MODELLING OF STICK – SLIP DRILLSTRING

VIBRATION IN OILWELL DRILLING OPERATION

ABDUL HAMID SULEIMAN

(2003328996)

BACHELOR OF ENGINEERING (HONOURS)

(MECHANICAL)
UNIVERSITI TEKNOLOGI MARA (UiTM)
NOVEMBER 2006
 AUTHOR DECLARATION

“I declare that this thesis is the result of my own work except the ideas and
summaries which I have clarified their sources. This thesis has not been accepted for
any degree and is not concurrently submitted in candidature of any degree”

Signed :………………………
Date :………………………..

ABDUL HAMID SULEIMAN

UiTM No: 2003328996
 SUPERVISOR CERTIFICATION

“I declare that I have read this thesis and in my point of view this thesis is qualified
in term of scope and quality for the purpose of awarding the

Bachelor of Engineering (Honours) (Mechanical)”

Signed :………………….
Date :…………………

Supervisor
DATO’ PROF. DR. IR. HJ. MOHAMED DAHALAN MOHAMED RAMLI
Faculty of Mechanical Engineering,
Universiti Teknologi MARA (UiTM),
40450 Shah Alam.
Selangor Darul Ehsan.
HEAD OF PROGRAM CERTIFICATION

Accepted:

Signed :………………….
Date :…………………

Head of Program
MR NIK ROSLI ABDULLAH
Faculty of Mechanical Engineering,
Universiti Teknologi MARA (UiTM),
40450 Shah Alam.
Selangor Darul Ehsan.
.
To My Lovely Family
MODELLING OF STICK – SLIP DRILLSTRING VIBRATION IN OILWELL
DRILLING OPERATION

ABDUL HAMID SULEIMAN

(2003328996)

A thesis submitted in partial fulfillment of the requirements for the award of

Bachelor of Engineering (Honours) (Mechanical)

Faculty of Mechanical Engineering

Universiti Teknologi MARA (UiTM)

NOVEMBER 2006

i
 ACKNOWLEDGEMENT

First of all, I would like to express my sincerest gratitude to my thesis

supervisor, Dato’ Prof. Dr. Ir. Hj. Mohamed Dahalan Mohamed Ramli, for his warm-
hearted encouragement and insightful guidance.
I would also like to express my appreciation to Associate Prof. Dr. Ahmad
Azlan Mat Isa and Mr. Ahmad Fakri Shaari for their valuable time, encouragements
and advises.
Special thanks are particularly due to the following persons, for without their
kind generosity and support in providing me with the various technical papers and
theses, it would not have been possible for me to complete this thesis.
Dr. A. M. Chevallier Universidad de Monterrey
Dr. Ir. A. F.A. Serrarens Eindhoven University of Technology
Prof. A. P. Christoforou Kuwait University
Prof. A. S. Yigit Kuwait University
Mr. C. Germay University of Liege
Prof. Dr. Ir. D.H. van Campen Eindhoven University of Technology
Mr. Ernie Chavez Sperry Drilling Services
Dr. G. H. M. van der Heijden University College London
Prof. Iain Finnie University of California, Berkeley
Prof. Dr. Ir. J. D. Jansen Delft University of Technology
Mr. J. F. Brett OGCI Inc.
Prof. J. K. Vandiver Massachusetts Institute of Technology
Dr. M. L. Payne BP America
Dr. M. W. Dykstra Hughes Christensen
Dr. R. F. Mitchell Landmark Graphics

ii
 Special credit is due to fellow colleague, Mr. Mohd Asreen Ishak, for his kind
collaboration during the course of completing of our final year projects.
Last but not least, I would also like to express my sincerest gratitude to my
ASME International’s mentor, Mrs. Caitlin H. Foley-Sainio, for her kind and patient
support during the years 2000 – 2006. No words can describe my utmost and
profound gratitude to her and to her family.
All contributions have been most meaningful to me and your support gave me
the strength to persevere. Thank you all very much.

Abdul Hamid Suleiman

iii
 ABSTRACT

Torsional vibrations are widely recognized as the cause of decrease in the rate
of penetration and premature failure of downhole drilling equipment. In this study, a
classical wave equation is utilized in the mathematical modeling while finite element
method formulation is used to obtain approximate solution of the modes shapes and
natural frequencies of vibration. Numerical and theoretical results show close
comparison especially in the mode shapes of vibration. The significance of the
present study is that it may trigger interests in drillstring dynamics in Malaysia in the
near future.

iv
 TABLE OF CONTENTS

CONTENTS PAGE

PAGE TITLE i
ACKNOWLEDGMENT ii – iii
ABSTRACT iv
TABLE OF CONTENTS v - vii
LIST OF TABLES viii
LIST OF FIGURES ix
NOMENCLATURE x - xi

CHAPTER 1 – INTRODUCTION

1.1 Overview
1.1.1 Oilwell Drilling 2–3
1.1.2 Drilling Vibration 4
1.2 Project Background 4–5
1.3 Objective 5
1.4 Significance 5
1.5 Scope 6
1.6 Methodology 6

v
CHAPTER 2 – LITERATURE REVIEW

2.1 Vibration Perspective 8

2.2 Modes of Vibration 8–9
2.2.1 Vibrations Outgrowth 10
2.2.2 Axial Vibration 10 – 11
2.2.2.1 Bit Bouncing 11
2.2.3 Torsional Vibration 11
2.2.3.1 Stick-Slip 12
2.2.4 Lateral Vibration 12
2.2.4.1 Whirl 13

CHAPTER 3 – TORSIONAL VIBRATION

3.1 Introduction 15
3.2 Models 16
3.3 Stick-Slip Oscillations 16
3.4 Consequences and Remedies 17

CHAPTER 4 – ANALYTICAL APPROACH

4.1 Continuous Modeling 19 – 20

4.1.1 Stick Phase 21 – 22
4.1.2 Slip Phase 23 – 24
4.2 Finite Element Model 25 – 27
4.3 Rayleigh Damping 27
4.4 Numerical Simulation 28 – 29

vi
CHAPTER 5 – RESULTS AND DISCUSSIONS

5.1 Results
5.1.1 Stick Phase 31
5.1.2 Slip Phase 32
5.2 Discussions 33 – 35
5.3 Comparison 36 – 38

CHAPTER 6 – CONCLUSIONS AND RECOMMENDATIONS

6.1 Conclusions 40
6.2 Recommendations 40

REFERENCE

References 41 – 43

APPENDICES

Appendix A – Classical Wave Equation 44 – 46

Appendix B – Experimental Set – up 47 – 50
Appendix C – Results 51 – 54

vii
 LIST OF TABLES

TABLE TITLE PAGE

5.1 Conditions of Drillstring during Stick Motion 33
5.2 Conditions of Drillstring during Slip Motion 33
5.3 The Analytical Results against FEM Results during 36
Stick Phase
5.4 The Analytical Results against FEM Results during 36
Slip Phase

viii
 LIST OF FIGURES

FIGURE TITLE PAGE

1.1 Rotary Drilling Process 2
1.2 Onshore Drilling Rig 3
1.3 Offshore Drilling Rig 3
2.1 Modes of Vibration in Drillstring 9
4.1 Process Flow in MATLAB 20
4.2 Boundary condition for drillstring during stick-phase 21
4.3 Boundary condition for drillstring during slip-phase 23
4.4 Torsional beam element for finite element model 25
4.5 Process Flow in Finite Element Analysis using ANSYS 29
5.1 First Three Modes of Vibration during Stick motion 31
5.2 First Three Modes of Vibration during Slip motion 32
5.3 Modes of Vibration during Stick – Slip Motion at n=50 35
5.4 Experimental Result of Stick Phase 37
5.5 Experimental Result of Slip Phase 38

ix
 NOMENCLATURE

BHA – Bottom-Hole Assembly

E&P – Exploration and production
FEA – Finite element analysis
FEM – Finite element method
MWD – Measurement While Drilling
ODE – Ordinary differential equation
PDC – Polycrystalline diamond compact
PDE – Partial differential equation
ROP – Rate of penetration
WOB – Weight on bit
T n ( x, t ) - Angular displacement of drillstring
G - Shear modulus of the drilling assembly material
U - Density of drillstring element
c - Torsional (sound) waves propagation velocity in the drillstring material
M - System mass matrix for torsional vibrations
nxn

..
T t t - Second order polar coordinate
nx1

Ct - System damping matrix for torsional vibrations

nxn

.
T t t - First order polar coordinate
nx1

Kt - System damping matrix for torsional vibrations

nxn

x
Tt t - Polar coordinate
nx1

.
§ ·
Gt ¨ t ,Tt ,Tt ¸ - External torsional forcing function
nx1 © ¹

\ x - Torsional displacement
S t - Twisting moment
kij - Equivalent stiffness

J - Polar moment of inertia of the considered cross-section of drillstring

mij - Equivalent mass

M - Mass matrix
K - Stiffness matrix
F - Excitation matrix
C - Damping matrix in Rayleigh damping
a - Multiplicative coefficient for the mass matrix in Rayleigh damping
b - Multiplicative coefficient for the stiffness matrix in Rayleigh damping
Zn - Natural frequency
t - Time

xi
 Chapter 1 – Introduction

CHAPTER 1

INTRODUCTION

Vibrations in structures and machines are normally the unwanted phenomenal

behaviour which may cause fatigue, failure, unreliability and noise pollution. Various
approaches were carried out to study this phenomenon through the theoretical,
numerical or experimental method. In this project, the focus is on the vibration of
drilling operation on an onshore or offshore drilling rig. The motivation to pursue
this particular sector of the petroleum exploration and production (E&P) industry is
the author’s very early exposure to oil and gas well drilling through the mouth of his
step-father, who used to work on an offshore drilling rig. Personnel related to the oil
and gas well drilling industry are highly regarded and sought after, especially in the
current state of rising energy costs. Unfortunately, to the writer’s knowledge, no
prior research or any related work has been previously carried out in Malaysia with
regards to the study of drillstring dynamics. The need to drill deeper and in harsher
and in more hostile environment will necessitate a serious study on the subject.
Therefore the author hope that by this humble beginning on the basic aspects of
drillstring dynamics, considerable interests will be triggered and generated in this
particular field in the near future.

1
 Chapter 1 – Introduction

1.1 OVERVIEW
1.1.1 Oilwell Drilling

Derrick

Rotary Hose
Swivel
Stand Pipe

Bell Nipple
Blowout
Preventer
Earthen Pit
Emergency
Mud
Flow Line Pump

Conductor
Casing

Annulus

Drill
Collars
Drillbit

Figure 1.1 Rotary Drilling Process

2
 Chapter 1 – Introduction

Rotary drilling rigs are used for almost all drilling operations today. A sketch
illustrating the rotary drilling process is depicted in Figure 1.1. Basically, the well is
drilled by rotating the bit to which a downward force is applied. In general, the bit is
turned by rotating the entire drillstring, using a rotary table at the surface. The
downward force is applied to the bit by using sections of heavy thick-walled pipe
called drill collars. During the drilling operations, the cuttings produced are lifted to
the surface by circulating a fluid (mud) down the drillstring through the bit nozzle,
and up the annular space between the hole and the drillstring. The cuttings are
separated from the drilling fluid at the surface.

As shown in Figure 1.2 and 1.3, the rotary drilling rigs can be classified
broadly as onshore and offshore rigs.

Figure 1.2 Onshore Drilling Rig Figure 1.3 Offshore Drilling Rig
Reproduced from P.T. Barata Nusatama Prima Reproduced from Arabian Drilling Co.
http://www.bnp.co.id/images/Rig%203%20photos/Rig3%20compr.jpg http://www.arabdrill.com/mkt/photos/ad20_1.JPG

3
 Chapter 1 – Introduction

1.1.2 Drillstring Vibration

Field observations have revealed that severe vibration occurs during drilling
operation which severely affects the overall drilling performance. Currently, drilling
of deepwater and ultra-deepwater wells for oil and gas production have opened new
horizon for petroleum engineers and experts to try to mitigate the influence of
vibration during drilling operation. Even though new technology has been deployed,
such phenomena still occurs affecting considerably on drilling costs and daily
operations. Before 1960s, studies were focused on material strength of the drillstring
components but the trends have since changed to emphasise on its dynamic
behaviour [Jansen – 8].

Among engineering applications, drillstring dynamics offers an opportunity

for the study and analysis of drilling vibrations. Early field observation and research
has revealed that the main causes of the reduction of rate of penetration (ROP) and
the premature failure of Bottom-Hole Assembly (BHA) has been the presence of
vibration. Many studies have been conducted to identify and recognise the different
types of vibrations during drilling operation. These have led to the identification and
classification of vibrations into three separate and distinctive categories namely axial,
torsional and lateral vibration. Whether observable or not, these bad vibrations
significantly increase the costs of drilling an oil and gas well [Jardine et al. – 23].

1.2 PROJECT BACKGROUND

Due to time constraint, this project will focus only on the theoretical and
numerical method of drillstring vibrations analyses. Since torsional vibration is the
prominent mode of vibration, it is therefore the main concentration of this work.
Accordingly, an analytical approach to dynamic behaviour of torsional vibration in
the drilling operations is considered. The emphasis is on the natural frequencies
during stick and slip motion, which is believed to have inter-relationship with other
types of vibrations, coupled or otherwise.

4
 Chapter 1 – Introduction

This study has been carried out by applying the classical wave equation to
determine the responses as well as mode shapes of vibration present during rotation.
Apart from that, the natural frequencies during stick and slip motion have been
proposed to suit their boundary conditions.

As stated earlier, this project concentrates on torsional vibration without

paying attention to the effects of axial and lateral vibrations. However, at the end of
this project, the effects of torsional vibration will be discussed together with its
couplings, namely, axial-torsional and lateral-torsional.

1.3 OBJECTIVE

The main objective of this project is to formulate an analytical approach to

dynamic behaviour of torsional vibration as well as to simulate its natural
frequencies during stick and slip motion. The result obtained is then analysed in
relation to displacement and the mode shape of vibration according to their specific
boundary conditions. The analyses rely on the two conditions (stick and slip) and
their effects. This project is carried out essentially to determine the cause of vibration
that influences its dynamic behaviour.

1.4 SIGNIFICANCE

The significance of this project can be extended to the study of vibrational

couplings of drillstring with more complicated models. It can also be extended to the
study of suppression of torsional vibration, which can then be applied to petroleum
engineering particularly in drilling operations.

5
 Chapter 1 – Introduction

1.5 SCOPE

The study is limited to torsional vibration during drilling operation. Classical

wave equation is applied by imposing two boundary conditions. The boundary
conditions are fixed – fixed ends and fixed – free ends that represent stick phase and
slip phase respectively.

1.6 METHODOLOGY

The methods of study are:

x Mathematical modelling using wave equation
x Finite element method using ANSYS software.

The output from mathematical solution is simulated using MATLAB. The natural
frequencies and mode shapes of vibration from both MATLAB and ANSYS are
compared.

6
 Chapter 2 – Literature Review

CHAPTER 2

LITERATURE REVIEW

The previous chapter has provided a brief introductory description of rotary

drilling process and vibration in oil and gas well drilling operation. It has also
provided an overview and background of this project as well as its objective, scope,
significance and methodology. In this chapter, the general perspective of vibration
will be discussed including the various modes of vibration. The phenomena of
specific modes of vibration will also be further discussed with their specific subsets
or consequences.

7
 Chapter 2 – Literature Review

2.1 VIBRATION PERSPECTIVE

Field observation and research have revealed that vibration severely affects
drilling performance. The first analytical and experimental study of drillstring
vibration was conducted initially in the sixties [Finnie et al. – 6, 7]. Prior studies
have shown that the various modes of vibrations produce rather similar effects,
which are namely the decreasing rate of penetration (ROP) and the premature failure
of downhole equipment.

Whether observable or not, these unwanted vibrations significantly increase

the cost of drilling an oil/gas well. Although the exact figure has still to be assessed,
large portion – an estimated 2% to 10% of well costs – may arise from vibration
related problems such as lost time while pulling out of hole (tripping), fishing,
reduced ROP, poor quality wellbore and increased service costs due to the need for
rugged equipment [Jardine et al. – 23].

2.2 MODES OF VIBRATION

The downhole dynamics of drilling tools are complex. It involves a

combination of simultaneous vibrational phenomena which render the analysis of
drillstring vibrations extremely challenging. Three primary classifications of
vibration inducing loads apply to rotary oilwell downhole equipment. They are axial,
torsional, and lateral. Related dynamic aspects of these phenomenons are bit-
bouncing, stick-slip and whirling. Figure 2.1 depicts some of these prevalent
drillstring vibration modes. Their dynamic responses are either transient or steady-
state. Drillstring vibrations can be induced by an external source, such as the axial
oscillations generated by the interaction of a tricone bit with the formation, or can be
self-induced. Dunayevsky et al. [24] believed that most forced vibrations originated
from bit-formation and drillstring–borehole interactions. Self-excited vibrations are
also thought to be the cause of this downhole phenomenon [Finnie et al. – 6, 7].

8
 Chapter 2 – Literature Review

Torsional Vibration

Drill Pipe

Axial Vibration

Drill Collars

Drillbit

Lateral Vibration

Figure 2.1: Modes of Vibration in Drillstring

9
 Chapter 2 – Literature Review

2.2.1 VIBRATIONS OUTGROWTH

The consequences of drillstring vibrations are multiple and hinder the goal of
faster and safer drilling. Mason et al. [14] described that downhole vibration causes
premature wear and tear of drilling equipments, resulting in fatigue induced failures
such as pipe wash-out and twist-off. Furthermore, vibrations also cause significant
wastage of drilling energy [Macpherson et al. – 11]. Finally, vibrations may often
induce wellbore instabilities that can deteriorate the condition of the borehole while
at the same time reduce the directional control and its overall shape [Dunayevsky et
al. – 24].

2.2.2 AXIAL VIBRATION 1

Axial vibrations of a drillstring involve motions of its components along its

longitudinal axis. Axial vibrations also have indirect consequences due to downhole
coupling mechanisms which result in significant lateral displacements [Shyu – 22].
Axial vibrations cause bit bounce and rough drilling behaviour that destroys drillbits,
damages bottom-hole-assemblies and increase total drilling time. Axial vibrations
however may be detected at the surface.

Drillstring are subjected to both static and dynamic axial loadings. The
classical buckling theory provides the maximum static weight on bit (WOB) that the
assembly can sustain without buckling. It yields a certain operational static axial
constraint corresponding to the applied WOB that fulfills the appropriate criteria
(safety factor, rate of penetration requirements, etc.). On the other hand, dynamic
axial loads on the drilling assembly originate primarily from bit-formation
interactions. They give rise to time-dependent fluctuations of the weight applied to
the bit, and are rather erratic.

1
Longitudinal vibration
10
 Chapter 2 – Literature Review

Severe axial vibrations develop quite often when drilling with roller-cone
bits, owing to their type of interaction with the formation. The multi-lobes pattern
specifically generated by tricone bits at the bottom of the well is a major source of
axial excitations for vertical or near vertical wells, limiting drillstring-borehole
interactions and reducing effective damping [Skaugen – 3].

2.2.2.1 BIT BOUNCING

The bit-bounce pattern is likely to develop when drilling with roller-cone bits.
As the multi-lobes pattern develops at the bottom of the well, it results in erratic
interaction of the bit with the bottom of the well, which makes the bit to have loose
contact with the rock formation. Consequently, the severe axial stress waves induced
by the lift-off of the bit constitute a primary source of excitation to the entire drilling
assembly.

2.2.3 TORSIONAL VIBRATION 2

Downhole measurements show that applying a constant rotary speed at the

surface does not necessarily translate into a steady rotational motion of the bit. In
fact, the downhole torsional speed typically exhibits large amplitude fluctuations
during a significant fraction of the drilling time. This non-uniform rotational speed of
the bit is due primarily to the large torsional flexibility of the drilling assembly
[Skaugen – 3]. Torsional vibrations cause irregular downhole rotation which
exacerbates fatigue to drill collar connections, damages the drill bit and slow down
the drilling operation thereby prolonging the overall drilling process. This type of
vibrations is detectable at the drillfloor by fluctuations in the power needed to
maintain a constant rate of surface rotation.

2
Rotational vibration
11
 Chapter 2 – Literature Review

2.2.3.1 STICK-SLIP

An important subset of torsional vibrations is the stick-slip behaviour of the

drillbit. Chapter 3 provides a more thorough description of the stick-slip phenomena.
Briefly, the stick-slip phenomenon is a self-excited torsional vibration induced by the
nonlinear relationship between the torque and the angular velocity at the bit [Jansen
et al. – 10]. It can either produce rotational speeds as high as ten times the nominal
rotary table speed or a total standstill of the bit. Stick-slip is summarized separately
from harmonic models as it represents a special torsional behaviour for the system
distinct from the harmonics of the BHA [Payne – 17].

2.2.4 LATERAL VIBRATION 3

One of the most destructive drillstring vibrations is lateral vibrations which

may be unleashed with no indication at the surface. Deep in the hole, the rotating
BHA interacts with the borehole wall generating shocks from lateral vibrations as
high as 250g [Jardine et al. – 23]. The collisions with the borehole wall will produce
eccentric hole and the shocks can damage components of the BHA. Thus the
deployment of Measurement-While-Drilling (MWD) tools and jars able to survive
the severe downhole environment will increase the cost of the hardwares
significantly.

Furthermore, the lateral oscillations of the drillstring cause severe damage to

the borehole wall [Jansen – 9] and affect the overall drilling direction [Millheim et
al. – 15]. Mitchell [21] mentioned that the presence of lateral vibrations may cause
BHA failure. Lateral vibrations have higher frequencies than their torsional
counterpart and are more difficult to predict or anticipate based solely on surface
measurements.

3
Transverse vibration
12
 Chapter 2 – Literature Review

2.2.4.1 WHIRL

The most recognizable manifestation of lateral vibrations is the whirling

behaviour of the bottomhole assembly. Drill collars whirling are simply the
centrifugally induced bowing of the drill collar resulting from rotation. If the center
of gravity of the drill collar is not initially located precisely on the centerline of the
hole, then as the collar rotates, a centrifugal force acts at the center of gravity causing
the collar to bend [Vandiver et al. – 13]. Forward and backward whirling behaviors
can further intensify due to the combined effect of fluid damping, stabilizer
clearance, and friction of the drilling assembly against the borehole wall [van der
Heijden – 4].

13
 Chapter 3 – Torsional Vibration

CHAPTER 3

TORSIONAL VIBRATION

In Chapter 2, the general perspective on vibration was discussed together

with the other modes and their specific subsets or consequences. In this chapter,
torsional vibration will be discussed specifically in relation to its model, phenomena,
consequences and remedies.

14
 Chapter 3 – Torsional Vibration

3.1 INTRODUCTION

Torsional vibrations have been studied since early 1960 [Finnie et al. – 6, 7].
Most of the previous literature [Kyllingstad et al. – 1] found that torsional vibration
is due to drillstring static effects. These previous works show that if the static friction
coefficient is sufficiently higher than the dynamic friction coefficient, then torsional
vibration will occur. The normal cause of this vibration is a “stick-slip” phenomenon.

When the drillstring rotates, the drill pipe begins to store torsional energy
subsequently leading to twist off. The BHA will also then begin to rotate and as the
static friction is higher than the dynamic friction, the stored energy in the drill pipe is
transmitted to inertial energy in the BHA. It will then accelerate to a speed faster
than the steady-state rotational speed.

Torsional vibration can be classified into two categories; transient and

stationary. Transient vibrations correlate with variations in drilling conditions, for
instance, heterogeneity in the rock. On the other hand, stationary vibrations are
caused by the natural resonance of the drillstring which is most likely to cause
problems. The most recognizable manifestation of stationary torsional vibration
during rotary drilling operation is stick-slip. This is due to friction between bottom-
hole assembly and wellbore [Jardine – 23]. Even though this phenomenon is still
debatable, it has nevertheless received much attention from the engineering
community, as its effects and influences on drilling performance are considerable.
Previous literature reviews on the subject of torsional vibrations include Dykstra [19]
and van den Steen [16]

Torsional vibrations are detectable at the surface for a considerable time. A

reasonable articulation is that the large inertia of the rotary table dominates the
overall drillstring dynamics at the surface. Control on the torsional behaviour of the
string can be achieved practically by varying the rotary speed and the weight on bit
(WOB), while modifying mud properties will ensure changes in downhole friction.
Changing the bit type or the configuration of the BHA may significantly mitigate the
effects of torsional vibrations.
15
 Chapter 3 – Torsional Vibration

3.2 MODELS

In the 1960s, some analyses have been carried out using the classical wave
equation to describe the torsional behaviour of drillstring assemblies [Finnie et al. –
6, 7]. In 1982, another approach was by assuming a friction-induced torsional
drillstring vibration mechanism. Fourier transforms have also been used to compute
torsional resonant frequencies [Halsey et al. – 5].

3.3 STICK-SLIP OSCILLATIONS

Stick-slip is a phenomenon where nonlinear wellbore friction induces a

torsional pendulum motion in the drillstring. The frequency of this oscillation is
lower than fundamental torsional frequency of the drillstring [Payne – 17].

During drilling the bit comes to a standstill, as the result of a sudden increase
in WOB, or the combined effects of significant drag, a tight hole, severe dog-legs
and key seats. The static friction that must be overcome to re-initiate the motion of
the bit may be significantly higher than the Coulomb (dry) friction normally applied
to the drilling assembly. As the rotary table is constantly turning, the drillstring
begins to store ever-increasing torsional energy. When torque is finally sufficient to
overcome the static friction, the torsional energy is suddenly released and the bit
comes loose. It whips at very high revolution-per-minute, sometimes ten times the
nominal rotational speed [Dufeyte et al. – 18]. As the torsional energy of the
assembly decreases its rotational speed and the bit ultimately comes to a standstill
again. The cycle may repeat several times. However, the phenomenon requires
specific drilling conditions to develop.

16
 Chapter 3 – Torsional Vibration

3.4 CONSEQUENCES AND REMEDIES

Stick-slip may result in extensive bit wear, backward rotation, severe torque
and shock loadings of the string, fatigue and eventually failure of drilling equipment
[Dufeyte et al. – 18]. It also decreases the rate of penetration (ROP) by typically 25%
due to the nonlinear relationship between the drilling rate and the bit rotational speed
[van den Steen – 16]. These torsional vibrations are an important factor for
consideration as they can cause drill pipe fatigue and may be extremely severe as to
damage the drillbit [Brett – 12]. Moreover, the whipping and high speed rotation of
the bit in the ‘slip’ phase can generate severe axial and lateral vibrations of the
bottom-hole assembly which may result in drillstring connection failures. At the
surface, the stick-slip phenomenon is sometimes characterized by a groaning noise
and saw-tooth like variations of large amplitude of the applied torque [Dufeyte et al.
– 18]. Recent Measurement-While-Drilling (MWD) tool enhancements have made it
possible to identify the stick-slip phenomenon and its severity while drilling [Robnett
et al. – 2], thus making it possible to adopt a corrective counter measure. Possible
solutions to stick-slip include higher drillstring stiffness, higher bottom-hole-
assembly inertia, increased rotational speeds, and a reduced difference between static
and dynamic frictions [van den Steen – 16].

In a parallel direction, an increasingly popular solution to stick-slip

(introduced in the late eighties) is to increase the drillstring damping by means of a
so-called soft torque system. The system aims at significantly reducing torque
fluctuations and torsional drillstring vibrations, thus influencing the stick-slip
conditions. The underlying concept is to reduce the amplitude of the downhole
torsional oscillations by means of a closed circuit providing torque feedback. The
feedback is to purposely slow down the rotary rate and directed to rotational drive.
This system is also commonly referred to as an impedance control or an active
damping system as presented by van den Steen [16].

17
 Chapter 4 – Analytical Approach

CHAPTER 4

ANALYTICAL APPROACH

The previous chapter discussed specifically the torsional vibration in relation

to its models, phenomena, consequences and remedies. This section formulates the
continuous equation for torsional motion. The dynamic behaviour of drillstring,
which has been mentioned several times earlier, has been investigated since the early
sixties [Finnie et al. – 6, 7]. This section introduces the finite element method (FEM)
as applied to the analysis of drillstring vibration.

18
 Chapter 4 – Analytical Approach

4.1 CONTINUOUS MODELING

Briefly, the stick-slip phenomenon consists of the repeated winding and

unwinding behavior of the assembly about its axis. Due to the damaging
consequences of the stick-slip phenomenon, it is desirable to be able to fully
understand and predict it, thereby justifying the study and investigation of torsional
modes of drillstring vibration. In the last four decades, significant effort has been
devoted to gaining insight into this problem by describing it with mathematical
models. One approach is to examine the torsional behaviour of the drilling assembly
using a continuous model. The equation governing the undamped torsional motion
T x, t of sections of the assembly, idealized as a linearly elastic bar, is a second-
order partial differential equation (PDE) which Finnie et al [6, 7] called the
undamped classical wave equation

w 2T x, t 1 w 2T ( x, t )
, (4.1.1)
w2 x c2 w2t

of which the general solution involves the superposition of terms to form

ª Zn x Zn x º
T x, t « An cos c  Bn sin c »>C n cos Z n t  Dn sin Z n t @ , n = 1, 2,… (4.1.2)
¬ ¼
Zn x
where An, Bn , Cn , Dn are arbitrary constants, and are dimensionless parameters
c
(phase angle) that describe the angular coordinate in terms of the vibration wave
length [Bailey et al. – 7]. The constants An, Bn and Cn, Dn are determined so that the
boundary conditions and the initial conditions be satisfied. The torsional wave
velocity c, can be expressed in terms of the shear modulus of rigidity, G and the
density U of the material as

G
c2 (4.1.3)
U

19
 Chapter 4 – Analytical Approach

Early investigations often considered the two vibration modes, axial and
torsional simultaneously [Finnie et al. – 6]. Therefore, it is possible to use Equation
4.1.1, with minor modifications to describe the torsional behaviour of the assembly.

Flow chart below illustrates process flow of numerical simulation using MATLAB.
Further equations can be referred to Section 4.1.1 and 4.1.2.

Start

Key – in the equations

x stick or slip motion

Input no. of mode, n

Output

Figure 4.1 Process Flow in MATLAB

20
 Chapter 4 – Analytical Approach

4.1.1 STICK PHASE

This approach utilizes an expansion of classical wave equation to determine

the natural frequencies of stick oscillations. Figure 4.2 represents the boundary
condition during stick phase.

l T 0, t T l, t 0

Figure 4.2 Boundary condition for drillstring during stick-phase

It is assumed that during the stick-phase, the drillstring is fixed at both ends
(rotary table and borehole). Therefore the boundary conditions are T 0, t T l, t 0
for all time t t 0 . Hence, from Equation E.9 (Appendix A), the followings are
obtained
T 0 0, (4.1.1.1) and T l 0, (4.1.1.2)

In order to satisfy Equation 4.1.1.1, An must be zero in Equation E.9

(Appendix A). Equation 4.1.1.2 requires that

Zl
B sin 0 (4.1.1.3)
c
Since Bn cannot be zero for a nontrivial solution, therefore
B

Zc
sin 0 (4.1.1.4)
l

21
 Chapter 4 – Analytical Approach

Equation 4.1.1.4 is called the frequency or characteristic equation and is

satisfied by several values of Z . The values of Z are called the eigenvalues (natural
frequencies of characteristic values) of the problem. The nth natural frequency is
given by
Znl
nS , n = 1, 2,..
c
or
ncS
Zn , n = 1, 2,.. (4.1.1.5)
l

The solution șn(x, t) corresponding to Zn can be expressed as

nSx ª ncS ncS º

Tn T n x Tn t sin «C n cos t  Dn sin t» (4.1.1.6)
l ¬ l l ¼

where C n and Dn are arbitrary constants. The solution T n x, t is called the nth
mode of vibration or nth harmonic or nth mode of the drillstring. In this mode, each
point of the drillstring vibrates with amplitude proportional to the value of T n at that

ncS
point, with the circular frequency Z n . The function T n x is called the nth
l
normal mode, or characteristic function. The first three modes of vibration are shown
in Figure 5.1 in Chapter 5. The mode corresponding to n = 1 is called the
fundamental mode, and Z1 is called the fundamental frequency.

The general solution of Equation E.2 (Appendix A), which satisfies the boundary
conditions of Equation 4.1.1.1 and 4.1.1.2, is given by the superposition of
all T n x, t :
f f
nSx ª ncS ncS º
T x, t ¦ T n x, t
n 1
¦ sin
n 1 l ¬«C n cos
l
t  Dn sin
t
t»
¼
(4.1.1.7)

22
 Chapter 4 – Analytical Approach

4.1.2 SLIP PHASE

This approach uses an expansion of classical wave equation to determine the

natural frequencies during slip oscillations. Figure 4.3 represents the boundary
condition during slip phase.

T 0, t 0,
l
wT
l, t 0,
wx

Figure 4.3 Boundary condition for drillstring during slip-phase

It is assumed that during the slip-phase, the drillstring is fixed at the rotary
table and free at the bottom (borehole). Therefore the boundary conditions for all
time t t 0 can be expressed as

T 0, t 0, (4.1.2.1)
wT
l, t 0, (4.1.2.2)
wx

The use of Equation 4.1.2.1 in Equation E.9 (Appendix A) gives An 0 while the
use of Equation 4.1.2.2 gives the frequency equation

Z Zl Zl
Bn cos 0 or cos 0 (4.1.2.3)
c c c

23
 Chapter 4 – Analytical Approach

The eigenvalues are given by

Znl S
2n  1 , n = 0, 1, 2,…
c 2
or
2n  1 cS
Zn , n = 0, 1, 2… (4.1.2.4)
2l

The solution T n x, t is called the nth mode of vibration or nth harmonic or

nth mode of the drillstring. In this mode, each point of the drillstring vibrates with
amplitude proportional to the value of T n at that point, with the circular

2n  1 cS
frequency Z n . The function T n x is called the nth normal mode, or
2l
characteristic function. The first three modes of vibration are shown in Figure 5.2 in
Chapter 5. The mode corresponding to n = 0 is called the fundamental mode, and Z 0
is called the fundamental frequency.

Thus the total solution of Equation E.2 (Appendix A) can be written, using the mode
superposition method, as

f f
2n  1 Sx ª 2n  1 Sct 2n  1 Sct º
T x, t ¦ T n x, t
n 0
¦ sin
n 0 2l «
¬
C n cos
2l
 Dn sin
2l »¼ (4.1.2.5)

24
 Chapter 4 – Analytical Approach

4.2 FINITE ELEMENT MODEL

Sl

So

Twisting moment

Figure 4.4 Torsional beam element for finite element model.

The finite element technique is selected as the solution procedure in the

ensuing analysis. This requires the discretization of the system, a set of discrete
equations of motion characterizing the torsional response of the assembly may be
derived in the form:
.. . .
§ ·
M T t t  Ct T t t  K t T t t Gt ¨ t , T t , T t ¸ , (4.2.1)
nxn nx1 nxn nx1 nxn nx1 nx1 © ¹

where T t is a vector monitoring the kinematics of the nodes of the model. Figure 4.4
shows the quantities that are necessary to adequately represent the torsional
behaviour of an element. They are twisting moments So and Sl. The rotation within
one element may be expressed in terms of the torsional displacements of its two end
nodes [Spanos – 20]. That is,

T x, t \ o x So t \ l x Sl t , (4.2.2)

25
 Chapter 4 – Analytical Approach

In this form, T x, t is called a Langrange interpolation polynomial in which

provides a set of boundary conditions that the shape functions must satisfy

\o 0 1,\ o l 0
, (4.2.3)
\l 0 0,\ l l 1

where l is the element length. Next, considering the torsional deformations due to a
static load permits deriving shape function, yields

x
T x c1  c 2 , (4.2.4)
l

Substituting the boundary conditions 4.2.3 into Equation 4.2.4 allows the
determination of a set of appropriate trial functions

x x
\o x 1  , and \ l x , x  >0, l @ (4.2.5)
l l

As in the analytical analysis, damping is neglected. Therefore, the components of the

stiffness and mass matrices are respectively given as:

l
d\ i x d\ j x
k ij ³ GJ
0
dx dx
dx , (4.2.6)

and
l
mij ³ UJ\
0
i x \ j x dx , (4.2.7)

26
 Chapter 4 – Analytical Approach

Substituting Equation 4.2.5 into Equation 4.2.6 and Equation 4.2.7 yields the
element stiffness and mass matrices

GJ ª 1  1º
K te « 1 1 », (4.2.8)
2u2 l ¬ ¼
and

UJl ª2 1º
M te , (4.2.9)
2u2 6 «¬1 2»¼

With K te and M te the local element stiffness matrix is formulated and hence, the
global element stiffness matrix.

4.3 RAYLEIGH DAMPING 4

If damping is to be considered, Rayleigh damping as suggested by Spanos [20] is as

below. In structural dynamics, the systems are usually in the form of

Mx  Cx  Kx F (4.3.1)

where C is the viscous damping matrix. A particularly convenient form of C is the

Rayleigh damping matrix

C aM  bK 4.3.2)

where a and b are parameters for the desired damping characteristics.

4
Viscous damping
27
 Chapter 4 – Analytical Approach

4.4 NUMERICAL SIMULATION

Finite element analysis (FEA) has become more prominent in recent years.
Numerical solutions to even very complicated stress problems can now be obtained
routinely using FEA, so its principle features are outlined.

The finite element method is a technique for obtaining approximate solutions

to ordinary differential equations (ODE) or partial differential equations (PDE). It
finds application in situations where either the equation to be solved is complex or
the geometry on which the desired solution is complex. The method exchanges
solving a single ODE or PDE on a large complex domain, for solving a much larger
number of algebraic equations on smaller simply shaped pieces that make up the
actual solution domain.

ANSYS is a general purpose finite element modelling package for solving a

wide variety of problems. The problems included static/dynamic structural analysis
(linear and non – linear), heat transfer and fluid mechanics problems, acoustics,
electromagnetic and even coupled field problems involving thermal – structural
interactions, fluid – structural interactions, etc. In this project, modal analysis is
utilized to determine the vibrational characteristics (natural frequencies and mode
shapes) of the drillstring.

28
 Chapter 4 – Analytical Approach

Block diagram below illustrates the process flow using ANSYS

Start

Pre – processing
x Reads or generates nodes and elements
(e.g. ANSYS) Problem
x Reads or generates material property data Definition
x Reads or generates boundary conditions
(loads and constraints)

Processing
x Generates element shape functions
x Calculates master element equations
x Calculates transformation matrices
x Maps element equations into global
system
x Assembles element equations
x Introduces boundary conditions
x Performs solution procedures

Post – processing
x Prints or plots contours of displacements Analysis and Design
Decisions
x Evaluates

Stop

Figure 4.5 Process Flow in Finite Element Analysis using ANSYS

29
 Chapter 5 – Results and Discussions

CHAPTER 5

RESULTS AND DISCUSSIONS

Even though the various approaches in obtaining solutions to stick – slip

vibration have been discussed, only classical wave equation with specific boundary
conditions is considered. The result obtained is then numerically simulated using
MATLAB in order to analyze the responses of stick and slip motion. Discussions
will then be made according to the results obtained. Table 5.1 and Table 5.2 are
summaries of both stick and slip motions. Comparison between analytical analysis
and numerical simulation has been discussed. Experimental results also will be
presented at the end of this chapter.

30
 Chapter 5 – Results and Discussions

5.1 RESULTS
5.1.1 STICK PHASE

The first three modes of vibration are shown in Figure 5.1. The mode corresponding
to n = 1 is called the fundamental mode, and Z1 is called the fundamental frequency.

Figure 5.1 First Three Modes of Vibration during Stick motion

Note: In Figure 5.1, axis – y represents the length of the string while axis – x
represents displacement.

31
 Chapter 5 – Results and Discussions

5.1.2 SLIP PHASE

The first three modes of vibration are shown in Figure 5.2. The mode
corresponding to n = 0 is called the fundamental mode, and Z 0 is called the

fundamental frequency.

Figure 5.2 First Three Modes of Vibration during Slip motion

32
 Chapter 5 – Results and Discussions

5.2 DISCUSSIONS

Table 5.1 Conditions of Drillstring during Stick Motion

Conditions of Stick Phase
Drillstring
Boundary T 0 0
Conditions T l 0
Frequency Zl
sin
Equation c

Natural ncS
Zn , n = 1, 2, 3..
Frequencies l

Total Solution f
nSx ª ncSt ncSt º
T x, t ¦ sin
n 1 l ¬«C n cos
l
 Dn sin
l »¼

Table 5.2 Conditions of Drillstring during Slip Motion

Conditions of Slip Phase
Drillstring
Boundary T 0, t 0
Conditions wT
l, t 0
wx
Frequency Zl
cos
Equation c

Natural 2n  1 cS
Zn , n = 0, 1, 2…
Frequencies 2l

Total Solution f
2n  1 Sx ª 2n  1 Sct 2n  1 Sct º
T x, t ¦ sin
n 0 2l «
¬
C n cos
2l
 Dn sin
2l »¼

33
 Chapter 5 – Results and Discussions

In this work, the classical wave equation has been applied to determine the
natural frequencies and illustrate the mode shapes of vibration of both stick-phase
and slip-phase during the torsional vibration. In this approach, it is assumed that
there is no influence of axial and lateral vibration as well as coupled vibrations in the
system – drillstring. It is also assumed that the approach is based on single string.
During the stick phase it is believed that there is no displacement at both ends –
rotary table and borehole. As a result, the system is considered as fixed at both ends.
Therefore, the natural frequencies during stick phase are derived based on boundary
conditions – Equation 4.1.1.1 and 4.1.1.2.

Similarly, during slip phase, the bottom part of drillstring has no contact
between drill-bit and borehole. Again, the natural frequencies of the system are
derived based on boundary conditions – Equation 4.1.2.1 and 4.1.2.2. It is believed
that the stick-slip phase depends on the friction – contact between drill-bit and
borehole.

Equations derived are then simulated using computational numerical tool

(MATLAB) to evaluate the modes of vibrations – first three modes of vibrations
(Figure 5.1 and 5.2). MATLAB was chosen because of its capability in solving the
non-linear equation. Apart from that, it is also user-friendly.

Table 5.1 and Table 5.2 show that the natural frequencies during slip phase
are comparatively higher than during stick phase – Equation 4.1.2.4 (Slip Phase) and
Equation 4.1.1.5 (Stick Phase). Hence this indicates that during slip phase, the drillbit
may cause whirling during the higher modes of vibrations.

Based on Figure 5.1 and 5.2, it is shown that the alternating series of stick-
slip phase may cause other forms of vibrations – coupled vibrations. As stick-slip
motion results in harmonic, there will be interaction between torsional and axial or
torsional and lateral or combination of three types of vibrations.

34
 Chapter 5 – Results and Discussions

Besides, the response of the natural frequencies may also vary depending on
the modes of vibrations. Figure 5.3 shows the condition for both stick-slip phases
when the nth order is 50. Figure 5.1 to 5.3 indicate that the response of vibrations
depend on the number of modes.

Stick Slip
Figure 5.3 Modes of Vibration during Stick – Slip Motion at n=50

In Figure 5.5 also indicates that at higher modes of vibrations during stick
phase the vibrations may transmit longitudinally and can be detectable at the surface
whereas during slip phase the bottom part of the drillstring may slip and cannot be
detectable from the surface and may cause lateral vibration. The high speeds
achieved during the slip phase are also one of the causes of the third mode of
vibration - lateral vibrations. Furthermore for deviated wells, it may be the cause of
pipe buckling. Unfortunately, analysis of lateral vibration cannot rely on a harmonic
model.

35
 Chapter 5 – Results and Discussions

5.3 COMPARISON

The material properties in this work are

G = 74Gpa
ȡ = 2710kg/m3
l = 100m
where G is the shear modulus of rigidity, U is the density of the material and l is
length of the element.
Equation 4.1.1.5 and 4.1.2.4 are computed to determine natural frequencies of stick
and slip respectively. The results of both analytical and finite element method are
tabulated below.

Table 5.3 The Analytical Results against FEM Results during Stick Phase
Mode Shape Analytical Solution (Hz) FEM (Hz)
First 100.66 127.72
Second 201.31 194.36
Third 301.97 291.89

Table 5.4 The Analytical Results against FEM Results during Slip Phase
Mode Shape Analytical Solution (Hz) FEM (Hz)
Zero 50.33 69.584
First 150.98 161.80
Second 251.64 197.19

With reference to Table 5.3 and 5.4, note that there are differences between both
results. The difference can be attributed to the accuracy limitation of the FEM
software. Moreover, the software has limited space capacity in order to be able to
generate the required numbers of meshes. Nevertheless, both approaches have
produced similar mode shapes. Hopefully with this initial work, companies from
related industry may be encouraged to embark on future researches/studies by
providing resources and/or financial grants in order that more precise and accurate
results may be obtained.

36
 Chapter 5 – Results and Discussions

To complement the result of this study, the result of the experimental work
that has been carried out by fellow colleague [Asreen – 2006] is solicited. Asreen has
recorded the mode shapes and natural frequencies from the experimental setup using
an accelerometer. The results are shown in Figure 5.6 and 5.7. Similarly, the results
display similar mode shapes but different values of natural frequencies.

Figure 5.4 Experimental Result of Stick Phase

37
 Chapter 5 – Results and Discussions

Figure 5.5 Experimental Result of Slip Phase

38
 Chapter 6 – Conclusions and Recommendations

CHAPTER 6

CONCLUSIONS AND RECOMMENDATIONS

Chapter 5 provided discussions and explanations of the results obtained

through numerical simulation using MATLAB and ANSYS. In this chapter, the
conclusions together with recommendations for future studies are presented.

39
 Chapter 6 – Conclusions and Recommendations

6.1 CONCLUSIONS

The stick and slip phase have been discussed analytically in section 4.1.1 and
4.1.2. The main difference between those phases is their boundary condition due to
friction. It can therefore be concluded that the friction induces torsional vibration.
This indicates that the study of friction is very crucial in understanding the dynamics
behaviour of drillstring.

It can also be concluded that this work can be applied in suppressing the
torsional vibration and may be useful for further development especially in the oil
and gas well drilling operation.

6.2 RECOMMENDATIONS

Although an extensive study on the subject has been conducted, it is beyond

doubt that further interest and development is necessary. Areas for further studies are
summarized as follows:
ƒ The analytical approach developed was based on a simple string. The
limitation of assumption is restricted to other modes of vibration such as axial
and lateral vibration as well as their coupled vibrations.
ƒ In this study, it is assumed that the drillstring has similar torque throughout
the whole Drillstring whereas in real-life situation the surface torque differs
significantly from the bottom torque.
ƒ ‘Unlimited’ software should be provided to achieve more precise and
accurate results.

40
 References

REFERENCES

1. Å. Kyllingstad & G. W. Halsey. A Study of Slip/Stick Motion of the Bit,

SPE Drilling Engineering, Vol. 3, No. 4, 369–373, 1988.

2. E. W. Robnett, J. A. Hood, G. Heisig & J. D. Macpherson. Analysis of the

Stick-Slip Phenomenon Using Downhole Drillstring Rotation Data,
SPE/IADC 52821, 1999 SPE/IADC Drilling Conference held in
Amsterdam, Holland, 9–11 March 1999.

3. E. Skaugen. The Effects of Quasi-Random Drill Bit Vibrations upon

Drillstring Dynamic Behavior, SPE 16660, Society of Petroleum
Engineers, 1987.

4. G. H. M. van der Heijden. Nonlinear Drillstring Dynamics, Ph.D Thesis,

University of Utrecht, 1994.

5. G. W. Halsey, Å. Kyllingstad, T. V. Aarrestad & D. Lysne. Drillstring

Torsional Vibrations: Comparison between Theory and Experiment on a
Full-Scale Research Drilling Rig, SPE 15564, 1986.

6. I. Finnie and J.J Bailey. An Experimental Study of Drill-String Vibration,

Journal of Engineering for Industry, Transactions of the ASME, 82(2),
May 1960.

41
 References

7. J.J. Bailey and I. Finnie. An Analytical Study of Drill-String Vibration,

Journal of Engineering for Industry, Transactions of the ASME,
82(2):122-128, May 1960.

8. J.D. Jansen. Nonlinear Dynamics of Oilwell Drillstrings. Delft University

Press, 1993.

9. J. D. Jansen. Whirl and Chaotic Motion of Stabilized Drill Collars, SPE

Drilling Engineering, June 1992.

10. J. D. Jansen & L. van den Steen. Active Damping of Self-Excited

Torsional Vibrations in Oil Well Drillstrings, Journal of Sound and
Vibration, 179, No. 4, 647-668, 1995.

11. J. D. Macpherson, J. S. Mason & J. E. E. Kingman. Surface Measurement

and Analysis of Drillstring Vibrations While Drilling, SPE/IADC 25777,
Society of Petroleum Engineers, 1993.

12. J. F. Brett. The Genesis of Torsional Drillstring Vibrations, SPE Drilling

Engineering, September 1992.

13. J. K. Vandiver, J. W. Nicholson & R. J. Shyu. Case Studies of the

Bending Vibration and Whirling Motion of Drill Collars, SPEDE, SPE
18652, pp. 282-290; 1989, December 1990.

14. J. S. Mason & B. M. Sprawls. Addressing BHA Whirl: The Culprit in

Mobile Bay, SPE Drilling and Completion, December 1998.

15. K. K. Millheim & M. C. Apostal. The Effect of Bottomhole Assembly

Dynamics on the Trajectory of a Bit, Journal of Petroleum Technology,
Vol. 33, No. 12, 2323–2338, 1981.

42
 References

16. L. van den Steen. Suppressing Stick-Slip-Induced Drill-string

Oscillations: A Hyper Stability Approach, Ph.D Thesis, University of
Twente, 1997.

17. M. L. Payne. Drilling Bottom-Hole Assembly Dynamics, Ph.D Thesis,

Rice University, 1992.

18. M. P. Dufeyte, H. Hcmneuse & E. Aquitaine. Detection and Monitoring

of the Slip-Stick Motion: Field Experiments, SPE/lADC 21945, 1991.

19. M. W. Dykstra. Nonlinear Drillstring Dynamics, Ph.D Thesis, University

of Tulsa, 1996.

20. P. D. Spanos & A. M. Chevallier. A Primer for Deterministic and

Stochastic Drill-String Vibrations Analyses, Springer-Verlag, 2002.

21. R. F. Mitchell & M. B. Allen. Case Studies of BHA Vibration Failure,

SPE 16675, Society of Petroleum Engineers, 1987.

22. R. J. Shyu. Bending Vibration of Rotating Drill Strings, Ph.D Thesis,

Massachusetts Institute of Technology, 1989.

23. S. Jardine, D. Malone & M. Sheppard. Putting a Damper on Drilling's

Bad Vibration, Oilfield Review, January 1994.

24. V. Dunayevsky, F. Abbassian & A. Judzis. Dynamic Stability of

Drillstrings under Fluctuating Weight on Bit, SPE Drilling and
Completion, Vol. 8, No. 2, 84–92, 1993.

43
 APPENDIX A

CLASSICAL WAVE EQUATION

44
The free vibration equation, Equation E.1, can be solved by the second – order partial
differential equations. In this method, the solution is written as the product of a
function T x , which depends only on x and a function T t , which depends only on
t

w 2T x, t 1 w2T ( x, t )
, (E.1)
w2 x c2 w2t

T x, t T xT t, (E.2)

Substitution of Equation E.2 into Equation E.1 leads to

c 2 d 2T 1 d 2T
, (E.3)
T dx 2 T dt 2

Since the left hand side of this equation depends only on x and the right hand side
depends only on t , their common value must be a constant – say, a so that

c 2 d 2T 1 d 2T
a, (E.4)
T dx 2 T dt 2

The equation implied in Equation E.4 can be written as

d 2T a
 T 0, (E.5)
dx 2 c 2

d 2T
 aT 0, (E.6)
dt 2

45
Since the constant a is generally negative, we can set a Z 2 and write Equation
E.5 and Equation E.6 as

d 2T Z 2
 T 0, (E.7)
dx 2 c 2

d 2T
 Z 2T 0, (E.8)
dt 2

The solution of these equations are given by

Zx Zx
T x A cos  B sin , (E.9)
c c
T t C cos Zt  D sin Zt , (E.10)

where Z is the frequency of vibration and the arbitrary constants A, B, C and D can
be evaluated from the boundary and initial conditions.

Therefore general solution for Equation E.2

ª Zx Zx º
T x, t « A cos c  B sin c »>C cos Zt  D sin Zt @
¬ ¼

46
 APPENDIX B

EXPERIMENTAL SET – UP

47
 Experimental Set – up Instruments

Exciter
In this work impact hammer has been utilized as an exciter to stimulate the loadings.

Transducer
The piezoelectric type of transducer has been used as means of measuring parameters
of interest in vibrational testing. In this work, the accelerometer ACC 3715 typed
(0.01138 V/PU) was used. In an accelerometer, transduction is indirect and achieved
using auxiliary or seismic mass.

Analyzer
In this project for dbRTA software, Spectrum (Fourier) Analyzers through 4 –
Channel FFT analyzer was used.

Excitation Signal
In this project allows the real – time analysis of transient phenomena.

48
Experimental Set – up for Stick Phase

Set – up for Stick Phase

The rod is divided into ten sections which represent the ten nodes for the
analysis. Therefore it will be eleven nodes altogether. In preparing the test, the rod is
placed at the clamped by using clamper. The setups are base either on fixed-free or
fixed-fixed condition which represent stick phase and slip phase respectively. Figure
above and below illustrate the setup for each condition. Then the accelerometer is
placed on the rod. One important consideration is the placement of accelerometer.
Since the specimen has been divided the plate into 11 points (nodes) for excitation, it
is agreed to place the accelerometer at point 3 at the rod that is attached by glue.
After that, setup of dbRTA software play as major rule. The setup of dbRTA must
follow into four major rules as hardware configuration, general parameter,
processing and visualization respectively. Then, the rod is excited by using impact
hammers. The impact hammer will deliver force to the accelerometer with a
controlled amplitude and frequency using its feedback controlled amplifiers.
49
Experimental Set – up for Slip Phase

Set – up for Slip Phase

50
APPENDIX C

RESULTS

51
Comparison: Stick Phase Results
Natural Frequencies vs. Mode Shapes

400

350

300
Natural Frequencies (Hz)

250

Exact
200 Numerical
Experimental

150

100

50

0
1 2 3
Mode Shapes of Vibration

Comparison: Slip Phase Results

Natural Frequencies vs. Mode Shapes

300

250

200
Natural Frequencies (Hz)

Exact
150 Numerical
Experimental

100

50

0
1 2 3
Mode Shapes of Vibration

52
Stick Phase
Exact Numerical Experimental
First Mode of Vibration

Second Mode of Vibration

Third Mode of Vibration

Mode Shapes of Vibration

53
Slip Phase
Exact Numerical Experimental
Zero Mode of Vibration

First Mode of Vibration

Second Mode of Vibration

Mode Shapes of Vibration

54