomaly.

9 Electromagnetic (EM) homing-in systems

The general principle of homing-in with electromagnetic, current injection tools has been given in Section 8. Current flowing on the pipe in a target well will generate a magnetic field according to Ampere's law. The fluctuating magnetic field is detected by AC magnetometers, and the direction to the blowing well is at right angles to the direction of this field.

9.1 Definition of terms

Before discussing electromagnetic homing-in tools and interpretation in greater depth, it is necessary to define the terminology used throughout the report:

Highside direction The direction in the sensor plane having the greatest upward gravity component. The unit vector in this direction is denoted by iC.

Highside-right (HSR) A direction in the sensor plane 90° clockwise from highside. This direction is perpendicular to the Earth's gravity vector and the relief well axis. The unit vector in this direction is denoted by jC.

Zero current point (ZCP) That point on the target for which the line connecting it with the electrode position is perpendicular to the target well axis.

Signal point That point on the target for which the line linking it with the sensor position is perpendicular to the target well axis.

Target field vector A vector in the direction of the resultant field at the sensor position. For a straight target well, the direction of this vector is perpendicular to the target well axis and the line connecting the sensor position to the signal point.

Hxy The component of the target field vector lying in the sensor plane, also called the cross-axial component of the field.

HSDIR The angle measured clockwise from the Highside direction to Hxy.

Apparent target direction (ATD) The angle measured clockwise from Highside to a direction 90° clockwise from Hxy, i.e. the apparent target direction (ATD) = HSDIR + 90°. Note that, only for parallel wells, this is is the true direction to the signal point on the target.

Closest approach point When a relief well passes a target well the point of closest approach is defined as that point on the relief well axis for which the distance ot the signal point is minimum. At this point the line linking the signal point to the sensor position is perpendicular to both the axis of the relief well and that of the target well.

x-direction In an Earth fixed reference system, the xE-direction lies towards the North. North may be True, Grid or Magnetic, providing it is used consistently. In the Highside reference system, the xC-direction is along the Highside direction. (The subscripts are E and C for Earth and Casing fixed reference, respectively. Not yet mentioned is x B, the probe body fixed x coordinate.)

y-directionIn both the Earth fixed and the Highside reference system the y-direction is defined such that x,y,z form a right-handed orthogonal set. Therefore in the Highside system the y-direction coincides with the HSR direction.

z-direction In an Earth fixed system the z-direction is in the same direction as the Earth's gravity vector. In a Highside system, z lies along the well axis pointing and increases with along hole depth. The unit vectors in x, y and z direction are indicated by i,. j and k and their subscript indicates the reference frame.

Background signal The AC magnetic field due to currents in the formation, measured at the sensor.

Intercept angle If the target well at the point of closest approach is projected into the plane through the relief well, being perpendicular to the line connecting signal point and sensor position at this point of closest approach, the intercept angle is that angle between the relief and target axis at the point of intersection. For a straight vertical target well the intercept angle is simply the inclination angle of the relief well.

Toolface (TF) A reference mark on the sensor, relative to which the AC magnetometer axes are defined. The angle between the toolface and Highside is required in order to convert the raw tool data to the Highside reference system. The angle from Highside clockwise to the toolface is referred to as HSTF. The angle between the TF and the yB-axis is given the label WC. Similarly from TF to the xB-axis is the angle EC. Hence, EC + 90° = WC.

Equivalent radius (ro) The radius of a cylindrical column of earth having the same resistance per unit length as the target casing. This may be shown to be $\sqrt {(\rho _e.2r_ct_c/\rho _s)}$ where: re resistivity of earth (Wm) rc radius of the target sheet tc wall thickness of the target sheet and rs resistivity of the target steel (Wm) [1725].

9.2 The sensor package

9.2.1 DC magnetometers

Two orthogonal DC magnetometers measure only the components of the Earth's magnetic field in a plane perpendicular to the tool axis, i.e. the sensor plane. One magnetometer axis is aligned with an external reference mark on the tool housing, referred to as "toolface". The tool outputs the angle ET, of the toolface relative to the Earth's magnetic vector (projected in the sensor plane). The Earth's magnetic field can be used as a fixed reference for correcting for tool rotation during logging, provided there is no magnetic interference from the target casing or elsewhere.

9.2.2 AC magnetometers

The AC detection system consists of two orthogonal induction type magnetometers MAGx, MAGy lying in a plane perpendicular to the tool axis, i.e. the sensor plane (Fig. 1290). These magnetometers respond to the AC magnetic field (H) in that plane, (Fig. 1291). The field vector occurs in a plane perpendicular to the blowout well (i.e. the target plane) according to Ampere's law. For co-planar well paths the magnetic field vector H falls also in the sensor plane. In case of non-coplanar wells only a projection of H in the sensor plane is measured. The target direction is defined as being 90° clockwise from the field vector produced by a positive (down-going) current according to the 'right-hand rule'.

The output of an AC magnetometer in a sinusoidally changing magnetic field is a sine wave. The voltage applied to the current injection electrode is used as the reference sinusoidal signal to define the phase shift of the magnetometer output. The phase shift is interpreted as the orientation of H relative to the position of the AC magnetometer., Hence relative to the injected signal each magnetometer has a signal comprising a 'real' component which is in phase with the injected signal, and an imaginary component that lags by p/2.

In practice, phase shifts also occur in the formation and tool electronics so that the magnetic field is already phase shifted by an unknown amount relative to the injected current. The electronic phase shift f remains constant and is known at calibration. However, the formation phase shift, or background signal (if any) can only be determined and vectorially subtracted from the total signal once sufficient downhole measurements have been made. The phase shift on the x and y magnetometers may not be the same due background effects. Laws of background signals are described in Section 9.3. Background corrections are dealt with in Section 10. Fig. 1292 illustrates the complex representation (i.e. real and imaginary components) for various positions of MAGx in an AC sine wave field.

EC is another tool calibration constant (Section 10) to correct for mechanical misalignment of the x-magnetometer axis with the toolface. EC must be added to qH in order to get the field direction relative to the toolface. The AC field intensity Hx, Hy is given in mA/m, and H is normalised to a 1 Ampere current injection to give units of mA/m/A.

Due to the low intensity of the AC magnetic field (typically 20-3000 mA/m/A, and the high sensitivity of the AC magnetometers it is necessary to make stationary measurements with an electromagnetic homing-in tool. Any movement of the tool in the Earth's field will give an induced voltage in the induction type AC magnetometers. The WELLSPOT tool has a motion sensor which will give a warning of this at surface.

9.2.3 Magnetic and highside reference system

In the magnetic reference system, target and toolface angles are given relative to Magnetic North. Magnetic North is defined as the direction of the horizontal component of the Earth's field. Therefore the magnetic vector measured by the DC magnetometers corresponds to the definition of Magnetic North only when the tool is vertical, i.e. the sensor plane is horizontal.

If the sensor plane is not horizontal as is the case for deviated relief wells, ET is an angle relative to 'Apparent North'. Apparent North owes its name to the fact that it is not the actual direction to Magnetic North, but due to the projection appears to be so. ET has to be corrected to give the angle from Magnetic North. This correction is achieved by removing the effect of the Earth's vertical component and taking the projection of the resultant vector into the horizontal plane. This correction is often referred to as "magnetic compass correction". The magnitude of this correction depends on hole azimuth, inclination and magnetic dip.

To give the target direction relative to magnetic north (not apparent north), the AC magnetic vector must also be projected into the horizontal plane. Therefore the magnetic system is only convenient for wells with low inclination where the sensor plane is almost horizontal.

Highside is defined as a direction 180° from the projection of the gravity vector in the sensor plane. The highside reference is used in directional drilling at inclinations in excess of 5°. The advantages of using a highside reference system are:

1.Ease of visualisation of the target position relative to the relief well.

2.The same reference is used by directional drillers.

3.It is possible to use the highside toolface output directly from the tool without any conversion to the magnetic system.

4.Because of possible DC magnetic interference (e.g. from the target casing) affecting ET (due to local distortion of the Earth's field), this is the preferred mode of operation.

Directions in the highside frame of reference are illustrated in Fig. 1296.

Guideline: for wells with an inclination over 5°, the highside reference system should be used. Local distortion of the Earth's magnetic field due to the target casing will affect ET and may lead to erroneous interpretation of tool response when using the magnetic reference system.

9.3 Background signals

Any signal measured by the AC magnetometers in the absence of the target signal is called "background". In the presence of a target the vector sum of this background signal and the target signal is recorded. If the background signal strength becomes comparable to the target signal strength, it cannot be neglected in the interpretation of target position without causing large errors (see Section 11.1). Background signals are generated by the flow of current from the electrode through the formation or the borehole in the direction of the sonde. However, when the electrode is situated in a direct line with the tool axis in a homogeneous medium and in the absence of a target then contours of constant current density in the sensor plane will be circles centred around the sonde (In a homogeneous medium the surfaces of constant current density around the electrode will be spheres. Intersection of these spheres with the sensor plane gives circular contours.) The flow of current around the sonde is in this situation exactly symmetrical. Consequently the tool does not see a net current in any direction and thus no (background) signal.

Background is caused mainly by wireline curvature, dipping formations with electrical anisotropy and flow of current in the borehole.

9.3.1 Wireline curvature

If the electrode is not in a direct line with the tool axis, e.g. in a build up or drop off section, the contours of constant current density in the sensor plane will be shifted as shown in Fig. 1298, resulting in a net current crossing the sensor plane on one side of the sonde. This would result in a signal with the same direction as the well bearing in a build up section (High Side) and opposite to the well bearing in a drop off section (Low Side).

9.3.2 Formation dip

In an anisotropic formation which is dipping relative to the relief well axis, resistivity anisotropy causes more current to flow in the direction of the dip than across bed boundaries. This is depicted by the length of the current vectors in Fig. 1300. Surfaces of constant current density will thus be ellipsoids around the electrode at the dip angle with two long axes in the bedding plane and a short axis perpendicular to this plane. Contours of constant current density in the sensor plane will be the intersection of these ellipsoids with the sensor plane as shown in Fig. 1300. Again there is a net current on one side of the sensor. Thus the sensor will record a signal with the direction of the relative formation dip.

9.3.3 Borehole currents

At high ratios of formation resistivity to mud resistivity, a considerable part of the emitted currents will flow in the mud column. If this 'borehole current' has not completely died out at the sonde, 91 m (300 ft) below the electrode, it will cause another background signal if the sonde is not exactly centralised. Minimisation of borehole currents is also the reason why the relief well hole diameter should preferably be small. However, this is often neglected due to preferred larger hole diameters required for dynamic kill operations.

9.4 Signal reduction effects

The signal that the EM homing-in system can detect is reduced by a number of effects. These effects can significantly reduce the range at which the homing-in tools can detect a target well and should be considered when planning a relief well programme. Signal reduction effects are particularly problematic when combined with conditions which are liable to give large background signals (i.e. high relative formation dip angles). Three major causes of signal reduction are:

1.Non conductive fluids in the target borehole.

2.Pipe-end effects.

3.High inclination of relief well relative to the target well.

9.4.1 Non conductive fluids

If a drill string in an open hole is the target, the homing-in tool signal will be reduced by non-conductive fluids such as gas and/or oil flowing in the annulus between drill string and open hole. The amount of current which can get onto the target from the current injection electrode is reduced. Nothing can be done to eliminate this problem with the usual downhole current injection system. The only solution may be to use the surface current injection system (Section 9.5).

Non conductive fluid (e.g. oil based mud) in the relief well will severely limit the amount of current that can be injected into the formation, thus also reducing the target signal. Equally background signals due to dipping formation or relief well curvature will proportionally be reduced.

9.4.2 Pipe-end effect

The theory used to evaluate the current distribution on the target pipe for the case of an infinite pipe and a semi-infinite pipe. It is known that when approaching the end of a target (or a break in the pipe) the AC magnetic field is drastically reduced. This is due to the reduced current flowing on the target pipe near/at the pipe end. Interpretation software used both by contractors and Shell assume zero current to be flowing at the pipe end.

The electrode/sensor spacing of 90 m equals an elevation of 56 m for an intersection angle of 52°. From the various signal levels at sensor depth of 56 m the signal reduction due to the pipe-end effect can be derived as a function of the length of pipe below the point of closest approach. It clearly demonstrates how severe the signal reduction can be, even for considerable lengths of pipe.

The reduced signal length also implies that the contrast between maximum and minimum signal level will not be as pronounced and therefore more difficult to interpret (i.e. there will be reduced character in the tool response with depth along hole in the logged interval).

When planning to use electromagnetic homing-in tools in an interval near a pipe end, the esnsitivity of tool response to proximity of the pipe end can be examined by using program ELMODFSP of the 'relief' package. Poor signal character, weak signal strength and AC magnetic field reversals can thus be anticipated.

9.4.3 Non-parallel approach

If the relief and target well paths are not parallel in the interval of interest then for a given distance between sensor and target the signal will be reduced for three reasons:

1.The larger distance between the current injection electrode and the target compared to the parallel case means less current will arrive at the target.

2.The sonde will be closer to the zero current point so that the measured field H, in this situation associated with a target section carrying relatively less current, will be weaker.

3. For coplanar well paths the field H is always in the sensor plane along the x-axis. If the wells are not coplanar, however only a projection of H is measured in the sensor plane.

9.5 Surface current injection

In cases where the detectable AC magnetic field is reduced to the level where background is a problem or where the target well is filled with non-conductive fluid, the surface excitation system (i.e. current injection at the blowout wellhead) can be useful. Background signals produced, although they can be large, will vary in a regular way with logging depth, because no current is injected near the sensor. This is not always guaranteed with electrode current injection because the electrode faces formations with different electrical conductivities at the various depths of operation.

9.5.1 Methods and problems

The power supply equipment can be located either at the blowing well if conditions permit (depending on well conduction/availability of power/location etc.) or at a remote (relief well) location. The dimensions of the system can vary considerably.

The surface current injection electrode can be used in various ways:

1.Attached directly to the wellhead of the target well.

2.Buried in the ground close to the wellhead.

3.If there is no wellhead left but only a crater, it may be possible to lower the electrode several tens of metres deep into the crater.

There are two most important considerations concerning the position of the return electrodes:

1.The return electrodes are to be sufficiently remote from the target wellhead to ensure that the surface formation will not act as a short circuit for the injected current. The current should not preferentially travel from the source electrode directly to the return electrodes along the surface.

2.The return electrodes should be placed such that the current leaking off the target casing returns to surface in a pattern which is symmetric about the line joining the relief well to the target well in order to reduce the background level.

There are a number of practical problems associated with surface current injection:

1.Surface current injection may not be feasible due to inaccessibility of the blow-out location.

2.Land borders and topographic barriers may restrict the location and geometry of the return electrode positions.

3.Cable laying (especially for offshore locations) may present considerable difficulties since not all homing-in companies provide this or any surface current injection equipment as part of their service. In this case power supplies, cables and manpower have to be obtained locally.

The main technical problem concerns the physics of the current injection system. The current leaks off the casing fairly quickly with depth and returns via the formation to the return electrodes. The current on the casing also decreases because of the electromagnetic skin effect. Both these effects give a fairly rapid decrease in current with depth. Thus the source current required is a function of depth and the detection range one needs to achieve.

Guideline: if surface current injection is required consult closely with the homing-in service company to determine the power requirement and availability of such power source. Sensitivities can be calculated using the 'RELIEF' package to determine the surface current requirement. Target current distribution can be determined using ELMODTC and the resulting AC magnetic field over the interval of interest in the relief well can be modelled with ELMODFS2.

9.5.2 Background signal

The main advantage of the surface injection system is the predictability of the background signal produced. The background signal is caused mainly by the currents in the surface cables and to a lesser extent by return currents in the formation. With the symmetrical configuration given in Fig. 1309 equal currents should flow in the wires at surface in opposite directions on either side of the target. These currents produce equal and opposite magnetic fields along a plane half way along and perpendicular to the line joining the two return electrodes. Hence while the relief well path lies approximately on this plane, the background signals due to both cables cancel.

In practice the system is not balanced simply by geometry so that a low adjustable resistor with a high power rating is required in one branch of the circuit to balance the return currents. Also if the horizontal projection of the relief well path over the logging interval is not perpendicular to the line joining the return electrodes, marginal background signals may be induced.

9.6 Remanent magnetic effects on magnetic compass of homing-in tool

As the sensor package of the homing-in tool approaches the target, the compass will be affected by the remanent magnetism of the target pipe. For a strong magnetic pole this perturbation could begin up to 15 m (50 ft) away from the target. The output toolface angle ET would be erroneous causing an error in the target direction. The WELLSPOT tool has a 3-axis accelerometer which permits measurements to be made in the Highside frame of reference hence alleviating the problem of static magnetic interference with the DC magnetometers. The ELREC tool must be fitted with a gyro or steering tool (i.e. accelerometer) system, calibrated for toolface and then run in hole.

10 The "relief" package for interpreting EM homing-in tools

10.1 Process of interpretation

The process of interpretation involves establishing the position of the target well path relative to the relief well path using the direction and intensity (AC magnetic field strength) data from the homing-in tool.

11 Relief well planning

The intersection of a relief well is often determined by the killing requirements, total depth of the target well and surface location of the relief well. The configuration of wells can have detrimental effects on the homing-in tool response. The most important aspect is the detection range that can be expected from homing-in tools with the planned configuration. Two general well configurations, i.e. parallel approach and high relative angle approach, and their effect on homing-in tool range and response are worth discussing.

11.1 EM homing-in tool range

The sensitivity of EM homing-in tools in general is sufficiently high to respond to a change in signal level of about 10 mA/m. Hence, the range is not limited by the system sensitivity. The detection range must be considered in terms of the error associated with determination of the distance and direction as separation from the targert increases, i.e. target signal level decreases. At large distances this error is mainly governed by the effects of background signal and the inherent system noise level. At close range it is determined mainly by the absolute accuracy of the system in measuring the AC magnetic field strength.

EM homing-in tools are generally calibrated at ±10% of the maximum signal. Direction error is determined by the calibration and alignment of the AC and DC magnetometers, the general criterion is to predict direction within ±2°. The inherent system noise is very small and has an effect at very long range only. The magnitude of system noise can be determined by making acquisitions with the tool, whilst not injecting any current into the formation, hence all environmental AC magnetic fields will be zero and only tool noise is present. Measurements of this nature gave a mean value for noise of 10mA/m. The noise direction is assumed to be random, and for a downhole current electrode injecting a 4 A current, this becomes a signal of ±2.5mA/m/A. Because of its random nature, the effect of tool noise on direction should, on average, be limited.

The main source of error is in the interpretation and subtraction of the background signal. The background signal has to be subtracted vectorially and its effect on the interpretation of intensity and direction data will vary with its direction relative to the target signal. The maximum error in distance will be caused when the background is in the same direction or 180° opposite to the field induced by the target. The maximum error in direction will occur when the background is at right angles to the target field.

The main aspects of homing-in tool range can be illustrated by looking at the error in measured distance due to background as a function of the actual distance from the target. Therefore all errors in intensity (absolute calibration, tool noise, background estimate) are summed and expressed as a fraction of the target signal.

11.2 Possible relief well paths

Several possible relief well designs exist.

Well (a) relies predominantly on a pass-by and triangulation of direction data to locate the target well and calibrate the intensity data. The well is subsequently sidetracked for an interception at a deeper objective.

Well (b) initially uses a pass-by to locate the target and calibrate the intensity measurements, followed by a parallel section where intensity is used to determine the separation between wells until an interception is made at a much deeper objective.

Well (c) aims for a direct hit without a pass-by and hence relies on good directional data. While approaching the target at some distance the relief well should be turned such that the AC field direction (HSDIR) is at 270° relative to highside. A good intensity model which cannot be calibrated is used for predicting distance to the target.

The parallel approach of well (d) does not enable triangulation of directional data. This approach relies upon an intensity model which cannot be calibrated for predicting distance to the target.

11.2.1 Parallel approach

Parallel wells have the advantage that maximum signal strength is measured by the Homing-in tool. Hence background signals become less significant. Pipe-end effects are less likely to cause problems since well paths will probably be parallel over a large depth interval so that signal reduction due to proximity of pipe-end is not critical.

The main disadvantages of the parallel approach are that there is no direction calibration of the directional data by triangulation, since there is no point of closest approach as seen in a passing situation. Another disadvantage is, that it is difficult to make an estimate of the background signal directly from measurements. Since the sensor cannot increase its distance from the target except in the drop-off section where background levels will not be the same, background cannot be directly measured. Therefore it is necessary to rely upon modelling to obtain a magnitude and direction for the background signal.

11.2.2 High relative angle approach

The major limitation of the high relative angle approach is the reduction of signal level and pipe-end effects. As such, these effects are not as severe as when there is a background signal present, since this can cause a sharp reduction of the tool range due to the low target signal levels.

The main advantage of a high angle approach is that it allows a good determination of distance, even when pipe-end effects are present. Triangulation of directions in a passing situation permits a good fix of target position. Also the signal intensity profile can be used to determine distance at the point of closest approach. It is no longer necessary to rely upon modelling the magnitude of the recorded intensity as the only method for determining distance.

12 Magnetostatic homing-in theory

12.1 Remanent magnetisation of steel casing/drillstrings

The steel used for casing, drillpipe or accessories (e.g. jars, drillcollars or bits) usually acquires a degree of magnetisation, as a result of the manufacturing process or from subsequent magnetic inspection or mechanical shocks. The simplest model of the field due to this magnetisation is by superposition of magnetic monopoles of either 'North' or 'South' polarity. A North monopole is defined as one in which the lines of magnetic flux flow symmetrically and spherically towards the pole. For a South pole the lines of flux flow similarly away from the pole.

Although free monopoles are never observed in nature, in the vicinity of a pole, for which the opposite pole is far removed, the field will be dominated by that of the near pole, and the far pole may be ignored. When two or more poles are of similar distance from the point of measurement, the field is a linear superposition of the monopole field due to each of the individual poles.

12.2 The overlay technique

When the trajectory of the relief well is virtually parallel to that of the target well, measurements made in the relief well may show the influence of multiple poles along the axis of the target well. The signature of these poles may interfere with one another, and make interpretation difficult. BSP introduced a technique called the 'overlay technique', to enable the determination of the distance to the target in such a situation.

The basic weakness of the overlay method is that there can be no absolute certainty that the monopole signature is placed in the correct position, without a prior knowledge of the distribution of the poles along the target axis.

The overlay method is valid for parallel relief and target wells, irrespective of the distribution, strength or polarity of the poles. For some pole configurations it may be difficult to correctly position the monopole signature over the measured signature. The method should therefore be used with caution, and not without a minimum understanding of the mathematical principles involved.

12.3 Artificial magnetisation of casing/drillstring

One of the main limitations of magnetostatic homing-in is that it cannot be guaranteed that a magnetic pole of significant strength can be found at a specific depth in the target well. In order to remove this uncertainty artificially magnetised sections of casing or drillpipe can be included as part of the casing design or bottom hole assembly. In this way, the depth of the magnetic poles and resulting magnetic field is well defined. Moreover the pole strength can be made much higher than commonly encountered with remanent magnetisation, therefore increasing the detection range.

The procedure for magnetising sections of pipe involves passing a coil, carrying a DC current (of roughly 2A) which results in a pole-strength of the order of 10,000 mWb on the pipe.

Experiments have shown that artificial magnetisation of casing and drillstring components significantly improve their detectability. The detection distance can be increased to approximately 30 m, instead of the 15 m commonly quoted for magnetostatic techniques relying on remanent magnetism.