Influence of fatigue on casing strength


Casing failure can have various causes. Casing may fail after exceeding, even once, the ultimate tensile or compressive strength, but also after repeated load cycles below the ultimate tensile or compressive strength. This phenomenon is known as fatigue, and practically all materials are subject to it.

1 Introduction

The effects of surface condition, corrosion, temperature, etc., on fatigue properties are well documented. The microscopic mechanism of fatigue damage has been identified as cyclic plastic deformation of the material at the source of a fatigue crack (crack initiation) or at the tip of an existing fatigue crack (crack propagation)..

2 Fatigue failure parameters

below is an overview of the parameters which can influence fatigue failure.

2.1 Number of cycles to failure

Failure data are presented in the form of an S/N curve where the cyclic stress amplitude is plotted on log-log paper versus the number of cycles to failure. Ferrous metals in air show a lower limit to the stress amplitude called the fatigue limit, or endurance limit. This generally occurs after 105 to 107 stress-reversal cycles. Stress reversals below this limit will not cause failure regardless of the number of repetitions. Ferrous metals in seawater, however, do not show this cut-off: S tends to zero with increasing N.

Several modified forms of the Goodman diagram are used for predicting the stress levels at which cracks will form, but other more extensive plots such as the Haigh diagram can be used to predict in addition the stress level for which cracks, once formed, will propagate.

2.2 Stress history

A very important question concerns the influence of previous stressing on fatigue strength. One theory generally accepted is the linear damage law (miner's law); the damage produced by repeated stressing at any level is directly proportional to the number of cycles. It implies that the effect of a given number of cycles is the same, whether they are applied continuously or intermittently, irrespective of the order in which high- and low-stress cycles occur. The linear damage law is not reliable for all stress conditions. For truly random loading, i.e. random distribution of smaller and larger cycles such as due to waves, the theory is reasonably reliable. The fatigue accumulation rules become unreliable for cases where cycles of different magnitudes appear in blocks. For example, the theory cannot distinguish between cases where first all larger cycles and later the small ones occur, as opposed to the reverse.

2.3 Stress concentrations

In a situation of stress concentration, fatigue failures will occur at stress levels below those at normal stress level. For example, badly dressed welds form a potential source of stress concentrations. It is important to minimise the number of notches and sharp edges on the casing to maximise to endurance limit. It is also important to treat the object with care as improper handling can also lead to deformation and thus stress concentrations.

2.4 Residual stress

Since residual stresses, whether deliberately introduced or merely left over from manufacturing processes, will influence the mean stress, their effects can be accounted for. The knowledge regarding residual stresses is very limited. Their magnitude heavily depends on the manufacturing process.

2.5 Range of stress

Stressing a ductile material beyond the elastic limit or yield point in tension will raise the elastic limit for subsequent cycles but lowers the elastic limit for compression. A single extreme load, causing plastic deformation at the tip of a crack, leaves a residual compressive stress after unloading. This has a beneficial effect on the endurance as the average and effective peak stress level may be reduced during subsequent stress cycles.

2.6 Loading method and sample size

A uniaxial stress as created in a bench test can be produced by axial load, bending or a combination of both. Since fatigue properties of a material depend upon the stochastic distribution of defects throughout the specimen, it is apparent that the three methods of loading will produce different results. Similarly the size and geometry of a specimen influences the maximum endurance limit.

2.7 Combined stress

Uniaxial stress is not a common feature. Usually an object will be subjected to triaxial or biaxial stress.

2.8 Surface conditions

Surface roughness constitutes a kind of stress raiser. Even particle-size unsmoothnesses can act like notches or sharp edges and thus cause stress concentrations.

2.9 Corrosion fatigue

Under the simultaneous action of corrosion and repeated stress, the fatigue strength of most metals is drastically reduced, sometimes to a small fraction of the strength in air, and a true endurance limit can no longer be said to exist.

3 Specific issues

Casing fatigue failure is related to casing dimensions, material properties, number of load cycles and types of load amplitudes. The last two are related to other parameters like the movements and mechanical properties of all components connected to the casing.

Fatigue related issues can separated in: externally generated loads and internally generated loads on the casing. External loads are usually caused by waves and currents; while internal loads are induced by the internal fluid flow.

Below these loads and possible solutions to avert casing fatigue failures will be addressed.

3.1 Externally generated loads

External loads are usually associated with offshore structures. The current design procedures covering fatigue loading is sufficiently developed for onshore wells enabling the casing designer to work separately on the casing string design, this is not the case with complex offshore wells in hostile environments. An integral design of the whole marine drilling and production system is necessary, requiring a coupled analysis approach to the total system.

Wave and current loads induce stresses, either directly in the casing assembly (above the sea bed, incl. the marine conductor), or indirectly in the suspended strings (below the sea bed) through forces being transferred by the subsea wellhead.

a. Directly generated loads

With respect to direct loads two causes of fatigue can be distinguished:

1.action of waves and (tidal) currents;

2.vortex shedding.

a.1. Waves and (tidal) currents

Waves and tidal currents apply a direct force on the marine riser/conductor, by intermittently exerting a sideward force on the marine riser/conductor surface.

If these loads exceed the actual technical and/or economic design limits, alternative measures can be taken, for example by defining user limitations. Drilling activities may be limited to spring and summer in order to reduce the environmental loading on the structure. For permanently installed production facilities such limits obviously do not apply, as the structure will be exposed all year round.

a.2. Vortex shedding

During fluid flow around a body laminar and turbulent flow patterns can be observed. Usually the flow will become turbulent at the downstream side of the body and vortices will then be created. At low Reynolds numbers (i.e. relative low flow velocities) vortices are simultaneously shed from each side of the cylinder causing forces in line with the flow. At higher Reynolds numbers the vortices are shed alternately. This will cause forces perpendicular to the main flow. When the frequency of shedding becomes equal to the frequency of the oscillating cylinder, i.e. a structural natural frequency, large hydrodynamic forces will arise.

Possible solutions to avert fatigue failures consist of:

  • detuning of the natural frequency, by for example changing the diameter or applying top-tension;
  • disruption of the excitation by streamlining the flow for example by shaped buoyancy material, i.e. fairings;
  • disruption of the excitation by breaking the vortex pattern by for example mounting helical strakes.

b. Indirectly generated loads

Waves, winds and currents do not only act directly on the casing assembly, but also indirectly by exerting loads on other structural components, being the subsea wellhead or mudline suspension system to which the marine riser/conductor is connected. The unit's type will determine the severity of the load:

  • floating unit (semi-submersibles, drill ships);
  • fixed unit (fixed platforms, jack-up rigs).

b.1. Floating-unit movements

Again, a subdivision can be made into:

1.lateral movement;

2.vertical movement.

b.1.1. Lateral movement of floating unit

The floating unit and riser system will oscillate due to the combined action of currents, waves and winds. This results in loading of the subsea wellhead, foundation pile and casing system.

The movement of the subsea wellhead is limited due to the presence of a template, the presence of soil and cement around the foundation pile and the weight of the casing strings. If the soil is relatively strong, the location of the effective reactions of the foundation pile will be near the mudline. If the soil is weak these effective reaction points will be lower and thus the maximum internal loads in the subsea wellhead/casing string assemblies will be higher due to the longer effective lever arm. These conditions will create bending moments (M) and shear forces (F) on the subsea components. These bending moments and shear forces can be transferred to the seabed through the following mechanisms.

b.1.1.1. Reactive shear at the template

A template which is connected to seabed by a long spudcan or driven piles can exert a significant reactive shear on the subsea wellhead(s). This shear reduces the effective movements of the casing assembly below the template. A single well completion without this template or a non rigid template will not have this reactive shear.

b.1.1.2. The conductor string and foundation pile load paths

There are two possible load paths. The first load path is through the wellhead housing, into the foundation pile suspension joint, and then into the seabed. This is the strongest and most desirable load path. The second load path is through the wellhead housing via the conductor string below, and then through the cement to the foundation pile to the seabed.

A solution to reduce the conductor string loading consists of introducing a two-point contact between the subsea wellhead/foundation pile combination and the conductor string allowing for load transfer through the foundation pile to the seabed.

Apart from the above solution the movement of a floating unit may be reduced by increasing the number of anchors. Floating units are often equipped with dynamic positioning systems, aiming to minimise lateral movements.

Once the well is taken in production the relatively stiff marine drilling riser may be replaced by a flexible production riser. This will reduce the loads exerted on the foundation pile and wellheads. On the other hand the flexible riser strength will have to be evaluated.

b.1.2. Vertical movement of floating unit

This movement induces a movement of the marine riser which will subsequently cause cyclic axial stresses in the connected components.

A solution is to apply constant tension to the marine riser by applying active heave compensation systems and decoupling the rig motion from the riser by application of a telescopic riser joint. Another possibility is to limit the vertical platform movement by the fixing the floating unit to the sea bottom with tension cables.

b.2. Fixed-unit movements

A fixed platform is usually connected to the sea floor by piles running through the jacket legs. The same environmental loads act on marine conductors as on the floating unit. Expert advice should be sought for detailed studies if required. The platform's structure will also be subjected to a cyclic movement. As a platform is not floating there is only a lateral movement. In case of shallow waters this movement will be small. In case of the deepwater fixed platforms, this lateral movement can be considerable and may have to be accounted for in the design of the subsea components. Although the lateral movements of deep water platforms may be more as compared to shallow water platforms, this does not necessarily imply that the situation is less favourable for the marine conductor. The induced stresses do not depend on the overall height of the platform, i.e. the distance between marine conductor guide frames.

3.2 Internally generated loads

Contrary to external loads, the internal loads are not a function of the location type, but are caused by fluid movement and temperature cycles inside the casing. We can distinguish the following possible load causes on a casing:

a. Fluid-flow-induced vibration

Fluid particles of the internal flow experience a dynamic force generated by centrifugal and Coriolis accelerations as they travel inside the curved path along the deflected fluid path, eg. deflected production riser. Riser deflection is caused by the combined forces of the currents, winds, and waves. Internal fluid flow and external vortices may occur simultaneously, causing vibrations in the riser. Several documents exist about the riser equation of vibration.

b. Pressure surges

A casing may be subjected to cycles of pressure changes through the intermittent use as an injection and production well. This is the case for cyclic steam injection during steam soaking operations. Another, though minor, cause for pressure surges of very small amplitude can be the production of oil with sucker rods through non sealed-off tubings. The casing is then continuously subjected to oil level changes. The effect will be minimal and may generally be disregarded.

c. Thermal stresses

Thermal stress cycles are imposed on casing during steam soaking operations. Large temperature fluctuations may occur.

Similarly to pressure surge-related stresses, thermal stresses may be imposed to a casing due to production shutdowns.



#1 Ward 2017-08-20 08:33
What i don't realize is in truth how you're now not really a lot more neatly-liked than you may be right
now. You're very intelligent. You realize thus significantly in relation to this topic, produced
me in my opinion imagine it from so many various angles.
Its like women and men aren't fascinated unless it's something to accomplish with Girll gaga!

Your individual stuffs excellent. At all times deal with it up!

Also visit my homepage spring fashion '11: