SPE/IADC-194167-MS

Tortuosity: The Rest of the Hidden Story

Robello Samuel and Yuan Zhang, Halliburton

Copyright 2019, SPE/IADC International Drilling Conference and Exhibition

This paper was prepared for presentation at the SPE/IADC Drilling International Conference and Exhibition held in The Hague, The Netherlands, 5-7 March 2019.

This paper was selected for presentation by an SPE/IADC program committee following review of information contained in an abstract submitted by the author(s).
Contents of the paper have not been reviewed by the Society of Petroleum Engineers or the International Association of Drilling Contractors and are subject to correction
by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers or the International Association of Drilling Contractors,
its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers or
the International Association of Drilling Contractors is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations
may not be copied. The abstract must contain conspicuous acknowledgment of SPE/IADC copyright.

Abstract
Conventional methods calculate tortuosity based on the predetermined shape of the trajectory using the
minimum curvature method; however, this discontinuity vanishes after the casing is run, and thus the well
path becomes smooth. A fundamental flaw to calculate drag force, hook load, stretch, and casing wear
using openhole surveys inside the casing is overestimation or underestimation of these values, depending
on whether the torque-and-drag model used is soft or stiff string. This paper describes a mathematical
formulation that considers flexural and torsional rigidities of casing pipes to estimate the vanishing
tortuosity. It also presents research findings concerning the structural shape of wellbore tortuosity when
combined with the data-driven approach from measurement while drilling (MWD)/gyroscope data, and an
objective approach of smoothing using a new tortuosity calculation method.

Background
This paper discusses the method of quantifying wellbore undulation based on mechanical methods, rather
than subjective methods. It also discusses the pitfalls of estimating tortuosity to quantify borehole quality
and provides answers to several important questions, such as whether the claimed benefits (i.e., enhanced
drilling performance, improved hole cleaning, ease of running casing, and superior cement operations) can
be fully attributed to reduced borehole tortuosity. The paper provides an estimate of the wellbore quality
change after casing/cementing and the minimum tortuosity that a wellbore can handle, including vanishing
tortuosity. Results demonstrate that the borehole curvature and torsion at survey stations determine the
shape of wellbore trajectory of the adjacent intervals and, consequently, the new tortuosity inside the
casing. The use of soft string and stiff string calculations provide additional results and compare with
smooth and nonsmooth wellbores. Results confirm that without this vanishing tortuosity, estimating drag
force may not always be conservative. In some cases, results can significantly underestimate frictional
drag effects, resulting in casing not reaching its target depth and underestimating completion string reach.
Extensive simulations have been performed and are reviewed in this paper. The recoil factor provides a
quantifiable method of estimating the severity of undulation or spiraling. This paper compares the predicted
mathematical simulation results with a description of the vanishing tortuosities and the insignificance of
micro tortuosity after casing and cementing.
2 SPE/IADC-194167-MS

Introduction
The term "tortuosity" is widely used to describe fluid flow in porous medial and river meandering. It is
also used in the medical field to characterize arteries and retinal vessels. In oil and gas, aside from the flow
fluid in the formation, tortuosity is used in drilling to characterize the well paths. It is used interchangeably
with wellbore undulation, wellbore oscillation, hole spiraling, wellbore smoothness, wellbore quality, and
wellbore profile energy. Tortuosity can be broadly classified as micro tortuosity, mezzo tortuosity, and macro
tortuosity, and can refer to 2D undulation or 3D spiraling. It has become important in the determination of
how the wellbore quality is maintained when different tools are used in the bottomhole assemblies (BHA).
When a mud motor is used between the sliding and rotary modes, the well path becomes a spiraled hole.
Early researchers have described borehole spiraling and attempted to quantify the extent of the spiraling.
With the advent of borehole imaging techniques and logging tools, borehole spiraling can be clearly mapped
and characterized. Some of the claimed benefits of smooth wellbore include the following:

• Enhanced drilling performance

• Torque and drag; low tortuosity, friction

• Weight transfer; rate of penetration (ROP), friction

• Vibration; ROP

• Improved hole cleaning

• Hole cleaning; gauge hole

• Ease of running casing

• Better-quality logging tool response

• Superior cement jobs

• Smooth completion runs

These claimed benefits may be common for any types of tools because there is no industry standard to
quantify them. Consequently, an essential need exists to standardize the model to quantify it uniformly.
This standardization will help the industry to accurately characterize complex wellbore trajectories and to
study the effects of different BHAs and associated downhole tools. Several researchers (including Oag and
Williams 2000; Williams et al. 2001; Dodson and Dodson 2003; and Mason and Chen 2005) describe the
quality of the hole subjectively, rather than provide qualitative quantification of the hole. Samuel and Liu
(2009) introduced the energy-based approach (Samuel's Criterion). Consequently, there is no clear criterion
for either defining or estimating the quality of the wellbore.

Wellbore Tortuosity
As the tortuosity increases, the friction force on the casing or wellbore wall concentrates the friction force
at several peak points and valley points, partly because of the tortuosity shape of the borehole. When
cementing, the straight casing is run into the open wellbore; it may be bent by the snake-like shape of the
wellbore or maintain its straight shape. If the casing pipe maintains its straight shape, the tortuosity of the
open hole will be removed after cementing. Currently, however, most drilling engineers still use the survey
data, which is the tortuosity of the open hole, to estimate the sideforce and, consequently, the drag force.
Obviously, this estimate includes a substantial amount of bias error.
Wellbore tortuosity provides the designer with an option to observe the natural undulation or wellbore
oscillation. Tortuosity is given as (Samuel et al. 2005):
SPE/IADC-194167-MS 3

(1)

Wellbore Torsion and Well Profile Energy

In the well path design, curvature or dogleg is widely used to describe how the well path is bending. It is
used to calculate the total curvature between the survey stations or local dogleg, depending on the spread
of the survey measurements. Wellbore torsion is another important quantifiable parameter that is neglected
in the well plan.
The tangent and normal vectors are used when describing a well path, as shown in Fig. 1. The vector n
is orthogonal to t and lies in the osculating plane at the point P. The binormal vector b is the product of the
unit tangent and the unit normal. The Frenet-Serret (t-n-b) frame can be used to show the three orthonormal
unit vectors at any given survey point on the well path.

Figure 1—Tangent, normal, and binormal vectors.

Wellbore torsion provides the extent of departure of a well path from a plane geometrical path and can
be given as (Fitchard and Fitchard 1983; Shan et al. 1993; Xiushan and Zaihong 2001, Xiushan 2005; and
Xiushan 2006, Samuel and Liu 2009):

(2)

where
α = Inclination angle (°)
ϕ = Azimuth angle (°)
κα = Rate of inclination change (dropping off is a negative value), (°)/30 m or (°)/100 ft
κϕ = Rate of azimuth change (decreasing azimuth is a negative value), (°)/30 m (°)/100 ft
κ = Curvature of wellbore trajectory, (°)/30 m or (°)/100 ft
= First derivative of inclination change rate, viz. second derivative of inclination angle
= First derivative of azimuth change rate, viz., second derivative of azimuth angle
● = First derivative
4 SPE/IADC-194167-MS

The borehole torsion can be either positive or negative, whereas the borehole curvature is always positive.
The average borehole torsion over a survey interval can be calculated as shown in Eq. 3.

(3)

where,

Using the curvature and torsion as a method to quantify the twisting and turning of the well path, the
complexity of a well can be easily quantified with physical reasoning, as shown in Fig. 2.

Figure 2—Wellbore undulation.

The non-linear curve modeling of a thin elastic beam is known as the minimum energy curve; it can be
used as a basis for combining these parameters and is given as (Samuel 2010):

(4)

It can be further normalized with the drilled depth and given as:

(5)

The minimization of the total energy of the curve will result in less torque and drag and casing wear.

Actual Shape of the Well Path

To describe the actual well path shape, several questions must be addressed, including the following:
SPE/IADC-194167-MS 5

• What are the assumptions in describing the shape between the course length?

• What is the predetermined shape of the well path?

• What parameters are used to determine the shape of the well path?

• Is the shape the same between all the survey intervals?

• What is the actual shape of the wellbore when it is changed after cementing?

It is not clearly understood how the shape of the well is changed after casing and cementation with
and without centralizers. It is also not known how to quantify the effect of the well path after casing and
cementing the hole. In addition, there is a minimum tortuosity limit that a casing string can withstand while
running in because of the bending stiffness of the pipe so as to eliminate the micro-tortuosity that has been
created during drilling.
It has been found that the original survey or the original tortuosities may vanish after the casing is landed
and cemented. Fig. 3 shows some of the conditions that may exist in the undulated portion of the wellbore.
As shown in Fig. 3, the casing pipe can take different shapes with varied deflection of the casing inside
the wellbore because of the stiffness of the pipe. The following section provides additional details about
the modeling.

Figure 3—Vanishing tortuosities.

Vanishing Tortuosity as a Result of Casing

To estimate the tortuosity and the new wellbore survey after casing has been landed, a mathematical model
with underlying engineering mechanics has been developed. When a casing pipe goes through a tortuous
wellbore, the straight casing pipe will bend as a result of its self-weight. It is reasonable to assume the
boundary condition of the casing pipe as pin-supported on two peaks of a tortuosity wave. The self-weight of
the casing pipe will cause the casing pipe to bend between two peaks, and the deflection can be calculated as:

(6)

Where
E = modulus of elasticity
I = moment of inertia
6 SPE/IADC-194167-MS

q = unit weight of the string

L = distance between two peak points
The tortuosity of a wellbore, TW, can be defined as:

(7)

Where
∆w = wellbore deformation which can be calculated form the original survey data
L = distance between two peak points
Furthermore, three scenarios can be analyzed, as described in the following paragraphs.
Scenario 1. When ∆p>∆w, the casing pipe can be taken as a soft pipe that lies on the bottom of the wellbore
like a rope. In this scenario, the shape of the casing pipe is the same as the shape of the wellbore; its tortuosity
is same as the tortuosity of wellbore, and can be written as:

(8)

Fig. 4 shows this condition. The grey dashed line is the casing pipe shape, whereas the solid line is the
wellbore calculated from the survey data. The actual shape of the casing pipe resulting from self-weight is
supported by the wellbore. Under this scenario, the casing pipe lies on the low side of the wellbore.

Figure 4—Casing deflection on self-weight larger than the wellbore deflection.

Scenario 2. When ∆w-CL<∆p<∆w, the casing pipe bends downward, but does not reach the bottom of the
wellbore, as shown in Fig. 5. At the same time, its top also does not reach the top of the wellbore. In this
scenario, the shape with self-weight is the actual shape of the casing pipe after cementing. The tortuosity
of the casing pipe can be calculated as:

(9)

Figure 5—Casing deflection on self-weight less than the wellbore deflection.

Scenario 3. As shown in Fig. 6, when ∆p<∆w-CL, the casing pipe bends downward as a result of both its
self-weight and the sideforce from the top of the wellbore. In this scenario, the deflection of the casing pipe
is limited by the clearance CL between the wellbore and the casing pipe. Under this condition, the tortuosity
of the casing pipe can be calculated as:
SPE/IADC-194167-MS 7

(10)

Figure 6—Casing deflection on self-weight less than the wellbore clearance.

In this scenario, the wellbore top adds deflection to the casing pipe, causing additional sideforce and
friction force on the casing pipe.
These analyses illustrate that when a casing pipe is run into the wellbore, its shape will be determined by
its self-weight. The new tortuosity of the wellbore depends on the actual shape of the casing pipe. In this
process, the stiffness of the casing pipe is a key factor. If the casing pipe is strong (higher EyI), the deflection
of the casing pipe will be smaller than the deformation of the wellbore. Consequently, the stiffness of the
casing pipe will help to reduce the tortuosity of the wellbore. Conversely, if the casing pipe is soft (lower
EyI), then the deflection of the casing pipe under self-weight is greater than the wellbore deformation. Under
this condition, the casing pipe can only lie on the bottom of the wellbore, which further results in no change
in the tortuosity.

Survey Distance Calculation

The actual effect of the casing pipe stiffness depends on the distance of a tortuosity wave. For this survey
interval or course, length is important. If the survey interval is too large (for example, 100 ft), then the
wave distance L will be large. Under this condition, the casing pipe deflection calculated is larger than the
wellbore deformation. In addition, this will make the actual shape of the new wellbore difficult to quantify.
Consequently, it is important to determine the minimum survey interval that is needed with the given casing
pipe stiffness and the wellbore diameter.
Assuming the tortuosity of the wellbore to be Tw, when the deflection of the casing equal to the deflection
of the wellbore, the distance between two support point can be given as:

(11)

Where
Z = coefficient, 4.8 for one span
Dh = diameter of the wellbore
 d = outside diameter of the casing pipe
ρ = density of the pipe
 g = acceleration as a result of gravity
Eq. 11 provides the condition needed to calculate the actual shape of the new wellbore; i.e., the survey
interval must be less than this critical interval.
Under actual wellbore conditions, the tortuosity varies, and the casing pipe is continuous (Fig. 7).
Consequently, the calculation must be changed with a modified coefficient Z, depending on the number of
the segments analyzed.
8 SPE/IADC-194167-MS

Figure 7—Casing shape in a continuous tortuous wellbore.

Pitfalls
Several pitfalls associated with estimating tortuosity to quantify borehole quality can be elucidated,
including the following:

• Survey station interval: When the survey station interval is large, the position of the casing
calculation based on stiffness will be different when the station intervals are small.
• Casing centralization: Casing centralization will change the composite stiffness of the pipe and,
consequently, the position in the wellbore.
• Casing cementation: If the casing is not fully cemented, the casing section above the top of cement
may change.
• Casing cementation: The effect of roughness created by micro tortuosity may enhance the bonding
between the casing and formation.
• The use of openhole survey after casing and cementing for the remainder of the calculations has
been found to result in the following:

◦ Low/high torque and drag values

◦ Lower/higher swab and surge values

◦ Fluid and mechanical forces are coupled

◦ Low weight transfer calculations

◦ Decreased hole cleaning

◦ High friction and sideforce values

◦ Incorrect friction factor calibration

◦ Higher casing wear estimation

◦ Higher stretch values and thereby positional error

◦ Higher fatigue calculations

◦ Increased well profile energy

◦ Overestimation of safety factors for casing design

◦ Lower estimation of reach

◦ Incorrect calculation of packer setting forces

Conclusions
The following conclusions can be drawn from this study:

• Stiffness of the casing pipe can change the original tortuosity of the wellbore. Consequently, it is
essential to recalculate the tortuosity of the wellbore to enable more accurate estimates of doglegs
and sideforces.
SPE/IADC-194167-MS 9

• A critical interval (minimum survey interval) is needed to accurately estimate the new tortuosity
and, consequently, the vanishing tortuosity after the casing pipe has been landed.
• Based on the study results, a course survey interval data is not always needed. Obviously, the survey
distance interval depends on the stiffness of the casing pipe and wellbore diameter.
Consequently, it is essential to correct the survey after the hole is cased and cemented; the theoretical
method presented helps to understand both the quality of the wellbore before and after casing and the overall
effect across all phases of drilling, rather than only local observations. The use of curvature and wellbore
torsion are needed to completely describe the quality of the wellbore.

Acknowledgement
The authors would like to express their appreciation to Halliburton for the opportunity to present this paper.

Nomenclature
α = Inclination angle, (°)
ϕ = Azimuth angle, (°)
κα = Rate of inclination change (dropping off is a negative value), (°)/30 m or (°)/100 ft
κϕ = Rate of azimuth change (decreasing azimuth is a negative value), (°)/30 m (°)/100 ft
κ = Curvature of wellbore trajectory, (°)/30 m or (°)/100 ft
θ = Torsion angle, (°)
ρ = density of the material,
= First derivative of inclination change rate, viz. second derivative of inclination angle
= First derivative of azimuth change rate, viz. second derivative of azimuth angle
τ = Torsion of wellbore trajectory, (°)/30 m or (°)/100 ft
= Average borehole torsion, (°)/30 m or (°)/100 ft
Δp = Deflection of the casing pipe with self-weight load
ΔW = wellbore deformation
b = Unit binormal vector of wellbore trajectory
CL = Clearance, in.
D = Depth, ft
Dh = Diameter of the wellbore, in.
d = Outside diameter of the casing pipe, in.
E = East coordinate (west is negative), m or ft
Ey = Young's modulus
Es = Energy
g = acceleration resulting from gravity
L = Tortuosity wave length
N = North coordinate (south is negative), m or ft
n = Unit principal normal vector of wellbore trajectory
q = Self-weight load of the casing
t = Unit tangent vector of wellbore trajectory
Tc = Tortuosity of the casing
Tw = Tortuosity of the wellbore
x = x-coordinate, ft
y = y-coordinate, ft
z = z-coordinate, ft
• = First derivative
10 SPE/IADC-194167-MS

Subscripts
abs = absolute
i = survey station index
n = index

References
Dodson, J. and Dodson, T. 2003. Drilling Efficiency Numbers Static. Offshore 55 (9): 26–28.
Fitchard, E.E. and Fitchard, S.A. 1983. The Effect of Torsion on Borehole Curvature. Oil & Gas Journal 81 (3): 121–124.
Mason, C.J. and Chen, C-K. 2005. The Perfect Wellbore. Presented at the SPE Annual Technical Conference, Dallas,
Texas, USA, 9-12 October. SPE-95279-MS. https://doi.org/10.2118/95279-MS.
Oag, A.W. and Williams, M. 2000. The Directional Difficulty Index - A New Approach to Performance Benchmarking.
Presented at the 2000 IADC/SPE Drilling Conference, New Orleans, Louisiana, USA, 23-25 February. SPE-59196-
MS. https://doi.org/10.2118/59196-MS.
Samuel, G.R., Bharucha, K., and Luo, Y. 2005. Tortuosity Factors for Highly Tortuous Wells: A Practical Approach.
Presented at the SPE/IADC Drilling Conference, Amsterdam, The Netherlands, 23-25 February. SPE-92565-MS.
https://doi.org/10.2118/92565-MS.
Samuel, R. and Liu, X. 2009. Advanced Drilling Engineering – Principles and Designs. Houston, Texas: Gulf Publishing.
Samuel, R. 2010. A New Well-Path Design Using Clothoid Spiral (Curvature Bridging) for Ultra-Extended-Reach
Drilling. SPE Drill Compl 25(03: 363–371. SPE-119459-PA. https://doi.org/10.2118/119459-PA
Shan, W., Xiushan, L., Daqian, Z. et al. 1993. The Shape of the Space Curve of Borehole Trajectory. Journal of Daqing
Petroleum Institute, 17 (3): 32–36.
Williams, C., Mason, J.S., and J. Spaar. 2001. Operational Efficiency on Eight-Well Sidetrack Program Saves $7.3
Million vs. Historical Offsets in MP 299/144 GOM. Presented at the SPE/IADC Drilling Conference, Amsterdam,
The Netherlands, 27 February-1 March. SPE-67826-MS. https://doi.org/10.2118/67826-MS.
Xiushan, L. and Zaihong, S. 2001. Numerical Approximation Improves Well Survey Calculation. Oil & Gas Journal 99
(15): 50–54.
Xiushan, L. 2005. Average Borehole Curvature Calculation of Hole Trajectory. Oil Drilling & Production Technology
27 (5): 11–15.
Xiushan, L. 2006. New Technique Calculates Borehole Curvature, Torsion. Oil & Gas Journal 104 (40): 41–49.