Вы находитесь на странице: 1из 13

A Compendium of Directional

Calculations Based on the Minimum


Curvature Method
S.J. Sawaryn, SPE, and J.L. Thorogood, SPE, BP plc

Summary Industry Requirements


The minimum curvature method has emerged as the accepted in- Over the years, various algorithms based on the minimum curva-
dustry standard for the calculation of 3D directional surveys. Using ture method have been published for the construction of increas-
this model, the wells trajectory is represented by a series of cir- ingly complex trajectories and tasks, such as interpolation. Be-
cular arcs and straight lines. Collections of other points, lines, and cause these algorithms have emerged piecemeal, they have tended
planes can be used to represent features such as adjacent wells, to use different nomenclatures and mathematical techniques for
lease lines, geological targets, and faults. The relationships be- their solution. The result of this piecemeal development is dupli-
tween these objects have simple geometrical interpretations, mak- cated and inefficient computer code and a poor understanding of
ing them amenable to mathematical treatment. The calculations are the engineering integrity of the systems.
now used extensively in 3D imaging and directional collision
scans, making them critical for both business and safety. However, Safety and Business Criticality. An undetected fault in the cod-
references for the calculations are incomplete, scattered in the ing or use of directional-surveying and collision-scanning software
literature, and have no systematic mathematical treatment. These has been classified as having the potential to cause property dam-
features make programming a consistent and reliable set of algo- age, environmental damage, personal injury, or loss of reputation.5
rithms more difficult. Increased standardization is needed. The integrity of these business- and safety-critical drilling systems
Investigation shows that iterative schemes have been used in is, therefore, a concern. Modern 3D imaging and directional-scanning
situations in which explicit solutions are possible. Explicit calcu- packages execute thousands of calculations for each task. Increased
lations are preferred because they confer numerical predictability automation of the workflows associated with these tasks means that
and stability. Though vector methods were frequently adopted in
most calculations must be taken for granted and will pass un-
the early stages of the published derivations, opportunities for
checked. Sawaryn et al.6 have described a process for managing
simplification were missed because of premature translation to
these systems, and identified a number of requirements related to
Cartesian coordinates.
the equations they use. Specifically, equations must be traceable
This paper contains a compendium of algorithms based on the
minimum curvature method (includes coordinate reference frames, back to the source documentation, which must clearly explain their
toolface, interpolation, intersection with a target plane, minimum purpose, limitations, and use. The general characteristics of the
and maximum true vertical depth (TVD) in a horizontal section, published algorithms can be assessed against these requirements.
point closest to a circular arc, survey station to a target position Angular Change. Like other survey calculation methods, the
with and without the direction defined, nudges, and steering runs). minimum curvature algorithm was originally developed to calcu-
Consistent vector methods have been used throughout with im- late a wells position from directional surveys. The spacing be-
provements in mathematical efficiency, stability, and predictability tween the survey stations was normally 30 to 500 ft. At that time,
of behavior. The resulting algorithms are also simpler and more with typical build rates, the total-angle change over a 100-ft course
cost effective to code and test. This paper describes the practical length would rarely be allowed to exceed 5 and the final inclina-
context in which each of the algorithms is applied and enumerates tion of most of these early wells was below 90. When creating
some key tests that need to be performed. directional well plans, the total-angle change between adjacent
stations in the plan may be considerably larger. These days, in
designer wells, the angular change between two adjacent points on
Introduction a well plan may exceed 90, and the final inclination often exceeds
The first reference to the minimum curvature method is credited to 90. One well7 is recorded as having reached 164.7 inclination.
Mason and Taylor1 in 1971. In the same year, Zaremba2 submitted Many of the published algorithms do not contain an explicit defi-
an identical algorithm that he termed the circular arc method. In nition of the maximum permitted angle change. The multiple so-
the minimum curvature method, two adjacent survey points are lutions arising from periodicity of the trigonometric equations in-
assumed to lie on a circular arc. The arc is located in a plane, the volved makes this a serious concern.
orientation of which is defined by the known inclination and di- Mathematical Behavior. The possibility of multiple solutions
rection angles at the ends. By 1985, the minimum curvature means the results of the calculations may not always be as in-
method was recognized by the industry as one of the most accurate tended, unless great care has been paid to their implementation.
methods, but was regarded as cumbersome for hand calculation.3,4 Some algorithms employ iterative schemes so that even if the
The emergence of well-trajectory planning packages to help man- scheme converges, there is no guarantee that it converges to the
age directional work in dense well clusters increased its popularity. correct solution. Ideally, iterative schemes should be accompanied
It was natural to use the same model for both the surveys and the by proof of convergence. At the very least, they should be thor-
segments of the well-plan trajectories. Today, with the widespread oughly tested over some specified range of variables. Additionally,
use of computers, computational power is no longer an issue, and there are cases for which no solution exists and extra code is
the method has emerged as the accepted industry standard. needed to trap this condition. Explicit expressions are more pre-
dictable and usually confer advantages in speed and maintainabil-
ity of the computer code.
For certain values (for example, in geometrically straight hole),
Copyright 2005 Society of Petroleum Engineers
expressions may be indeterminate. One solution, adopted by Za-
This paper (SPE 84246) was first presented at the 2003 SPE Annual Technical Conference remba,2 is to define a suitably small number at which the expres-
and Exhibition, Denver, 58 October, and revised for publication. Original manuscript re-
ceived for review 27 January 2004. Revised manuscript received 15 January 2005. Paper
sion jumps abruptly to the asymptotic value. However, this can
peer approved 29 January 2005. give rise to random differences between software packages.6 A

24 March 2005 SPE Drilling & Completion


better method is to develop series expressions that enable a smooth vertical coordinates that comprise a right-handed set, as shown in
transition to be maintained. Fig. 1. A point N,E,V can be represented by the vector p in Eq. 2.


Common Constructs. The absence of consistent mathematical
methods and nomenclature may hide common constructs and po- N
tential simplifications in the coding of the algorithms. For ex- p= E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (2)
ample, an equation in the form CA sin()+B cos() appears in V
many of the geometric constructions associated with the minimum
curvature method. This equation can be solved explicitly for , and A unit direction vector t can be represented in terms of the local
several mathematically equivalent forms exist. Zaremba2 proposed inclination and azimuth , as shown in Eq. 3. The inclination and
the form (see Eq. 1) that is used throughout this paper: azimuth values can be calculated from the vectors components
using the expressions in Eqs. 4 and 5.


1
N sin cos
A A2 + B2 C 2 2
= 2 tan1
. . . . . . . . . . . . . . . . . . . . . (1) t= E = sin sin . . . . . . . . . . . . . . . . . . . . . . . . . . . (3)
B+C
V cos
When presented in this way, it can be seen that no real solutions

1
exist if C2>A2 + B2. This inequality has simple geometric interpreta- N2 + E22
tions and several examples of its use are highlighted in this paper. = tan1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (4)
V
Inconsistent nomenclature also leads to implemenational diffi-
culties. Review shows the nomenclatures used in the literature are
neither consistent with each other, nor consistent with accepted
mathematical practice. One example is the definition of the normal
= tan1
E
N
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (5)

vector that mathematical convention has pointing towards the cen- By using this reference frame, an implicit assumption is made
ter of curvature. This is opposite to the convention used in the that the Earth is flat. For moderate distances from the origin, this
earlier drilling literature. Because of the expansion in directional- assumption holds. For larger distances, such as those encountered
drilling applications, symbols inevitably conflict. The Industry in extended-reach wells, the Earths curvature is important and
Steering Committee on Wellbore Survey Accuracy has proposed corrections to the coordinates must be made. Williamson and
some standards,8 but with limited scope. We assert that the SPE Wilson10 discuss the matter in detail.
documentation standards9 associated with this subject area are no Borehole Reference Frames. Two reference frames are asso-
longer adequate and need revising. ciated with the borehole, see Fig. 2. The first frame is formed by
the highside, rightside, and tangent unit vectors h, r, and t, respec-
Directional Calculations tively. These form a right-handed, mutually orthogonal set. In
curved hole, the second frame comprises the normal, binormal, and
A consistent vector notation is used throughout this paper. This tangent unit vectors b, n, and t, respectively. These also form a
simplifies the development of the 3D equations and improves the right-handed, mutually orthogonal set. The angle between the
clarity and presentation of the results. For convenience, the main highside vector h and normal vector b is the toolface angle, .
vector operations are summarized in the Appendix. In some cases, The highside, rightside, and vertical unit vectors are represented in
series expansions have been used to ensure the smooth transition Eqs. 6, 7, and 8. Expressions for the normal and binormal vectors
of an expression into what would otherwise be an indeterminate b and n can be found in the Appendix (Eqs. A-8 through A-12).


form. The thresholds at which the series approximations should be
used depend on the machine precision. The constants used in this cos cos
paper assume calculations are good to at least nine significant h= cos sin , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (6)
digits. The angle subtending the arc may assume values such that
0<. Throughout, it is also assumed that the start and end sin
points of the arc are not coincident. Until such time as the stan-
dards are officially revised, we have chosen to maintain common-
ality with earlier papers on this subject and use a vector b pointing
r=
sin
cos , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (7)
0


away from the center of the arc. For comparison with mathematical
texts, the normal vector is, therefore, b. 0
Reference Frames. Coordinate Reference Frame. The tradi-
tional reference frame for directional work uses north, east, and v= 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (8)
1

Dogleg Severity. Dogleg severity is a measure of the change in


inclination and/or direction of a borehole, see Fig. 3. The change

Fig. 1North, east, and vertical coordinate reference frame. Fig. 2Borehole reference frames, including the toolface angle .

March 2005 SPE Drilling & Completion 25


Fig. 4The geometry of the minimum curvature between two
adjacent survey points.
Fig. 3A dogleg defined by the two direction vectors t1 and t2.
sented in Horner13 form to minimize both the number of arithmetic
operations and the propagation of errors:
is usually expressed in degrees per 100 ft of course length in
oilfield units3 and degrees per 30 m in metric units. Dogleg se-
verity is used to determine stress fatigue in drillpipe, casing wear,
and casing design loads. It can also be a limiting factor in casing
f 1 +
2
12

1+
2
10
1+
2
168
1+
18

312
. . . . . . . . . (12)

running and directional-drilling operations. For the minimum cur- There is a second possible solution for Eq. 9, which is equal to
vature method, the expression for the dogleg severity takes the (2). The measured depth between the survey stations must be the
form of (18,000* /)/(D2D1) in oilfield units. The difference in same in both cases, implying the second solution has a greater cur-
measured depths, D2D1, between the points is referred to as the vature, see Fig. 5. When calculating directional surveys, the density of
course length, S12. survey stations, behavior of the bottomhole assemblies, and knowl-
Most expressions found in the literature involve the calculation edge of the toolface settings means this situation is of no practical
of an arc cosine. These have been used even though it is recog- concern. This may not be the case for adjacent points in a well-plan
nized that cosines of small angles are more difficult to handle than trajectory, which may be separated by considerable distances.
sines of small angles. Eq. 9, developed by Lubinski,3 does not
make any assumption about the actual path of the wellbore, yet it Interpolation. This is often required to identify the coordinates of
is mathematically equivalent to the expression traditionally used in a particular point, for example, p* on a trajectory, see Fig. 6. In all
the minimum curvature method. An expression for tan(/2) is cases, the problem reduces to one of interpolation or extrapolation
readily developed from it: on an arc, defined by the positions p1 and p2 and directions t1 and
t 2 of its end points. The algorithms presented here may be used for


1
both functions. The interpolation may be driven by one of several
2 1 2 1 2 parameters, such as measured depth, subtended angle, inclination,
= 2 sin1 sin2 + sin1 sin2 sin2 . azimuth, northing, easting, or vertical ordinate.
2 2
. . . . . . . . . . . . . . . . . . . . . . . . . . . (9) Before discussing each of these cases, it is worth reviewing the
properties of a circular arc. The subtended angle, inclination, and
Because the trigonometrical identity and computational advan- azimuth of a point on a circular arc can be determined solely by the
tages of Eq. 9 were recognized, it is surprising that it has not been attitude of the circle in the coordinate reference frame. Knowledge
adopted sooner. The dogleg severity can be related to both the of the size of the circle enables course lengths to be determined.
radius of curvature and the curvature, , of the arc, by use of the Finally, it is only if a north, east, or vertical ordinate is needed that
relationships (see Eq. 10): the absolute position of the circle in the reference frame must
be defined.
1 * Interpolation on Measured Depth. The association of events
= = = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (10)
R S12 S* or observations, such as formation tops and overpulls with points
on the wellbore, is a common requirement. Interpolations on mea-
Survey Calculation. Accurate determination of wellbore position sured depth determined from the pipe tally are, therefore, the most
is critical to well placement, collision avoidance, reservoir mod- common interpolation mode. If S* is the course length along the
eling, and equity determination. Though the accuracy of the mini-
mum curvature method is acknowledged, Stockhausen and Lesso11
showed that modern drilling practices could introduce systematic
errors, even with survey intervals as frequent as 100 ft.
The position of the next survey point, p2, is calculated from
p1 by use of Eq. 11, see Fig. 4. The shape factor f() equals
tan(/2)/(/2). Details are summarized in the Appendix (Eqs. A-6
through A-17).

p2 = p1 +
S12 f
2 sin1 cos1 + sin2 cos2

sin1 sin1 + sin2 sin2 . . . . . . . . . (11)
cos1 + cos2
Straight-Hole Conditions. When equals 0, the shape factor is
mathematically indeterminate, so for <0.02 radians, the series
expansion (Eq. 12) should be used instead.12 The series is pre- Fig. 5The two possible mathematical solutions.

26 March 2005 SPE Drilling & Completion


cos(*). The solution for * is determined using Eq. 1. In this case,
the constants A, B, and C are given by Eqs. 16, 17, and 18, re-
spectively. Choose the smallest root, unless it is less than or equal
to both 1 and 2, in which case, the largest root should be chosen.
A = sin1 cos* 1, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (16)
B = cos1, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (17)
C = cos*. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (18)
As the subtended angle was originally determined from an
azimuth on the arc, a solution must exist. Finally, the position p*
is determined using the minimum curvature equation (Eq. 11).
Generally, if C2>A2+B2, the orientation of the arc is too shallow
for the inclination at any point on it to reach the desired value. For
details, see the Appendix (Eqs. A-27 through A-29).
Straight-Hole Conditions. Azimuth varies linearly with mea-
sured depth in near-straight-hole conditions. For angle <104
radians, Eq. 19 should be used to determine the corresponding
Fig. 6Interpolation at a point p* on a circular arc. measured depth. If the azimuth values in either the numerator or
denominator straddle north, the minimum angular difference
arc at which the properties are required, the relationships * should be used. The remaining properties can be determined by
(S*/S12) and (*)(1S*/S12) can be used to reduce the interpolating on this depth:
interpolation on measured depth to the interpolation on subtended
angle (see Eq. 13).
Interpolation on Subtended Angle. Eq. 13 enables the direc-
S* S12 * 1
2 1
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (19)

tion vector, t*, at a point p* on an arc to be determined solely from Interpolation on Inclination. The determination of measured
the direction vectors of its start and end points and the angle depths corresponding to some inclination range appears in the
subtended from the first point p1 to the point of interest. Refer to automation of calculations for hole cleaning and rock mechanics.
the Appendix (Eqs. A-18 through A-26). First, check that both points 1 and 2 are not equal to /2. In this
condition, the arc would lie in the horizontal plane and no solutions
sin * sin*
t* = t1 + t . . . . . . . . . . . . . . . . . . . . . . . . . . (13) are possible. In the Appendix (Eq. A-30), it is shown that the
sin sin 2 subtended angle * can be calculated with Eq. 1. In this case, the
Many of the algorithms presented in this paper involve the constants A, B, and C are given by Eqs. 20 through 22:
determination of the subtended angle * as a first step. The rela- A = cos2 cos cos1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (20)
tionship in Eq. 13 provides a convenient means of determining the
other parameters at the point, once the subtended angle has been B = sin cos1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (21)
found. For example, once t* is known, the corresponding point p*
C = sin cos* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (22)
can be calculated using the minimum curvature equation (Eq. 11).
Straight-Hole Conditions. When the subtended angle, , equals If C2 > A2 + B2, then the orientation of the plane containing the
zero, both terms in Eq. 13 are indeterminate. For <0.02 radians, arc is too close to the horizontal plane to enable the desired incli-
these terms that contain factors of the form sin(c)/sin may be nation to be reached; no solutions exist. The corresponding azi-
expanded14 in another Horner series (see Eq. 14): muth value can be determined from Eq. 23:
sinc
sin

1 c2
c + 2 c
6 6
+ 2 c
7
360
+ c2
1 c2
+
36 120
* = tan1 sin1 sin1 sin * + sin2 sin2 sin*
sin1 cos1 sin * + sin2 cos2 sin*
.

. . . . . . . . . . . . . . . . . . . . . . . . . . (23)
31 7 1 c2
+ 2 c + c2 + c2 Straight-Hole Conditions. Inclination varies linearly with mea-
15,120 2,160 720 5,040 sured depth in near-straight-hole conditions. For the subtended

+ c2 127
604,800
+ c2
31

90,720
+ c2
7
43,200 angle, < 104 radians, Eq. 24 should be used to determine the
course length. The remaining properties can be determined by
interpolating on this depth:

+ c2 1
+
c2
30,240 362,880
. . . . . . . . . . . . . (14) S* S12 * 1
2 1
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (24)
It should be noted that this interpolation mode is impossible Interpolation at a Plane. A plane can be used to represent
when the subtended angle is identically 0, unless the constant, c, many geological features, such as formation horizons and faults. In
can be defined in some other way, such as by use of the ratio of the 3D visualization tools, collections of interlocking planes are used
course lengths. to represent complex geological features. Imaginary planes can be
Interpolation on Azimuth. Occasionally, it is necessary to used to represent lease boundaries, or the north, east, and vertical
truncate a well-plan trajectory at a depth so that it is lined up with limits, on directional plots. The point p* at which the well meets
a specified direction. First, check the condition sin1 sin2 sin(2 or crosses these features, is of great practical interest, see Fig. 7.
1)0 to determine that the arc does not lie in the vertical plane The plane is uniquely defined by its normal vector m and any point
and that a solution exists. The subtended angle is then determined px on it.
by use of Eq. 15: There are five possible relationships between the arc and

* = tan1 sin1 *sin sin1


sin* 2sin2 + sin1 *sin1 cos .
the plane:
An infinite number of intersectionswhen the plane contain-
ing the arc and the target plane are parallel. This condition should
. . . . . . . . . . . . . . . . . . . . . . . . . . (15)
be tested first by establishing if both (mt1) and (mt2) are equal
The Appendix shows that Eq. A-28, used to calculate the in- to 0.
clination * from * and *, is of the form CA sin(*)+B Two intersectionswhen the arc completely cuts the plane.

March 2005 SPE Drilling & Completion 27


Fig. 7The intersection (extrapolation) of a circular arc and a
plane at a point p*.
Turning Point. In horizontal wells, the trajectory is steered to
remain in the reservoir section, successively building and dropping
One intersectionwhen the arc just touches the plane. inclination to avoid breaching the bottom or top of the pay zone.
No intersectionswhen the curvature of the arc is too large. The calculation of the turning points, at which the well becomes
If curvature is very small, then, for practical purposes, the prob- horizontal, is of interest from a reservoir perspective (see Fig. 8).
lem is reduced to the intersection of a straight line with a plane. Because these points represent the maximum and minimum verti-
The Appendix (Eqs. A-31 to A-35) shows that the subtended cal depths, they are also needed to determine the numerical range
angle, *, to the plane can be calculated with Eq. 1. In this case, of axes in well plots.
the constants A, B, and C are given by Eqs. 25 through 27: This case is equivalent to interpolating on an inclination. In this
A = m t1sin, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (25) special case, direct determination of the azimuth * at the turning
point is possible according to Eq. 31:
B = m t1cos m t2, . . . . . . . . . . . . . . . . . . . . . . . . . . . . (26)

C=
m px p1 sin
S12
+ m t1cos m t2. . . . . . . . . . (27)
* = tan1 cos1 sin2 sin2 cos2 sin1 sin1
cos1 sin2 cos2 cos2 sin1 cos1
. . . . . . . (31)

In deriving Eq. 31, it is assumed that 1<2, and so represents


If C2>A2+B2, the curvature of the arc is too large to intersect the a minimum inclination. Should 1>2, then the inclination is a
plane and no solutions exist. If C2A2+B2, then the arc just maximum. In this case, the direction of t* will change by , and
touches the plane and there is only one solution. For C2<A2+B2, the azimuth becomes (*+). The course length S* to the turning
there are two intersections and, in this case, the subtended angles point can then be calculated from the relationship S*S12*/.
to both of them may be less than . The two solutions correspond Finally, its position, vector p*, can be calculated using the t1 and
to the plus and minus signs in Eq. 1. This situation may be en- t* vectors with the minimum curvature equation (Eq. 11); refer to
countered when landing in a pay zone and the assembly is not the Appendix (Eqs. A-44 through A-46).
building at the desired rate. Using this interpolation mode, it is
possible to determine if and at what point the bottom of the zone Position at Target Defined. Hogg and Thorogood15 described
will be breached and the point at which the well is expected to expressions for the minimum curvature and minimum distance
re-enter. The lost-production interval can then be calculated directly. from a point p1 on a wellbore, with direction t1 to a target p3 (see
Straight-Hole Conditions. When the subtended angle equals Fig. 9).
0, the above solution is indeterminate. For <104 radians, a series Their algorithms were iterative and depended on the explicit
expansion (see Eq. 29) is used to determine the measured depth; determination of the center of the curvature of the arc, which
see the Appendix (Eqs. A-36 through A-43). No solution is pos- causes problems for small angle changes. The Appendix (Eqs.
sible if the line is parallel to the plane when (mt1) equals 0: A-47 through A-64) shows how the minimum curvature and mini-
m px p1m t2 m t1 mum distance can be determined explicitly, without reference to
= , . . . . . . . . . . . . . . . . . . . . . (28) the center of the curvature. For a fixed radius of curvature, there
S12m t12

S*
m px p1
m t1


1 1 . . . . . . . . . . . . . . . . . . . . . . (29)
2
Orientation of the Target Plane. The normal vector m, defining
the orientation of the target plane, can be constructed in terms
of the dip angle, , and dip azimuth, , of the plane according to
Eq. 30:

m= sin cos

sin sin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (30)
cos
North, East, and Vertical Interpolation. The interpolations on
north, east, or vertical ordinates are particular cases of the more
general expression. The values of the vectors px and m to use in
each of these are listed in Table 1. For example, if the well is
building and approaching an eastern lease line, use the vectors
listed under the heading East, with E* set equal to the numeric Fig. 8The turning point p* is the point at which the well be-
value of the eastern boundary limit. comes horizontal.

28 March 2005 SPE Drilling & Completion


Fig. 10For a fixed radius of curvature, there are two possible
trajectories from p1 to p3.

Fig. 9In general, one curved and one straight section are re- sponding to the minimum distance to the target, are calculated with
quired if only the position p3 of the target is defined.
Eqs. 35 and 36, respectively:
1 1
are two possible trajectories from p1 to p3, shown by the solid and
dashed lines in Fig. 10. All the trajectories lie in a plane. The solid = 2 2R2 222 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (35)
line shows the minimum distance to the target. The other trajec- and


tory, shown by the dashed line, requires an angle change greater
than .
The changes in angle can be categorized according to the po- = 2 tan1 1
. . . . . . . . . . . . . . . . . . . . . . . . . (36)
sition of the target point p3, relative to the starting point p1 and the 2R 2

2 2

direction t1. These categories are represented by the Regions A


through F in Fig. 11. Targets falling in Region A can be hit with The course length is then calculated as S12R. In straight
angle changes less than /2 (90); B, with angle changes less than hole, and, as long as R is finite, Eq. 35 reduces to the
(180); C, with angle changes less than 3/2 (270); and D, with distance between the points p3 and p1. The straight-hole case,
angle changes less than 2 (360). Areas E and F cannot be hit at therefore, degenerates to a straight line, rather than an arc with an
the prescribed buildup rate. infinitely large radius, conferring stability to the calculation. Fi-
The appropriate region is determined by calculating the dis- nally, t2 is calculated using Eq. 37:
tance, , between the points p1 and p3; the perpendicular distance,
, from p1 in the direction t1; and the distance, , normal to t1, in S12 f
p3 p1 t1
Eqs. 32, 33, and 34, respectively; refer to Fig. 12: 2
t2 = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (37)

2 = p3 p1 2, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (32)
S12 f
2
+
= p3 p1 t1, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (33)
Minimum Curvature to Target. If R22+2, then the target
1 must lie in Region E and the build rate is insufficient to hit it. To
= 2 22. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (34) hit the target, the build rate must be increased to the critical radius
of curvature Rc, given by Eq. 38. The corresponding angle change,
From Fig. 11 it can be seen that if 0 and 2R, then , c, is given by Eq. 39.
which is outside the imposed limit on the angle change. With the
restriction 0<, only targets in areas A, B, and E apply. 2
Minimum Distance to Target. If R2<2+2, then the tangent- Rc = 1
, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (38)
section length, , is greater than 0. Because of the imposed re- 2
2
22
striction 0<, the target must be in Region A or B (see Fig. 11).
The values of the tangent-section length and angle change, corre-

1


2 2 2
c = 2 tan1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (39)

Fig. 11The regions A through F are defined by the total


change in angle. Fig. 12Graphical realization of the and variables.

March 2005 SPE Drilling & Completion 29


Fig. 13In general, two curved and one straight section are
required if both the position p4 and direction t 4 at the target
are defined.
Fig. 14The iterative scheme used to calculate the trajectory, if
both the position and direction at the target are defined.
The course length can be calculated as S12Rcc. Finally, t2 is
calculated from Eq. 37 with the tangent-section length, , equal
to 0. Four iterations are usually sufficient to reduce the error, ,
below 105 radians. Note that on completing the calculation, the
Position and Direction at Target Defined. Advances in survey- direction vectors of the second build-and-hold trajectory must be
ing and geosteering techniques have enabled multiple targets to be reversed before use.
penetrated as a matter of course. The targets may be at different From a safety-critical-systems perspective, neither the above
geological horizons or different fault blocks in the same horizon. scheme nor that proposed by Liu and Shi16 are completely satis-
In these cases, the wells trajectory must be lined up and its di- factory. In neither case is convergence proved, and no definitive
rection on entry to or exit from the target must be defined, as well statement is made regarding the conditions under which no solu-
as its position. The trajectory can no longer be achieved with a tions exist (e.g., when the radii of curvature are too large to hit the
simple build-and-hold profile. An additional curved section must target). Further work is required.
be added, and in general, the trajectory is 3D (see Fig. 13). How-
ever, because the target can be specified so that the trajectory lies Closest Approach. The calculation of the closest distance of a
completely in a plane, this calculation can also be used to design point px to a circular arc (see Fig. 15) is the key to the construction
nudge profiles to increase well separation directly beneath a well of the normal plane collision-avoidance diagram described by
cluster. Liu and Shi16,17 described an iterative scheme using co- Thorogood and Sawaryn.18
ordinate transforms for the solution of the equations. Different The closest approach is reached when the vector (pxp*) is
calculations were used for each of the two arcs. normal to the curve. The subtended angle, *, to this point is
We offer an alternate iterative solution on the basis of the calculated using Eqs. 43 through 45:
geometrical symmetry of the problem and the minimum distance- 1 = px p1 t1, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (43)
to-target scheme described earlier in this paper. The advantage of
this scheme is that only the subtended angles, 1 and 2, of the 2 = px p1 t2, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (44)


arcs need to be determined at each iteration. Inspection of Fig. 13
shows that, in general, each of the two sets of points (p1, p2, p3 and 1 cos 2 S12
* = tan1 1 + . . . . . . . . . . . . . . (45)
p2, p3, p4) are geometrically similar. This suggests applying the sin
minimum distance-to-target algorithm alternately for each of the
sets of points. The corresponding position p* on the arc is determined by
Let p1,j and p4,j be the targets and 1,j and 2,j be the subtended interpolating on the subtended angle *, and then using the mini-
angles at the jth iteration. To start the scheme, the subtended angle mum curvature equation. Finally, the distance is calculated from
1,0 is calculated using the minimum distance-to-target algorithm the magnitude of the vector (pxp*). Mathematical and practical
between the points p1 and p4. This corresponds to Trajectory 1 in difficulties arise when the distance to the point is of the same
Fig. 14. magnitude or exceeds the radius of the curvature of the arc. Further
A new target, p1,1, is calculated with Eq. 40: discussion on this topic may become available in future publications.

p1,j+1 = p1 + R1 tan
1,j
2
t1. . . . . . . . . . . . . . . . . . . . . . . . . . . . (40)

The subtended angle, 2,0, is now calculated for the second arc,
using the points p4 and p1,1. When performing the calculation, the
direction of the borehole at p4 must be set to t4, to calculate the
correct angle. This corresponds to Trajectory 2 in Fig. 14. A
new target, p4,1, is calculated with Eq. 41, which completes the
first iteration:

p4,j+1 = p4 R2 tan
2,j
2
t4. . . . . . . . . . . . . . . . . . . . . . . . . . . . (41)

The next iteration is started by calculating the subtended angle,


1,1, using the points p1, and p4,1, and so on. Iteration continues
until the desired precision is achieved. A suitable criterion for
convergence is given by Eq. 42:
1

1,j+1 1,j + 2,j+1 2,j


2
.
2 2
. . . . . . . . . . . . . . . . . . . (42) Fig. 15The closest approach of a point px to a circular arc.

30 March 2005 SPE Drilling & Completion


Straight-Hole Conditions. Eq. 45 is indeterminate in straight of these expessions are commonly referred to as the build rate and
hole when equals 0. For <104 radians, small-angle approxi- walk rate, respectively. For further details, see the Appendix (Eqs.
mations are used, and Eq. 46 is used to calculate the course length, A-79 through A-81). These two components satisfy Wilsons19
S*, directly: expression (Eq. 53) for the total curvature :

S*
S121
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (46)
* = cos* , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (51)


1 2 + S12
sin*
The point p* is calculated by interpolating on measured depth, * = , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (52)
sin*
assuming straight-hole conditions. For these very small, subtended
angles, the value of the shape factor f(*) in the minimum curva- and
ture equation equals unity. Again, the distance is calculated from 1
the magnitude of the vector (pxp*).
= *2 + *2 sin2*2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (53)
Toolface Angle. Liu and Shi16 provided useful expressions for the Eqs. 51 and 52 can be used to derive the equivalents (Eqs. 54
toolface at the start and end of the arc, in terms of the inclination and 55) of Liu and Shis16 expressions. The representation of the
and azimuth values at its ends. The Appendix (Eqs. A-66 through azimuthal component is simplified:


A-79) shows how the general vector equation provided by
Thorogood and Sawaryn18 may be expanded in terms of the incli- cos *cos1 cos* cos2
* = . . . . . . . . . . . . . . . . (54)
nation and azimuth values at any intermediate point on the arc, as sin sin*
well as at its ends (Eq. 47):
and
tan * =
sin*sin2 cos* sin2 * + sin1 cos *sin* 1
cos *cos1 cos* cos2
.
* = sin2 cos* sin2 * + sin1 cos *sin* 1
sin sin*
. . . . . . . . . . . . . . . . . . . . . . . . . . (47) . . . . . . . . . . . . . . . . . . . . . . . . . . (55)
At the ends, this expression reduces to those presented by Liu Straight-Hole Conditions. Again, for <104 radians, small-
and Shi.16 Eqs. 48 and 49 have a small advantage because they are angle approximations to Eqs. 54 and 55 are used, and Eqs. 56 and
not singular when the inclination at either of the ends is zero. 57 are used to calculate the curvatures. If the azimuth values in Eq.
Dividing both numerator and denominator by a factor that can be 57 straddle north, the minimum angular difference should be used:
zero causes the singularity.
* 1 , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (56)

1 2
sin2 sin2 1 *
1 = tan 1
. . . . . . . (48) S* S12
sin2 cos1 cos2 1 sin1 cos2
and

* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (57)
and
1


1 2
sin1 sin2 1 *
2 = tan 1
. . . . . . . (49) S* S12
sin2 cos1 sin1 cos2 cos2 1
Implementation and Testing
Straight-Hole Conditions. For <104 radians, small-angle
The algorithms in this paper are presented in logical order with the
approximation to Eq. 47 is used, giving Eq. 50. The toolface is
later, more-complex cases using results of earlier ones. It is rec-
constant over the arc and is undefined when 0. If the azimuth
ommended that the routines are coded and tested in this order. All
values in Eq. 50 straddle north, the minimum angular difference
the equations required for successful implementation are contained
should be used:
in the body of the paper.

tan* sin1 2 1
2 1
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (50)
Once coded, a good test procedure is to calculate values in two
different ways. For example, a point can be interpolated using each
of the following: measured depth, inclination, and azimuth deter-
mined from the previous calculation in cyclic order. The results
Curvature. The total curvature, , is a constant 1/R on a circular
should be identical. The four-point trajectory presented in Table 2
arc (see Fig. 16). Liu and Shi16 presented expressions for the
is constructed with both the wellbore position and direction at the
inclination and azimuthal components of the curvature at any point
target defined. This example can be used to test most of the algo-
on an arc, in terms of the toolface angle at its start.
rithms presented here. Station numbers with alphabetical suffixes
Using vector methods, the inclination and azimuthal curvatures
indicate interpolated points. Williamson8 presents other trajecto-
can be shown to be Eqs. 51 and 52, respectively. The signed values
ries that may be used as tests in both oilfield and metric units.

Conclusions
1. Previously unpublished algorithms have been presented for
small-angle approximations associated with a circular arc, the
determination of a turning point, a general expression for tool-
face angle, and the minimum distance to a target with and with-
out the direction at the target defined.
2. Vector methods are useful tools for 3D-directional calculations
and often result in simpler expressions compared to other
means. Their use is recommended.
3. A standard nomenclature is required for all directional work that
is compatible with other related subject areas, such as survey-
instrument design and terrestrial surveying. Consistency with
accepted mathematical conventions in vector calculus should
also be reviewed.
4. The point-to-target algorithm, with both position and direction
Fig. 16The principal curve, inclination, and azimuthal components. at the target defined, is the only construction presented in this

March 2005 SPE Drilling & Completion 31


paper for which an explicit solution has not been found. Also, Subscripts and Superscripts
the conditions under which no solutions are possible should be c critical value
identified (e.g., when the radii of curvature are too large to hit j iteration counter
the target). Further work is needed. x position of a defining point
5. Increasingly complex trajectory plans emerging from work on
1,2,3,4 first, second, third or fourth point, arc, or property
designer wells means segments of the well plan may exceed
180 and the alternative solutions may need to be considered. inclination component
6. The simple representation afforded by the circular-arc construct azimuthal component
allows for a consistent treatment of the associated mathematical * component to be determined
operations. Representation of the wellbore path in a different
form (for example, a spline or polynomial) would necessitate Acknowledgment
rederivation of all the constructs in this paper. This may not be The authors thank BP plc for permission to publish this paper.
a simple task.
References
Nomenclature 1. Mason, C.M. and Taylor, H.L.: A Systematic Approach to Well Sur-
A constant veying Calculations, SPEJ (June 1971).
b negative unit normal vector 2. Zaremba, W.A.: Directional Survey by The Circular-Arc Method,
B constant SPEJ (February 1973) 5; Trans., AIME, 255.
c a function of the ratio of the course lengths S*/S12 on 3. Bull. D20, Directional Drilling Survey Calculation Methods and Ter-
an arc minology, first edition, API, Dallas (December 1985).
C constant 4. Walstrom, J.E., Harvey R.P., and Eddy H.D.: A Comparison of Vari-
ous Directional Survey Models and an Approach To Model Error
D measured depth, L, ft
Analysis, JPT (August 1972) 935.
E easting, L, ft
5. Sawaryn, S.J., Sanstrom, W., and McColpin, G.: The Management of
f geometrical shape factor Drilling Engineering and Well Services Software as Safety Critical
h unit highside vector Systems, paper SPE 73893 presented at the 2002 SPE International
m unit vector normal to a plane Conference on Health, Safety, and Environment in Oil and Gas Explo-
n unit binormal vector ration and Production, Kuala Lumpur, 2022 March.
N northing, L, ft 6. Sawaryn, S.J. et al.: Safety Critical Systems Principles Applied to
p position vector in N,E,V coordinates, L, ft Drilling Engineering and Well Services Software, paper SPE 84152
r unit rightside vector presented at the 2003 SPE Annual Technical Conference and Exhibi-
R radius of curvature, L, ft tion, Denver, 58 October.
S course length, L, ft 7. Delivery UpdateAutotrack Milestones, Baker Hughes Inteq,
Tananger, Norway (2002) 18.
t unit direction vector
8. Williamson, H.S.: Accuracy Prediction for Directional Measurement
v unit vertical vector in N,E,V coordinates, L, ft While Drilling, SPEDC (December 2000) 221.
V vertical, L, ft 9. SPE Letter and Computer Symbols Standard, SPE, Richardson,
z dummy angle, radians Texas (1993).
subtended angle, radians 10. Williamson, H.S. and Wilson, H.F.: Directional Drilling and Earth
length of the tangent section, L, ft Curvature, SPEDC (March 2000).
angle between the binormal and vertical vectors, 11. Stockhausen, E.J. and Lesso, W.G.: Continuous Direction and Incli-
radians nation Measurements Lead to an Improvement in Wellbore Position-
angular-error tolerance, radians ing, paper SPE/IADC 79917 presented at the 2003 SPE/IADC Drilling
substituted variable Conference, Amsterdam, 1921 February.
substitution for a dot product, L, ft 12. Thorogood, J.L.: Well Surveying Data, World Oil Magazine (April
inclination angle, radians 1986) 100.
13. Knuth, D.E.: The Art of Computer Programming Volume 2: Seminu-
curvature, radians/L, radians/ft
merical Algorithm, third edition, Addison Wesley (1997) 485488.
substitution for a dot product, L, ft
14. Wolfram Research, Mathematica 4 Standard Add-On Packages, Wol-
toolface angle, radians fram Media Inc., Champaign, Illinois (1999) 34.
azimuth angle, radians 15. Hogg, T.W. and Thorogood, J.L.: Performance Optimization of Steer-
substitution for a dot product, L, ft able Systems, ASME Energy Resources Technology Conference,
a difference in a parameter New Orleans (January 1990) 4958.
dip angle, radians 16. Liu, X. and Shi, Z.: Improved Method Makes a Soft Landing of Well
dip azimuth, radians Path, Oil & Gas J. (October 2001) 48.

32 March 2005 SPE Drilling & Completion


17. Liu, X. and Jun, G.: Description and Calculation of the Well Path with Finally, recalling the trigonometric identity tan(/2)(1
Spatial Arc Model, Natural Gas Industry, Beijing (2000) 20, No. 5, cos)/sin, and that S12R, we obtain the now familiar expres-
4447. sion for the minimum curvature equation:


18. Thorogood, J.L. and Sawaryn, S.J.: The Travelling-Cylinder Diagram:
S12
A Practical Tool for Collision Avoidance, SPEDE (March 1991) 31. p2 = p1 + tan t + t . . . . . . . . . . . . . . . . . . . . . . . . . (A-15)
19. Wilson, G.J.: An Improved Method for Computing Directional Sur- 2 1 2
veys, JPT (August 1968) 871. Straight-Hole Conditions. For small angles, tan(z) can be ex-
20. Weatherburn, C.E.: Elementary Vector Analysis, G. Bell and Sons, panded in a Taylor series,21 Eq. A-16. Using the first five terms of
London (1967). the expansion, the factor tan(/2)/(/2) can be put into Horner12,13
21. Abramowitz and Stegun: Handbook of Mathematical Functions, Dover, form, giving Eq. 12 in the body of the paper:
New York City (1972) 75.
z3 2z5 17z7 62z9
Appendix tanz = z + + + + + . . . . . . . . . . . . . . . . . (A-16)
3 15 315 2835
Summary of Vector Methods. Weatherburn20 presents details of
all vector methods used in this paper. The key constructions are the For small angles, the shape factor can be treated as unity and
dot, or scalar, product () and vector cross product (). Let a1, a2, Eq. A-15 reduces to Eq. A-17, which is recognized as the balanced
a3; b1, b2, b3; and c1, c2, c3 be the N,E,V components of vectors a, tangential-survey calculation method3:
b, and c, respectively. Let be the angle between the a and b S12
vectors, and n be a unit vector normal to both a and b. The dot and p2 p1 + t + t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-17)
2 1 2
cross products are then described by Eqs. A-1 through A-3:
a b = a1b1 + a2b2 + a3b3, . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-1) Interpolation. Referring to Fig. 6, interpolation involves deter-
mining the position p* at some point on the arc, given a criterion.

a b = a b cos, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-2) The corresponding direction vector at the point is t* (Eq. A-18):


a b = a b sin n, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-3)
a b c = a cb a bc, . . . . . . . . . . . . . . . . . . . . . (A-4)
a b c = a1b2c3 b3c2 + a2b3c1 b1c3 + a3b1c2 b2c1.
t* = sin* cos*

sin* sin* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-18)
cos*
. . . . . . . . . . . . . . . . . . . . . . . . (A-5) The binormal vector, n12, can also be written in terms of the
direction vectors at the start point and the point at which the
Minimum Curvature. Zaremba2 presented the following deri- interpolation is to take place:
vation (refer to Fig. 4, the direction vectors t1 and t2 at the arcs t1 t*
ends are given by Eqs. A-6 and A-7, respectively): n12 = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-19)


sin*
sin1 cos1
Equating Eqs. A-8 and A-19, and taking the cross product of
t1 = sin1 sin1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-6) both sides of the equality with t1, gives Eq. A-20:
cos1
t1 t1 t* t1 t1 t2
= . . . . . . . . . . . . . . . . . . . . . . . (A-20)
and sin* sin

t2 = sin2 cos2

cos2

sin2 sin2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-7)
Using Eq. A-4 to expand the triple-cross products, and rear-
ranging for t*, gives Eq. A-21:

t* = t1 cos*
sin*t1 cos t2
. . . . . . . . . . . . . . . . . . . . (A-21)
The binormal vector, n12, and vectors b1 and b2 at the arcs sin
ends, are given by the Eqs. A-8, A-9, and A-10, respectively: Multiply the numerator and denominator of the first term in
t1 t2 A-21 by sin and collect terms in t1 and t2. After simplification,
n12 = , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-8) this gives the important relationship in A-22 that is the foundation
sin for all the interpolation formulae:
b1 = t1 n12, . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-9) sin * sin*
t* = t1 + t . . . . . . . . . . . . . . . . . . . . . . . . (A-22)
and sin sin 2
b2 = t2 n12. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-10) Substituting Eqs. A-6, A-7, and A-18 into Eq. A-22 gives
Eq. A-23:


Substituting n12 from Eq. A-8 into Eqs. A-9 and A-10, and using
Eq. A-4 for the vector triple products, gives Eqs. A-11 and A-12: sin* cos* sin1 cos1
sin *
t1 cos t2 sin* sin* = sin1 sin1
sin
b1 = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-11) cos* cos1


sin
sin2 cos2
and sin*
+ sin2 sin2 . . . . . . . . . . . . (A-23)
t1 t2 cos sin
b2 = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-12) cos2
sin
Interpolation on Measured Depth. Because the radius of the
The position p2 is calculated from p1 using Eq. A-13. Sub- arc is fixed, the ratio of the subtended angles is identical to the ratio
stituting b1 and b2 from Eqs. A-11 and A-12 into Eq. A-13 of the course lengths to the same points, as shown in Eq. A-24:
gives Eq. A-14:
* S*
p2 = p1 + Rb2 b1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-13) = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-24)
S12
1 cos Substituting * from Eq. A-24 into Eq. A-22, gives Eq. A-25
p2 = p1 + R t1 + t2. . . . . . . . . . . . . . . . . . . . . . . (A-14)
sin for t*, in terms of the course lengths:

March 2005 SPE Drilling & Completion 33


t* =
sin 1
S*
S12

t1 +
sin
S*
S12

t2. . . . . . . . . . . . . . (A25)
Eq. A-34 can be used to eliminate p* from Eq. A-33. After
rearranging, the resulting expression Eq. A-35, is of the form CA
sin(*)+B cos(*):
sin sin
m px p1 sin
For small values of , Eq. A-25 can be expanded in a Taylor + m t1cos m t2 = m t1sin sin*
S12
series. Evaluation of the terms is tedious, and computer assis-
tance14 was used to establish Eq. 14, presented in the body of the + m t1cos m t2cos*. . . . . . . (A-35)
paper. For small angles, the following simple expression can be used:
Straight-Hole Conditions. For small angles, equation Eq. A-35


t* 1
S*
t +
S12 1
S*

t . . . . . . . . . . . . . . . . . . . . . . . . . . (A-26)
S12 2
is badly behaved. Small-angle approximations must be used, and
the interpolation must be conducted with respect to measured
depth. The expression Eq. A-17 can be used to calculate the po-
Interpolation on Azimuth. Dividing the easting and northing sition of p* from p1, as shown in Eq. A-36:
components of Eq. A-23 eliminates sin*, giving the expression
S*
Eq. A-27: p* = p1 + t + t*. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-36)
2 1
sin* sin1 sin1 sin * + sin2 sin2 sin*
= . Sustituting Eq. A-26 into Eq. A-36 gives Eq. A-37:
cos* sin1 cos1 sin * + sin2 cos2 sin*
. . . . . . . . . . . . . . . . . . . . . . . (A-27) S*2
t t + S*t1 p* p1 = 0. . . . . . . . . . . . . . . . . . . (A-37)
Next, expand the terms sin(*) in both numerator and the 2S12 2 1
denominator of Eq. A-25 by use of the trigonometric identity: Taking the dot product of Eq. A-37, with the normal vector m
sin(*)sin cos*sin* cos. The terms involving sin* of the plane, and substituting (mpx) for (mp*), as before, gives a
are then collected on the left side of the equals sign, and terms quadratic equation (Eq. A-38) in the course length, S*. Eq. A-38
involving cos* are collected on the right side. The azimuth terms has the solution Eq. A-39:
are then combined by use of the same trigonometric identity to
give Eq. 15 for *, which was used to interpolate on azimuth in the m t2 m t1 2
body of the paper. The traditional form of the dogleg-severity S* + m t1S* m px p1 = 0 . . . (A-38)
2S12
equation3 Eq. A-28 can be used to determine *. This is in the
form of CA sin(*)+B cos(*): and


1
cos* = sin1 cos* 1sin* + cos1 cos*. . . . . . . . (A-28) 2m t2 t1m px p1 2
m t1 m t1 + 2

If the arc lies in the vertical plane, then (vn12) equals 0, and the S12


S* = .
solution is single valued. The vector n12 is given by Eq. A-8. Ex- m t2 t1
panding the scalar triple product using Eq. A-5 gives the expression S12
. . . . . . . . . . . . . . . . . . . . . . . (A-39)
sin1 sin2 sin2 1
= 0. . . . . . . . . . . . . . . . . . . . . . . . . (A-29) To simplify the manipulation, define a variable, , according to
sin
Eq. A-40. Eq. A-39 can then be written as Eq. A-41:
Interpolation on Inclination. Extracting and rearranging the
vertical component of Eq. A-23 gives Eq. A-30, which is of the m px p1m t2 m t1
= . . . . . . . . . . . . . . . . . . . (A-40)
form CA sin(*)+B cos(*). Once * has been found, the azi- S12m t12
muth component is determined from Eq. A-27:
1
sin* cos* = cos2 cos cos1sin* + sin cos1 cos*. m t1 m t11 + 22


. . . . . . . . . . . . . . . . . . . . . . . (A-30) S* = . . . . . . . . . . . . . . . . . . . . (A-41)
m t2 m t1
Interpolation at a Plane. Refering to Fig. 7, Eq. A-14 can be S12
used to calculate the point p* from p1. The radius, R, can be
Because 2 is much smaller than unity, the square root may be
expressed as S12/ to give Eq. A-31:
expanded in a series using the first four terms of the binomial
S121 cos* expansion,21 Eq. A-42, to give Eq. A-43:
p* = p1 + t1 + t*. . . . . . . . . . . . . . . . . . . . . (A-31)
sin* 1
z z2 z3 z4
1 + z2 = 1 + + + . . . . . . . . . . . . . . . . . . . . (A-42)
Now, use Eq. A-21 to substitute for t* in Eq. A-31 to give Eq. A-32: 2 8 16 128

p* = p1 +
S121 cos*
sin*
1 + cos*t1
sin*t1 cos t2
sin
. m t1 m t1 1 + 2 3
+
2 2


S* . . . . . . . . . . . . (A-43)
. . . . . . . . . . . . . . . . . . . . . . . (A-32) m t2 m t1
S12
Now, take the dot product of Eq. A-32, with the normal vector
m of the plane, and rearrange slightly: From Eq. A-43, it can be seen that the positive root must be
chosen so that the expression degenerates to the straight-line case
m p* p1 =
S121 cos*
sin*
1 + cos*m t1
when equals 0. Factoring Eq. A-43 gives Eq. 29 in the body of
the paper.


Turning Point. From Fig. 8, the vector t* can be written as Eq.
sin*m t1cos m t2 A-44. The angle between the n12 and v vectors is :
. . . . . . . . (A-33)
sin
n12 v
The equation of the plane20 is given by Eq. A-34, showing t* = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-44)
sin
(mpx)(mp*):
Using A-8 for n12, and expanding the resulting vector triple
m px p* = 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-34) product, gives Eq. A-45:

34 March 2005 SPE Drilling & Completion


v t2t1 + v t1t2 2 2R sin + R2 sin2
t* = . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-45) 2 = . . . . . . . . . . . . . . . . . . . . . . . (A-59)
sin sin cos2
Note that (v t2)cos2, (v t1)cos1, and that */2 at The substitution of from Eq. A-52 and 2 from Eq. A-59 into
the turning point. Using these values and Eqs. A-18, A-6, and A-7 Eq. A-53 results in Eq. A-60:
for t*, t1, and t2, respectively, in Eq. A-45 gives Eq. A-46:


2R sin
cos* sin2 cos2 2 = 2R21 cos + R sin
cos1 cos
sin* = sin2 sin2
sin sin 2 2R sin + R2 sin2
0 cos2 + . . . . . . . . (A-60)


cos2
sin1 cos1
cos2 Multiplying by cos2 and rearranging terms, gives Eq. A-61.
sin1 sin1 . . . . . . . . . . . . . (A-46)
sin sin Completing the square on the right side of Eq. A-61 gives
cos1 Eq. A-62:
Dividing the easting by the northing components gives Eq. 31
2 cos2 = 2R sincos 1 + 2 + R21 + cos2
in the body of the paper, for the azimuth * of the turning point.
2R2 cos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-61)
Position at Target Defined. In the most general case, the target p3
can be hit with one curved section of radius R and one straight 2 2cos2 = sin R1 cos2. . . . . . . . . . . . . (A-62)
section of length (see Fig. 9). Using Eq. A-15, the position p3 of Taking the square root of Eq. A-62 and rearranging for R
the target can be written as Eq. A-47: results in equation Eq. A-63 in the form of CA sin()+B cos().

The solution of such an equation is given by Eq. 1 in the body of


p3 = p1 + R tan t + t + t2. . . . . . . . . . . . . . . . . . . . . (A-47) the paper:
2 1 2
1
Taking the dot product of A-47 with t1, and also with itself, R = sin + R 2 22cos. . . . . . . . . . . . . . . . . . . (A-63)
gives A-48 and A-49, respectively:
After substituting the values of the constants A, B, and C, and
p3 p1 t1 = R sin + cos . . . . . . . . . . . . . . . . . . . . . . (A-48) comparing the result with Eq. A-57, we obtain the important result
and A2+B2C2 2.
Finally, the expression for /2 is given by A-64:


|p3 p1|2 = 2R21 cos + 2R sin + 2. . . . . . . . . . . . (A-49)
Let tan = 1
. . . . . . . . . . . . . . . . . . . . . . . (A-64)
2
2R 2
22
2 = |p3 p1|2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-50)
Setting 0 and RRc in Eq. A-64, gives Eq. 39 in the body
and of the paper for the subtended angle, c, for the minimum curva-
= p3 p1 t1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-51) ture to target.
Substituting Eq. A-50 into Eq. A-49 and Eq. A-51 into Eq. Toolface Angle. The toolface angle is determined by the dot prod-
A-48 gives Eqs. A-52 and A-53, respectively. The variable is ucts between the b, h, and r vectors (see Fig. 2).
determined using Pythagorass theorem (see Fig. 12).
b* h* = cos* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-65)
= R sin + cos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-52)
and
and
b* r* = sin*. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-66)
2 = 2R21 cos + 2R sin + 2. . . . . . . . . . . . . . . . . . (A-53) Eqs. A-65 and A-66 lead directly to the vector equation Eq.
Multiplying Eq. A-52 by 2R gives Eq. A-53. Rearranging Eq. A-67 for the toolface angle presented by Thorogood and Sawaryn18:
A-54 gives Eq. A-55:
2R sin + 2R cos = 2R . . . . . . . . . . . . . . . . . . . . . . . . (A-54)
2
tan* = b* r*
b* h*
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-67)

and The rightside unit vector lies in the horizontal plane, normal to
both the v and t* vectors, as shown in Eq. A-68. Evaluating the
2R sin 2R2 cos = 2 2 2R2. . . . . . . . . . . . . . . . . (A-55) expression gives Eq. 7 in the body of the paper.
Squaring both Eqs. A-54 and A-55, and adding the results, v t*
r* = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-68)
eliminates and results in a quadratic in 2, as shown in Eq. A-56. sin*
Solving the quadratic gives an explicit expression for
The highside unit vector lies in the vertical plane normal to
(Eq. A-57).
both the r* and t* vectors, as shown in Eq. A-69. Combining Eq.
4 222 + 4 4R22 + 4R22 = 0, . . . . . . . . . . . . . . . (A-56) A-68 with Eq. A-69, and expanding the triple vector product gives
Eq. A-70. Evaluating the expression gives Eq. 6 in the body of
1 1
the paper.
= 2 2R2 222. . . . . . . . . . . . . . . . . . . . . . . . . . . (A-57)
h* = r* t* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-69)
Inspecting Eq. A-57, the minimum curvature to target will oc-
cur when 0 and the radius equals Rc, as shown in Eq. A-58. t* cos* v
h* = . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-70)
The negative root of Eq. A-57 is the correct one: sin*
2 The unit vector, b*, is normal to both the t* and n12 vectors.
Rc = 1
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-58) Combining Eqs. A-8 and A-18 with Eq. A-71 gives Eq. A-72:
22 22 b* = t* n12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-71)
To find the angle , rearrange Eq. A-52 for and square the cos *t1 cos* t2
b* = . . . . . . . . . . . . . . . . . . . . . . . . (A-72)
result to give Eq. A-59: sin

March 2005 SPE Drilling & Completion 35


Taking the dot product of Eqs. A-72 and A-70 gives an ex- dt*
pression for (b*h*), as shown in Eq. A-73. Taking the dot prod- = * h* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-80)
dS*
ucts gives Eq. A-74:
1 dt*


and = * r*. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-81)
t* t2t1 t* t1t2 t* cos* v sin* dS*
b* h* =
sin sin* Eliminating the differential between Eqs. A-79 and A-80 and
. . . . . . . . . . . . . . . . . . . . . . . (A-73) taking the dot product of both sides with h* gives Eq. 51 for * in
cos* cos2 + cos *cos1 the body of the paper. The use of Eqs. A-65 and A-74 for b*h*
b* h* = . . . . . . . . . (A-74) gives Eq. 54 in the body of the paper.
sin sin* Eliminating the differential between Eqs. A-79 and A-81, and
taking the dot products of both sides with r*, gives Eq. 52 for *
Taking the dot product of Eqs. A-72 and A-68 gives an ex-
in the body of the paper. The use of Eqs. A-66 and A-78 for b*r*
pression for (b*r*), as shown in Eq. A-75:
gives Eq. 55 in the body of the paper.
cos*t2 v t* cos *t1 v t*
b* r* = .
sin sin* SI Metric Conversion Factors
. . . . . . . . . . . . . . . . . . . . . . . (A-75)
ft 3.048* E01 m
By use of Eq. A-5, the scalar triple products in Eq. A-75 are /100 ft 0.984252 E00 /30 m
given by Eqs. A-76 and A-77. Inserting these expressions into *Conversion factor is exact.
Eq. A-75 gives the final form for (b*r*), as shown in Eq. A-78:

t1 v t* = sin1 sin* sin* 1, . . . . . . . . . . . . . (A-76) Steve Sawaryn is currently the Wells Team Leader, Engineering,
for BPs Mature Business Unit in Aberdeen and is also the Drilling
t2 v t* = sin2 sin* sin2 *, . . . . . . . . . . . . . . (A-77) and Completions Adviser, specializing in drilling systems. During
the last 26 years, he has held a variety of posts in drilling op-
b* r* = erations and consultancy, as well as in projects in Aberdeen,
sin2 cos* sin2 * + sin1 cos *sin* 1 London, Kuwait, Alaska, and Norway. He is also a chartered
. engineer and fellow of the British Computer Soc. Sawaryn
sin holds an MA degree in chemical engineering from Cambridge
. . . . . . . . . . . . . . . . . . . . . . . (A78) U. He is currently serving on the SPE Digital Energy Committee.
John Thorogood is Chief Engineer for the BP Sakhalin explora-
The general expression (Eq. 47) given in the body of the paper tion program, based in Yuzhno-Sakhalinsk, Russia. In addition,
for toolface angle *, is obtained by substituting Eqs. A-74 and he is actively working on the issues associated with the opera-
A-78 into Eq. A-67. tional command and control of drilling operations. He project
managed the BP deepwater exploration operations in the
Curvature. Referring to Fig. 16, Frenets20 formula for total cur- Faroe Islands, U.K. sector, and in Norway from 1996 to 2001. He
vature, , gives Eq. A-79: has 30 years of experience in drilling operations and technol-
ogy, including his participation in the original deepwater wells
dt* west of Shetland in the early 1980s, and is the author of more
= b*. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (A-79) than 40 papers on directional drilling, surveying, and deepwa-
dS* ter operations. Thorogood holds BA and PhD degrees from
Cambridge U. and is a member of the U.K. Inst. of Mechanical
The definitions for *
and * as the inclination and azimuthal Engineers. He has served on the SPE Board of Directors with
components of the curvature, respectively, give Eqs. A-80 and A-81: special responsibility for Drilling and Completions.

36 March 2005 SPE Drilling & Completion