5.3.1
Four operating modes are normally encountered in offshore drilling operations.
— Drilling mode—The drilling mode refers to normal drilling activities, including drilling ahead, tripping, under-reaming, circulating, etc.
— Connected nondrilling mode—In this mode, the only drilling operation that should be conducted is circulating and tripping drill pipe. The drill pipe should not be rotated.
— Disconnected mode—If environmental conditions exceed the limits for safe operation in the connected nondrilling mode, the riser should be disconnected to avoid possible damage to surface or subsea equipment. Either the LMRP or the entire BOP stack may be suspended using the riser.
— Running and retrieval—A disconnected mode when deploying or retrieving the riser and the BOP stack or the riser and the LMRP. During riser running and retrieval operations, consideration should be given to full-depth deployment as well as varying partial-depth deployments (as the riser string is deployed or retrieved) since riser dynamic behavior is sensitive to deployed riser length.
Special operations, such as running casing, cementing, or well testing, can dictate more restrictive operating limits.
Recommended Guidelines for Design and Operation 5.3.2
Recommended guidelines for design and operation include the following.
a) Environmental Conditions and Hydrodynamic Coefficients.
Selection of the appropriate combination of environmental conditions and hydrodynamic coefficients for the analysis involves judgment, experience, and an understanding of the type of riser analysis being employed. Design and operating limits for the key riser parameters—upper and lower flex/ball joint angles, maximum and alternating stress, and the appropriate factor of safety for the tensioner system design constraints—are selected based on sound engineering principles and successful operating experience.
Table 1 defines recommended design and operation practices for the operating modes. In cases of extended drilling in a harsh environment, a fatigue analysis of the riser should be considered. The mean and maximum flex/ball joint angle limits given for the normal drilling mode are intended to prevent wear
and keyseating damage to the riser and flex/ball joint. Prudent operational procedure should strive to maintain these angles as small as possible (e.g. 1° or less), and consider 2° (mean) and 4° (maximum) as upper bounds. The maximum flex/ball joint angle limits for the connected nondrilling mode and disconnected mode are intended to prevent damage to the riser, flex/ball joints, BOP stack, subsea wellhead, and casing.
b) Predicted Maximum Values.
For a steady state condition, the magnitude of individual peak responses varies randomly. Consequently, the largest peak response (e.g. stress, angles, displacements, loads) that is likely to occur during an event increases with duration.
It is common to use the “most probable maximum” associated with a specific duration or number of oscillations for reporting riser analysis results and setting operating guidelines. The user should be aware that a “most probable maximum” represents the most likely maximum to occur over “n” cycles, not an absolute maximum value. The most probable maximum has a 63 % probability of being exceeded. If appropriate, a lower probability of exceedance, which corresponds to higher predicted extreme values, can be used.
Ochi (1981) showed how to predict extreme values based on the mean and standard deviation of a response, the exposure time or number of cycles, the specified exceedance threshold, and other statistical considerations.
c) Wear Considerations.
The rate of wear at flex/ball joints is dependent on:
— contact force,
— relative velocity,
— material parameters,
— properties of the surrounding fluid,
— travel distance per contact cycle,
— number of contact cycles.
For a drill string in a riser system, the relative velocity is dictated by operations, either rotation rate or running/pulling speed, the material properties are usually specified (steel), and the surrounding fluid is the drilling fluid returns (more of an abrasive than a lubricant). The only parameter subject to control is the contact force.
The contact force is of concern where the drill string is forced to conform to the bore of the riser system (primarily near the flex/ball joints). The force is amplified when the annulus between the drill string and the riser bore is small (smaller ID risers, larger OD drill pipe) and when the drill string tension at the point of contact is high (drill string is less flexible). The tool joints are usually responsible for much of the wear.
Examples of high-wear cases include the following.
— Wells deep below the mud line: high tension in the drill string at the lower flex/ball joint;
— Use of large-diameter drill pipe [6 5/8 in. (168 mm), for example]: smaller annulus, together with higher tension due to a heavier drill string.
An example of a low-wear case is a shallow well drilled with small-diameter drill string.
Table 1—Marine Drilling Risers—Maximum Design and Operating Guidelines a
Design Parameter Riser Connected
Riser Disconnected Drilling Nondrilling
Mean upper flex/ball joint angleb 2° N/A N/A
Max. upper flex/ball joint angleb 4°
90 % of flex/ball joint capacity or other
geometric limit
90% of flex/ball joint capacity or other
geometric limit
Mean lower flex joint angleb 2° N/A N/A
Max. lower flex joint angleb 4°
90 % of flex/ball joint capacity or other
geometric limit
N/A
Mean intermediate flex joint angle (if present)b 2° N/A N/A
Max. intermediate flex joint angle (if present)b 4°
90 % of flex joint capacity or other geometric limit
90% of flex joint capacity or other geometric limit Stress criteria: d e
maximum nominal stress f 0.67σyc 0.67σyc 0.67σyc
Significant dynamic stress range: g
nominal stresses 10 ksi N/Ah N/Ah
local peak stresses 15 ksi N/Ah N/Ah
Tension setting limits:
minimum top tension (Tmin) i k i k N/A
maximum tension setting (Tmax) j k j k N/A
a These guidelines apply to global riser response.
b In some circumstances (such as elevated risk of wear, large OD components in or near a flex/ball joint, etc.) more restrictive limits on mean and/or maximum angles may be appropriate.
c σy is the minimum yield strength of the material.
d Maximum nominal stresses are calculated according to von Mises stress failure criterion, σvm, where:
( ) (2 ) (2 )2
2 1
vm 2 1 2 2 3 3 1
σ = σ −σ + σ −σ + σ −σ
where
σ1, σ2, and σ3 are the principal stresses.
Horizontal and torsional shear stresses should be considered when appropriate.
e Allowable stress and significant dynamic stress ranges are illustrated in Figure 2.
f This refers to the static stress plus the maximum dynamic stress amplitude in the main riser pipe away from any stress concentrations. The effects of the external lines on the riser main tube should be considered in the global riser analysis.
g These values refer to largest variation in principal stress (not von Mises). Also, they are not limits but are indicators that further fatigue analysis may be necessary.
h “N/A” should not be interpreted to mean that fatigue should be neglected.
i The minimum top tension, TMIN, required to prevent global buckling of riser in the event of the sudden loss of pressure in a tensioner or tensioner pair. This is discussed in more detail later in this section (5.3.2).
j See 4.4.3.a.
k Other considerations, such as riser or wellhead loads, recoil, and minimum tension required to separate the LMRP connector in an emergency disconnect, may place additional limits on tension setting. These are typically addressed separately from TMIN and TMAX.
Key
X time Y stress
1 significant dynamic stress range (SDSR) 2 maximum stress
Figure 2—Allowable Stress and Significant Dynamic Stress Range
d) Stress Analysis.
The purpose of the maximum stress analysis is to ensure that the riser is strong enough to support the maximum design loads. This is accomplished by requiring the riser to support the maximum design loads while keeping the maximum stresses below the allowable stresses. Maximum nominal stresses in Table 1 refer to the von Mises stress criterion, σvm, as defined in Table 1. Local peak stresses are not considered for the maximum load analysis; however, these peak stresses are of concern for evaluating the fatigue life of the riser. Fatigue analysis is discussed in Annex C.
e) Significant Dynamic Stress Range.
The significant dynamic stress range limit should also be used in conjunction with the maximum load analysis. This limit is intended to provide some indication of the fatigue life of the riser. This limit in the maximum load analysis is intended to limit dynamic stresses that can lead to accelerated fatigue.
0 5 10 15 20 25 30 35 40 45 50
2200 2300 2400 2500 2600 2700 2800
Stress (KSI)
Time (s)
RESPONSE TIME TRACE
Response Mean Min Max. Bottom of SDSR Top of SDSR Significant Dynamic
Stress Range
2
1
f) Operating Modes.
Additional operating modes that can influence the design, i.e. all relevant load cases, should be considered. Specifically, running-tool interfaces, storm hangoff on either spider or riser hangoff structure, special running situations, and emergency conditions should be reviewed for their impact on riser system design.
g) Minimum Tension Setting.
A minimum tension setting is required to ensure the stability of the riser. The tension setting should be sufficiently high so that the effective tension, as addressed in 5.4.5, is always positive in all parts of the riser even if a tensioner or tensioner pair should fail. In most cases, the minimum effective tension is encountered at the bottom of the riser. In some cases, the minimum effective tension may occur at another location.
The minimum tension setting, TMIN, also accommodates inefficiencies in the tensioner system and the possibility of a sudden loss of tension. For a typical wire-rope tensioner system in which sudden tensioner failure causes tension from one or two tensioners to be lost, TMIN can be determined by Equation (1):
( )
min srmin f
T =T ×N R N−n (1)
where
N is the number of tensioners supporting the riser;
n is the number of tensioners subject to a single sudden failure (typically one or two depending on tensioner plumbing arrangement);
Rf is the reduction factor, relating minimum vertical tension at the tension ring to tensioner setting to account for fleet angle and tension variation due to mechanical and hydraulic effects in the tensioner system. The amount of tension variation can increase as vessel motion increases.
Tension variation as a percentage of the mean tension can increase as mean tension is reduced (typically for shallow water and/or low mud weights);
Tsrmin is the minimum vertical tension on the tension ring that prevents the riser from buckling. For this criterion to be met, effective tension throughout the riser must remain positive. Equation (2) shows the formula for determining Tsrmin, for one particular case:
— the OD and ID of the main tube and external lines is constant throughout the riser;
— the drilling fluid weight is the same in the main tube, C/K, and boost lines;
— surface pressure in each line is either zero or is capped inside the riser, thus creating an internal pressure end loads that increase total “real” tension in the riser [e.g. pressure testing of a C/K line, MPD pressure trapped by a rotating control device (RCD) in or below the telescopic joint outer barrel].
srmin s wt n bt i m m w w
T =W f −B f +Ar gH −r gH (2)
where
Ws is the submerged riser steel weight above the point of consideration;
fwt is the submerged weight tolerance factor (minimum value of 1.05, unless accurately weighed);
Bn is the net lift of buoyancy material above the point of consideration;
fbt is the buoyancy loss and tolerance factor resulting from elastic compression, long-term water absorption and manufacturing tolerance (maximum value of 0.96, unless accurately known by submerged weighing under compression at rated depth);
Ai is the internal cross-sectional area of riser, including external lines that contain drilling fluid;
ρm is the drilling fluid density;
Hm is the height of the drilling fluid column from the point of consideration to the top of the fluid column;
ρw is the seawater density;
Hw is the height of the seawater column from the point of consideration to the waterline;
g is the gravitational acceleration.
For cases that do not conform to the assumptions listed above, Equation (2) may require modification. In such cases, Tsrmin should still be high enough to maintain positive effective tension throughout the riser.
Equation (2) for Tsrmin, the exterior pressure, ρwHw, is multiplied by the internal cross-sectional area of the riser, Ai, rather than the exterior cross-sectional area, Ao. This is because the buoyancy of the riser pipe walls, ρwHw(Ao − Ai), has been included in the submerged riser weight, Ws.
It is not typically necessary to include the internal area of the hydraulic lines in Ai because hydraulic lines contain hydraulic fluid, which has a density that is very similar to seawater, thus its contribution to the riser’s effective weight is negligible.
It is important to recognize that the tension value that is displayed on the tensioner panel is typically a value that has been calculated from pressures measured in the tensioner system. The method that is used to calculate this displayed tension varies from one rig to another and may also include adjustments that cause the mean tension that is applied to the riser to be either higher or lower than the value that is displayed on the panel. This should be accounted for when defining minimum and maximum tension settings for actual or anticipated operating conditions. If the displayed tension does not account for the weight of the piston, rod, and/or ring or if displayed tension includes other factors that cause it to differ from the mean that is applied to the riser, these should be accounted for, either by modifying Equation (1) or by making a conversion from the actual tension requirement to the corresponding displayed tension.
Rf should account for mechanical and hydraulic inefficiencies that are consistent with actual or anticipated operating conditions, so that effective tension remains positive at all elevations above the subsea BOP stack.
Tension variation can be determined by simulation or measurement. The full amount of the tension variation will not typically be visible on the tensioner instrumentation.
In a typical direct-acting tensioner system, a sudden loss of all pressure on the rod side of a tensioner can produce a downward-acting force on the tension ring. This downward-acting force is created by the weight of the rod and piston and also the blind side pressure acting on the piston. In addition, the displayed tension in some direct-acting tensioner systems is calculated at the interface between the riser tension ring and the outer barrel of the telescopic joint. Equation (1) should be modified to match the design of the system, meeting the intent of this sudden tensioner-loss criterion. For example, Equation (1) can be modified as
follows to account for pressure acting on the blind side of the piston as well as the weights of the piston, rod, and the portion of the ring that had been previously supported by the failed tensioner:
( ) min Ring
min sr Piston/Rod BSP
f
N nW
T T nW nF
R N n N
= − ⋅ + + + (3)
where
Tsrmin is as defined in Equation (2) (in this example, it is not necessary to include the weight of the tension ring in Tsrmin because it has been accounted for separately);
N is the total number of tensioners on line;
n is the number of tensioners subject to a single sudden failure (typically one for direct-acting tensioners);
WPiston/Rod is the weight of the failed tensioner piston and rod;
N WRing
is the portion of the tension ring that had been previously supported by the failed tensioner;
FBSP is the force created by pressure acting on the blind side of the piston.
Other failure scenarios, such as the failure of the upper connection of a direct-acting tensioner to the rig, asymmetric loading on the tension ring, or other tensioner system configurations, may create a different set of loads that the remaining intact tensioners must support. If so, such loads should be accounted for so that minimum tension settings are set high enough that the failure scenarios will not produce momentary effective compression in the riser. In such a case, Equation (3) may require additional modifications to meet the intent of this sudden tensioner-loss criterion.
Tension Setting Limitations 5.3.3
The following factors (individually or in combination) can cause the riser’s capacity to be exceeded even if the tension displayed on the tensioner control panel indicates otherwise:
— bending moments from all sources, including unbalanced pressure end loads, dynamic response, and wave loading;
— wave-frequency-motion-induced tension variation due to hydraulic and mechanical efficiencies in the tensioner system;
— changes in tensioner stroke due to rig offset (such as during a drift-off);
— pressure end loads in the C/K, boost, hydraulic, and/or other external lines (all of which cause the riser to bear additional tension that is not applied by the riser tensioners).
The tension and bending moment carried by all components in the riser system due to top tension, pressure, current, waves, and rig motion should be accounted for. In some cases, individual components may have lower tension ratings than the rest of the riser. In many cases, these considerations require a lower maximum allowable tension setting than Tmax.
Motion-induced tension variation from the tensioner system is caused by gas compression and expansion, pressure drop due to flows in the tensioner system hydraulics, and mechanical friction in the cylinder seals
and sheaves. The magnitude of this variation depends on the performance characteristics of the tensioner system and vessel motions.
Changes in tensioner stroke (such as during a drift-off) can cause significant changes in pressure in the tensioner system. This can lead to a significant increase in tension above the tension that had been set at the tensioner panel.
Pressure at the top of an external line creates a pressure end load that is resisted by the riser. The magnitude of this load is typically calculated from the surface pressure and the area formed by the seal diameter of the pin/box connection in the termination gooseneck. For example, 15,000 psi acting on a 6-in. seal diameter creates a pressure end load of 424 kips. Unbalanced pressure end loads (e.g. different pressure on the choke side than on the kill side) induce a moment on the riser at the top and bottom termination of those lines that should be accounted for. The influence of these pressure end loads on all parts of the riser should be accounted for.
Tension fluctuations and pressure changes occur simultaneously with riser dynamic response induced by waves, current, and rig motion. Consequently, relevant combinations of all these factors should be accommodated.
Bearing in mind that the tension rating of a drilling riser may be governed by either the riser stresses or the connector capacity, it is not appropriate to allow loads to exceed the riser’s tension rating even if calculated riser stresses fall within the stress limits that are shown in Table 1. The riser tension rating should account for many factors (e.g. connector preloads, flange separation, connector allowable stresses, sensitivity of load sharing, etc.) that may not be captured in a global riser analysis.
Tension ratings per API 16F and API 16R are for pure tension. Additional restrictions on tension settings may be needed when significant bending and/or external line pressures are present. The riser main tube stress is not the only relevant consideration. Deflections, forces, and stresses throughout each component may also be important. Additional clarification from the riser manufacturer may be required for specific riser designs and/or loading scenarios. For example, many drilling riser designs rely on “load sharing,” in which a portion of the tension and bending moment is borne by the external lines rather than the main tube. This complicates the determination of loads and stresses that develop in the riser (in the main tube, external lines, and the connector) as a result of tension, pressure, thermal growth, and bending moment.
Limits on tension setting should account for all relevant well-kill, pressure testing, and drift-off scenarios including appropriate combinations of tension fluctuations due to rig motions and offset, tension, and moment induced by pressure end loads and riser dynamic response. These considerations apply to the entire riser system, not just the top joint. For example, the bottom joint may have less tension than the top joint, but it may have the largest bending moments (e.g. when lower flex joint angles are large) and/or the highest external line pressures.