Design of a Testing Facility for Investigation of Drill Pipes Fatigue Failure

Drillstring and down-hole tool failure usually results from failing to control one or more of the vibration mechanisms. The solution starts with the ability to measure different modes of vibration, hence identifying different vibration mechanisms. Lateral, torsion and axial are vibration modes that take place when drill pipes run into problems downhole. Due to the three modes of vibration mechanisms such as bit bounce, stick-slip, lateral shocks, bit and bottom hole assembly (BHA) whirl, parametric and torsional resonance occur. Understanding the causes of the destructive loads is the main step towards developing approaches to prevent or reduce their effects, hence improving drilling performance. Vibration modes and mechanisms lead to failure of the drill pipes, BHA and drill bits. Drill pipes fatigue failure is very common due to capability of producing all vibration modes and mechanisms. Drill pipe and downhole tool assembly failure usually result from failing to have power over one or more of these vibration mechanisms. A novel in house experimental setup has been developed to mimic downhole axial, lateral and torsional vibration modes and mechanisms in drilling operations. In this paper, we focus on the design and construction of the testing facility. A number of tests were conducted to validate the capability and performance of the test setup. Drill pipe fatigue failure due to lateral cyclic stresses induced in the drill pipe has also been investigated and presented in this paper. The results show that operating on a rotation speed higher than 90% of the drillstring critical speed leads to yielding in the


Introduction
Drillstring consists of a drill pipe and Bottom Hole Assembly (BHA).The drill pipe is a long hollow shaft that drives a bit at the bottom of the wellbore.Drillstring failure occurs in approximately one out of seven drilling rigs in low to moderate well depth and three out of seven in deep drilling wells.The cost of each failure ranges from $100,000 to $200,000 depending on well depth.Due to the frequent and high cost of drillstring failure, investigations have focused on raising awareness of the factors causing drillstring failure, and eliminating or reducing the occurrence of such failure incidents.
During drilling operations, drillstring interacts continuously with the rock formation, which results in severe shock and vibrations.Drillstring shock and vibrations are identified as the principal sources of performance deterioration in oil and gas drilling operations (Macdonald and Bjune 2007;Moradi and Ranjbar 2009;Reid and Rabja 1995).Drill pipe, BHA and drill bit are affected by the vibration induced as a result of high input impact energy.This energy introduces different states of stresses, which translate into excess vibrations that may lead to failure.The severity of shock and vibrations depends on three parameters: shock magnitude, duration or length of time and frequency or number of shocks.The main causes of these vibrations include: contact and friction at the borehole/drillstring and bit/rock formation interfaces, eccentricity, imbalance, initial curvature in the drill collar sections, and various linear or nonlinear resonances.Severe shock and vibration may subsequently produce fatigue and abrasive wear in drill pipes, drill bit damage and reduction of penetration rate ROP (Jardine et al. 1994;Spanos et al. 2003;Khulief and Al-Naser 2005).As a consequence, the drilling process becomes inefficient and costly.In a drilling operation, drillstring exhibits one or more of three common modes of vibrations: axial, torsion and lateral.Preventing drill pipe failure calls for addressing all possible factors that can cause failure.Sometimes this is as simple as managing loads and load capacities, and having a good inception of the drilling operation.However, with complex causes, much more is involved in order to prevent failure (Jardine et al. 1994).Regardless of vibration mechanisms, drill pipes happen to deteriorate as a result of flaws in one or more of the following areas: design, inspection, operation and characteristics of drilling operation.The mechanical and chemical environments surrounding the drillstring can have a major effect on whether or not failure may take place.Vibration, mud type, dissolved gas content, salinity, and other factors play key roles in many drill pipes failure.
Lateral vibrations are recognized as a leading cause of drillstring and BHA failures.Lateral vibration also damages the well when BHA large shocks impact the wellbore.Several attempts have been made to theoretically study drill pipe lateral vibration in order to overcome the difficulties encountered during drilling process (Berlioz et al. 1996;Yigit and Christofourou 1996).Most of the mathematical models report slight progress in order to determine causes of drill pipe failure.However, not much has been reported about the procedures to accurately account for the effects of induced lateral vibration in drillstrings during drilling operations.Very few modeling attempts have been undertaken in order to address the effects of lateral vibration on drill pipes, drill bit, and BHA.It suffices to say that nearly all models are very much simplified and far away from addressing the real problem.One should note that modeling of such a long rotating element is non-linear and complex due to the fact that drillstring dynamics involve broad vibration profile, which includes axial, lateral and torsional modes (Baryshnikov et al. 1997;Wang at al. 2016).In addition, drilling process is very complex as many influencing parameters are not under drilling engineers' control (Abdo and Al-Sharji 2015).Some attempts have been made to study the non-linear dynamics of the drilling string.Non-linear treatments were proposed in (Al-Hiddabi et al. 2003, Abdo 2006, Abdo et al. 2015, Farhang and Lim 2007).
The authors undertook such a procedure to suppress torsional vibrations and reduce lateral vibrations of drilling sting non-linear modes.
This paper aims at gaining a deeper understanding of drill pipe complex behavior under vibrations and its detrimental effect on drilling operation.The present work focuses on the design and development of an in-house testing facility.The developed facility is proposed to be capable of generating various vibration modes that a drillstring experiences when it runs into hard rock formation during drilling operations.The facility intends to provide further expansion for more comprehensive drillstring testing.It can readily accommodate other relevant effects such as wellbore and drillstring contact, drillstring interaction and stick-slip phenomenon.The controlled testing facility also aims to offer oil and gas industries with satisfactory answers and justifications on the drillstring performance under various loading and operating conditions.The investigations are expected to lead to better understanding of drill pipe behaviors.It also accounts for the number of cycles to fatigue failure due to induced lateral vibration in the drill pipe under various loading conditions.

Drillstring Testing Facility
Drilling processes are complex as many of their influencing parameters are not in control of drilling engineers.In many situations, mathematical formulations do not necessary lead to accurate explanations of failure causes.Drill pipe fatigue failure is one of the most frequent failures that occur due to rapid and continuous cyclic stresses induced by vibration and high shock loads.Mathematical and computational models have important roles in the investigation of vibrations in drillstrings.Such models enable configuration of drillstring behaviour under range of loading, physical and boundary conditions.However, a combination of the drillstring geometry (long thin-walled rotating object) as well as forces and torques acting on it during drilling operations lead to complications in drillstring dynamic response and broad vibration profile that includes axial, lateral and torsional patterns.Hence, there is a need to develop an inhouse testing facility that is capable of imitating drilling process and has the potential of generating the three different types of vibrations.The testing facility is also utilized to investigate drillstring connections behavior under all modes of vibrations.
In the current setup, a braking mechanism was developed and assembled to the lower side of the facility in order to introduce torsional vibration in the drill pipe.The braking mechanism is employed to induce torsional vibration of various amplitudes in the drill pipe.Lateral vibration is created due to the eccentric rotation of the drill pipe.The frequency and amplitude of the lateral vibration are controlled by varying the drillstring rotational speed.The following sections briefly discuss the testing facility design and construction.In subsequent sections, the effect of inducing lateral vibration of the drill pipe on its lifetime fatigue failure is experimentally investigated utilizing the developed testing facility.

Experimental Setup
The experimental setup shown in Fig. 1 consists of the following: a testing pipe that represents the drillstring; support system that includes supporting plates fixed to the ground to provide stability for the facility, C-Channels beams to connect the supporting tables; loading and rotation mechanism which includes motor, pulleys, belts, chucks, pillow bearings, connections, flanges and housing pipes; mild steel cylinders representing the wellbore and also used as stabilizers for the drillstring; a control unit and speed controller; sensing mechanism, that includes load cells and sensors for measuring top and bottom torques, rotation speed, number of cycles, and vibration amplitude; braking mechanism and pressure mechanism utilized for inducing torsional vibration in the drill pipe at various frequencies and amplitudes; and a data acquisition system which includes twelve channels data logger and computer.Figure 1a shows an image of the facility while Fig. 1b shows a schematic of the facility and its individual components and systems.The representations for the numbers indicated on Fig. 1b are listed in Table 1.
The facility structural support consists of two 16 mm thick mild steel tables, two ring shape stabilizers and four C-channel beams connecting the tables to the stabilizers and providing stability to the facility.The C-channels are fixed to a concrete floor.The structural support is responsible for providing safe working environment and holding other mechanisms such as rotation and braking mechanisms.The rotation mechanism is responsible for rotating drill pipe and generating lateral vibration.Three phase 12 hp electric motor rotates the pulley whereas the controller regulates the drill pipe rotational speeds.The V-belts rotate the housing pipe at a specified rotational speed hence the drill pipe rotates at the same rotation speed as it is connected to a chuck as shown in Fig. 1.The three- Caliper and pads 20 Brake disc jaw chuck is connected to the housing pipe.A similar housing pipe and a chuck are fixed on the other end of the testing facility.The housing pipe guides the drillstring through these for about 0.5 m on each end of the facility to provide stability and rigidity to the drill pipe.The housing pipes are also guided through two pillow block bearings.Such configuration allows for smooth rotation.A number of test specimens (mild steel pipes) representing drillstrings of various external diameters (1", 2", and 3") were used during testing.The drill pipe is 6 m long; however, the length between the two chucks is 5 m.A total of 1 m of the drill pipe is inserted inside the two sides of housing pipes using three-jaw chuck and pillow block bearing.Fig. 1a illustrates a zoomed-in image of the drill pipe, chuck and flange to housing pipes connection.

Braking Mechanism
A brief description of the braking mechanism is discussed in this section.The braking mechanism is used to provide time varying resistance to simulate wellbore rock formation resistance and prompt torsional stress in the drill pipe.Such engagement should also simulate stick-slip phenomenon in drilling operations.Table 2 includes     The braking mechanism consists of a steel hosting structure and a linear pneumatic actuator that is connected to the brake pedal to initiate braking action.Booster and master cylinder are both bolted into a C-channel, which is welded to the testing facility base.Pneumatic double acting actuators with controlling valves are used in the system.Required pressure needed to actuate the braking mechanism is provided by 12 bar compressor.A control unit, as shown in Table 2, has been utilized to control stroke pressure, the braking time and the time of engagement.Figure 2 configures exploded view of the brake mechanism parts using 3D CAD modeling.Table 2 also illustrates parts of breaking mechanism.

Results and Discussion
The experiments were carried out under controlled loading and unloading conditions to examine the facility performance and investigate the effects of inducing lateral vibration on the drill pipe fatigue failure.One of the useful characteristics is to rotate the drill pipes near their resonances.This allows for quick testing using constructed facility despite the high strength of the component.The typical drill pipes diameters used in the tests are 1", 2" and 3".Two steel rings are placed to constrain the drilling pipe in case of catastrophic failure for safety reasons.Ultrasonic displacement sensors are used to measure the displacement at the middle of the drill pipe.Two Hall Effect sensors are placed; one near the first chuck and the other near the second chuck.The sensors are utilized to record the pipe's rotational speed near the motor and the braking mechanism.
Displacement, deformation and rotational speed have offered a complete description of the drill pipe's dynamic vibration profile.The drill pipe rotational speed is constrained up to a certain critical rotation close to the drill pipe resonance.
Sets of 1", 2", and 3" diameters and 6 m long drill pipes were selected and utilized in the testing procedure.The first few experiments were performed by rotating the 1" diameter drill pipes at various rotational speeds.The lateral vibration amplitudes (deflection from the center axis of the drill pipe) were then measured using ultrasonic deflection sensors.The experiments were repeated using drill pipes of 2" and 3" diameters.The drill pipes' lateral deflections versus the rotational speeds for different pipe sizes is shown in Fig. 3 The maximum lateral vibration amplitude (donated as AMP) takes place at the middle of the drill pipe.It can be observed from Fig. 3 that the deflection is lower when a larger drill pipe diameter is used and is higher when a smaller diameter pipe is used for corresponding rotational speed.The lateral vibration frequency increases almost linearly as the drill pipe rotational speed increases.
Figure 4 illustrates a 1" diameter drill pipe that has been subjected to a rotational speed near its resonance or critical speed.The deflection suggests that any further increase in the rotational speed might lead to unacceptable deformation of the drill pipe.Theoretical resonance rotational speeds of the three drill pipes used in the experimental work were calculated using Eq. ( 1) and presented in Table 3. (1)   f: Natural frequency g: Gravitational constant E: Modulus of elasticity I: Moment of inertia : Span length λ: Frequency factor, dimensionless μ: Weight per unit length The second set of tests was performed to examine the fatigue failure due to induced lateral vibration in the drill pipes.Drill pipes of three sizes were rotated at various rotational speeds and the number of cycles to fatigue failure were recorded (crakes were also visually observed).Figure 5 shows number of cycles to fatigue failure for each drill pipe against various rotational speeds.It can be noticed from the figure that the number of cycles to failure is decreased for all drill pipes as the rotational speed is increased.Meanwhile, the number of cycles to failure are increased as the diameters of drill pipes are increased.Figure 5 also demonstrates that operating at a rotational speed near the resonance of each drill pipe, which is approximately at 10% less than the resonance rotational speed as it is given in Table 3, has significantly reduced the number of cycles to failure.The reduction in the number of cycles to failure in the vicinity of resonance is due to the high lateral vibration induced in each drill pipe as the rotational speed is about 90% of the drill pipes resonance rotational speed.The 90% resonance rotational speeds for the three-size drillstrings are presented in Table 3.It is worth mentioning here that rotating drillstrings at a speed higher than 90% of their critical speeds leads to yielding of drillstrings.Figure 6 shows a 1" diameter drillstring that has been yielded due to its operation at about 94% of its critical speed (620 rpm).The drillstring yielded after rotating the drillstring for a few thousands of cycles.Measured lateral deflections (amplitude, AMP) at 620 rpm were around 11.5 cm.
Figure 7 shows fatigue failure of a 1" diameter drill pipe when it was rotated near its resonance rotationed speed.The pipe failed at 53110 cycles due to operation at a rotational speed of 594 rpm which is corresponded to 90% of the drill pipe resonance rotation speed.It is apparent that operating near resonance rotational speed is a useful characteristic in testing as it allows quick test results and provides important information about the number of cycles to failure if operated near resonance rotation speed.

Figure 1 .
Figure 1.Drillstring testing facility (a: image of the testing facility and loading mechanism, b: schematics shows individual components).

Table 2 .
Braking mechanism along with deliverables of the sub-mechanism.Ultrasonic sensor and Halleffect sensors have been installed  RPM at both ends of the string and deflection in the middle of the string are measured.

Figure 2 .
Figure 2. Configuration of parts of the braking mechanism.

Figure 4 .
Figure 4. Large deflection of the 1" diameter drill pipe at rotational speed close to its resonance.

Figure 5 .
Figure 5. Drillstrings rotational speeds versus number of cycles to fatigue failure.

Table 1 .
Main components of the drilling facility.

Table 3 .
Theoretical resonance rotational speeds of the drillstrings.