Professional Documents
Culture Documents
TABLE OF CONTENTS
DRILLING SYSTEMS
1.0 CATEGORIZATION OF DRILLING
2.0 ROTARY DRILLING OPERATIONS
3.0 DRILLING PLAN
4.0 OFFSHORE DRILLING
WELL COMPLETION
1.0 CASING DESIGN AND CONNECTORS
2.0 WELLHEADS - PLATFORM AND SUBSEA
3.0 TUBING PACKERS
4.0 DOWNHOLE SAFETY VALVES
PIPELINES
1.0 CATEGORIZATION OF PIPELINES
2.0 DESIGN OF PIPELINES
3.0 INSTALLATION OF PIPELINES
4.0 PIPELINE OPERATIONS
PRODUCTION FACILITIES
1.0 INTRODUCTION
2.0 DESCRIPTION OF VARIOUS COMPONENTS
WELL OPERATIONS
1.0 INTRODUCTION
2.0 WELL PRODUCING NEEDS
3.0 WELL FLOWING CHARACTERISTICS
4.0 TUBING FLOW CONSIDERATIONS
5.0 THEORETICAL CONSIDERATIONS FOR VERTICAL TWO PHASE FLOW
6.0 “OPTIMUM” TUBING SIZE
7.0 CHOKE PERFORMANCE CURVES
8.0 GAS LIFT
9.0 WELL START-UP OF “KICK-OFF”
10.0 WELL MONITORING
11.0 WELL KILLING
12.0 WELL WORKOVERS
13.0 ROUTINE WELL SERVICING
MARINE SYSTEMS
1.0 THE SHIPSHAPE
2.0 STABILITY
3.0 MOTION
4.0 MODEL TESTS
5.0 CLASSIFICATION SOCIETIES
TABLE OF CONTENTS
LIST OF FIGURES
FIGURE 1. Geologic Time
FIGURE 2. Basic Rock Types
FIGURE 3. Depositional Environment
FIGURE 4. Porosity
FIGURE 5. Permeability
FIGURE 6. Variations of Horizontal and Vertical Permeabilities
FIGURE 7. Structural Traps
FIGURE 8. Salt Dome Traps
FIGURE 9. Seismic Survey
FIGURE 10. Seismic Record
FIGURE 11. Physical Basis of Well Logging
FIGURE 12. Subsurface Maps
FIGURE 13. Estimation of Total Reserves
FIGURE 14. Well Testing
FIGURE 15. Material Balance
FIGURE 16. Darcey’s Law
FIGURE 17. Oil Reservoir Depletion Characteristics
FIGURE 18. Dissolved Gas Drive Reservoirs
FIGURE 19. Water Production Problems
FIGURE 20. Gas Cap Drive Reservoirs
FIGURE 21. Gas Production Problems
FIGURE 22. Combination Drive Reservoir
FIGURE 23. Volumetric Gas Reservoir
FIGURE 24. Advantage of Early Waterflood Initiation
FIGURE 25. Waterflood
FIGURE 26. Water and Gas Injection
FIGURE 27. New Recovery Methods
Clastics are formed from the physical and chemical breakdown of preexisting rocks when deposited in new
arrangements after being moved by water, air, or ice. They are classified based primarily on composition
and grain size. Sandstones are mostly quartz grains from .05 to 2 mm in diameter. Shales, clays, and
muds are characterized by extremely small particle size and are important in both sourcing and sealing
hydrocarbon reservoirs. Shales form the bulk of the volume of sediment in the world.
Carbonates and evaporites are formed in place by biological and chemical action. Limestones are usually
formed by biologic activity such as coral reefs. Most of the Middle East production is from limestones.
Other examples are dolomite and salt.
The two other rock types, besides sedimentary, are igneous and metamorphic as summarized on Figure 2.
They are rarely reservoir rocks. Igneous rocks are formed by the colling and consolidating of complex
siliceous solutions from a molten or partially molten state, for example, granite and basalt. Metamorphic
rocks are formed when high temperature and/or pressure alter pre-existing rocks; examples are quartzite,
marble, and slate.
As previously discussed, sedimentation produces the source material for the hydrocarbons, the reservoir,
and the seal (impermeable shales). Digenesis, or change due to heat, pressure, and mineral rich solutions
generates the hydrocarbon and alters the reservoir by compaction, chemical solidification, and other
physical changes such as channeling and fracturing. Structuring frequently forms or increases the closure
necessary to trap the hydrocarbons and prevent them from continuing to migrate updip. The hydrocarbon
traps may be formed during deposition, i.e., stratigraphic traps, or later due to structural movement
2.3 Logs
Logs are considered the basis for modern petroleum geology. They are a record of a tool run in the hole
which measures certain formation properties and they allow very reliable qualitative and quantitative
evaluations as summarized in Figure 11. One of the most important functions they serve is clear definition
of the formations and other strata which allows subsurface mapping of reservoirs. Other logs can verify the
presence of hydrocarbons and allow calculation of critical reservoir parameters such as porosity and water
saturation. These critical parameters are usually the basis for reserve estimates which can decide field or
reservoir development feasibility.
geologists consider regional field production histories when assessing potential new fields. In existing fields
the production history is carefully monitored as it can indicate opportunities for additional reserves.
Normally production history is used in conjunction with subsurface mapping to predict new reservoirs.
Contour mapping employs contour lines connecting points of equal value. Reservoir structure maps are
prepared where the contours represent lines of intersection between a series of parallel horizontal planes
(depth) and the non-horizontal reservoir top surface. The structure maps indicate the direction and angle of
the sloping surface. Structure types, such as anticlines, synclines, and faults can be read on the map.
Fluid contacts (i.e., gas-oil, gas-water, oil-water) are horizontal planes and are mapped like contours on a
structure map.
Recoverable reserves are assumed to be some percentage of the reserves in place. The recovery efficiency
is based on the expected recovery mechanism. This subject will be covered later.
When a volume of oil is produced from a reservoir, the space once occupied by this oil must be filled with
something else. Unless fluid is injected, the production of oil must result in a decline in reservoir pressure.
This pressure decline can cause the influx of fluids from an adjoining gas cap or aquifer, the expansion of
fluids originally in the reservoir, and the expansion of reservoir rock grains.
Consider an oil reservoir, where the original pore volume contains only oil and connate water. After some oil
has been produced, the pore volume can be divided into six discrete volumes to account for all the possible
ways that the produced oil space can be refilled. The six items are: (1) the expansion of an adjoining gas
cap; (2) The volume of gas released from oil; (3) the volume of oil still in the reservoir; (4) The expansion of
the reservoir rock grains; (5) the original connate water which has expanded; and (6) the influx of water from
an adjoining aquifer. Only item (3), (4), and (5) need to be considered in every oil reservoir. All the items
are graphical shown in Figure 15.
Gas form the gas cap will tend to occur near the top of the reservoir and water from the aquifer will generally
be near he reservoir base. Under some conditions, gas released from the oil will gravitate to the top of the
reservoir. More often the released gas will be evenly distributed throughout the reservoir, as are the
expansions of rock and connate water.
For multiphase fluid flow, the permeability and pressure terms of Darcy’s law require additional explanation.
In general, the presence of one phase reduces the ease of flow for other phases. This leads to the concept
of relative permeability. Relative permeability to a given phase is defined as the effective permeability of that
phase, at a given saturation, divided by the single phase permeability. The nature of relative permeability
curves depends on local distribution of fluid within pores, which is controlled by capillary pressures.
Capillary pressures dominate local fluid distribution in all porous media, but may or may not be important in
a macroscopic displacement.
For the equations in this form, there is no analytical solution to describe the movement of oil and water.
Viscous, capillary, and gravitational forces all can be important in the displacement process. They can be
solved by computer using difference equation techniques.
When viscous and gravitational forces are much larger than capillary forces, the capillary forces can be
neglected. This allows the second-order partial differential equations to collapse to a first-order partial
differential equation. This resulting equation can be solved to give a so-called Buckley-Leverett
displacement. The Buckley-Leverett equation is an expression for the rate at which water saturations move
across a system when water is injected into one end of the system. A major use of the Buckley-Leverett
simplified displacement theory is in waterflood predictions.
Dissolved gas drive (DGD) is the most common depletion mechanism; practically every reservoir that
experiences pressure decline will have some portion of its drive contributed by DGD. A DGD reservoir relies
principally on the liberation and expansion of gas dissolved in the oil to supply the energy necessary to
displace the oil to the wellbore. This “dissolved gas” actually refers to light hydrocarbons which are liberated
as pressure declines due to production. Figure 18 illustrates this mechanism.
Normally a predominant DGD mechanism will be associated with closed or volumetric type oil columns. An
adjoining aquifer and /or gas cap will be small or non-existent.
DGD reservoir wells have a characteristic behavior during their producing life. As oil is produced from a well,
pressure drops in the immediate vicinity. This causes fluids to expand from out in the reservoir driving oil
toward the pressure sink. As reservoir pressure declines below the bubble point gas evolves from solution,
occurring as small separate bubbles in individual pore spaces. When enough gas is released such that
critical gas saturation is reached, relative permeability to gas exists, and gas begins flowing to he well along
with the oil.
This well behavior results in a characteristic reservoir performance during the producing life. The initial gas-
oil ration is the dissolved ration, and the gas-oil ration declines slightly until critical gas saturation is
reached. Once gas begins to flow, the gas-oil ration increases steadily. The reservoir pressure decline is
very rapid during the fluid the rock expansion phase above the bubble point pressure. The pressure declines
slowly after the bubble point is reached, then more rapidly as more and more gas is produced per unit of oil.
The most important characteristic is that oil recovery is low at abandonment pressure. It is far below the
recovery expected for the two other types of primary drive mechanisms; water drive or gas cap expansion.
Rock properties have the most important influence on DGD recovery. Relative permeability has the most
important effect, since this determines the critical gas saturation. In general, oil recovery increases as
absolute permeability and connate water saturation increases. Porosity has no effect on recovery efficiency.
Reservoir oil properties affect recovery, but to a lesser degree than rock properties. High solution gas-oil
ration results in low recovery because the oil shrinkage more than offsets the greater available displacing
action. Low oil recoveries also are obtained from reservoirs where oil viscosity is high. The high viscosity oil
tends to be bypassed by the low viscosity gas.
Gravity drainage is really a separate drive mechanism, but it is a part of all types of drive mechanisms. In
DGD reservoirs, it can have a profound effect on overall recovery. The driving force for gravity drainage is
simply the density difference between gas and oil. In thick or steeply dipping reservoirs, gravity can cause
oil to move downward if replaced by gas moving upward. Gravity drainage can result in much higher
recoveries and should be recognized early because the recovery may be rate-sensitive. DGD recovery is not
rate-sensitive.
In the absence of gravitational effects and competitive problems, oil recovery under DGD is independent of
well spacing. The optimum spacing should be determined based on an economic study which indicated the
optimum production rate. Location of the development wells should give consideration though to the
Because of very low recovery, a pure DGD reservoir of any extent should be recognized early so that it can
be evaluated for application of a more efficient secondary drive mechanism. If adequate well control exists,
basic reservoir geology provides the earliest clues to the presence of a DGD reservoir. The DGD process is
associated with closed reservoirs formed by stratigraphic traps, fault traps, and structural traps. The
performance of nearby reservoirs and fields in a more advanced stage of depletion would also provide a clue.
As reviewed earlier, early production performance will offer evidence of DGD.
Water drive reservoirs are those in which the predominant source of energy for producing oil is water
encroachment from surrounding or underlying formations. For effective water drive a very large aquifer is
required to provide the expansion energy, because the water and rock are not very compressible. The
reservoir and aquifer must be reasonably permeable and continuous, and production rates must not greatly
exceed the rate at which the aquifer can supply water.
Reservoir performance is characterized by a high degree of pressure maintenance, high recovery efficiency,
low gas-oil ratios, and continuously increasing water cut. The original pressure is usually close to
hydrostatic head. Ultimate recoveries from water-drive reservoirs of 30-80% have been reported.
High production rates have potentially an important influence on performance and recovery. High rates may
(1) lower pressure rapidly causing DGD, increased oil viscosity, and excessive oil shrinkage, (2) reduce
gravity benefits, (3) prevent imbibition of water into tight zones, and (4) cause water coning and fingering as
shown in Figure 19. Although the high production rates may have quite an influence on early performance, it
has substantially less effect on ultimate recovery. This is because rate effects are reduced by typical long
periods of relatively low production rates as the reservoirs approach depletion.
Other factors influence recovery to a greater extent. Water-oil relative permeability dictates the residual oil
saturation in the water contacted areas. High permeability improves recovery, however, high permeability
layers in heterogeneous reservoirs can reduce recovery because water tends to channel through these
regions.
Water drive should be detected early so that development and producing operations can be tailored to take
maximum advantage of available energy. Wells should be located so as to remain uninvaded as long as
possible. Early detection can be achieved by observing static oil-water interface rise in observation wells,
and water cut behavior of producing wells near the reservoir periphery. Material balance calculations can
also indicate presence of a water drive.
Volumetric balance can be used to calculate water influx for the known history since pressure and
production are known. To predict future reservoir behavior, water influx must be determined independent of
volumetric balance. For this an unsteady state solution of the radial diffusivity equation is used assuming a
constant terminal pressure. The equation, however, requires that the aquifer properties be evaluated
empirically from the history using volumetric balance. This is because aquifer properties are difficult to
evaluate individually. Water influx predictions can then be made for any given forecast of production rates for
oil, gas, and water.
Simultaneous solution of the volumetric balance and water influx equations is possible only for a stipulated
schedule of production. Independent equations or relationships are necessary to predict production rates
and ultimate recovery. The equations or relationships needed are those which describe the distribution of
oil, gas, and water in the reservoir in relation to the positions and completion depths of all wells drilled or to
be drilled. In general, an accurate determination of fluid distribution requires that the reservoir be treated in
at least two dimensions. Fortunately, meaningful answers can be obtained by approximating reservoir
behavior with simplifying assumptions in one dimension.
A gas cap occurs in a reservoir when more light hydrocarbons are present than can be dissolved in the oil.
The energy for producing oil is provided by the expansion of the gas cap into the oil zone as the pressure
drops as shown in Figure 20. DGD is always associated with gas cap drive.
Pressure drop in a gas cap drive reservoir is usually proportional to cumulative production unless excess
free gas is produced. The rate of pressure drop depends on the size of the gas cap and the oil zone. If the
cap is large compared to the oil zone, the pressure drop will be small. Producing gas-oil ratios (GOR)
remain low until the expanding gas cap invades upstructure producing wells through coning and fingering as
shown in Figure 21. High GOR wells should be shut-in or recompleted lower to prevent waste of the
reservoir energy.
Oil recovery from gas cap drive ranges from 25% to 60% of the oil in place. The requirements for an effective
gas cap drive are: (1) Large gas cap relative to the size of the oil zone; (2) Continuous uniform reservoir; and
(3) Good gravitational segregation characteristics such as high dip angle, high permeability, low oil viscosity,
and low production rate.
Most reservoirs produce by a combination of the drive mechanisms as shown in Figure 22. In developing a
depletion plan for a combination drive reservoir, it is important to determine the degree to which each
mechanism is affecting oil displacement and recovery efficiency.
Response or reaction time of each drive mechanism must be considered in the analysis of a combination
drive reservoir. Gas cap drive has a quick reaction time and is very responsive to a pressure drop in the
reservoir. This is due to the high expansive property and low viscosity of the gas and will result in rapid
movement of the gas-oil contact.
Water drive response is slow compared to gas cap drive and DGD. The factors which contribute to the slow
response are (1) its relative high viscosity compared to gas; (2) its low expansive properties; and (3) the
distance that the pressure transient must travel. DGD response is quick but limited because of the amount
of gas liberated and is tendency to be produced.
Reaction times explains why gas cap and/or DGD will usually dominate in early life and anytime oil zone
withdrawals are accelerated. Water drive will become important later in life after sufficient time to react to
the pressure drop. If gravity is effective, DGD becomes less important as released gas recharges the gas
cap and retards pressure drop.
Recovery efficiencies can be calculated from actual reservoir performance after a period of production history
has been experienced. This is done using portions of the volumetric balance equation. If there are
insufficient data to calculate displacement efficiencies from production data, the Welge equation can be
used to calculate theoretical displacement efficiencies. The Welge equation estimates the average
saturation of the displacing fluid behind the advancing front. It does include conformance or pattern
efficiency, which would have to be estimated from geologic data and well drainage pattern.
To improve reservoir performance, major reservoir problems must be recognized as soon as possible and
then steps should be taken to correct the problem. For example, gas cap shrinkage is one of the most
prevalent problems associated with multi-drive reservoirs. Shrinkage occurs when free gas production
exceeds the expansion of the gas cap. A loss in ultimate recovery occurs as a portion of the oil entering
the gas cap never be recovered.
Means of detecting gas cap shrinkage include: looking for high ration production while reservoir pressure is
constant or increasing; volumetric balance calculation; premature watering out of wells near the oil-water
contact; and nuclear logging programs for wells near the gas-oil contact. Means of arresting gas cap
shrinkage and improving reservoir performance include: initiating a gas injection program; shutting in high
ration oil wells: and increasing production rate if possible to cause a reduction in reservoir pressure and
permit gas cap expansion.
drive reservoirs and 80% to 95% for pressure depletion reservoirs. The main reasons for lower efficiencies in
water drive reservoirs are (1) encroaching water does not contact all of the reservoir: (2) in portions where
water does invade, some gas is left as residual gas saturation ranging from 15% to 35%.
In the case of gas reservoirs with effective water drive, the pressure will usually be maintained as water
sweeps through the reservoir. Ultimate recovery can be increased by producing the reservoir at a sufficiently
high rate to reduce
the pressure at abandonment to a low level. This is possible due to the response times relationship
between gas the aquifer. Strategic location of producing wells, of course, may increase ultimate recovery
significantly.
3.6.1 Waterflooding
Waterflooding is most often used in reservoirs whose dominant primary recovery mechanism is dissolved
gas drive. The reservoirs would become pressure depleted before much of the oil is recovered. As shown in
Figure 24, water injection is an efficient method of maintaining reservoir pressure and sweeping the
recoverable oil to producing wells. The recovery of oil by waterflooding is a combined function of the
reservoir invaded and the displacement efficiency in the invaded area.
Reservoir properties and mechanics of waterflooding are not known in sufficient detail to allow rigid prediction
calculations. Reasonable predictions can be made with simplifying assumptions. The Buckley-Leverett
Theory is considered the best technique.
The most important controllable factor for an efficient waterflood project is the location of injection and
production wells to maximize areal sweep efficiency. Figure 25 shows several waterflood strategies. Areal
sweep efficiency is the fraction of the waterflood pattern area contacted by water at a given time. Production
streamlines develop between injectors and producers, and a straight line connecting them is the shortest
streamline and has the highest pressure gradient. Injected water moves more rapidly through the shortest
streamline and therefore reaches the producing wellbore when only a portion of the pattern area has been
contacted by water.
For optimum areal sweep efficiency, waterfloods are frequently conducted in repeated patterns such a 5-
spot, line drive, inverted 9-spot, and others. For these common patterns, much experimental and theoretical
work has been conducted to determine the most efficient waterflood patterns for given applications. In these
studies, pattern efficiency is usually expressed by curves as a function of either cumulative water injection
or producing water cut. Uniform and homogeneous reservoirs are almost always assumed.
Injection wells in many water injection projects fit no classical pattern. Some of these irregular patterns are
peripheral, crestal line drive, and down dip injection into an aquifer. Accurate predictions of areal sweep
efficiency for irregular patterns can only be made with computer reservoir simulators.
Like waterflooding, gas and water injection is also used for pressure maintenance and improved
displacement efficiency. Unlike waterflooding, the fluid is injected into the gas cap or aquifer and not into
the producing zone as shown in Figure 26. In addition, gas can be injected to (1) displace unrecoverable oil
that is trapped upstructure of the highest penetration available in a strong waterdrive reservoir, and (2)
prevent oil from migrating into an original gas cap.
One important aspect of gas injection which must always be carefully considered is that the gas has value.
Injection of a saleable commodity must discount to future income. In the case of gas injection for
upstructure drainage, the commodity is unrecoverable.
Generally, the displacement efficiency of gas injection is less than that of water injection. This is because
of the high relative permeability to gas and low gas viscosity. However, where gravity effects are important,
gas displacement may be more efficient than water due to the benefits of gravity drainage. In thick, highly
permeable, dipping reservoirs, typical of salt dome fields, gas displacement efficiency is very high.
Several tertiary recovery methods, summarized in Figure 27, have been used but no universally applicable
technique has been found. The objective of several of the methods is to decrease oil viscosity to increase
will productivity or to increase the displacement efficiently of waterflooding. The most promising tertiary
processes are usually grouped into three categories:
Refers to processes where the injected gas or fluid is soluble with the residual oil. The methods include
high pressure gas, enriched gas, and gas-driven LPG banks. The miscible methods solve the common
problem of limited recovery by gas drive. They introduce a buffer zone of solvent between the gas and the oil
and sweep clean the rock invaded by solvent.
Refers to a process in which heat is applied to the reservoir. The methods include underground burning, hot-
water flooding, steam flooding, and stimulation by steam.
Refers to techniques which involve both water additives to improve recovery efficiency and those involving
banks of solvents driven by water. These methods include the use of surfactants, viscous water, CO2, and
alkaline water. Surfactant and alkaline water injection can mobilize the residual oil remaining in the reservoir
after a conventional waterflood and build a “secondary” oil bank.
DRILLING SYSTEMS
TABLE OF CONTENTS
LIST OF FIGURES
The only way to locate and evaluate hydrocarbon reserves is to drill an exploratory well. The initial
exploratory well in a field is called a wild cat. Once the reservoir is located, delineation wells are drilled at
the perceived lateral boundaries of the reservoir to confirm its size and producing characteristics.
In order to produce the discovered reserves, development wells must be drilled into the reservoir. The
reservoir engineer dictates the location of these wells after studying the results of the exploratory drilling
program. Development wells can be used to produce hydrocarbons, or used to inject water or gas in order
to maintain pressure in the producing reservoir.
Exploratory wells are drilled with systems leased for the gathering of information; offshore, temporary
platforms are used to support the drilling systems (i.e., mobile drilling units). Development wells are drilled
with either leased or purchased systems, but the platform supporting the drilling system is usually
purchased. Regardless of the intended use of the well, the methods and systems used in drilling the hole
do not vary.
Though the product of the entire drilling operation is a hole, only the drill bit actually makes the hole. The
roller cone rock bit is the most commonly used drill bit. Each cone is equipped with teeth that chip the
formation as the bit is rotated. The chips, or cuttings, are removed from the teeth and the bottom of the hole
by the action of the circulating drilling fluid. The design of the majority of roller bits is to have teeth of varying
length and spacing on the three cones, the spacing and length of the teeth depending upon the formation to
be drilled. Long, widely spaced steel teeth are used in soft formations; harder formations use shorter teeth
made from harder materials such as tungsten carbide insets laid into the cutting faces of the cones.
A widely used variant of the standard roller bit is the jet bit in which the drilling fluid exits jet nozzles forming
high velocity streams direct at the bottom of the hole. These jets of fluid better lift the cuttings off the face of
the hole, allowing the drill teeth to strike new formation rather regrinding loosened cuttings. Figure 2 is a
visualization of the downhole action of a roller jet bit. The pressure losses through these jetting nozzles are
considerable and require both extra pump capacity and pump horsepower.
Poly-crystalline diamond (PDC) bits have become widely used in the last 10 years. These bits incorporate
synthetic poly-crystalline diamond and/or tungsten carbide cutters into a steel bit body. The bits cut with a
shearing action rather than the chipping action of the roller cone bit, and are used in moderate to soft
formations with less weight on the bit and good penetration rates.
Diamond bits are most widely used as core heads. A core head is a bit with a cut out in the center which
allows a core or spine of rock to left when the hole is cut. The core is recovered by a core barrel.
The upper connection of the rock bit is normally a tapered right hand tool joint pin. This tapered thread
makes up into a drill collar sub which is usually a short, heavy-walled section of drill collar with each end
threaded with a tool joint box.
The next section of the drill string above the bit and drill collar sub is comprised of drill collars. Drill collars
are lengths of heavy weight steel pipe threaded with a right hand tool joint box up and pin down, which is the
conventional way for oilfield tubulars to be connected. The drill collars come in varying lengths and sizes.
Most commonly they are 30 ft. long.
The purpose of the drill collars is to provide the weight which will act on the bit as it is rotating in order to
improve the penetration rate. The bit weight required will dictate how many drill collars are included in the
drill string. A certain length of drill collars above the bit is under compressive loads while drilling is in
progress. This explains why the drill collars are very rugged components of the drill string and why the
threaded connections are so strong. The drill pipe and the balance of the drill collars are in tension during
the drilling operation. Normal practice is to keep the ‘neutral point’, the transition point between
compression and tension, in the drill collar section the drill string. The neutral point appears to be a
common place for failure of the drill collar tool joints.
Above the drill collars will be the drill pipe section of the drill string. Drill pipe is lighter weight steel pipe
threaded with a right hand tool joint box up and pin down. The drill pipe comes in varying lengths and sizes.
Typically, the drill pipe would be in 31’ lengths, 5” OD equipped with 4 ½” API internal flush tool joint
threads, box up and pin down. In normal drilling operations the drill pipe is under tension as the drill string is
being rotated to the right.
At the top of the drill string is a kelly saver sub and the kelly. The kelly saver sub is a short section of
heavy-walled pipe with drill pipe tool joint connections, box up, pin down. The purpose of the kelly saver sub
is essentially to protect the lower threaded connection of the kelly. The saver sub remains permanently
connected to the lower end of the kelly until the thread on the lower end of the saver sub needs replacement
or repair. The kelly is constructed from high grade steel pipe with either a square or hexagonal outside
profile. On land drilling rigs the kelly is normally about 40’ long. However, on floating drilling vessels the
kelly will normally be about 54’ long; the extra length being required to accommodate the up and down
motion of the drilling vessel.
The kelly is the component of the drill string to which the right hand rotation is applied. This in turn rotates
the drill string and drill bit to the right. The right hand rotation is applied to the kelly by way of the kelly
bushing. The kelly bushing has a square or hexagonal hole through it which corresponds to the outside
profile of the kelly. The kelly bushing in turn sits and locates in the rotary table and receives its right hand
rotation when the rotary table is rotated.
The rotary table sits on very large steel beams, called ‘rotary beams’, at the rig floor level. It is centered in
the middle of the rig. The main function of the rotary table is to apply rotation and transmit torque into the
drill string by way of the kelly bushings and kelly. The center section of the rotary table is the part that
rotates and this center section typically has a 37 ½ in. opening without the master bushings in place. The
master bushings are split and sit into recesses in the center section of the rotary table. The master
bushings have a tapered conical hole into which drill pipe slips can be set when it is necessary to support
the drill string weight, as would be the case when another joint of drill pipe is added or when tripping the drill
pipe in and out of the hole. As the kelly is rotated during drilling operations, it moves up and down by
means of rollers in the kelly bushing. In offshore drilling operations however, it is necessary to lock the kelly
bushings in the rotary table. This is because the friction bind between the kelly and the bushing (as the
vessel moves up and down) is sometimes sufficient to lift the kelly bushing drive pins out of the matching
recesses on the rotary table. When the kelly is picked up sufficiently to clear the rotary table, as is the
case when adding another length of drill pipe, it will also pick up the kelly bushings out of the rotary table.
The kelly and kelly bushings are then handled together. The upper end of the kelly is equipped with a left
hand thread, as the right hand rotation of the drill string tends to disconnect right hand threads above the
rotary table.
The swivel is the next component of the drilling system and it is connected to the kelly by way of the left
hand thread at the top of the kelly. The swivel supports the total weight of the drill string and allows it to
rotate on a large bearing contained within the swivel body which does not rotate. The swivel also has the
inlet on it through which the circulating drilling fluid is pumped into the bore of the kelly and the drill string.
The swivel body has a large heavy bail at the upper end which is engaged by the drilling hook of the traveling
block.
An alternative to the Kelly and rotary table is a top drive system. Top drive systems transmit the torque
directly to the top of the drill string. The top drive systems are a packaged unit consisting of hydraulic or
electric motors driving a reduction gear which is directly connected to the lower end of the swivel. The
advantages of the power swivel is that it allows new sections of drill pipe to be installed in the drill string
more quickly than the Kelly and rotary table systems, and also allows the operator to simultaneously trip
the pipe and circulate during trips.
Top drive systems are being retrofitted on many existing rigs as they are now proven to improve drilling
speeds and provide more flexibility in the drilling operations. Rotary and Kelly systems are still retained for
back-up to the top drive systems.
The traveling block and hook are part of the hoisting or lowering mechanism by which the drill string is pulled
out of, or lowered into, the hole. Figure 3 shows the typical arrangement while tripping the drilling pipe. The
traveling block usually houses 6 large sheaves, or pulley wheels, and the drilling hook is normally integrated
into the traveling block housing. The travel block is strung up to the stationary crown block at the top of the
drilling derrick or mast. The traveling block, drilling line and crown block are in essence a giant block and
tackle.
The crown block also has 6 large sheaves and the drilling line can be strung over 4, 5, or 6 of the sheaves.
This gives rise to the terminology; ‘8-line stringing’ (4 sheaves), ‘10-line stringing’ (5 sheaves), and ‘12-line
stringing’ (6 sheaves). The number of drilling lines strung is selected according to the drill string loads are
being supported, and the desirability of hoisting and lowering the drill string in and out of the hole at
acceptable speeds. As in any block and tackle system, one end of the drilling line is anchored at the foot of
the drilling derrick or mast by way of a ‘dead line’ clamp. The moving end of the drilling line, the ‘fast line’, is
wound onto the drum of the hoist or drawworks on the rig floor.
In offshore drilling operations the floating drilling vessels are subjected to different motions due to wave
action, notably heave, pitch, roll, yaw, etc. These movements are transferred into the traveling block and
could cause the block to move about. However, a rigid tracking system is installed in the mast or derrick
which stabilizes the traveling block in order to prevent any swinging action of the block.
The hoisting unit which winches the drilling line in, and lets it out, in order to raise or lower the traveling
block and hook load, is called the drawworks. Because of the high loads carried on the hook, the
drawworks is equipped with a very powerful brake system. Most rigs are equipped with a dual braking
system, one mechanical and the other hydraulic or electric. Inside the drawworks is a system of gears,
which permits the selection of different operating speeds. The gear selection will depend on the load of the
pipe being pulled from the hole.
2.1.6 Derrick
The structure that supports the crown block and all the weight suspended on the drilling lines below is called
the derrick or mast, and offshore is typically rated at 1,100,000 lbs. load capacity. Most derricks (or
‘Standard derricks’ as they are known) are open steel structures made with four supporting legs from the
corners of a more or less square base.
Masts are more slender and tend to occupy less base area. Masts are usually cantilevered, and can be
lowered to a horizontal position during rig movements, whereas the Standard derrick is usually erected at a
drilling site and dismantled at the end of the hole.
Masts and derricks are supplied in various heights and can range from 60’ to 187’ in height. The most
popular height is about 147’ to accommodate motion compensators and power swivels in the drilling
system. Within the structure of the mast or derrick is provision to stand and rack the drill string as it is
pulled out of the hole. Normally the drill pipe is ‘stood back’ in stands of three interconnected joints called
‘thribbles’, which have an approximate length of 90 ft. The masts and derricks also have to withstand side
loads caused by high winds, and most offshore masts and derricks have wind resistance capabilities in the
region of 100 - 165 mph.
One of the essential support systems for rotary drilling is the circulating system. Figure 4 illustrates the
key features of the system. High pressure drilling fluid from the mud pump (A) is conducted into the
standpipe usually attached to a derrick or mast leg, and then into the flexible rotary hose. The rotary hose
is connected to the inlet of the swivel (B) through which the drilling fluid then enters the kelly (C) and drill
string (D). The drilling fluid is pumped down inside the drill string, through the drill bit (E) and back up the
annulus (F) between the drill string and the wellbore. The returning drilling fluid, with its entrained formation
cuttings, passes over a vibrating screen called a ‘shale shaker’ (G). The shale shaker removes the cuttings
from the drilling fluid and the drilling fluid drops into a settling pit (H). The action of the settling pit allows
finer formation particles to drop to the bottom of the pit as the drilling fluid moves through this settling pit to
the suction pit. The suction line of the mud pump or pumps takes the treated drilling fluid from this suction
pit and the circulation cycle starts again.
The pump used to circulate the drilling fluid is called the mud pump, or slush pump. Typically these pumps
are large positive displacement duplex or triplex pumps, and depending on the liner sizes, piston stroke, and
strokes per minute, can have capacities up to 900 gallons per minute and output pressures up to 5600 psi.
The rotary hose is very important and is exposed to extremely rugged service conditions. It has to be
flexible in order to accommodate the mechanical action of raising and lowering the drill string with the
drawworks. On floating drilling vessels equipped with motion compensators the rotary hose also has to
accommodate the up and down motion of the vessel. The rotary hose has to withstand both high pressures
and the highly erosive action of the drilling fluid. Rotary hoses are usually rated at 5000 psi working
pressure. Usually the rotary hose is 75 ft. long and 3 ½ in. ID.
After the drilling fluid has passed through the bit, it returns to the surface up the annulus between the drill
string and wellbore. On a land drilling operation the ‘returns’ pass through the BOP stack installed on the
wellhead and finally exit through the mud return flowline in the bell nipple installed on top of the BOP stack.
In offshore drilling operations the drilling fluid ‘returns’ have to travel further and pass through the marine riser
system (Section 3.3.1), attached to the subsea BOP stack before they finally exit through the mud return
flowline in the bell nipple attached to the inner barrel of the telescopic joint.
Offshore drilling systems generally use diesel-electric generator sets to produce all alternating current (AC)
power for the rig and the support equipment. The drawworks, mudpumps, rotary, and other systems which
require high power and variable speeds generally use direct current (DC) electric motors with the DC power
being derived from the AC electrical system through silicon controlled rectifiers (SCRs).
To clean the bottom of the hole so that the bit can drill.
Remove the cuttings generated by the bit action and transport them to the surface.
Provide cooling and lubrication for the bit and drill string during drilling operations.
Control formation pressures by exerting an overpressure on the formation. This is achieved by controlling
the hydrostatic head of the drilling fluid in the hole.
Prevent the wall of the open wellbore from caving or collapsing into the wellbore.
Most drilling fluids are water based. Oil based systems are used in deep wells where downhole
temperatures are above the boiling point of water; mineral oil based systems are used when cuttings must
be disposed of in environmentally sensitive areas.
In many areas water is the drilling fluid for the shallow sections of the hole being drilled. However, although
water is excellent in terms of penetration rates, it does not have all the properties to provide the formation
and pressure control mentioned above.
When drilling and circulating water as the drilling fluid, the water picks up a good deal of naturally occurring
clay minerals, and in fact, starts to form a natural mud, which is how the term originated. The mud formed
is thin and slurry. This natural mud slurry has improved properties over plain water, in that it has a higher
viscosity which improves its cutting load carrying capacity, and it has an increased density which means
that the hydrostatic head of the drilling fluid in the wellbore is also increased.
The naturally occurring clays when hydrated in the drilling water tend to be unstable and particularly prone to
contamination; for instance, salt water entering the drilling fluid from one of the formations penetrated by the
wellbore can cause the gel to flocculate and lose its suspending power. This problem has been solved by
using specially prepared bentonitic clays. They form very stable gels and provide a much greater support for
cuttings, and also greater support for the powdered weighting materials.
The most common weighting material used is barytes (barium sulphate). It is supplied in finely powdered
form that is introduced into the circulating drilling fluid and remains suspended in the drilling fluid due to the
gel strength of the slurry. The specific gravity of barytes is about 4.2 which means that the density of the
drilling fluid can be increased greatly by addition of the powdered barytes. This in turn provides the control
of the hydrostatic head exerted by the drilling fluid on the formation.
All drilling fluids or muds can be tested in order to determine their physical and chemical properties, and this
is done on a regular basis at the rig, either on land or offshore. The most regular tests performed measure
the viscosity and weight of the drilling fluid or mud. Other tests show the acidity or alkalinity of the mud in
terms of its pH value; the amount of mud cake or filter cake the drilling fluid or mud deposits with a known
pressure differential; the amount of solids in the mud as a volumetric percentage, and the amount of fluid (be
it water or oil), in the mud as a volumetric percentage. These properties provide a guide for the mud
engineer who can then prescribe a treatment for the mud, if required, to improve the muds performance in
the hole.
The actual plan for removing rock and creating a hole is the drilling program. Though the hole is drilled from
the surface, the drilling program is designed from the target depth up. The planning of a hole starts with the
determination of objectives for both the depth of the target formation and the hole size in that formation. The
drilling program then is the strategy for attaining these objectives with the minimum hole and casing sizes
possible, while allowing for contingencies.
Before discussing the selection of casing sizes and their individual lengths, it is instructive to review the
mechanics of controlling formation pressures.
The hydrostatic pressure of the drilling fluid is a function of the mud weight in the annular column (i.e., the
amount of baryte in the drilling fluid mixture). In deciding on the appropriate mud weight, the driller must
take into account overburden pressure, formation pressure and fracture pressure.
The overburden pressure is the measure of the stress in the formation rock caused by the weight of the
material above it. The density of rock varies with depth but is generally assumed to be about 94 #/ft3. The
rock stress gradient in a formation is generally assumed to be about 1 psi per foot of vertical depth.
The fracture pressure is the pressure of the fluid in the rock pores that will cause the formation rock to
fracture and separate, opening fissures in the formation. This pressure is usually equal to the overburden
pressure plus the tensile strength of the rock.
Open fissures in the rock increases the rock porosity and permeability. This is sometimes a desirable
Fracturing a formation is not a desirable occurrence while drilling a hole because it creates a sink for the
drilling fluids to flow into, leading to the loss of the circulation fluid. With sufficient lost circulation fluids,
also called lost returns, the hydrostatic head of the drilling fluid column can decrease to a point where the
formation pressure of a deeper zone can cause a kick.
The fluid pressure in the rock pores is the formation pressure. A formation is considered normal if the
formation pressure equals the normal hydrostatic gradient pressure.
Formations are not typically solid rock but rather are porous, containing primarily salty water called brine.
The water pressure in the rock pores increases with depth by the hydrostatic gradient, typically .465 psi per
foot of vertical depth.
A formation is abnormally pressured if the formation pressure is above the hydrostatic gradient pressure due
to the effect of the rock stress gradient. This is generally the case in most regions of the world. Figure 5
illustrates an abnormally pressured formation.
A formation is subnormally pressured when the formation pressure is less than the normal hydrostatic
gradient pressure. This occurs infrequently.
Figure 6 shows the formation, fracture and overburden pressures plotted for a hypothetical hole. The
formation pressure is normal down to 7000 ft where it becomes abnormal.
A drilling engineer does not refer to such composite profiles in order to keep the mud weight between the
formation and fracture pressures. Instead, they generally work with gradients for mud weight and the
resulting pressure profile in the hole, as shown in Figure 7.
To provide a structural member to support the wellhead and deeper set casing strings.
To isolate shallow, weaker formations from lower, higher pressured formations.
To isolate troublesome formations from the working area of the hole.
Figure 8 shows an example of a typical casing program. The shallowest casing string, or conductor pipe,
generally serves as a structural member, and keep the less consolidate rock from sloughing off into the
hole. Prior to drilling into the deeper and abnormally pressured formations, a string of surface casing is set
to permit increasing the mud weight. As drilling progresses deeper towards the target formation, additional
strings are set to isolate the lower strength zones higher up in the hole.
Most casing programs allow for the use of ‘trouble strings’ which may be set if an unanticipated condition in
the hole requires a zone to be isolated before the target formation can be reached. The need for trouble
strings is more prevalent in exploratory drilling where little is known.
A production casing string is run in development wells in order to isolate the production zone from the rest of
the hole. This string will support the production tubing, with the production zone isolated by packers. In
some cases, the production casing serves as the production conduit.
There is no basic difference in the BOPs used onshore and on a platform and there is only detail differences,
primarily in the control system, between those used on the surface and those used on the seabed when
drilling from a floating rig. The principles of operation and the internals of the BOPs are the same.
BOPs are identified by their bore and pressure rating. The bore limits the maximum tool size which can be
run through the BOP. Standard BOP pressure rating are 2000, 5000, 10,000, or 15,000 psi. The required
BOP pressure rating is generally determined by the depth of the well being drilled and the anticipated
formation pressures in the area.
Floating drilling units which drill and complete the majority of subsea wells generally use 18-3/4” - 10,000 psi
working pressure BOP systems. In the past 10 years the higher pressure 18-3/4” - 15,000 psi systems
have become available and rigs equipped with these units are in demand for high pressure/high temperature
(HPHT) wells and for deepwater drilling operations.
There are two types of BOPs: Annular, and Ram. Figure 10 shows a cut away of a typical subsea stack
with both types of BOPs. The annular consists of a ring of rubber with metal inserts which, when activated,
is forced inward towards the center of the wellbore closing around whatever is in the well. The ram type have
two separate blocks of metal, with rubber seals which, when activated, meet in the center of the wellhead
closing around the pipe that is in the hole or sealing if there is nothing in the hole.
In an annular preventer, as pictured in Figure 11, the actuating cylinder is around the outside of the wellbore
and the inside of the cylinder has a tapered side, against which rests the outside of the BOP rubber, which
has a taper the reverse of the cylinder. On being actuated, the cylinder moves upward (the oil above the
cylinder being returned to an oil reservoir) forcing the rubber (with metal inserts) inward to close around
whatever is in the hole, be it drill pipe, tool joint, kelly or nothing.
In the ram type preventers there are two types of rams: Pipe Rams and Blind Rams. The pipe rams will
only fit one specific size of pipe as the rams have to be cut out so as to fit around the pipe size in question;
therefore, for every size of pipe used, there is a different size ram.
To close on an empty hole, blind rams are used. These have no cut-outs and are made so that they will
mesh into each other when closed, sealing the hole completely. A variant of this type is the shear ram,
which has the same function as the blind ram but also includes a scissor action cutting shear which will cut
through any pipe that is in the hole.
The number of BOPs that are used on any hole will vary with the type of hole being drilled. Onshore the
number will be three or four, consisting of one annular preventer, one pipe ram and one blind ram. If a fourth
preventer is added, it will also be a pipe ram. The same configuration will apply on a platform or on a jack-
up rig. On a floating rig with a seabed system, there will be five or six BOPs, consisting of one or two
annular preventers, three pipe ram preventers and one blind/shear ram.
All BOPs are worked in the same manner, by hydraulic pressure. When it is desired to shut the BOP, a
valve is operated either by the driller or at a remote position, and allows hydraulic pressure to enter the BOP
closing it. For opening, pressure is applied to the opposite side of the operating mechanism and the BOP is
opened.
The ram type BOPs have cylinders on both sides fixed to the ram carrier. To close the ram, hydraulic fluid
is admitted to the outside of the piston in the cylinders on either side closing the ram. Opening is the
reverse, with the fluid being admitted to the center of the cylinders. Changing the rams is in all cases a
relatively simple matter though the method does vary.
The control panels for the BOPs will be by the drillers position and there will be remote panels away from
the rig so that the BOPs can be closed even if the rig has had to be evacuated (this is of great use if H2S is
present). The hydraulic fluid for actuating a BOP system is kept in accumulators, which are pressure
vessels with nitrogen and hydraulic fluid. The hydraulic fluid is pumped into the accumulator until the
required pressure is reached when the pump automatically shuts down. On opening the valve to actuate the
BOP, the nitrogen expands forcing the fluid into the BOP and the surplus fluid on the other side of the
operating mechanism in the BOP is returned to the fluid reservoir in the accumulator system.
Once the kick has been controlled by the use of BOPs, other equipment is required to control the annulus
pressure while circulating the kick out of the hole. This equipment includes kill and choke lines, and a
choke manifold. A heavy drilling fluid, called a kill fluid, is pumped into the annulus to regain hydrostatic
pressure control over the formation fluids. The fluids already in the annulus are vented through the choke
line.
The annular fluids are vented to the choke manifold, which consists of a remote control choke that is
controlled from the drillers position, a hand operated adjustable choke, and fixed chokes. Figure 12 is a
schematic of a typical manifold for a subsea BOP. By adjusting these chokes it is possible to control the
back pressure in the annulus. The remote control adjustable choke enables the driller to be able to hold the
desired pressures on the annulus while being able to control pumping rate, etc. With the hand operated
choke it is necessary to have an operator at the choke manifold and communication is a possible problem.
Fixed chokes can be used if the rate at which the fluid is coming out of the well is too great for the
adjustable choke and then the fixed choke is used in parallel with the adjustable choke.
The submersible drilling unit is sometimes called the swamp barge or posted barge. This type of unit is
used in shallow waters such as rivers and bays, usually in waters up to 50 feet deep. The semisubmersible
is floated to location like a conventional barge and is then ballasted to rest on the river bottom. The lower
hulls are designed to withstand the weight of the total unit and the drilling load. The semisubmersible has
two hulls. The upper hull, sometimes referred to as the “Texas” deck, is used to house the crew quarters
and equipment, and the drilling is performed through a slot on the stern with a cantilevered structure. The
lower hull is the ballast area and is also the foundation used while drilling.
Jack-up designs can generally be classified into two basic categories: independent leg jack-ups and mat
supported jack-ups. Each unit has its particular value.
The independent leg jack-up depicted in Figure 13, will operate anywhere currently available, but it is
normally used in areas of firm soil, coral, or uneven sea bed. The independent leg unit depends on a
platform (spud can) at the base of each leg for support. These spud cans are either circular, square, or
polygonal, and are usually small. Spud cans are subjected to bearing pressures of around 5,000 to 6,000
pounds per square foot, although in the North Sea this can be as much as 10,000 psf. Allowable bearing
pressures must be known before a jack-up can be put on location.
The mat supported jack-up is designed for areas of low soil shear value where bearing pressures must be
kept low. The mat is connected to all the legs. With such a large are in contact with soil, bearing
pressures of 500 to 600 psf usually exist. The mat type jack-up penetrates the sea bed perhaps 5 or 6 feet.
This compares with a penetration of perhaps 40 feet on an independent leg jack-up. As a result, the mat
type unit requires less leg than the independent jack-up for the same water depth. The mat type unit
requires a fairly level sea bed. A maximum sea bed slope of 1 ½° is considered to be the limit. Mats are
designed for uniform bearing. In areas where there is coral or large rock formations, the uneven bottom
would probably cause a structural failure.
Jack-ups can be either self-propelled, propulsion assisted, or nonpropelled. The majority of jack-up rigs are
nonpropelled. The self-propelled unit requires a specially trained crew of seamen as well as a drilling team.
Jack-ups have been built with as many as 14 legs and as few as 3 legs. As the water depth increases and
the environmental criteria become more severe, more than 4 legs is impractical. The prime forces on a jack-
up are generated from the waves and currents, hence, the less exposure to the waves and currents the
fewer the forces being developed on the unit.
The early jack-ups were designed to operate in the U.S. Gulf of Mexico in water depths up to 200 feet.
Wave heights in the range of 20 to 30 feet with winds up to 75 mph were considered as design criteria for
these units.
Newer units have been designed to operate in 360 ft water depth in the North Sea with design criteria of 90
foot wave, 125 mph wind and 2 to 2 ½ knot currents.
The semisubmersible evolved from the semisubmersible. The general configuration of a semisubmersible
consists of two longitudinal lower hulls which are used as ballast compartments to achieve the necessary
drilling draft. The semisubmersible is ballasted to float on these hulls during transit. This last unit does not
offer the towing capabilities of twin hull units, but it does provide good drilling characteristics.
Semisubmersibles are held on location either by a conventional mooring system or by dynamic positioning.
The conventional mooring system usually consists of 8 anchors placed in a spread pattern and connected
to the hull by chain or wire rope, sometimes even a combination of both. The dynamic positioning method is
an evolution of the ship sonar system whereby a signal is sent out from the floating vessel to transducers
set out on the ocean floor. Dynamic positioning becomes a greater necessity as the water depth increases.
The motion that causes problems for any floating drilling vessel such as a semisubmersible is heave (i.e.,
vertical motion). Heave is generated in response to exposed waterplane area. The lower submerged hulls
supply the primary displacement. The columns connecting the hulls to the topside deck have a small
waterplane.
With the loss of waterplane area to reduce heave response, a reduction in stability follows. A compromise
between acceptable heave response and adequate stability is needed.
Some semisubmersibles are self propelled. Propulsion is a large initial expense which can be recovered in
a reasonable period of time if mobility is required. Considering that once a unit has reached location it is
generally in that area for a long time, propulsion units become not only unnecessary but they also use up
valuable consumables (weight) capacity.
4.1.4 Drillships
The drillship is a shipshape vessel used for drilling purposes. Earlier drillships were converted vessels,
either barges, ore carriers, tankers, or supply vessels.
Drillships are the most mobile of all drilling units, but are also the least productive. The configuration that
permits mobility results in bad drilling capabilities. The drillship, because of its surface contact with the
sea, develops very large heave response compared to the semisubmersible. It is possible, by means of
stabilizing tanks and other methods, to reduce roll on drillships, but heave cannot be reduced. A
subsequent increase in “rig downtime” or “lost” time occurs. This increases the demand for the use of
compensation devices.
Mooring for drillships is very similar to the methods previously discussed for semisubmersibles. There is
one additional system that has been developed on a drillship - the “Turret” system. This is a spread mooring
tied to a turret at the mid section of the vessel. The moonpool is in turret. The rest of the vessel can rotate,
or weathervane, around the fixed mooring.
A riser can be simply described as a conduit from the platform to the ocean floor which transmits the drilling
mud and serves as a guide for the drill string. There are two broad classes of risers, those used for mobile
drilling operations and those used for production operations.
In mobile drilling, the riser tube running from the vessel down to the wellhead is normally called a riser while
the first casing joint running from the wellhead slightly above the mudline down through the template to 30-
100 ft below the mudline is called a conductor or surface pipe. In production operations, the riser tube
running from the platform deck down to the wellhead is also sometimes called a conductor.
In mobile drilling operations conducted with a jack-up rig (See Figure 13), the riser runs from the drilling deck
down to the wellhead. The drilling fluid is pumped down the drill string. The mud returns in the annulus
between the string and the riser in the water or between the string and the casing for that part of the drill
string below the subsea wellhead. Ordinarily, when oil is struck in mobile drilling the well is tested and
capped and the rig moves on to another drilling site. In most areas the regulations require that all equipment
be removed from the ocean floor. The riser string can then be reused.
Figure 15 shows a perspective view of a marine riser system for a semisubmersible rig. Figure 16 illustrates
the arrangement of the riser system. It is important to note that the riser is not a simple piece of equipment
but rather a complex system of component parts. The riser string for a floating exploratory vessel is usually
made up of 50 ft long joints which can be stacked and stored on deck during transit from one drilling location
to another. The ends of each joint have quick-disconnect couplings permanently attached to the joint. The
telescoping joint which is at the upper end of the riser string is usually designed for a maximum heave of
between 15 and 30 ft.
Figure 17 shows the general arrangement on a floating drilling rig. A constant force tensioner system is
attached to the top of the fixed outer barrel of the telescoping joint to provide enough axial force in the riser
string to prevent buckling. The outer barrel and the riser string have lateral movement with vessel surge and
sway but essentially no vertical movement with vessel heave. The vessel and the inner barrel of the
telescoping joint move together vertically with vessel heave. The optimum tension is a function of water
depth and operating conditions.
Ball joints on each end of the riser allow for rotation in any direction up to about 7-10 degrees. Actually,
only a few operators insist on two ball joints, which offer more reliability than a single ball joint, because the
use of two ball joints incurs greater costs and greater running time. The usual arrangement for floating
drilling operations is a gimbal under the drilling deck and one ball joint attached to the top of the subsea
BOP stack, which sits on the wellhead. The wellhead attaches to the base template that is set with the
conductor type at the beginning of the operation.
The first riser systems had the choke and kill lines strapped to the riser pipe. Most riser systems now use
integral choke and kill lines which are permanently attached to opposite sides of the riser and have their own
connectors. When the riser joints are stabbed and quick-connected the receptacles allow the choke and kill
lines to be stabbed and automatically connected at the same time.
The choke and kill lines run from the deck along the riser string down to the wellhead. At the lower riser ball
joint there are various schemes, such as looped pipes, to get the required flexibility in a jump line
arrangement running from the bottom of the riser string (top of ball joint) around the ball joint to the BOP
stack. The choke and kill lines control kicks in order to prevent them from developing into blowouts.
Figure 18 shows the general arrangement at the wellhead. When a potential blowout is detected, mud is
pumped down the kill line at the BOP stack to restore pressure balance in the hole. When excess gases
occur the annular and ram type BOPs are closed around the drill string. The gas is relieved at the choke
manifold on the BOP stack by running up the choke line on the riser string. As the gas expands it proceeds
up the choke, displacing more mud and traveling faster as the gas bubble or gas-entrapped mud approaches
the mean water line. Without the choke, the gas would push out the annulus mud between the drill string
and the riser control from the weighted mud would be lost.
There are two basic types of telescoping joints used with marine risers. The constant tension system
(remote axial tensioning system) is most often used because maintenance is easier. This method uses a
linkage system at the base of the drilling floor to maintain equal force on the several wire ropes attached to
the outer barrel of the telescoping joint. Figure 19 illustrates a general arrangement.
An alternate design of telescoping joint uses the direct axial tensioning method. This is a procedure where
the seals and guide rings on the telescoping joint are designed to compensate for internal pressure so that
the telescoping joint has the dual function of allowing vessel heave and acting as a direct tensioning piston.
A diverter is located at the top of the telescoping joint, as shown in Figure 20. Depending upon the
magnitude of the kick, the gasified mud is valved either onto the shale shakers or to port.
Most floating exploratory vessels are now outfitted with a riser system for a maximum water depth of about
1,200 ft. Several floating rigs in operation will have riser systems for the extreme water depth of 3,000 ft.
One drillship has a riser system for 7,500 ft, and semisubmersibles have been built to accommodate 10,000
ft water depth.
A drawback with risers for over 6,000 ft water depths is that the static head of heavy drilling fluid (e.g. 20 lb.
mud) could be about 7,500 psi at the mudline and proportionately greater with drilling depth compared to the
normal water pressure of about 3,000 psi at the mudline. This extraordinarily high pressure can cause
severe fracturing of the formation. Very light drilling fluids are used to avoid fracturing the shallow
formations, necessitating additional casing strings as the drilling goes deeper.
The maximum tension in the riser occurs at the top of the riser string and decreases with water depth. In
very deep water, some type of added buoyancy is required to stay within the practical limits of the tensioner
system. Figure 21 shows a riser joint equipped with a air can system. However, it must be remembered
that the current drag force goes up as the square of the velocity and increases with riser diameter. This
means that the outside diameter of the added buoyancy cylinders should be minimized. The optimum
design riser has the smallest diameter and lowest possible wall thickness consistent with allowable stress
levels and allowances for wear and/or abuse.
There are two means to guide objects from the drilling floor to the seafloor. The most common system is
guidelines, either 2 or 4, run between a base template and the moonpool. The second system, used in very
deep water, is guidelineless. This system incorporates a baseline system of acoustic transponders to
determine the relative positions of the object being lowered, and the seafloor target, usually the well location.
The guideline system is established by running a temporary guide base to the seafloor on a running, as
shown in Figure 22. This guide base provides to guidelines which are used to guide the drillstring for
creating the hole for the 30-inch surface casing. The 30-inch case is run with the permanent guide base as
shown in Figure 23. Sometimes the 30-inch casing is jetted, or washed, into the seafloor by pumped
seawater down through the casing as it is lowered.
The permanent guidebase creates a four guideline system which is used to drill the hole for the 20-inch
structural casing. This operation can be done with or without the drilling riser, as shown in Figure 24. If the
returns are not brought to the surface, consideration must be given to possible interference problems with
the return pile that will be created around the wellhead. The running of the 20-inch casing, as shown in
Figure 25, establishes the wellhead on which the BOP is landed.
In guidelineless systems, the temporary guidebase is typically run on the hole opener for the 30-inch
surface casing as shown in Figure 26. Note the acoustic targets on the guidebase. The 30-inch casing with
the permanent guidebase attached is lowered into and through the temporary base by use of the acoustic
location system and a wireline sonar and T.V. tool located at the bottom of the 30-inch as shown in Figure
27. The long baseline acoustic system gets the vessel over the target, while the ultra-short baseline sonar
guides the casing towards the acoustic targets on the temporary base. The camera system then is used to
stab the casing into the well guide.
The 30-inch casing can be directly jetted into the seafloor, as shown in Figure 28 if the soils are appropriate.
This avoids use of the temporary guidebase.
The 20-inch casing is run in a similar manner as shown in Figure 29. This establishes the wellhead on
which the BOP is landed using similar guidance.
Relative motion must be accommodated in the riser, guideline and drill pipe systems. The drill pipe
compensation system is usually designed into the drawworks, as shown in Figure 30. The riser and
guidelines systems use external compensators called tensioners, as shown in Figure 31.
A riser on a floating rig is supported by axial tension applied to the top of the riser and possibly by buoyancy
along the length of the riser. Tension controls the stress level in the riser pipe and affects the riser
straightness during drilling operations. As rise size, water depth, sea conditions, mud weights, etc.
increase, the axial tension requirements for providing proper support also increase. A rough order of
magnitude of the nominal tension required to maintain the bottom joint of the riser in positive tension is:
Standard tensioner sizes utilized today are 60,000 lbs. or 80,000 lbs. per tensioner. These units are utilized
in systems containing 4 to 10 tensioners. Riser tensioners are operated in pairs so that the two tensioners
connected on diagonally opposed sides of the riser are always at the same tension level. Additional pairs of
tensioners are used to achieve both redundancy of tension availability and high levels of tension by utilizing
all the tensioners on the riser at the same time. A well designed system will provide the maximum required
tension level with at least one tensioner out of service.
The tensioner wireline travel or motion compensation capability must exceed the maximum vessel heave
expected to occur while the riser is connected to the wellhead. The tensioner motion capacity must not
only exceed vessel heave but must also account for tidal motion, connection adjustment, and changes in
vessel ballast position. Excessive travel requirements will normally be detrimental to overall operational
efficiency of any tensioner. The tensioner must have the capability to respond at the maximum peak
response of rig vertical heave motion. This response must equal or exceed the instantaneous maximum
vertical speed of the vessel heave, which exceeds the average vertical speed of the vessel.
This linear force is transmitted by a wire rope to the top of the marine riser to provide axial tension. One end
of the wirerope is connected to the tensioner or rig and the other end is reeved across an idler sheave and to
the top of the marine riser. Force in the cylinder is thus translated to wirerope tension and then to tension in
the riser. Air pressure is maintained at an approximately fixed level to maintain the desired tension level in
the riser. As the rig heaves upward, the top of the marine riser remains fixed (as it is fixed to the earth by
being connected to the wellhead).
Wellhead guidelines must also be taut to be effective. In order to maintain guidelines taut at a preselected
tension level, hydra-pneumatic tensioners are applied to each of the four wellhead guidelines and normally to
the blowout preventer control lines. Guideline tensioners operate exactly the same as riser tensioners and
are designed the same, except they are smaller in size. Figure 33 illustrates their operation.
The major application for a drill string compensator (DSC) is to nullify rig heave that would be imposed on
the drill string. This motion nullification significantly improves the operation of the following procedures:
All present drilling rig DSC’s are an air spring tensioning device. They are passive devices that function
based on the difference in the suspended weight of the drill string and the tension level set in the DSC.
During drilling, the drill string weight is supported by the hydraulic-pneumatic cylinder of the DSC and the
drill bit weight on the bottom. The cylinders are interconnected to the air pressure vessels (the same as the
riser tensioners). Control of the air pressure in the air pressure vessels determines the tension level. Proper
DSC drilling techniques always require a DSC tension setting to be less than the weight of the drill string.
There are two basic types of DSC’s: motion compensate the crown block, or the traveling block.
A crown block DSC, shown in Figure 34, is a tensioning device that supports the crown block and thus the
drill string. By supporting the crown block with controlled tension, the compensator becomes a motion
nullification device by raising and lowering the crown block. This then raises and lowers the traveling block
and hook to nullify motion or to isolate rig motion from the drill string.
A traveling block DSC, as shown in Figure 35, is connected between the traveling block and the hook. Its
tension level is controlled by techniques identical to a riser tensioner. The tension device can be either a
tension type or a compression type.
WELL COMPLETION
TABLE OF CONTENTS
LIST OF FIGURES
FIGURE 1. Typical Casing Programs
FIGURE 2. Engagement Illustration
FIGURE 3. Typical Wellhead Assembly
FIGURE 4. Typical Platform Wellhead Assembly
FIGURE 5. Christmas Tree Assemblies
FIGURE 6. Casing Heads
FIGURE 7. Casing Spools
FIGURE 8. Casing Hangers
FIGURE 9. N Tubing Spool
FIGURE 10. Tubing Hangers
FIGURE 11. Xmas Tree in Production Mode
FIGURE 12. Cross-Section of Seafloor Completion
FIGURE 13. Marine Wellhead Stack-ups
FIGURE 14. Wellhead Assembly for a Two Stack System
FIGURE 15. SG-5 High Pressure Housing
FIGURE 16. Completion Riser
FIGURE 17. Running Tubing
FIGURE 18. Vetco H-4 Connector
FIGURE 19. Retainer Type Hookwall Packer
FIGURE 20. Mandrel Type Hookwall Packer
FIGURE 21. Anchor Packer
FIGURE 22. Cup Packer
FIGURE 23. Inflatable Packer
FIGURE 24. Mechanically Set Packer
FIGURE 25. Otis Type F Pressure - Differential DHSV (Storm Choke)
FIGURE 26. Wireline Retrievable Flapper and Ball Type DHSV’s
FIGURE 27. Tubing Retrievable Flapper and Ball Type DHSV’s
FIGURE 28. Balance Type DHSV
LIST OF TABLES
TABLE 1. Summary of Casting Design Criteria
TABLE 2. Typical Casing Design Factors
TABLE 3. Reasons for Using Production Packers
1.1.1 Caisson
A normal offshore platform well will include a caisson that is usually 30 inches O.D. This conductor is driven
into the sea bottom to allow spudding the well. In recent years, curved conductors have been used to help
maximize horizontal reach for the directionally drilled wells.
The conductor pipe is the first string that is run and cemented. For offshore wells, it is usually run to 1000
feet deep and it can be from 7 to 30 inches in diameter. For most offshore wells, the conductor is usually
20 inches O.D. and cemented as deep as 1000 feet. This string provides a fluid course for deeper drilling by
casing off the shallow unconsolidated formations. It is commonly cemented back to the surface and
occasionally the cemented conductor pipe acts as a support for the well.
The surface casing can vary in depth from a few hundred feet to more than 4000 feet. It can act as the
foundation pile if this function is not provided by the conductor pipe. Also, the surface casing protects the
freshwater sands from contamination by wellbore fluids. Sizes typically vary from 7 to 16 inches with 10-3/4
and 13-3/8 inches being the most commonly used. For offshore directionally drilled wells, the cement top is
limited to the kick-off point to simplify platform abandonment.
For normally pressured wells, drilling can proceed without the need of further casing provided normal mud
weights can be used (up to about 12 ppg). If lost circulation zones exist, special considerations may be
necessary to minimize drilling problems. Once the well is drilled through the interval to be completed, the
production string is set. Sizes vary from 2-3/8 through 9-5/8 inches and extend to or through the producing
formations. These strings are cemented to an adequate distance above the producing formation. This may
be established by regulations and usually is about 1000 feet. However, this string is sometimes cemented
back to the surface.
If geopressures are encountered, high weight mud must be used to control the well. This requires installing
a protective string to prevent losing returns to shallower formations. These strings vary from 7,000 feet to as
much as 15,000 feet in length. Large pipe of 7 inches to 11-3/4 inches is used. At least the lower 1000 feet
of the string is cemented and it is hung from the surface casing head housing. Longer cement intervals are
frequently used to prevent buckling.
1.1.6 Liners
If the abnormal pressure increases at depth such that the fracture pressure of the shallower abnormal
pressure zones is exceeded, the shallower zones must be cased off to prevent loss of circulation. The
protective string is then extended with a liner. The liners are usually cemented full length, with 5-inch and 7-
inch sizes being common.
1.1.7 Tubing
The tubing string is installed to allow handling well fluids through a removable conduit. This conduit protects
the casing strings from erosion and corrosion damage, which is important because the casing strings
cannot be removed if they are damaged. Tubing size is selected based on the rates which must be
handled.
If H2S gas is encountered in the produced fluids, the production can be “sour”. This is the term given to fluid
streams containing hydrogen sulfide gas. If the concentration exceeds an H2S partial pressure of 0.001
atmosphere, tubular steel must be selected to have a specific cracking susceptibility value (10 minimum).
This normally requires a Rockwell C hardness of less than RC22 which limits the ultimate strength to about
100,000 psi. API grade C-75 and L-80 casing are manufactured especially for this service and a C-90 grade
is being developed.
Whenever unusually high or low temperature conditions can exist during either normal producing operations,
well killing operations, or workover conditions, the casing/tubing designs must account for the thermal
expansions and contractions which will occur. If the design does not account for these conditions, large
thermal stresses can result and failure is almost certainly assured.
The doglegs in directional wells bend the casing and induce added axial stress. Except for the larger
casings, this is usually not critical. API Bulletin 5C4 is available to provide design guidance for this effect on
tension. Also, the casing tends to deform (oval) when bent. This reduces its collapse resistance and must
be considered.
1.3.1 Depth
The well depth objectives can usually be predetermined to within a few hundred feet. This is done through
subsurface and seismic control techniques.
1.3.2 Size
The tubing size is established on the basis of design rates and acceptable pressures. Deep wells with high
rates will require large tubing sizes. The casing sizes are determined to accommodate the tubing without
“running out of hole” before reaching depth. Deeper wells will require larger casing programs.
1.3.3 Pressures
Casing design pressures include both those imposed by the drilling mud gradients and the well fluid
gradients.
The casing design should consider the type of well being drilled. If it is a development well with a high
change of success, the design should be conservative without being wasteful. If the well is an exploratory
well, the design should be minimal because of the low change of success. This design requires
considerable judgement on the part of the operator because the cost consequences of losing control of an
exploratory well can be many times the cost savings resulting from a minimal design.
These criteria will apply for most casing designs including low to medium pressures and shallow to medium
depth wells. The criteria include effects of buoyancy but generally neglect axial load effects.
These criteria are intended for high pressured deep casing designs. More thorough analysis of loads and
their interactions on designs are warranted. Buoyancy, effect of tension on burst, and the effect of
Sometimes even more specific analysis warranted. These criteria are directed to those special wells and
include considerations for triaxial stress analysis, thermal stresses, and temperature effects.
Factors of safety are used in all engineering designs. These factors are used to allow for conditions which
methods of calculation are not available, such as uncertainties of materials strength or loading conditions.
Use of factors of safety for casing design can reduce the complexities of the designs. However, because
numerous simplifying assumptions are made, the resulting factors are not true factors of safety as the term
normally implies. They are actually design factors which can be applied to assure that resulting designs are
adequate with confidence that safety levels are acceptable. Typical design factors are shown on Table 2.
Engineering judgement can be used to determine which type of design factors should be applied.
Regardless of the type that is used, the designer should be confident that casing design integrity results.
A casing string must be designed for basic tension, burst, and collapse loads. Also, if they can occur,
loads due to bending, buckling, acceleration, or torque must be considered.
The tension design must allow the casing string to be run into the well without parting. The casing string
must be capable of withstanding any internal pressures that exist during drilling or producing operations.
Also, the casing string must be able to withstand external collapse pressures which can exist if the
pressure inside the casing is reduced or when the casing is immersed in well fluid. If combinations of
tension, burst, or collapse can exist simultaneously, the casing design must consider these interactive
loads. When appropriate, possible buckling failure should be considered.
A casing string may also be exposed to other loading conditions which should be considered.
Slip crushing at the wellhead slip area is possible because the casing is subjected to severe collapsing
loads. Controlled friction slips can be used to keep the strain from these loads within drift tolerance for most
casing strings. Heavy casing strings should be reviewed carefully. For normal drilling practices, it is
common to install the uppermost “gage” joint which is as heavy as the heaviest weight of casing in the
string. This will assure adequate collapse resistance of the casing.
The effects of temperature changes should be considered. Very large tensile or compressive stresses can
be imposed on the tubing string because it is fixed at both ends of the casing string.
Surface casing is subjected to large compression loads just below the wellhead. All of the hanging weights
of the protective or production string is transmitted to the earth through the surface casing. If large loading
circumstances warrant, the stability of the surface casing as a freestanding column should be checked.
Drilling wear of the casing string (or key “seating”) can be caused by the drill pipe rotation. If this wear can
be quantified, reduction in burst or collapse pressure should be considered.
Perforating the casings can damage the casing and reduce its collapse strength. Appropriate perforating
techniques is the best method to use to prevent this from becoming a problem.
Corrosion can be caused by produced fluids or groundwater. If necessary to assure long term casing
integrity, appropriate corrosion allowances or protective measures may be necessary.
During some cementing processes, the casing is rotated rather than reciprocated. This can place a large
torque on the casing and its joint. The torque value should not be allowed to exceed the make-up torque
value for the connections. Sometimes, rotating torque requirements can be reduced by using a combination
of lowering and torquing the string.
These methods are tedious and time-consuming compared to design chart methods. Thus, it is rarely used.
These methods are frequently used, especially if computer facilities are not available. The design chart
method is frequently used as a learning tool to demonstrate how the computer methods work.
Once the designer understands how to apply the proper design methods, the computer is the best way to
perform many complex casing designs quickly and easily.
Their use tend to inflate casing costs because they are usually based on loads which are greater than
necessary. They can be used when necessary, but designs are usually very conservative.
1.7 Connectors
Three types of API joints are typically used for casing connectors. As shown on Figure 2, these include
round thread, buttress thread, and extreme line connectors.
The connector is threaded and coupled with conventional rounded crest and root V-type 8 pitch threads
tapered ¾ inch per foot on diameter.
The connector is threaded and coupled with special buttress form 5 pitch threads tapered ¾ inch per foot on
diameter.
The connector is an integral type joint with special modified acme form threads and a specially designed
positive metal-to-metal pressure seal.
The tubing/tubing hanger equipment provides the pressure tight producing conduit from the producing zone
to the surface. A manually operated master valve is mounted immediately on top of the tubing string, with
an operated master valve immediately above the manual master. A flow fitting allows mounting a swab valve
on top of the operated master valve and choke and manual wing valve to the side. This arrangement allows
access to the tubing string for wireline work and positive isolation for operated valve testing/repair or choke
size changing.
Immediately below the tubing master, a side annulus access valve is provided. The most common function
for this is to allow checking for tubing/packer failure or gas lifting. Also located in this section is the ¼ inch
control line penetration to the downhole safety valve (DHSV).
An additional valve is located below the production annulus valve to provide annulus access between the
surface casing and protective casing. Normally, this is only used to monitor for downhole failures.
Wellhead working pressure rating must be greater than the maximum expected operating pressures.
Wellhead working pressures and pressure testing requirements are defined in API Spec 6A as follows:
It is not uncommon for the working pressure rating to vary from the bottom to the top of tree. For example, a
3000 psi casing head can crossover to a 5000 psi tubing head and master valve. This is acceptable since
the lower portion of the wellhead may be subjected to lower pressure conditions. Crossover seals must be
used to isolate the lower pressure equipment from the higher pressure equipment.
The casing head is selected to withstand the maximum pressure which will be encountered during drilling.
Its working pressure should be only somewhat above the casing burst pressure. The casing head also
serves as a support for BOP drilling equipment.
The effect of temperature should be accommodated. API Spec 6A identifies applicable steel parts for
temperatures between -20° and 250°F. this range should handle the vast majority of wells.
Materials should be selected in accordance with API 6A and the product should be obtained from an API
qualified manufacturer. Close evaluation will be necessary only for special applications. If CO2 content
exceeds more than about 15 psi partial pressure, 410 stainless should be used unless H2S is present. 316
stainless may then be a better choice.
Wellheads and components should be coated to prevent rusting. Several acceptable coating
systems/methods have been applied successfully.
If the well heads are located on a congested, manned platform, wellhead selection should consider integrity
under fire conditions for a few hours. This choice may result in either the saving or loss of major
investments.
A casing head, as shown in Figure 6, should be installed on the first string of casing which requires blow-out
preventers for further drilling. Connection to the casing can be either by a female thread or a slip-on welding
connection. The latter eliminates space-out problems but requires availability of a welder.
A casing head spool, as shown in Figure 7, is normally used for the running and hanging of each additional
casing string. Careful attention to crossover connections and pressure ratings must be given.
Casing hangers, as shown in Figure 8, are required for all subsequent strings of casing after installation of
the casing head housing. The hanger supports and centralizes the inner string and provides a seal between
the two strings. The hanger must be designed not to mash or crimp the inner string. Careful attention
must be given to this area and consideration provided for getting the casing through the BOP, landing, and
Design of the tubing head spool, as shown in Figure 9, is much like that of the casing head spool. The
bottom flange must be the same size and rating of the mating flange.
The most suitable all around hanger is the mandrel type, as shown in Figure 10, that is internally prepared
for a back pressure valve. They should be threaded with the same thread on bottom as the tubing joint. The
top thread should normally be EUE and the mandrel size and type must be compatible with the tubing head
spool.
A typical wet tree subsea diverless system is shown on Figure 11. This particular arrangement is equipped
for through flowline (TFL) operations. TFL permits performing many of the downhole operations which would
normally be done by wireline methods if the wellhead were located on a platform.
A cross-section of the seafloor completion is shown on Figure 12. As evident from this figure, although the
functional needs for the platform and subsea wellheads are similar, the methods for meeting these needs
are quite different.
Numerous marine wellhead stack-ups are possible and some of these arrangements are shown of Figure 13.
When ever the completion needs can be met with a single stack system, it is best to do so. Dual stack
arrangements require that two subsea BOP’s and marine drilling risers be used. This is not desirable from a
drilling standpoint.
An extended view of two stack wellhead assembly is shown on Figure 14. This information demonstrates
the interfaces between the temporary guidebase, permanent guide structure, low pressure wellhead housing,
high pressure wellhead housing, and the two casing hanger assemblies.
A Vetco SG-5 high pressure housing system for 13-3/8 inch/9-5/8 inch/7 inch casing strings is shown on
Figure 15. This arrangement reveals the intricate mechanisms which must be interfaced remotely during the
well drilling operations.
After setting the final casing string, the well is ready for running the tubing string(s). This is done with a
completion riser, as shown in Figure 16. The tubing and tubing hanger is run as shown on Figure 17,
landed, and locked into place.
In the Vetco System, the means for positively connecting the marine riser and tree to the wellhead is an H-4
connector as shown on Figure 18. The final arrangement can be seen on Figures 11 and 12.
Sometimes, regulatory requirements demand that a packer be used for production, injection, or salt water
disposal wells. When warranted by conditions or sound engineering principles, waivers can usually be
obtained.
Hookwall packers are the most frequently used packer types. As shown in Figures 19 and 20, it depends
on slips to engage the casing which develops the capability to resist forces imposed on the packer and
allows the pressure seal to be effected.
Hookwall packers can be set mechanically, hydraulically, or on an electric line. The setting procedure
causes the slips to move outward along an inclined plane to engage the casing. The inclined plane is
usually called a cone because of its circular configuration. Continued movement between the packer
mandrel and body compresses the resilient seal to effect the pressure seal. Seal compression is
maintained by weight applied from the tubing string, tension pulled in the tubing string, or opposing slips.
The hookwall packers are either the retainer type, as shown in Figure 19, or the mandrel type as shown in
Figure 20. The retainer type allows independently pulling the tubing. The tubing connects to and is an
integral part of the tubing string. Retainer type packers may either be drillable or retrievable. The drillable
type are called permanent packers and must be destroyed, destructively retrieved, or destructively released
and pushed to bottom. Mandrel type packers are designed to be retrievable. Release is accomplished by
manipulation of the packer to slide the inclined plane (cone) from beneath the slips.
Anchor Packers, as shown in Figure 21, require support from some other well component. When the
supporting member has been contacted, additional weight compresses and expands the seal into contact
with the casing. The additional weight must be maintained on anchor packers in production service to keep
an effective seal. Anchor packers are relatively inexpensive and are especially advantageous for low
pressure, multiple completion applications.
Cup Packers are run with the seals in contact with the casing. There are no supporting features other than
the tubulars on which they are run as shown in Figure 22. Differential pressures cause the seals to expand
and conform to the confining casing. Application is usually limited to shallow depths, low differential
pressures, and multiple completions where another packer serves as both a separation packer and a
support for the cup packer.
The inflatable packer has a seal which may be expanded with pumped liquids to contact the casing as
shown in Figure 23. A check valve system is used to confine the inflating liquids in the packer. The packer
can be released by tubing movement to release the confined liquids. Inflatable packers can provide a more
positive seal for open hole completions.
Mechanical set packers are actuated by manipulation of the drill pipe or work tubing, either rotation or
reciprocation. Drag blocks or springs, which ride the wall of the casing and allow rotational motion or
vertical travel between the packer components, provide means of transmitting a sequence of events. A
setting mechanism, which holds the packer in an unset position during placement at the proper depth, is
activated by rotation and/or reciprocation to allow setting the packer. Setting mechanisms are usually of the
jay-latch or split-nut assembly in hookwall packers as shown in Figure 24. The seal is effected by
continuation of the setting procedure until the desired results are obtained. In weight set hookwall packers,
continued “slack-off” will apply tubing load to the packer, thereby applying compression to the packer seal.
In tension set hookwall packers, upward movement of the tubing to a desired strain will compress the seal.
Hydraulic set packers are actuated and the seal compressed by liquid pressures. These offer advantage in
directional wells since no tubular movement is required (vertical or rotational); thus, the packer can be set
after the Xmas tree is installed. Also, these are useful for having “buttoned-up” condition while displacing
completion fluid in abnormally pressured wells. Hydraulic set hookwall packers have essentially the same
components as mechanical set hookwall packers (except drag flocks or springs). The setting mechanism
is a hydraulic cylinder which is actuated by plugging the tubulars below the packer and applying pressure
through the tubulars. Forces transmitted by the hydraulic cylinder cause the slips to engage the casing and
compress the seal.
Electrically set packers are run on conductor cables. Surface initiated ignition of a powder charge in a
subsurface setting device actuates the packer and compresses the seal. The final event in the sequence of
setting a wireline packer is release of the setting device, which is then removed from the well bore.
Electrically set packers are retainer type packers (opposed slip hookwall packers).
An anchor packer is mechanically set by applying tubing load after the supporting well member has been
contacted. The supporting member may be a stub liner placed in the hole or, more frequently, tailpipe run
below the packer and set on bottom. As weight is set on bottom, a collet moves and snaps over a retaining
ring (or a shear pin is sheared) releasing the packing element. Additional weight compresses the seal. A
modified or hybrid anchor packer may include slips or a hydraulic holddown.
3.4.1 Mandrels
A mandrel is the tubular member around which the packer is built. This member holds the various
components together and is a component of all packers. The mandrel is connected to and becomes an
integral part of the tubing string in all packers except the retainer type hookwall packer, and therefore, must
be as strong as the tubing.
3.4.2 Seals
Since the primary function of any packer is to provide a seal, this component is found on all packers. Seals
may be run into the well in contact with the casing, i.e., cup packers. Other seals are expanded by
compressive forces or inflated to effect a seal. Most of the sealing elements in the industry are nitrile rubber
products, the trade name Hycar being the most familiar. For wells having a temperature beyond the range of
nitrile rubber (350°F) or wells with an adverse fluid composition (alcohol, methanol, sour gas), other seal
elements must be employed.
3.4.3 Slips
Slips are packer components which engage the casing and provide packer support in all hookwall packers.
Slips for retrievable hookwall packers should be heat-treated to a Rockwell hardness grater than the casing
expected to be gripped and should have sufficient surface area hookwall packers must be frangible (break up
easily during milling operations to remove the packer); however, support must be provided until removal of
the packer is desired.
Many retrievable packers depend on hydraulic holddown slips, or buttons, for preventing packer release or
movement resulting from high pressure below the packer. These slips are actuated by pressure within the
tubing or, preferably, from below the seals. Hydraulic holddown slips may be built into the head of the
packer or attached as an accessory in the tubing string immediately above the packer.
3.4.4 By-Pass
A by-pass is an opening or channel which permits fluid movement through the packer during placing or
retrieving operations. An equally important function is that the by-pass serves to equalize the pressure
above and below when attempting to unseat the packer. Before pulling a packer with hydraulic holddown, a
by-pass must be opened to equalize pressure across the holddown slips or buttons.
Packers of one category may have components usually associated with another category; e.g., an anchor
packer on which an abnormal load will be applied may be equipped with slips to support a portion of the
applied load. Such hybrid packers play an important role in the industry.
Multiple zone completions may be made with a series of packers permitting production through a single
tubing string or multiple tubing strings with packer(s) having more than one opening. Packers for multiple
completions have secondary mandrels dependent on the number of well zones to be produced
simultaneously through separate tubes. Multiple packers may be designed and built with characteristic
features of single well packer categories. Multiple well packers are normally mandrel type packers and, as
such, are designed to be retrievable.
Packer accessories are those pieces of equipment which may be run in conjunction with or placed and
removed at selected times to extend the range of packer functions. Some common accessories are
extension tubes, on-off connectors, plugs and flapper valves.
Extension Tubes are four to six foot length, large internal diameter tubular run below drillable retainer type
packers as a part of the total packer assembly. It is of larger diameter than the packer bore, permitting a
spring-loaded catch sleeve on the packer milling tool to pass through the packer and open into the
extension tube. Once the upper slips of the packer are milled through, the packer can be pulled with the
catch sleeve. Special nipples may be run blow the extension tube for testing or for isolating the producing
zone while the well is being pulled. The extension tube, nipples, subs, etc., make up the tail section on a
packer.
On-Off Connector (tubing seal divider) is a device run above an opposed slip hookwall type packer which will
allow removal of the tubing for leak repairs or other problems without disturbing the packer setting.
Plugs are used in conjunction with retainer type packers and are either retrievable or expendable. Plugs
serve to isolate the zone below the packer and protect it from operations being conducted above the packer.
Retrievable packer plugs are installed in the sealing bore (or in the packer tail section) of a retainer type
packer after the packer has been positioned in the well. The retrievable plug may be recovered to allow re-
entry into the zone below the packer.
Expendable plugs are placed in the sealing bore of the packer (or in the packer tail section) and run at the
time the packer is positioned. Upon completion of desired operations, the plug is pushed out of the packer
and allowed to fall to the bottom of the well.
Flapper Valves are spring loaded closures mounted immediately below the sealing bore of a retainer type
packer. The valve is opened and displaced by insertion of the seal assembly run on the tubing string.
Removal of the seal assembly will allow the flapper to close and retain pressure from below the packer. One
of the important uses of a flapper is to isolate formation pressure below the packer while doing work above,
e.g., pulling tubing.
Other devices which should be considered accessories are available and my be attached to the packer for
special application or for localized well problems. Included in this category are a multitude of multiple
completion accessories. Consultation with manufacturer representatives can usually resolve modifications
useful for optimum application of such devices.
Metal Components of retrievable packers are generally AISI type 4140 steel or equivalent. Selection of this
grade material was made because the chemical composition is compatible with the usual well effluent, the
strength properties are adequate (similar to N-80), and the machinability is excellent. The metal
components of drillable packers are most frequently cast materials because of the frangibility requirement.
Heat Treatment is used to control the hardness or wear resistance of components subjected to rigorous
service. In addition, controlled hardness of some components may be desirable because of well effluent
properties. In a sour gas environment where the packer operating temperature exceeds 200°F, the heat
treatment of 4130-4140 steels should be such that the resultant minimum yield of the material should not
exceed 100,000 psi. If the packer operating temperature is below 150°F and the packer is exposed to a
sour service environment, absolute controlled hardness is essential. The above sour gas considerations
may be modified by BRC; therefore, an investigation should be made to review the latest technology. In all
cases, hardness of slips or hydraulic holddown buttons should exceed the hardness of the casing in which
the packer is to be set.
3.5.3 Seals
Seals are usually a nitrile rubber compound. Special seals may be required for elevated temperatures and
are required for sour service environment above 150°F. Seals for sour service above 150°F should consist of
Teflon (or special elastomers). However, at elevated temperatures Teflon (or special compounds) is subject
to deformation and/or flow and must be contained. The containment mechanism usually consists of steel
rings. The steel rings, of course, must be heat treated in accordance with other packer steel components
outlined above.
Seals are available in different durometer hardness specified according to ASTM D1706-6. Although there is
no unit of measurement, readings of 1 to 100 are assigned. Low durometer seals are easily deformed and
should be used in shallow wells where compressive forces may be difficult to achieve. High durometer seals
should be used in deeper, high pressure, high temperature wells where contained forces may tend to
extrude the seal over long periods of time. Some packer manufacturers use multiple seals with low
durometer to insure a seal and high durometer to minimize distortion under high compressive forces. Both
tubing stress conditions and tubing movement at the packer under given conditions should be considered.
Tubing movement data permits accurate selection of the amount of seals needed for retainer type hookwall
packers.
Plastic Coating is frequently used on packers to provide corrosion protection in water injection and SWD
wells. Corrosive fluids, other than H2S gas, may justify plastic coating protection. Generally, when the
tubing is coated for protection, consideration should be given to coating all packers and accessory
equipment.
A brief list of acceptable packer materials for sour gas serice follows:
API Grade C-75 material (API Specification 5AC, March 1973). C-75 to be Type I with less than 1.7
percent manganese and having a hardness of RC22 or less.
Carbon steels and low alloy steels having a hardness of RC22 or less. Low alloy steels should have a
nickel content less than 1 percent.
300 series stainless steels fully annealed and having a hardness of RC22 or less.
K-Monel hot rolled and age hardened having a hardness of RC35 or less. Cold drawn K-Monel is not
acceptable.
Inconel X-750 having a hardness of RC35 or less.
Stellites and colmonoys - cemented carbides. Base material having a hardness of RC22 or less.
ASTM A-286 having a hardness of RC35 or less.
The following materials are not acceptable for use in most sour service:
Stimulation Treatments can be Selectively Placed and casing Protected from Excessive Treating Pressures
by Using Packers.
The hostile environment in which this type of DHSV is used, combined with its inherent sizing difficulties has
reduced the application frequency for this type of valve.
The need for a more reliable DHSV led to development of surface operated DHSV’s. Both ball and clapper
type wireline retrievable DHSV’s are shown on Figure 26.
This type of DHSV has several advantages over the earlier storm chokes, but its reliability is still not
generally acceptable. Also, since the tubing bore is restricted through the DHSV, it can be susceptible to
sand or erosion cutting.
The tubing retrievable clapper and ball DHSV’s, as shown in Figure 27, have helped to overcome many of the
wireline retrievable device shortcomings. They provide full tubing opening, they can be controlled from the
surface, and they are highly reliable.
Both types of DHSV’s are vulnerable to failure. If the flapper or ball is opened with a large pressure
difference across it, it can be broken. Thus, pressure equalizing before opening is very important. Repair
requires pulling the tubing string.
Deep setting the single control line DHSV requires use of very strong spring systems. This is because the
hydrostatic head in the control line acts against the spring. Smaller springs and possibly better reliability
can be obtained by using a second line, or balance line, as shown on Figure 28. The balance line equalizes
the hydrostatic across the DHSV. This method requires more seal assemblies and two control lines to the
DHSV, and whether improved reliability truly occurs will be established with time. The single line DHSV
manufacturers have successfully tested their DHSV’s to settings several thousand feet deep.
TABLE OF CONTENTS
LIST OF FIGURES
FIGURE 1. Simple Schematics of Dual and Single Trees
FIGURE 2. Non-TFL Tree
FIGURE 3. TFL Tree
FIGURE 4. Diver Operated Manual Injection Tree
FIGURE 5. ROV with Tree Intervention Tool Package
FIGURE 6. Running with Guidelineless Tree with Workover Tool
FIGURE 7. Sub-Mudline Completion Systems
FIGURE 8. Wellhead Connectors
FIGURE 9. Vetco H-4 Connector
FIGURE 10. Mudline System with Wellhead Adapter and Casing Hanger Adapter and Tieback Adapter
Installed
FIGURE 11. Gate Valve and Actuator
FIGURE 12. Ball Valve
FIGURE 13. Wye Spool and Diverter
FIGURE 14. Diverters
FIGURE 15. Tree Cap Assembly with Guide Frame
FIGURE 16. Flowline Pull-in, Alignment, and Connection Tool System
FIGURE 17. Open Direct Hydraulic Control with Subsea Dump Valve
FIGURE 18. Discrete Piloted Hydraulic Control
FIGURE 19. Sequential Piloted Hydraulic Control
FIGURE 20. Direct Electrohydraulic Control
FIGURE 21. Electrohydraulic Multiplex Control with Subsea Hydraulic Power Unit
FIGURE 22. Electrohydraulic Multiplex Control
1.1 Function
Like a surface tree, a subsea tree is an assembly of components whose purpose is to contain reservoir
pressure and permit access to the reservoir for maintenance and measurement. The only difference is that
the subsea tree is located underwater.
Subsea trees are used to control gas and/or oil production from a well, or water or gas injection into a well.
The outward appearance is similar for production and injection wells, though the latter tend to be less
complex.
Subsea trees can control dual or single tubing strings. A dual completion tree will have three vertical bores,
one each for the tubing strings and on for vertical access to the annulus. Likewise, a single completion tree
has two vertical bores.
The subsea tree can be configured for only vertical downhole access with conventional wireline tools, or for
both vertical wireline access and Through Flow Line (TFL) access with pumpdown tools. The differences
between wireline and TFL methods will be discussed in a later section. Figure 1 shows simple schematics
of dual and single trees, and TFL and non-TFL trees. Figures 2 and 3 show a dual Non-TFL tree and a dual
TFL tree, respectively. Note the large (5 ft) bend radius flowloops and the wye spool required on the TFL
trees for passage of pumpdown tool strings.
Subsea trees typically come in just two rating categories of 5,000 and 10,000 psi. These represent the
maximum operating pressures. An off size rating is possible to design and build (e.g., 15,000 or 7,500 psi),
but is rarely done since it effects numerous components supplied by different vendors.
The subsea tree is rated by its lowest pressure rated component that is exposed to internal pressure. For
example, a tree with a 10,000 psi wellhead connector and 5,000 psi valves is rated at 5,000 psi.
In addition to pressure rating, a subsea tree is typically specified by the tubing size and the number of
vertical bores. For example a 3x3x2 is a tree for two 3 inch tubing strings and a 2 inch annular access
passage (annular access always being the last size given). Actually, the proper way to specify a subsea
tree is by the valve size. This is because there are more tubing sizes in different pressure ratings than there
are subsea tree valves.
Subsea trees are typically controlled and monitored remotely with the use of a controls system. Controls
systems will be presented in a later section. For specific applications, systems operated by divers can be
suitable. Figure 4 shows a subsea tree configured for diver operation of all tree valving. In this case, the
tree is for a water injection well where rapid actuator response is not needed.
A growing trend is the use of robotics for maintenance and certain operational functions. Specifically,
Remotely Operated Vehicles (ROV’s) are being outfitted with tool systems that can routinely interface with
appropriately equipped subsea trees. The interventions can be mechanical such as turning valve stems, or
hydraulic for direct actuator control (i.e., ‘hot stab’). Figure 5 shows an ROV with a tool package interfacing
with a tree equipped with docking ports for the ROV’s docking arm and hydraulic valves with mechanical
override by the ROV’s manipulator. Note that this tree also design includes a vertically run hydraulic crane
for retrieving/replacing other tree components such as chokes and control pods.
1.4 Installation
Two installation related designations for subsea trees are diver assisted versus diverless, and guideline
versus guidelineless. These designations refer to the installation of the tree on the wellhead and/or to the
connection of the flowline to the tree.
Saturation divers can typically operate to 900 ft water depth; divers have worked deeper for very special
applications but their work efficiency drops off quickly below that depth. Due to the high cost of a saturation
diver spread, and to prove it can be done, diverless subsea trees have been used in shallower depths than
900 ft. For both diver assist and diverless systems, tool systems are run down to the wellhead or tree on
guidelines, and are typically controlled remotely through an umbilical by a surface operator. The principle
purpose of the diver is for back-up if something goes wrong. Hence, the major difference between the two
installation approaches is the cost of assuring that remote systems operate correctly, and/or the cost for
providing non-diver back-up, typically with ROV’s equipped with special work tools.
In deepwater, wells are routinely drilled without use of guidelines to control the running of the BOP and
drilling riser. Subsea trees have also been installed without use of guidelines for landing the tree on the
wellhead or for tieing-in the flowlines. Figure 6 shows a guidelineless subsea tree. Note the use of docking
cones to located the tree connector with the wellhead.
1.5 Location
Typically a subsea tree is located on the seafloor. There are cases where the tree needs to be below the
mudline of the seafloor, for instance in iceberg prone areas and in heavily traveled shipping lanes where
anchors could be frequently dropped. For those situations, the tree could be placed in a caisson set below
the mudline. Figure 7 shows two applications. In the first case, the caisson is large enough to place a
more conventionally configured tree; the large caisson hole requires special drilling technology borrowed
from the mining industry. In the second case, the structural casing of the well serves as the caisson
requiring a radically different tree configuration.
Subsea trees can be used on an isolated satellite well, tied back to another facility with flowlines. Or the
tree can be used on a well drilled through a template structure with direct connection to a manifold also
located on the template. A hybrid of these two arrangements is a clustered well system where two or more
wells are drilled as satellites and tied with short flowline ‘jumpers’ to a manifold template located central to
the wells. Functionally the trees are exactly the same; however, components can be arranged differently to
accommodate installation and/or maintenance.
2.3 Valves
Functionally, there are three types of valves on the tree. The Master Valves are located between the
wellhead connector and the flow loop(s). They serve as the principle control devices for the reservoir. The
Swab Valve(s) are located above the flowloop junction and the top of the tree. They serve as a pressure
containing barrier between the well and the sea. The Wing Valve(s) are located on the flowloop between its
junction with the vertical bore and the flowline connector. They also serve as a pressure containing barrier
between the well and the sea when the flowline is not present or damaged. The wing valves are typically
used as the principle flow control device in order to preserve the master valves.
API 17D specifies that there must be at least two fail-closed pressure barriers along all production (injection)
and annulus access passages between the wellhead and any potentially open port on the tree. Along the
production (injection) flow path (i.e., from wellhead through flowloop), both pressure barriers must be
hydraulically operated valves. Along the vertical production (injection) bore (i.e., from wellhead to top of
tree), one pressure barrier must be a hydraulically operated valve. Along any annulus passage, one
pressure barrier must be a hydraulically operated valve. For a practical example, the simplest single
completion tree would have single hydraulically controlled fail-closed master valves on the production
(injection) and annulus access bores, manual swab valves (or wireline crown plugs) on both bores, a single
hydraulically controlled fail-closed wing valve on the production (injection) flowloop, and a manual valve on
the annulus access flow loop.
The two main types of valves used in subsea trees are gate and ball valves. Figure 11 shows an illustration
of a typical gate valve. Newer gate valve designs allow an ROV to remove and replace the actuator portion of
the valve. The ball valve, as shown in Figure 12, is generally not used as a master valve. It is most widely
used in applications where the valve would be opened or closed without flow in the line, and where clean
fluids are expected. The significant advantage of a ball valve is its compact nature, even with a hydraulic
operator. Each manufacturer’s valve has a slightly different design. Subsea valves used in trees should
meet API specifications.
Valve blocks are made of high strength alloy steels. Special materials such stainless steel are used where
carbon dioxide corrosion is a problem. The high strength alloys are heat-treated to accommodate hydrogen
sulfide problems. Cladding is also used for special corrosion and erosion situations.
2.7 Flowloops
Flowloops on the tree serve to connect the vertical bores of the tree to the typically horizontal flowlines.
Loops must be constructed with a minimum 5 ft radius for TFL applications. Flowloops typically must be
flexed to accommodate the flowline connection mechanism which can require as much as 14 inches.
The tool used to make up the horizontal connection can be the same regardless of how the flowline is
aligned. Flowline connections can also be made vertical wherein the flowline connection is typically made-
up at the same time as the wellhead connection. This complicates the initial running of the tree, but saves
on special tooling for the flowline connection.
Control systems require communication to select a control function and power to actuate it.
Communication is by hydraulic pressure or electric signal. Power to move a subsea actuator comes from
hydraulic fluid provided from a remote source through a high pressure hydraulic line, or from a local low
pressure reservoir powered by an accumulator. Subsea actuators are usually designed to “fail safe” by
automatically placing the well in a preferred control condition if communication, or hydraulic pressure is lost
(e.g., fail-closed valves).
A variety of control systems using different combinations of communication and power are in use. Selection
of the best system depends on the requirements of the specific application. The design criteria for a control
system typically includes the number of remotely controlled actuators, the number of monitored items, the
distance to the surface control station, regulatory requirements (i.e., response time to close a valve), and
environmental constraints (i.e., ability to vent control hydraulic fluids into the sea). Frequently, selection is a
balance of reliability, cost, and flexibility.
There are four basic types of control systems available: 1) direct hydraulic, 2) piloted hydraulic, 3)
sequenced hydraulic, and 4) multiplexed electrohydraulic. Two other types exist: multi-wire electrohydraulic
and hybrid (systems which combine features of the other types). These last two are infrequently used due
to the outdated technology associated with the first and the general absence of applications for the second.
If data acquisition of any type is required, a simple direct hydraulic system will not be suitable. Either a
more complex electrohydraulic system will be required or a hybrid system with a basic hydraulic system
coupled with a data acquisition package.
Because each subsea actuator in a direct hydraulic system has a dedicated flow path from the surface, the
number of functions which can be controlled is limited. This is due in part to space and weight associated
with the surface components required for each control function and in part due to interface umbilical size
limitations.
Direct systems are difficult to modify for future system growth unless major new topside components are
added or existing components are extensively modified. As a result, other system types are often superior
technical solutions where future expansion is a concern.
Straight hydraulic control is satisfactory until the distance between the surface station and well becomes
excessive (greater than 3 miles as a general rule). Subsea valve actuators designed to operate with 3000
psi hydraulic fluid may require as much as 2 gallons or more of fluid. To prevent erosion of the subsea valve
seats during opening or closing, the valve must move quickly. Actuation times of 18 sec or less are normal.
When the distance becomes long (due either to vertical or horizontal distances to the well), moving 2
gallons of hydraulic fluid in 18 seconds for both opening and closing requires an impracticality large and
consequently expensive hydraulic line. To address these and other concerns more advanced system types
were developed.
Piloted hydraulic control systems utilize a common hydraulic line between the surface and the seafloor well
to supply hydraulic fluid to all functions and one hydraulic signal line per control function for communication.
To operate a valve, pressure is applied to the signal line at the surface for the selected subsea valve. This
pressure activates the pilot valve which allows fluid to flow from the main power supply line or accumulator to
the desired subsea valve.
The pilot system still utilizes a multi-line hydraulic umbilical as in the direct hydraulic system. However,
hydraulic line sizes are reduced and response time is improved over direct only systems. The system is
still relatively simple and is applicable to wells where a few valves are to be controlled individually, no data
acquisition is required and distances are intermediate. The only equipment required in addition to that of a
direct system is a subsea control module to house pilot valves and a local accumulator (is used). Control
distances for piloted systems can extend up to 8 miles before response time begins to suffer appreciably.
Piloted systems suffer similar limitations to direct systems for future expansion and topside weight and
space concerns.
This system is applicable to wells located long distances from the surface station where multi-line hydraulic
lines would be expensive. The pilot valves on the seafloor are more complex because of the need for a
reference pressure (from the signal line or spring discussed above). Seafloor well control systems must
operate reliably without pilot valves and regulators. Thus, the number of sequence positions which can be
reliably used with a 3000 psi hydraulic system is about eight. A 100-200 psi range between each selected
pressure level allows for inaccuracy in the pilot valve setting and for “drift” after the unit is on the seafloor.
With only eight select positions, the number of individual production valves which can be controlled is
limited. A second liability is that once sequencing is designed no operational modifications are possible.
Direct, or hard-wired, E-H systems are applicable where minimizing expensive hydraulic control lines is
important and where the number of control functions is small. The major disadvantage is the need for
multiple electric connections between the cable and controls on the wells. In shallow water, it is common to
leave sufficient slack in the cable to allow the subsea controls to be brought to the surface without
disconnecting the cable to obtain maximum connector integrity by making all possible connections dry.
As with MUX E-H systems the topside support components for a hard wire system are lightweight and
compact making the systems ideal for cramped floating production applications.
A typical MUX system will have a dedicated control module for each major subsea system component (tree,
manifold, chemical injection system, etc.). Because each module in the system has its own address,
future expansions are relatively simple and straightforward. Dependent on its location in the overall system,
the module may be equipped with an accumulator to provide near instantaneous response of valve actuator
to a control signal. Figures 21 and 22 show the flow schematics of MUX E-H control systems with and
without subsea accumulator.
Use of a common signal wire for all controlled functions raises the possibility of an incorrect valve being
actuated. With the multiplexed electronic system, signals normally include “message security” coding to
minimize the chance of wrong commands being executed. These have historically been very successful
and no major operational problems have arisen due to incorrect commands.
As with direct wire systems, the presence of electrical power provides a convenient and economic means of
instrumenting the subsea system. This capability is becoming increasingly important as subsea systems
become more complex and regulatory bodies become more active.
A final aspect of the MUX E-H system of recent years is its associated compact topside support system.
This typically consists of hydraulic and electric power packs and a computer based master control station.
Interfaces to the subsea system consist of single or multiple junction boxes/or reels (depending on the
production facility type). The compact nature of all these components makes them ideal for restricted
space/weight applications which often arise in floating production application.
A second hybrid type has been the addition of a hard or multiplexed data acquisition system to one of the
hydraulic only systems. This can be an economic solution to the need for subsea data acquisition when the
expense of an E-H system cannot be justified and the parameters to be monitored are few.
PIPELINES
TABLE OF CONTENTS
LIST OF FIGURES
The term riser refers to primarily vertical tubulars that transport fluids from the seafloor to a surface facility.
Though functionally identical to pipelines, risers are typically designed to more stringent codes (e.g. larger
satiety factors) because of a higher potential of damage due to increased local activity, and because of their
proximity to humans. Also, short lengths of pipeline immediately adjacent to a surface facility are some
times designed to riser codes for the same reasons.
1.2 Applications
Pipelines have different designations depending on their application. An export pipeline transports a fluid
product from an offshore surface facility to an onshore location. A particularly large diameter export pipeline
that transports a like product from multiple offshore facilities is typically called a trunkline. Product lines
connecting offshore facilities are typically called intra or interfield pipelines depending on whether the
facilities are in the same field or not. Collectively, a system at such lines can be called a gathering system.
Product lines going to an offshore loading terminal are called offloading or offtaking pipelines depending on
whether the product is being exported or imported from the loading terminal. Similar names can be applied
to other special application pipelines (e.g. lines to a flare tower); the thing to remember is that they are
functionally the same.
1.3 Material
Most offshore pipelines are constructed of carbon steel. The steel is available in a range of grades (i.e.,
yield strengths), and in discrete sizes, (i.e. established combinations of outside and inside diameters)
information on line pipe can be found in API 5.L “Specification For Line Pipe.” Pipelines can also be flexible
pipe as shown in Figure 1. A flexible pipe is an armored hose that can withstand external hydrostatic
pressure. A spiral wound, interlocking, steel channel with a “Z” cross-section (called “Z” lock) provides the
resistance to hydrostatic collapse. Spiral wound flat wire provides axial tension capacity. The layers of
steel windings are typically separated by layers of extruded synthetic material. The outer layer is also
normally an extruded synthetic material, typically a nylon derivative . This outer synthetic layer is the
primary contributor to bending stiffness (i.e. the area moment of inertia is effectively confined to the outer
extruded layer). The combination of layers can be constructed to meet the users need for axial strength,
burst/collapse resistance, product chemistry, and even thermal conductivity.
The principle formula for calculating the external hydrostatic collapse pressure was developed by
Timoshenko, and presented in Figure 3. In its unaltered form, this formula chooses the lower of the external
pressures required to either 1) yield the pipe wall in compression (i.e., a compression hoop stress); or 2)
cause elastic instability in a circular cross section with an initial out-of-roundness (i.e., pipe ovality). A
manufacturing tolerance of 2% is typically used for initial out-of-roundness; this means that the maximum
difference of any two measurements of pipe outside diameter can be only 2% of the average outside
diameter. Various design codes and company design practices apply a range of safety factors to the
calculated Timoshenko pressure to account for all other uncertainties.
Once a pipeline has buckled due to hydrostatic pressure, the pipe will continue to collapse along the length
of the pipeline in both directions until the ambient external pressure is below a critical buckle propagation
pressure. This critical propagation pressure, as calculated in Figure 4, is lower than the pressure required
to cause initial hydrostatic collapse. Any buckle has the potential of causing a crack in the pipe wall and
flooding the pipeline. This is called a wet buckle, and could have disastrous consequences if it occurs
during installation when submerged pipe weight is critical.
Two design approaches are available to deal with buckle propagation. The pipe wall thickness can either be
designed to withstand buckle propagation along its length, or a series of buckle arresters can be
constructed into the pipeline that prevent the buckle for propagating further. Short, small diameter flowlines
usually increase wall thickness to resist propagation while long, large diameter pipelines usually use
arresters.
There are several types of arresters. The simplest is a thick section of pipe welded into the pipeline. Other
arresters are bolted on collars or a spiral wound bar. Figure 5 shows some of the typical arresters used.
After a WT is calculated for a specified internal and external operating pressure, an allowance is usually
added for corrosion and negative manufacturing tolerance. The DnV code specifies 2mm (.0787 in) for
corrosion while API recommends consideration of an allowance but does not give specific value. In the
U.S., a corrosion allowance of .0625 in is typically used. Both DnV and API recommend a 12.5% negative
manufacturer’s tolerance.
The most frequently used corrosion coatings are: Asphalt Enamel; Bitumen; Coal Tar Epoxy; and Film
Bonded Epoxy (FBE). The corrosion coat is applied directly to the pipe. FBE is the most resistant to
damage but also the most expensive due to surface preparation. Asphalt Enamel, or Coal Tar Epoxy are
more often used underneath a protective concrete coating.
Pipelines are typically laid empty and must remain stable on the seafloor after installation. Even if the line
is intended to carry a liquid, the empty pipeline must remain stable on-bottom in the expected worst
seasonal storm.
On bottom stability requires that a pipeline not lateral move under the action of the design wave and bottom
current. Figure 6 presents the modified version of the Morrison Equation used to calculate on-bottom
stability. The pipe can be acted upon by three hydrodynamic forces: a lateral drag (CD term) and inertial
force (CM term); and a vertical lift force (CL term). The lift force counteracts the submerged pipe weight.
The lateral resistance force is the product of the net vertical downward force and soil friction (µ term).
Many oil companies have derived their own values for CD, CL, CM, and µ from experimental programs.
Figure 6 presents the values recommended by DnV. Also shown are the typical values used by industry in
the Gulf of Mexico.
There are two ways to add weight to a pipeline: increase the wall thickness or apply a concrete weight
coating. If the weight need is slight, a lower grade material could be considered which would increase the
required wall thickness and hence increase submerged pipe weight. In general, increasing steel weight is
more expensive then adding a concrete coating.
The concrete can be either high density (190 lb./ft3) or low density (140 lb./ft3). The higher density material
is more expensive but has better impact resistance for damage protection (discussed later). Most oil
companies specify the higher density.
The concrete is reinforced with a steel wire mesh or a steel wire cage, the cage providing greater structural
integrity. The concrete is usually applied rotating the pipe with reinforcing wire around its axis while the
concrete is in blown into the reinforcing wire.
Thermal insulation is added as an outer coating when heat loss is critical. Some reasons for retarding heat
loss are: protection of perma-frost soils, improve viscosity of low API gravity crudes, or suppress hydrate
formation. Three common types of insulating materials are rubbers (e.g. neoprene), plastics (e.g.
polyethylene) and foams (e.g. polyurethane). Polyethylene has about half the thermal conductivity of
neoprene (i.e. twice as good an insulator); polyurethane foam has half again the thermal conductivity of
polyethylene. Polyurethane foam, cannot resist hydrostatic pressure and requires an external pressure
barrier. In shallow water, a polyethylene jacket can be used. Beyond around 100 ft. water depth, a steel
carrier pipe is typically used creating a pipe-in-a-pipe construction.
Figure 7 shows two pipeline cross sections. In Figure 7A, a simple concrete coat is applied over a coat for
epoxy (somatic) corrosion coat. Figure 7B combines several features for a shallow pipeline with a weight
coat over a polyurethane insulation layer encapsulated within a polyethylene jacket.
Manmade hazards are more difficult to guard against. The possibility of dropped objects increases around a
surface facility. Covering pipelines with mats or dumped rock around platforms protects the pipeline but
makes inspection/repair more difficult. Choosing a route into the platform that avoids going under the lifting
range of surface cranes can reduce the likelihood of dropped objects and eliminate the regulatory need for
additional coverings (e.g. come in under a flare boom).
Dragged anchors are a problem. There is no effective way of protecting a pipeline from the dragged anchor
of a construction or drilling vessel because of their weight and depth of penetration. The best way to
alleviate the problem is by carefully monitoring pipeline and anchor positions.
In shipping lanes, or areas prone to random anchoring by commercial vessels, pipelines must be buried (i.e.
trenched and backfilled) below the bottom 3-ft. or more depending on regulatory requirements. In the U.S.,
the 3-ft. depth requirement goes back to the depth land pipelines had to be buried to avoid farm plows.
Certain countries require rock dumping to protect against anchorage damage.
Fishing equipment also collides with pipelines. In many areas of the world, fishermen pull trawl nets along
the bottom using trawl boards to keep the nets open. Figure 8 depicts a typical scenario were a trawl net is
dragged over the pipeline. Pipelines less than 6-inch diameter are prone to snagging by trawl boards; trawl
boards tend to go over pipelines 8-inches in diameter and greater.
Burying the pipeline in fishing areas is usually required. Trenching a pipeline below the bottom, but not
backfilling it, actually creates a worse situation where the trawl board actually can drop onto the pipeline.
Coating the pipeline with high density concrete in sufficient thickness (1-inch or more) has been shown to
absorb the impact of even the heaviest trawl boards. Typically if an operator can convince the local fishing
community that their concrete coated pipeline will damage neither the fishermen’s nets nor their pipeline,
the appropriate regulatory bodies will give a variance on burying the line.
In order to safeguard against damaging platforms or subsea templates, many operators require a weak link,
or breakaway joint, be located near the pipeline tie-in point. Figure 9 shows a typical safety joint that acts
as a weak link that will separate under a predetermined axial tension while still maintaining design operating
requirements. The safety joint shown also has pressure activated check valves in order to isolate the
separation.
Typically these two effects happen concurrently as shown in Figure 10. The chart shown is for a 24-inch
buried pipeline coming into a vertical riser. The elongation shown was the product of the combined effects.
Figure 11 shows the expression for axial force created by the two effects.
Any constrained termination must be designed to resist the combined load. Even if the termination resists
movement, thermal expansion will probably cause some section of the pipeline to expand; the actual
amount of elongation becomes a function of the length of pipeline free to expand. It is prudent design to
consider any pipeline length laid straight on the seafloor to try to expand in-line, usually towards the hotter
end. This is why dog leg bends are laid into the line as it approaches a termination point. Also, expansion
spools are placed just outboard of a termination.
Because break-away soil friction is greater than sliding friction, the pipeline will not return to its original
position when cooled (i.e. hysteresis). This propensity for the pipeline to ratchet across the seafloor must
be taken into consideration when designing a pipeline tie-in.
The way to avoid iron oxidation is to provide a supply of available electrons. The electrons can be supplied
by an impressed current or from sacrificial anodes (i.e., cathodic protection). Pipelines and risers typically
use anodes made of zinc or aluminum alloys. Risers are sometimes protected by impressed currents.
The anodes are sized to provide a minimum current to cover an area expected to be exposed over the life of
the pipeline. Typical minimum values are a 5 ma current over 10% of the pipeline surface over a 20 year
period. The anodes must be exposed to the seawater (i.e. not covered by any coatings), and electrically
attached to the pipe.
A new pipe joint is positioned at the start of the firing line where the ends are prepared for welding. The joint
is aligned to the previous joint, usually with an internal alignment clamp. The initial circumferential weld is a
full penetration weld called the root pass. Additional passes made at subsequent stations along the firing
line fill in the beveled profile gap between the two pipe joints. The final pass is the cap weld, and may or
may not be ground flush with the pipe surface.
At a station near the end of the firing line, an X ray is taken of the weld in order to detect cracks. If cracks
of an unacceptable length are detected, the region of the weld with the crack must be ground out, new weld
material inserted, and the entire weld reexamined. Needless to say, in-line repairs are very expensive and
warrant the use of highly qualified welders using proven methods and equipment. There are society codes
for qualifying welders, materials and methods; some clients have their own standards.
The last station on the firing line is where the external corrosion coating is applied to the welded area. If
concrete coating is used, a quick set material is cast in forms positioned around the weld area in order to
provide a continuous outside surface.
Along the firing line are tensioners. These are hydraulically driven devices that grip the pipe with long rubber
tractor treads. The purpose of a tensioner is maintain a minimum tension on the pipeline. If the tension
goes below the threshold, the tensioner draws the pipeline back onto the barge until the threshold is
achieved. If the tension exceeds the threshold, the tensioner pays out the pipeline from the barge until the
tension decreases to the threshold. In order to prevent the tensioner from continuously drawing in and
paying out the pipeline, a distraction to the welders, a deadband is set between the upper and lower tension
thresholds.
The rate at which the pipeline can be constructed, the lay rate, is usually a function of how fast the
preceding steps can be done. A rate of 2 miles/day is considered fast, with 1 mile/day more typical. As a
way of speeding up the process, some lay vessels can handle 80 ft. double joints where two 40 ft. joints are
welded together on the barge or onshore. Theoretically, the lay rate is either determined by the time it takes
to fill the gap with weld material (i.e., number of weld stations and the rate of depositing weld material) or the
time it takes to perform the X ray. In deep water, the rate at which the lay vessel can advance, or move-up,
on its mooring lines could control the lay rate.
After the pipeline leaves the firing line, the curvature must be controlled as it descends by use of a stinger.
Figure 12 shows the three types of stinger. In very shallow water, the stinger can be essentially straight and
extend to near the bottom (Figure 12A).
As the water depth increases, the pipeline is allowed to assume a double reverse bend configuration as
shown in Figure 12B. In this position, the upper bend, or overbend, must be supported by a curved stinger.
An articulated stinger, as shown in Figure 12B, is a series of straight pontoon sections hinged together. By
matching pontoon buoyancy and tension, the articulated segments can be made to create the appropriate
overbend curvature. An alternative stinger design is a continuous curved stinger as shown in Figure 12C.
The curved stinger can be rigidly fixed to the lay vessel, or it can be hinged. Continuous curved stingers can
be constructed as wide truss frames to resist lateral deflection. Articulated stingers are more easily
adjusted to meet the design needs of water depth and pipe size.
The lower bend, or sagbend, begins with the point of bend reversal (i.e., the point of inflection) which is near
the point of departure from the stinger, and extends a little beyond the point of touchdown on the seafloor.
The sagbend curvature is controlled by the tension in the pipeline.
The early lay vessels were flat bottom barges; this is still the most common type. Figure 13 is ship shape
“ALLSEAS” that can operate with a traditional spread moored pattern, or by dynamic positioning (DP) with
thrusters, or a combination of the two. The ship shape has advantages of faster mobilization; the DP allows
it to maneuver in confined areas such as is likely near existing platforms.
Figure 14 shows one of the large semisubmersible lay vessel “SEMAC” that is sometimes referred to as a
third generation vessel. Note the wide truss stinger for rough weather operation. A lay vessel must follow a
predetermined path, regardless of weather. The semisubmersible design has a low response to the sea
from all directions, unlike a monohull vessel.
Lay vessels using the ‘S’ lay method face problems in laying pipe in deep water. The problems can be seen
in Figure 15. Shown is a plot of departure angle from the stinger versus laybarge tension for a 10 inch
pipeline in 2300 ft. water depth. The graph shows that the lay vessel must have at least 320,000lb. tension
to lay with a departure angle less than 45°. Nominal departure angles for conventional barges are 30°and
less. Only a few vessels have over 320,000lb. tension. Stingers can be lengthened to accommodate 45° to
60° departure angles which brings tension requirements back into the range of more vessels. Beyond 60°
departure angle, small tension variations can result in larger changes in departure angle. This makes the
setting of the deadband range on the tensioners very critical; too large a deadband and the pipe could
buckle over the tip of the stinger with a small amount of vessel surge movement. At the same time, the
mooring becomes softer in deep water causing longer surge amplitudes.
Pipelines can be laid by the ‘S’ lay method in deep water. However, the problems involved make it
achievable by only a very few vessels after significant modifications. For these reasons, other methods have
been sought that reduce the tensioning, stinger and mooring problems of the ‘S’ lay. One such method is
the ‘J’ lay.
Such a high firing line angle, or ramp angle as it is known, requires single station welding. In single station
welding, the pipe joint is aligned, welded from root pass to cap weld, X rayed and field wrapped all at the
same near vertical station.
A great deal of research has gone into finding a fast method for welding pipe at a single station in order to
have lay rates comparable to the ‘S’ lay methods. Laser and electron beam welding, friction welding, and
flash butt welding are a few of these methods. The only methods recognized to give a metallurgically
acceptable weld are still the same conventional methods used in ‘S’ lay methods.
One way of improving lay rate is to use triple joints, or 120 ft. joints. The problem of handling such a joint in
heavy seas is not trivial.
Several vessel types have been considered for use with a ‘J’ lay system. Figure 16 shows the concept of
using a ship shaped drilling rig. In order to maintain the sagbend, there must be an adequate pipe tension
at the touchdown point. This requires a lateral component in the pipe tension at the vessel. Either the
single station is maintained at an angle or a small vertical stinger is needed to impart the required later
component. The latter is assumed in Figure 16.
Figure 17 shows the same ‘S’ lay vessel from Figure 13 now configured for the ‘J’ lay. This design
incorporates a stalking system capable of handling triple joints. In the ‘J’ lay mode, the vessel would
operate exclusively on DP. The ramp is hinged so that the angle of departure can be adjusted during the
laying operation, which it must do in order to adjust for water depth. The vessel is shown laying backwards
so that the vessel can back into a surface facility and directly transfer the end of the pipeline.
rigid pipelines are bent beyond their elastic limit when coiled on the reels. When they are unreeled offshore,
they must be run through straightening rolls. Lay rates of 1 to 2 miles/hour are typical with reel methods.
In order to resist buckling, reeled pipes generally have a diameter to wall thickness ratio of less than 25.
The yield strength of the pipe material is generally 60,000 psi or less in order to allow plastic deformation
without excessive forces in the straightening rollers.
The reels can be horizontal or vertical, and built into the vessel or portable. The forces required to straighten
the pipe limits the size of pipe that can be installed by reel methods. The weight of the pipe generally limits
the length (i.e., the bearing limitation) though volume sometimes is a limit.
Horizontal reel vessels have the advantage of pipe storage due to the high load capacity of its lazy susan
type bearing. However, they must use a ‘S’ lay method for supporting the freespan since the pipe comes off
the reel horizontally. Vertical reel vessels have the advantage of ‘J’ lay since a high departure angle can be
used.
Concrete coated pipe can not be installed by reel methods. Likewise, anodes, insulation and other external
systems must be attached after the pipe has been straightened. Also remember that all unreeled pipe will
have some residual curvature; the straightening process is not perfect.
Figure 18 shows a portable reel mounted on the fantail of an ocean going tug. Portable reels are popular for
small size pipe (i.e., 4 inch diameter or less). They are used when laying multiple pipelines in a single
pass; for example installing a 4 inch line along side a larger sized line being constructed on the vessel.
Figure 19 shows the “APACHE” reel ship. It can lay up to 16 inch pipe with 280,000lb. tension capacity.
The high departure angle capability and DP stationkeeping make it suitably for deep water
Figure 20 shows the various tow-out techniques. Surface tows use simple floatation systems, offer the least
towing resistance, and require the least route survey; however, they are the most prone to weather and
surface collision. The near surface methods reduce these disadvantages at the expense of complicating the
floatation system and increasing the tow resistance. On bottom systems (not shown in Figure 20) eliminate
the floatation system but cause the highest tow resistance, and pose the greatest danger with on bottom
collisions that require the most extensive route survey. Near bottom methods reduce these disadvantages
at the expense of complicating the floatation system. Mid-depth tows offer the advantages of surface tows
with none of its disadvantages; however the method is limited in application by the catenary equation
relationship between pipeline length and weight, water depth, and
The ocean going tug is the principle offshore equipment used. The various towing techniques require
different horizontal, or bollard, pull loads from the tug. The surface and near surface techniques require the
least bollard pull per length of line. The mid-depth tow must have sufficient pull from the leading and trailing
tugs to keep the pipeline from sagging in the middle and dragging on bottom. The on/near tow methods
must overcome soil friction. Figure 21 shows the relationship between bollard pull and tug horsepower for
typical ocean going tugs. Towing speeds can range from 6 knots for surface tow down to 1 to 2 knots for on
bottom tows.
A principle attraction of tow methods is that multipipeline/umbilical systems can be fabricated and tested on
shore, and encased in a carrier pipe. Figure 22 shows the cross section of a possible bundle design. Use
of a carrier pipe has several advantages: hazard protection, buoyancy, and thermal insulation. The
disadvantage is the added expense.
Manifold assemblies have been built into towed system for tieing satellite trees or other pipelines. Again,
the advantage is land testing the system in conjunction with its pipeline/ umbilicals. Also, the cost of a
separate installation spread is avoided.
The most familiar first end pipeline termination is a shore crossing. Figure 23 shows the typical obstacles
faced in shore crossings. Shore line obstacles are typically avoided when possible. When not possible, a
tunnel is sometimes dug or drilled under the obstacles. The surf zone poses the problem of both seasonal
and long term erosion/deposition. Typically a trench lined on both sides with cofferdams is dug deep
enough to avoid the worse case erosion. Ice pack erosion and iceberg scour pose additional problems in
particular areas. Again the long term extent of the problem must be assessed so that the line can be
placed deep enough, either with trenches, or tunnels. Ice prone areas may also pose the additional problem
of permafrost in which case the line must be insulated as to avoid melting the permafrost and possibly
For long pipelines, one vessel is sometimes used to lay the offshore portion while a second vessel is used
to install the shore crossing. is such cases, the two lines are tied together with a mid-length connection is
shallow water. The steps in such a connection are shown in Figure 24. The lines are laid parallel to other
with sufficient excess line os that a single vessel can lift both lines to the surface along its side. A series of
small cranes called davits are typically used to the lift the lines, possibly with the aid of floatation on the
lines. After the two lines a cut and welded together, the vertical loop is laid over on its side by moving the
vessel to beam as the line is lowered.
A J-tube pull-in at a platform is a familiar fist end termination. An oversized riser pipe is installed in the
platform at the time of its fabrication. At the seafloor end, the riser has a right angle bend. At the time of
pipeline installation, a cable is run through the J-tube and out to the stationary lay vessel. As in a shore
crossing, winches draw the pipeline off the lay vessel, and into the J-tube. As shown in Figure 25, the
pipeline is plastically deformed as it passes through the right angle bend section in the J-tube on its way to
the surface. Special computer programs, supported by extensive test results, are used to predict the pipe
stresses and pull loads involved. As with reel installation methods, pipe used in J-tubes should have a
diameter to wall thickness ratio of 25 or less.
A J-tube pull-in can also be used as a second end termination. The pipeline is laid past the platform. A
pull-in cable is run through the J-tube to the bull nose at the end of the pipeline, and pulled towards the J-
tube sweeping the pipeline across the bottom. With sufficient excess pipe having been laid on bottom, the
pipeline is pulled through the J-tube to the surface.
A reverse J-tube method has also been used as a first end termination. In this method, the pipe is vertically
stalked and welded together on the platform, and lowered down through the J-tube. The weight of the pipe is
sufficient to force the pipe through the right angle bend shoe and the reverse bend section at the outboard
end of the tube near the seafloor. A recovery cable is run from the lay vessel A&R winch to the pipeline. As
new joints are stalked on the platform, the lay vessel takes in the recovery line, ultimately bringing the
pipeline onboard.
A first end termination to a seafloor structure (e.g., subsea tree) usually involves some type of lay-away
method. Figure 26 shows a typical lay-away sequence.
A cable from the first end pulling head is transferred to a vessel overhead the pipeline termination port. A
sheave is lowered down to the termination port with the pull cable. The pull cable is drawn in as the lay
vessel pays out the pipeline until it arrives at the termination port.
Divers remove the pulling head and make a flanged connection. In deep water, too much effective elasticity
in the pulling cable may require the pulling device to be located on the seafloor, rather than on the overhead
vessel. Also in deep water, other means besides divers must be used to make the connection. Various
suppliers have developed remote hydraulic pipeline connection tools such as the one shown in Figure 27.
These devices can make-up a mechanical connection with metal-to-metal seals once the pipeline ends have
been properly aligned and positioned.
First end terminations, whether to platforms or to subsea structures, have the advantage of positioning and
aligning the pipeline end before the remaining pipeline is installed on the seafloor. Second end terminations
have neither of these advantages since the pipeline exists on the seafloor making any adjustment to the
position or alignment of the pipeline end very difficult.
To understand second end connections, a clear distinction must be made between positioning, aligning, and
connecting. Portioning brings the pipeline end to the proper coordinates; aligning puts the pipeline in proper
register with the inboard piping; and connecting mechanically seals the pipeline end to the inboard piping.
Various methods may combine two of these steps but no method does all three simultaneously.
Figure 28 shows the sweep-in method, variations of which are known as the bend-around or bend-and-deflect
methods. In these methods the second end of the pipeline is lowered into a target area. The end is pulled
towards the pipeline port. In Figure 28 multiple lines are used to deflect the pipeline; this allows the end of
the pipeline to be both positioned and aligned at the same time. A variation of this approach uses a single
pull-in line between the pipeline end and the termination port; in that case, a special tool located at the
termination port must align the pipeline end after it has been positioned by the pull-in cable. Figure 29
shows one such alignment tool. Because of the moments generated in order to align the pipeline, these
tools are usually massive; the one shown weighs 55,000lb. in air. A tool similar to that shown in Figure 26
can then make the mechanical connection between the pipeline and the inboard piping.
The sweep-in technique is typically used with both ends of a near-bottom, on-bottom or mid-depth tow
method. In such a case, the entire towed pipeline is positioned in a target area prior to making the final
bend-around.
For near-surface and on-surface tow methods a draw-down technique is typically used. Figure 30 shows the
typical sequence of a draw-down technique. Note that the excess length is adjusted for with a bend in the
vertical plane due to buoys, rather than a bend in the lateral plane as in the sweep-in technique.
Figure 31 shows a typical jet sled in operation. High pressure pumps on the burial vessel drive water
through the nozzles located in the trailing sled. The jetting water fluidizes the seabed, effectively reducing
the bearing capacity of the soil beneath the pipeline allowing it to sink to its prescribed depth. The sled
uses the pipeline as a guide as the surface advances.
The fluidized soil eventually resettles, though much of the seabed material will be transported by the water
jet currents away from the pipeline. The jet sled can be used in consolidated soils in which the jets can cut
a steep walled trench as shown in Figure 31. In unconsolidated materials like sand, a jet sled system will
just blow a shallow trench requiring repeated passes to get the pipeline down to the prescribed depth.
Figure 32 shows a plow system in operation. The seafloor plow cuts a wide trench with stable slopes, even
for unconsolidated materials. The pipeline acts as the guide as the plow is pulled by a surface vessel. The
plow in turn guides the pipeline down into the trench as the plow is pulled ahead. The trench will backfill
naturally with soil movement on the seafloor, or it can be actively backfilled.
Mechanical trenchers have been used that cut a trench beneath a pipeline with a combination rotating
cutters and suction pumps to remove the spoils. All burial systems, particularly the mechanical systems
suffer when soil types change along the route, or boulders are encountered.
When pipelines are buried and covered, magnetometer type devices are used to detect and follow the
pipeline. Figure 33 shows the principle of the magnetometer where a the presence of a massive electrically
conducting object like a pipeline causes a measurable disturbance in the magnet field of the instrument.
The instrument can be towed near bottom, or made into an ROV workpackage.
Caliper pigs are mechanical devices that, as implied in their name, use caliper fingers, spring loaded to
remain in contact with the wall, to measure inside diameter and mechanically record it on a chart log.
Caliper pigs do not measure wall thickness, nor do can they detect pitting is the pipe wall.
“Smart” pigs are electronic devices that use magnetic fields to detect variations in the amount of metal in the
pipe wall caused by uniform corrosion and/or pitting. Figure 34 shows a typical “smart” pig. The pig has
electric power, magnetic transducers, odometer, electronic instrumentation, and fluid power cups, all of
which are packaged in a multi-segmented system. Use of “smart” pigs typically requires a minimum pipe
bend radius of five times the pipe diameter.
repair. In the first case a spool is used. In the second case, the two end fittings serve the role of the spool.
The system shown in Figure 35 requires the use of divers. In water depths beyond diver capability, special
tool systems have been developed to essentially replace the diver (e.g., ROV workpackages, submersible
systems, drill pipe running tools, etc.). The steps required for a repair, however, remain the same.
LIST OF FIGURES
FIGURE 1. Pressure-Temperature Diagram for a Pure Component
FIGURE 2. Pressure-Temperature Diagram for a Binary System
FIGURE 3. Examples of Retrograde Behavior
FIGURE 4. Phase Diagrams for Typical Reservoir Fluids
FIGURE 5. Flow Diagram for Two-Stage Separation
FIGURE 6. Viscosity as a Function of Temperature
FIGURE 7. Moody Friction Factor as a Function of Reynolds Number and Relative Roughness
FIGURE 8. Relative Roughness as a Function of Pipe Diameter and Material
FIGURE 9. Resistance of Valves and Fittings
FIGURE 10. Resistance of Valves and Fittings
FIGURE 11. Two-Phase Flow Patterns in Horizontal Flow
FIGURE 12. Two-Phase Flow Pattern Regions in Horizontal Flow
FIGURE 13. Conditions for Hydrate Formation of Natural Gas
FIGURE 14. Freezing Points of Aqueous Glycol Solutions
FIGURE 15. Viscosities of Aqueous Ethylene Glycol Solutions and Methanol
1.0 INTRODUCTION
The subject matter covered in this lecture is divided into three areas: hydrocarbon phase behavior, fluid flow
in piping, and operational considerations. These areas serve to describe the behavior of produced reservoir
fluids an dhow this behavior impacts the production system from the reservoir to the initial separation
facilities. A detailed discussion of well operations and production facilities will be covered in later lectures.
Knowledge of the vapor-liquid phase behavior of hydrocarbon reserves to be produced is essential for proper
design of the production system. Whether the fluid is all vapor, all liquid, or both can have a major impact
on reservoir performance and thus ultimate recovery, and the design of piping systems. Many times the
phase behavior can be controlled to the advantage of the operator, e.g., maintaining single phase production
to minimize pressure loss in long flowlines. The behavior and characteristics of the produced hydrocarbons
is always a critical factor in the design of separation, process, and pumping equipment.
Hydrocarbons are classified as open-chain and closed-chain compounds. They are subdivided further
according to the “hydrogen” saturation within a molecule. Molecules with single-carbon bonds are hydrogen
saturated; those with at least one double carbon-carbon bond are unsaturated.
Paraffins: These are saturated, straight-chain compounds. Paraffins comprise a larger fraction of most
petroleums than any of the other individual classes. (Example: n-butane)
Olefins: These are unsaturated, straight-chain hydrocarbons that contain one double carbon-carbon bond.
Similar hydrocarbons, but having two, three or more double bonds, are called diolefins or dienes, triolefins or
trienes, etc. Olefins occur only as trace compounds in crude oils but are common to refinery streams after
cracking operations.
Napthenes: These are saturated ring compounds and differ from paraffins by the closure of the carbon
chain. Napthenes constitute the second most abundantly occurring series of compounds in most crudes.
(Example: cyclohexane)
Aromatics: These are unsaturated ring compounds, and although fairly common, they occur less frequently
in crude oil than do paraffins and napthenes. (Example: benzene)
2.1.3 Isomers
Isomers are compounds having the same number of identical atoms in a molecule and therefore the same
molecular weight. However, these atoms are arranged in different molecular structures. Isomers are
common among the compounds of carbon but rare among the compounds of other elements. For example,
there are to isomers of butane C4H10.
Crude petroleum is a complex mixture of an almost infinite number of hydrocarbon molecules. Because
hydrocarbon mixtures are routinely separated by distillation and since the low boiling constituents are
usually paraffin hydrocarbons, these components can be precisely separated. Heavy components cannot
be separated completely and are usually removed in “cuts” over certain narrow boiling point ranges.
The four basic chemical classifications of crude oils are paraffinic, napthenic, aromatic and asphaltic. It
should be emphasized that crudes never contain entirely one type of hydrocarbon. The U.S. Bureau of
Mines has developed an elaborate method of classification of crude oils. The nine possible classifications
are based upon the API gravities of low boiling and high boiling fractions.
The following basic definitions are useful before discussing vapor-liquid phase behavior:
System - a body of matter that has finite boundaries and is isolated from its surroundings. A system
participates in equilibrium. An example: hydrocarbons in a reservoir.
Phase - a portion of a system which is (a) homogeneous in composition, (b) bounded by a physical surface,
and (c) mechanically separable from other phases that may be present. Examples: oil, gas, water, ice.
Properties - the characteristics of a system by which it can be evaluated quantitatively. Properties are
usually defined at a particular time. There are two types of properties - extensive and intensive. Extensive
properties depend on the extent of the system or amount of material involved; an example is volume.
Intensive properties are independent of the extent or size of the system; examples are temperature and
pressure.
Homogeneous System - one whose intensive properties vary continuously and uniformly from point to point
throughout the system and thereby constitute a single phase. Examples: helium in a balloon; reservoir
fluids above the saturation pressure (single phase).
Heterogeneous System - comprised of two or more homogeneous systems (two or more phases) and its
intensive properties change abruptly at surfaces of contact between the phases. Examples: reservoir below
its saturation pressure (vapor and liquid phases); barrel half full or water (air and water phases).
Equilibrium - the state of a system that exists when its intensive properties remain unchanged with time
while under a given and constant environment. Example: if at a given temperature the vapor and liquid
phases in a cell are brought into intimate contact by agitation, the system is at equilibrium if the present
remains constant with time.
A typical pressure-temperature diagram for a single component system is shown in Figure 1. The vapor
pressure curve (DC) is the backbone of vapor-liquid phase behavior. The vapor pressure is defined as that
pressure which is exerted on the surroundings of a pure component in the vapor and liquid phases in
equilibrium at a given temperature.
Referring to Figure 1, “C” is the critical point at which the intensive properties of liquid and vapor become
identical (highest temperature and pressure at which two phases can coexist—for single component only).
“D” is the triple point at which solid, liquid and vapor phases are in mutual equilibrium. (This region can be
neglected when discussing petroleum reservoirs.) “DC” is the vapor pressure curve (vapor-liquid equilibrium).
“AD” is the sublimation curve (vapor-solid equilibrium). “BD” is the melting curve (liquid-solid equilibrium).
“NBP” is the normal boiling point—that temperature at which the vapor pressure of a component is 14.7
psia.
The “dew point” is temperature and pressure conditions at which an infinitesimal amount of liquid is in
equilibrium with vapor (first drop of liquid). The bubble point is temperature and pressure conditions at which
an infinitesimal amount of vapor is in equilibrium with liquid (first bubble of vapor). Continuity of phases is
demonstrated in Figure 1 and below by a-b-c-d-e-b’. Vapor and liquid phases are distinguishable only along
the vapor pressure curve (a single phase exists elsewhere).
a - vapor
Vapor-liquid equilibrium at a given temperature occurs across a pressure range for a multi-component
system. This is unlike that for a single component system where at a given temperature, vapor-liquid
equilibrium occurs at one pressure only.
The qualitative behavior of multi-component systems is more easily understood by examining a two-
component, or binary, system. A schematic P-T diagram for a binary system is presented on Figure 2. The
two-phase envelope for a 50-50 mole % mixture of the two components is shown. The two-phase envelope
differs for different mole % mixtures of the same components. This is demonstrated by the vapor pressure
curves for the two pure components and the locus of critical points.
Referring to Figure 2, CA and CB are critical points for components A and B, respectively; CM is the critical
point for mixture M. Bubble point (d), dew point (b), and volume percent liquid lines (e.g., c) for any mixture
converge to the critical point. Critical pressures for mixtures are often above the critical pressures for both
pure components.
Generally, the size of the two-phase P-T envelope for any binary system depends on the difference between
the molecular weights of the constituents. The larger the difference in molecular weights, the larger the size
of the two-phase envelope.
Normally when pressure on a system is increased isothermally, condensation occurs. Also, when
temperature on a system is increased isobarically, vaporization occurs. When the reverse happens—i.e.,
when vaporization occurs during an isothermal pressure increase, etc.—these are known as retrograde
phenomena.
Figure 3 illustrates typical retrograde behavior for a binary system. The dark areas are the areas of
retrograde behavior. Retrograde regions are often more intense for systems with more than two
components. CB is the cricondenbar, the highest pressure at which vapor-liquid phases can coexist. CT is
the cricondentherm, the highest temperature at which vapor-liquid phases can coexist.
In a binary system, the equilibrium composition of the liquid phase and the vapor phase at given pressures
and temperatures can be determined from a knowledge of the dew and bubble point conditions for various
mixtures. The quantitative composition of systems with more than two components is more complex, and
the above technique cannot be used to determine phase compositions.
The distribution of each component between vapor and liquid phases for given pressures and temperatures is
represented by the use of equilibrium vaporization ratios, of K-values. the K-value is the mole fraction of the
component in the vapor phase divided by the mole fraction of the component in the liquid phase. K-values
for a given component depend on temperature and pressure and on the overall system composition.
The second concept is that of material balance. This concept requires that the total mass of the system is
equal to the sum of the mass of liquid and of vapor phases. Material balance can also be applied for each
component in the system.
With equations (1) and (2), we can determine the amounts of liquid and vapor phases and their compositions
at any condition of temperature and pressure. This can be done for a system of fixed composition or for one
whose overall composition can be described from condition to condition. More specifically, the following
types of problems lend themselves to analysis by the flash equation.
This refers to the planned separation of oil and gas that reach the wellhead. Flash calculations will
determine or help to determine the gravity, the shrinkage, the gas-oil ratio for stocktank liquid, and the
specific gravity of the gas. When these results are plotted against separator pressure—or combinations of
separator pressure—the conditions for maximum liquid recovery (or some other maximum or minimum
feature) can be determined.
Flash calculations are used to describe reservoir performance where the composition of the system is
known. Most wells known in this category are gas condensate and volatile oil reservoirs. Flash calculations
in the compositional material balance methods will provide the composition of produced fluid and of fluid
remaining in the reservoir, the amount of production, and the effect of injected fluids on these quantities—all
as a function of pressure.
In addition to the major applications of flash equations for stage separator and reservoir analyses, the
following are examples of other problems that have been handled:
(a) The effect of gas-lift or a higher than solution gas-oil ratio on vapor and liquid recovery and
composition.
(b) The effect on vapor and liquid recovery of heating the oil.
(c) Deduction of fluid characteristics and hydrocarbon analyses where complete laboratory data are not
available.
The basic data needed for flash calculations are an analysis of the fluid to be handled and reliable K-values.
There are several methods for determining K-values. The best data are obtained experimentally for a given
system, but this is often difficult and expensive. Many systems have been studied and many of these
results have been correlated so that they can be used to determine K-values. These published K-value
methods are in wide-spread use today. K-values can also be calculated using certain natural laws, but
these techniques are limited to application below specific temperatures and pressures.
There are two kinds of processes for liberating gaseous hydrocarbons from the liquid: flash separation and
differential separation. In the flash separation, the gas and liquid remain in contact as pressure is reduced.
In the differential separation, gas is removed as soon as it is liberated from solution. Less oil shrinkage will
usually occur with the differential process.
Stage separation of oil and gas involves a series of separators operating at sequentially reduced pressures.
Liquid is discharged from a higher pressure separator into the next pressure separator. The purpose of
stage separation is to maximize recovery of liquid hydrocarbons and to provide stable liquid effluent.
However, maximum liquid recovery does not always constitute the optimum economic operation.
Conventional stage separators, widely used in the field, provide relatively low liquid recovery. However, they
are inexpensive to operate. A schematic flow diagram for two-stage separation is shown in Figure 5.
Economic factors dictate the number of stages used in a separation process. As the number of stages
increases, the process becomes more differential in nature. The heart of stage separator performance
calculations is the flash calculation. Remember, though, that maximum liquid recovery may not be the
optimum economic criterion.
Fluid properties can be determined by three methods. The method to be used in obtaining fluid property
data for any analysis will depend on how critical the fluid properties are to the specific analysis.
1. Laboratory examination of reservoir subsurface samples is the most accurate and reliable method.
In this process, the fluid properties are actually measured.
2. Generalized correlations are an inexpensive way to obtain fluid property data and in many cases are
the only source available.
3. Phase behavior computations based on composition analysis provide reliable fluid property data if
careful consideration is given to the limitations of the calculation procedure.
Reservoir fluid properties that relate reservoir volumes to surface volumes are:
1. Bo, oil formation volume factor, relates volume of oil at reservoir conditions to volume of oil
measured at stocktank conditions, res bbl/STB. Bo is always greater than 1.0 because of oil
shrinkage.
2. Bt, total formation volume factor, relates the barrels of oil (including dissolved gas) plus the gas that
was liberated between the initial saturation pressure and the pressure of interest to a barrel of
stocktank oil, res bbl/STB.
3. Rs, the dissolved gas-oil ratio, is the scf of gas dissolved at a given reservoir pressure and
temperature, scf/STB.
4. Rsi, the initial dissolved gas-oil ratio, is the volume of gas released when a given volume of
stocktank oil is produced by a particular set of gas-oil separators, res bbl/scf.
5. Bg, gas formation volume factor, relates reservoir volumes of gas to standard surface conditions, res
bbl/scf.
6. Bw, water formation volume factor, is near unity but varies with salinities, pressure, temperature,
and the amount of dissolved gases, res bbl/surface bbl.
7. Vrf/Vbp, the flash shrinkage factor, is the volume of stocktank oil recovered when a given volume of
oil at the initial saturation pressure is processed through a particular set of gas-oil separators,
STB/bubble-point bbl.
Reservoir fluid properties that describe volumetric and physical properties as they vary with pressure in the
reservoir are:
1. Vo/Vbp, oil relative volume factor, compares the oil phase volume of a sample at various pressures
to the volume the same sample would occupy at the initial saturation pressure, res bbl/bubble-point
bbl.
2. Vt/Vbp, total relative volume factor, compares the total hydrocarbon phase volume (oil and liberated
gas) at various pressures to the volume the same sample would occupy at the initial saturation
pressure, res bbl/bubble-point bbl.
3. RL, the liberated gas-oil ratio, shows how much gas evolves from solution at reservoir temperature
as pressure is reduced, scf/bubble-point bbl.
4. Co, oil compressibility, shows how much a volume of fluid at a pressure above the saturation
pressure will expand per unit of pressure reduction, vol/vol/psi.
5. Ho, oil viscosity, is needed to describe fluid movement and is usually determined experimentally
from PVT tests, cp.
6. Hg, gas viscosity, is generally based on correlations and is used in fluid flow calculations, cp.
7. Hw, water viscosity, varies primarily with temperature and is used in fluid flow calculations, cp.
v=µ p
2. Specific gravity of a liquid is the ratio of density of the fluid at 60 degrees F to the density
of water at 60 degrees F. The API gravity scale is related to specific gravity by the following
equation:
degrees API = 141.5 - 131.5 spec. gravity
The specific gravity of a gas is the ratio of the density of the gas to the density of air. It may be related to
the molecular weight by the following equation:
S.G. = M * 28.96
µ v
Prediction of fluid flow behavior is based on the following fluid flow principles:
In most pipeline problems, this equation reduces to pressure head being the sum of elevation head and
friction loss.
The head loss for liquid flow due to friction in circular pipes is described by the following equation:
hf = f L V2
D 2g
where L is the length of pipe, D is the diameter, V is the average velocity, g is the acceleration of gravity,
and f is the dimensionless. Moody resistance coefficient. (The resistance coefficient is sometimes
expressed in terms of the Fanning friction factor, which is one-fourth of the Moody coefficient.) In theory,
this equation applies only to laminar flow, but in practice it may be used in all flow regimes by using the
correct value of f.
The resistance coefficient f is a function of the Reynolds number NR, and the relative roughness of the pipe,
e.
D
For laminar flow, f is a function of NR: f = 64 (NR < 2000).
NR
For turbulent flow, f is a function of both the pipe roughness and the Reynolds number; however, at high
values of NR, f is a function only of e.
D
Figure 7 is the Moody Resistance Diagram, which shows the relationship of f, NR, and e.
D
Figure 8 shows the relative roughness for various sizes and types of pipe.
Expressed in oil industry units, the liquid head loss equation becomes:
A disadvantage of the above equations is that both the head loss and f are functions of D, prohibiting a direct
solution for D without resorting to trial and error. The Hazen-Williams equation overcomes this
disadvantage:
However, the Hazen-Williams discharge coefficient “C” must be carefully chosen to reflect both fluid
viscosity and pipe roughness.
In many pipeline problems, pressure drop in pipe fittings (bends, expansions, contractions, etc.) and valves
must be considered. It is often simple to treat valves and fittings in terms of their equivalent length of pipe, L
= KD/f. Figures 9 and 10 summarize the equivalent length for various commonly used valves and fittings.
Q = 33.7 To E (P12-P22) D5 ½
Po GZafLTf
where: Q = gas flow rate, ft3/day (at To and Po)
To = temperature base, oR
Po = pressure base, psia
E = efficiency factor
P1 and P2 = upstream and downstream pressure, psia
D = internal diameter, inches
G = specific gravity (air = 1.0)
Za = average compressibility
f = Fanning friction factor (.25 x Moody friction factor)
L = length (miles)
Tf = average flowing temperature, oR
The average compressibility should be calculated at the average pressure, Pa:
Pa = 2 P13-P23
3 P12-P22
A number of empirical gas flow equations have been developed. These equations are patterned after the
above general flow equation, but they express E and f in empirical terms based on the experience from
which they were derived. Described below are two of the more common gas flow equations:
1. Weymouth equation:
Q = 433 To (D2.6667) P12-P22 ½
Po GTfLZa
In this equation, f ~ 1
D1/3
Since the friction factor is a function only of diameter, this equation is applicable for fully turbulent flow in
commercially smooth pipe. In 80 field tests of lines 8” to 30” diameter and pressures between 15 and 1400
psia, this equation predicted an average of 6% less throughput than measured. Industry experience
indicates that Weymouth’s formula represents average capacities of clean commercially smooth steel
pipeline; however, the friction factor used by Weymouth is generally too low for large diameter lines.
This equation is based on tests in a 24” line. In this equation, 1/f~ (GQ/D) 0.1461. Since this equation
depends on Reynolds number, it is not based on fully turbulent flow. No compressibility is included
explicitly in the equation; it is accounted for in the empirical constants and exponents.
The recommended efficiency factor E = 0.92. In practice, the 92% efficiency has been found to be too low
for large diameter lines and too high for small pipe.
The above equations apply only to level lines. For lines with significant elevation changes, the general gas
flow equation must be rewritten:
Q = 33.77 To E (P12-esP22) D5 ½
Po GZafLTf
1. Flowing (or input) liquid fraction (?) is the ratio of liquid input volume to total input volume,
both volumes being measured at pipeline conditions.
(?) = 1
1- 0.005 ZT (GOR)
BoP
where: Z = super-compressibility factor
T = flowing temperature (degrees oR)
GOR = input gas-oil ratio (Scf/STB)
Bo = oil shrinkage factor (bbls/STB)
P = flowing pressure (psia)
2. Superficial velocity ( µ ) is the apparent velocity of fluid flow, ignoring excess liquid.
µ = 0.012 Q1Bo ?D2
where: µ = superficial velocity (ft/sec)
3. Liquid holdup (a) is the actual liquid volume fraction in the pipe. It is normally larger than
? due to slippage of higher velocity gas past lower velocity liquid.
4. Excess liquid (e) is the liquid fraction in the pipe in excess of the input liquid fraction.
1 - ?Code?
The nature of two-phase (gas and liquid) flow may be illustrated by examining the flow patterns which have
been observed to occur as shown in Figure 11.
(1) Stratified flow: Liquid flows at the bottom of the pipe and gas at the top, with a smooth interface.
(2) Wave flow: gas slippage over the liquid begins to create waves at the interface.
(3) Slug flow: liquid moves in periodic slugs; the liquid may occupy the full cross-sectional area, or the
tops of the slug may be frothy. (Refer to slug, plug, and bubble flow on Figure 11.)
(4) Uniform flow: (also spray or dispersed flow) the liquid is dispersed in the gas.
(5) Annulus flow: the liquid flows as a film around the perimeter of the pipe, with gas in the center.
The most commonly used flow regime map was developed by Baker and correlates the various flow regimes
with liquid and gas flow rates and densities, liquid viscosity, and surface tension (Figure 12). Many other
flow pattern maps are available, but in general they give similar results. Since liquid holdup and pressure
drop are not dependent on flow regime, the main uses of these observations are (1) achieving a better
understanding of two-phase flow, and (2) predicting the effect on downstream facilities (e.g., foaming or
slugging in separators).
Pressure drop in two-phase flow is due to energy transfers between the two phases at the interface, and the
reduced cross-section available for each phase, in addition to the normal pipe-fluid friction losses.
The basic pressure drop equations for two-phase flow are also based on momentum and continuity balance
principles. They do not have direct analytical solutions since the terms are an empirical function of pressure
and other variables. The solution must be obtained by a numerical approximation and so is almost always
solved by computer.
As with single phase flow, empirical relationships must be developed, however, two parameters must be
determined instead of one: the liquid holdup and the two-phase friction factor.
Many correlations have been published for estimating the pressure losses which will result when gas and
liquid flow simultaneously in horizontal or inclined pipes. Few of these are reliable beyond the ranges of
variables used to develop the correlations into a single hybrid model to take advantage of correlation
strengths and avoid their weaknesses.
Vohra, et al. evaluated friction factor and liquid holdup correlations for horizontal two-phase flow using
measured data. They concluded that the Dukler and Beggs and Brill friction factor correlations were the
most reliable and that the Eaton, et al. and Beggs and Brill liquid holdup correlations were the most reliable.
Experience has shown that these correlations in general will bracket the correct answer. Normally the
Dukler correlation will tend to slightly under predict pressure loss. In some ranges of flow variables, the
opposite is true.
Pipelines are never precisely horizontal and frequently follow the natural contours of the ground or seafloor.
Elevation changes must be considered in predictions of two-phase pressure drop. Uphill flow is normally in
the intermittent (slug flow) regime, and pressure drop due to the elevation change is a function of fluid
densities, liquid holdup, and the amount of elevation change. Downhill flow is normally stratified, resulting in
a zero pressure loss, as the liquid layer adjusts to the critical depth at which the static head balances the
friction loss. However, slug flow may also exist in downhill lines, and may result in pressure recovery (net
pressure gain).
The Dukler and Beggs and Brill correlations will probably also perform better than other correlations for the
hilly terrain pipeline. Since the Dukler correlation was developed for horizontal flow, it is normally combined
with the Flanigan correlation to account for pressure loss due to hills. This approach will frequently over
predict pressure loss due to hills since it ignores any pressure recovery on the downhill slopes of hills. The
Beggs and Brill correlation accounts for inclination effects through corrections to the horizontal liquid holdup.
Two-phase systems have advantages of economy of scale through centralized separation, pumping, and
compression facilities and through use of one large pipeline instead of two smaller lines. Single-phase
systems may have advantages where there are size limits in pipelines, where lines are very long, or where
the gas or oil sales point is located in the field.
(1) Most wells make their allowable at pressures which allow flow to shore in a two-phase pipeline.
(2) Two-phase systems allow the substantial investment in pumps, compressors, separators, and the
second pipeline required for single-phase flow to be deferred until the field has been defined and
developed.
(3) The high cost of installing and operating offshore and subsea facilities favors centralization of
facilities.
The trend today is toward two-phase gathering to a central offshore facility, with single-phase lines to shore.
Conditions which favor this trend are:
Use of two-phase lines presents unique operating problems which are of no concern in single phase line
operation. Liquid accumulations in valleys, increased slugging tendencies, changing composition of “held-
up” liquid, and pressure surging all occur in hilly terrain, two-phase lines.
If the liquids form long slugs, separation and storage facilities must be capable of handling the largest
possible slug. Slug flow is also frequently characterized by large fluctuations in pressure at the outlet end of
a pipe. Although the size of slugs and magnitude of pressure fluctuations cannot be predicted with any
degree of reliability, the presence of slug flow or other flow regimes can be predicted. Experience has
shown that a positive pipeline inclination angle (uphill flow) tends to promote slugging, whereas negative
angles promote stratification of the gas and liquid. Liquid holdup is greatest at low flow rates, and
decreases as flow rate increases. This results in a phenomena known as “excess liquid slug,” and explains
the upsets (e.g., separator carryover) which can occur when the flow rate in a two-phase line is suddenly
increased.
Hydrates resemble ordinary ice when formed without agitation, and resemble wet snow otherwise. The
hydrates are not chemical compounds but rather chemical combinations where one component (gas) fits
into a cage-like cavity formed by the other (water). Hydrates can form at temperatures above the freezing
point of water, and because they behave like solid solutions rather than chemical compounds, they readily
decompose when conditions are unfavorable for their existence.
There is a physical limit to the size of the cavities within the water lattice framework, limiting the hydrate-
forming gases to those molecules small enough to fit into the voids (the normal butane molecule is the
maximum size along with carbon dioxide and hydrogen sulfide). Each hydrocarbon exhibits a predictable
pressure and temperature relationship that is favorable for its formation into a hydrate.
When a mixture of hydrocarbons exists, pressure and temperature conditions for hydrate formation have
been estimated based on the specific gravity (see Figure 13). A method is shown in the NGPSA Data Book
for the prediction of hydrate formation for any hydrocarbon mixture by the use of vapor-solid equilibrium
constants.
High velocities
Pressure pulsations
Agitation
As mentioned above, hydrates exhibit predictable formation characteristics and have no strong chemical
bonds, so they readily decompose when conditions are unfavorable for their existence. Therefore, it is
possible to reliably prevent hydrates by controlling the variables which influence their formation. The
The temperature of the gas can be controlled by using line heaters or by hot water tracing of the
transmission lines. Lowering the operating pressure is not normally used because of wasted horsepower.
Drying the gas at the wellhead is done using solid desiccants of hygroscopic liquids. The most common
method for preventing hydrate formation in lines is chemical injection. Various chemical when added to and
dissolved in water can lower the formation temperature of hydrates.
Methyl alcohol, the glycols, various salt brines and ammonia have been used to inhibit hydrate formation.
The corrosiveness of salt brines and drystallization from solutions rule them out as effective inhibitors.
Ammonia is an effective hydrate inhibitor, but if carbon dioxide is present, it forms solid compounds, causing
lines to become plugged. The chemicals used most extensively as hydrate inhibitors are methyl alcohol
(methanol) and ethylene glycol.
Methanol is used as a pure component and has a freezing point less than -100 degrees F. Ethylene glycol
is used in a 70% to 80% aqueous solution because of its freezing point characteristics (see Figure 14). The
viscosity of methanol is much lower than aqueous ethylene glycol solutions, as shown in Figure 15.
Efficient inhibitor
Low cost in comparison to the glycols
Readily available
Easily handled in cold weather due to low viscosity
Remains fluid at very low temperatures (-100 degrees F and below)
Some of the disadvantages normally associated with using methanol are:
Cannot be recovered
Consumption can be high if liquid hydrocarbons are present
Production must stop immediately if injection shuts down
There are several different chemical variations within the glycol family that have been used as a hydrate
formation temperature depressant: ethylene, diethylene, triethylene, and tetraethylene glycols. Ethylene
glycol is most widely used based generally on the following advantages of ethylene glycol compared with
other glycols:
Least expensive
Least soluble in liquid hydrocarbons
Lowest viscosity at the low temperatures involved.
Lowest freezing point of its water solutions
Some of the general advantages of using ethylene glycol for hydrate inhibition are:
4.3 Paraffin
Flow restrictions in tubing and lines can also be caused by the formation of paraffin waxes from crude oils.
These waxes typically contain 25-25% oil, 45-55% paraffin, and 10-15% gums and resins. Often a large
proportion of the deposition is made up of water, fine sand or shale, and precipitated salts. Certain
conditions of temperature and pressure as well as the roughness condition of the pipe walls cause
deposition of the paraffin wax. The solubility of the wax is commonly decreased during production because
of the drop in pressure, heat loss to the surroundings, and loss of lighter fractions of the crude.
Several methods can be used to prevent or remove paraffin buildup. Some of the more common ones are
discussed below.
Production Methods - A typical production method employs back pressure to prevent loss of lighter fractions
from the crude oil. This loss generally accompanies its transport from the hole pressure to atmospheric
pressure. The choice of tubing and line size and flow rate has an effect on wax deposition.
Mechanical Methods - These include all those methods which involve using some sort of tool to scrape the
paraffin off after it has been deposited. A number of such tools have been designed, including the “paraffin
hook,” the “paraffin knife,” and the “flow devil.”
Heat Application Methods - Various hot petroleum liquids, as well as hot gas, steam, and hot water, have
been circulated to raise the temperature and either melt the deposited paraffin or increase the solubility.
Solvent Methods - These involve the use of chemicals in which paraffin is highly soluble to dissolve the
deposited wax and to prevent more wax from precipitating. The use of pour-point depressants is related to
this method, although it is generally believed that such depressants do not actually affect the solubility.
Sand difficulties in oil and gas wells usually occur in areas where the producing sane is loosely consolidated
and poorly cemented. In such areas, special completion methods and equipment must be applied to
preclude the entry of sand without harmful restriction of oil production. In some circumstances, these
measures must be applied upon initial completion of the well, and in others, the sand difficulties develop
later in the productive life of the well after reservoir pressure and temperature have declined and, in particular,
when water invasion of the reservoir has occurred.
The exclusion of unconsolidated sand from wellbores is extremely difficult in some cases. Coarse, clean,
well-sorted sand presents the most difficult situation. Other factors, such as type of fluid produced, fluid
viscosity, and rate of flow, also have an effect on the severity of the problem. No one method of sand
exclusion has proved universally successful. The most commonly used methods are slotted liner, wire-
wrapped-screen liner, gravel packing, and sand-consolidating plastics. Sand control by gravel packing has
proved to be the most effective method in the majority of fields with severe sand problems.
PRODUCTION FACILITIES
TABLE OF CONTENTS
LIST OF FIGURES
1.0 INTRODUCTION
The general flow schematic shown in Figure 1 depicts a feasible flow diagram for an offshore oil and gas
process facility. It should be noted that not two facilities are identical due to the differences in:
Other fields may require special facilities to deal with production problems such as hydrogen sulfide, carbon
dioxide, salt, sand, wax, paraffin, hydrates, etc.
Another way to avoid extensive offshore areas is to combine the production of various platforms or fields onto
a central facility before various functions are performed.
2.1 Manifold
The purpose of the manifold is to channel the flow from various wells into the proper piping on the structure.
The manifold, quite typically, is broken down into the major operating pressures of the field. Depending on
the individual wells encountered in the field, header banks for high, intermediate and low pressure are
provided. In addition, one or more test headers are provided. Even if at time of installation all wells are
producing at the same pressure, it is good practice to provide additional header banks, or at least blind
flanges for additional banks, should some wells begin to deteriorate.
Wells are choked upstream of the manifold; however, during a shut-down, it is likely that the manifold will
realize the wellhead shut-in pressure. Because of this, it is necessary to provide valves and piping capable
of this high pressure. The velocities in the manifold can quite often become very high. Often the chokes
feed directly into the header itself. For these reasons, it is good practice to fit tees and crosses with bull
plugs or lead targets rather than ells in order to avoid erosion. In addition, pipe wall thickness for erosion
and corrosion allowance, internal velocities, work access, etc., should all be considered for a good manifold
design.
For reference and guideline, to design a manifold one should refer to API RP 14 E.
2.2 Separators
The major vessel of a production facility is the separator. Separators can either be used to separate the
fluids from the gases (two-phase) or the water from the oil, from the gases (three-phase). The internals of
separators vary from one manufacturer to the next and from one production composition to another.
Separators can be either horizontal or vertical in design.
Horizontal separators, as used in the flow schematic, are used for high to medium gas/oil ratio streams,
large volumes of gas and/or liquids, foam problems and liquid/liquid separations. A cut-away of a horizontal
separator is shown in Figure 2. This design is typical in that the initial momentum of the flow stream is
absorbed by a deflection baffle as the stream enters the vessel. This baffle directs the flow and reduces the
stream’s velocity while the predominantly gas phase rises to the fluid surface. In this design, the fluids pass
through multiple plates of a corrugated matrix. The theory is that in each plate section there are many
shallow inclined channels through which the oil/water mixture passes. Oil droplets then rise to the
undersurfaces of the plates. These small oil droplets coalesce to form progressively larger drops under each
plate. As these large drops leave the plate section, they rise rapidly to the fluid surface, while the water
sinks to the bottom.
Once the fluids reach the large holding area, some amount of retention time is available for more fluid
separation.
Gas rises and exits out of the gas outlet. Not shown in the figure, but quite common, is to have a fine wire
mesh “mist extractor” at the gas outlet that helps to eliminate fluid particles leaving with gas. As shown in
figure 1, a back pressure valve is located downstream of the gas outlet which controls the operating
pressure of the vessel.
A weir located well above the oil/water interface allows the oil to flow into a small chamber. In the chamber
some type of float is present to indicate when dumping is necessary.
Located at the oil/water interface is a displacement float that signals the required dumping of the water.
Another form of separator, not shown in the Figure 1 flow schematic, is a vertical separator, as shown in
Figure 3. A vertical separator is generally used for intermediate gas/oil wellstreams. They are more readily
cleaned if sand, paraffin or wax are present. In addition, vertical vessels generally take less deck space and
are therefore useful in limited areas.
Vertical separators work as follows: The mixture of oil and gas enters the inlet where it is given a swirling
motion by a spiral inlet baffle in the separator chamber. At this point there are two forces tending to
separate the oil from the gas. The first is gravity; the second is the centrifugal force caused by the whirling
action of the wet gas. This causes the heavy oil particles to collect on walls of the separator, coalesce and
fall to the fluid.
The gas, which still contains some oil in form of small drops and spray, rises through the chamber. As the
gas enters the swirl cylinder ©, it moves faster and is again caused to whirl so that the oil is forced against
the side of the deflector cone (e). This oil drains down through tubes to the fluid. The gas then passes
through a mist extractor and out the top outlet.
As shown in the figure, this vessel only separates the gas from the fluids (two-phase). However, it is
possible for a vertical separator to be three-phase by adding some form of weir in the fluid section and
providing two dump valves.
The major goal for a separator is to allow the different phases of liquids and gases to separate as thoroughly
as possible. The simplest way to achieve this is to allow the substance to stand still as long as possible.
However, in order to achieve such a long retention time requires a large volume of fluid to be contained, thus
driving up costs and space. For this reason, the various vanes, extractors, plates, etc., have been devised
to keep efficiency up but allow retention time to decrease.
As can be noted in Figure 1, the oil leaving the High Pressure Separators enters the Intermediate Pressure
Separator. This vessel, operating at a lower pressure, allows more gas to break out of solution and go into
the atmosphere. This allows the oil to stabilize further. From here the oil continues to the Low Pressure
Separator for even more stabilization at still a lower pressure. By cascading to lower pressures, the
process train is said to be 2 or 3 stages of separation. Some plants may actually have a first and second
stage of a High, Intermediate, or Low Pressure system before the fluids actually intermingle. This is all
determined by the volumes, properties and preferences of the field.
Normally, all the vessels in a field have identical internals and are all rated to operate at the highest
pressure. By having all high-pressure vessels, a great deal of flexibility can be achieved in the process
throughout the plant life. In addition, with bypass piping (normally installed but not shown in Figure 1 for
clarity), maintenance on any one vessel can be achieved without disrupting production.
In addition to the 3 stages of separation shown, there is also a test separator. The test separator is used to
give a regular test to each well to monitor wells’ performances. At the outlet of each phase there is a flow
meter that allows the operator to record the well’s production of water, oil and gas.
This film must be weakened or broken before further coalescence of the water can take place. Chemical
aides, demulsifiers and heat can be used to break the emulsion. In addition, by increasing oil temperatures,
oil density is decreased for more rapid gravitational settling of water droplets. Therefore, the oil flows
through a treater unit as shown in Figure 4.
The treater unit operates as follows: The oil, with the emulsified water, enters over the heating unit. Gas
evolved during the heating is removed through a vent pipe and the emulsion is broken up. The gas-free oil
then flows through a surge section and enters into the electrical coalescing section. Coalescing of the
small water drops dispersed in the oil is accomplished by the high-voltage alternating electrical field. As the
emulsion rises through the field, the water droplets are given an electrical charge. When charged, they
rapidly move about - repelling, attracting and colliding with one another. The droplets collide with sufficient
energy to overcome the now weakened emulsifying forces, and combine into larger and larger drops. This
growth in mass allows gravity settling of the larger drops into the vessel’s water phase. The lighter, clean oil
continues to rise to the top of the vessel where it is collected and removed.
The electrostatic treater, or often called a “Heater-Treater”, is often not used on an offshore platform.
Instead, any water remaining with the oil after separation is pumped ashore with the oil. Once on land, a
“Heater-Treater” is used to remove any of the remaining water. This eliminates the rather large, fixed vessel
that requires a good deal of electrical power on the offshore platform.
Oil from the treater continues to a surge vessel and then on to the production pumps through the pig
launcher and into the pipeline.
Facilities are often available, as shown in Figure 1, to recycle the oil should an upset occur in the process
train. The oil will then flow to a holding tank or “bad oil tank” and, once the upset is corrected, back into the
treater.
Wet gas enters a contractor tower at the bottom . High-purity glycol flows down from tray to tray, or through
some sort of low density packing. The glycol absorbs the water from the rising gas.
The now dehydrated gas exits the tower at the top and continues on to the compression package. The
water-rich glycol then enters a series of heat exchangers, is filtered and then finally enters as steam and the
glycol returns to the tower for additional gas dehydration.
trapped at the surface by skimmers. Offshore, various forms of scrubbers can be used. One in particular
froths the water into a foam. This foam carries with it minute particles of oil to the surface where they are
skimmed.
From here, on an offshore structure the water can go to a sump pile. This is a large tubular member, closed
at the top and open at the bottom, that extends well into the seawater. As the water is dumped into the
sump, any oil remains on the water surface and is trapped by the tube. The clean water is free to pass out
the bottom. A submerged pump is used to direct the oily water back into the system.
2.7 Desalting
Often crude oil is hindered with a high degree of salt contamination. This is typical of Africa, the Middle
East, South America, and Southeast Asia. The salt must be removed before the crude can readily be
handled by tankers, pipelines or storage tanks.
A Desalting Plant, as shown in Figure 7, is used to rid the oil of salt. In this operation, the first step is
typically water removal. This can be achieved with an electrostatic treater as previously discussed.
The second step in the process is desalting. Fresh water is injected to dilute the salty oil; this is then
followed by a second dehydration phase to reduce salt and water content.
The dilution water can be “fresh” or can simply be less salty than the brine remaining in the crude to be
processed. In offshore installations or other areas where fresh water is in limited quantities, seawater can
be efficiently used for dilution.
Dilution water is pumped into the crude stream prior to a controlled pressure reduction to create agitation
and dispersion of the water. Automatic differential pressure controllers, and motor-operated mixing control
valves with trim design to create violent agitation of the liquids through the valves, are used to achieve this
dispersion.
WELL OPERATIONS
TABLE OF CONTENTS
LIST OF FIGURES
FIGURE 1. Well Inflow Performance Relationships (IPR’s)
FIGURE 2. Types of Tubing Flow
FIGURE 3. Tubing Flow Regimes
FIGURE 4. Two Phase Flow Pressure Drop
FIGURE 5. Vertical Two-Phase Flow for 2.875 In. Tubing Set at 8,000 Feet (0 psig Tubing Head
Pressure)
FIGURE 6. Example of Graphical Method for Determining Tubing Pressure Traverse
FIGURE 7. Example of Selecting “Optimum” Tubing Size
FIGURE 8. Best Tubing Size
FIGURE 9. Gas Lift Methods
FIGURE 10. Continuous Gas Lift Principle
FIGURE 11. Gas Lift Valves
FIGURE 12. Gas Lift Valves
FIGURE 13. Gas Lift Valve Setting Depths
FIGURE 14. Shut-in Flowing Well with Dead Oil in the Tubing String
FIGURE 15. Wellhead Pressure vs. Time While Unloading Flowing Well
FIGURE 16. Weighted Fluid Solutions
FIGURE 17. Weighted Fluid Balancing Formation Pressure
FIGURE 18. Typical Well test Separator
FIGURE 19. Wireline Assembly
FIGURE 20. Mechanical Flowsheet for TFL Pumpdown Operations
FIGURE 21. Steady-State Friction Pressure Drop vs. Flow Rate (Measured Data)
FIGURE 22. Tool String Monitoring
FIGURE 23. Tool String Force Developed by Differential Pressure
FIGURE 24. UWC No. 6 TFL System Pressure Falloff Data Field Measurements
FIGURE 25. UWC No. 6 TFL System Calculated Pressure Data
FIGURE 26. Combined TFL Tools String Pulling Force and Impact Force
FIGURE 27. Shear Energy Required for ¼ Inch Steel Shear Pin/Shear Plan (Tensile Test Data)
FIGURE 28. Analysis of TFL Tool String with an Accelerator
FIGURE 29. Kinetic Energy Developed by a Tool String with an Accelerator
FIGURE 30. UWC No. 6 SSCSV Flow Test
LIST OF TABLES
TABLE 1. Summary of Terms
TABLE 2. Critical Multiphase Flow through Wellhead Chokes
TABLE 3. Bean Performance Formula for Oil-Gas Mixtures
TABLE 4. Bean Performance Formula for Gas Flow
TABLE 5. Relationship to Calculate Fluid Density Required at 600F to Provide Necessary Density at
Downhole Temperature
1.0 INTRODUCTION
This discussion provides basic information for producing well operations. The focus is on oil well operations
because they are of primary interest for subsea or floating production systems.
Methods of providing artificial lift are briefly reviewed. Flowing well characteristics and theoretical
considerations are discussed, followed by determination of “optimum” tubing size. Gas lift principles, well
start-up, monitoring methods, and killing operations are discussed. Also, the types of well workovers and
well servicing methods are discussed.
Producing rates plotted against pressure at the bottom of the well provides very useful relationship. This
information is called the well’s Inflow Performance Relationship (IPR). As shown on Figure 1, a generalized
curve type relationship can be determined. The curve is due to the presence of more than one phase
flowing. Also, a straight line relationship can be developed based on Darcy’s law for the steady-state radial
flow of a single incompressible fluid. Clearly, the more general curved relationship will exist for most flowing
wells of interest.
The IPR for each well can be expected to change with time as reservoir depletion occurs. This can be due
to decline of reservoir pressure, water encroachment into the reservoir, and several other possible factors.
Thus, well design should consider both near-term and long-term conditions.
Based on this very cursory review of well IPR’s, two important observations can be made. First, lower
surface pressure allows higher fluid flow rates. Second, careful consideration of the total system, including
wellbore completion, tubing selection, surface equipment selection, and the possible need for artificial lift
can result in maximum producing rates.
The type of flow which exists in the tubing string determines the well bottom hole pressure. As shown on
Figure 2, several types of flow can exist in a tubing string. Several major variables have been combined as
shown on Figure 3 to allow numerically predicting the type of flow regime that may exist. It is apparent that
several types of flow regimes are possible as production flows up the tubing. As demonstrated on Figure 4,
the total pressure gradient consists of two components, head loss due to the density of gas and liquid, and
friction loss due to the flowing gas and liquid. As shown on Figure 4, an important difference exists between
the head-loss controlling and the friction loss controlling portions of the total curve. When head-loss
controlled conditions exist, flow can be unstable. A decrease in velocity (due to downstream shut-in, choke
size reduction, etc.), can cause an increase in the pressure gradient. This can lead to the well “loading up”
and “dying” if it is a weak producer. If it is a strong producer (high shut-in wellhead pressure), this condition
is not of as much concern. Unfortunately, many wells tend to be weak producers.
Typical flowing well performance can be summarized as shown on Figure 5. One dimensional data shown
on Figures A and B represent useful information and are combined in three dimensions in Figure C to give a
good physical representation of vertical two phase flow conditions. In summary, for any constant gas liquid-
ratio, there is a rate of flow which requires minimum intake pressure. Also, for any constant rate of flow,
there is a gas-liquid ratio which provides minimum intake pressure. From this information, it is observed that
column pressure is greatest at low gas-liquid ratios.
Also, for any gas-liquid ratio and depth, there is a rate which requires minimum lifting pressure. This is
because of slippage, and higher rates requiring more lifting pressure because of resistance.
In general, flowing wells with a natural GOR of 1000-1500 or more are not conducive to lowering the bottom
hole pressure (and thereby increasing the fluid rate). Lower GOR wells are excellent candidates for gas
lifting.
__ __ ___
p gc gc
Energy (Friction)
For closed conduit flow, the external work team can be omitted. Also, for normal flow velocities, kinetic
energy is negligible. Irreversible loss can be described by equation (2):
The dimension factor f is a function of Reynolds number and tubing relative roughness.
Based on the law of continuity for a compressible fluid flowing through tubing of constant diameter, equation
(3) becomes:
pV + G = constant
so that G, the mass velocity, is an invariant for the flow process. Putting expressions 2 and 3 in Equation 1,
simplifying as indicated, and rearranging gives
dp = g p sin o + fG2
dL gc 2gcpD
Equation 4 states that the pressure gradient in single-phase flow may be considered to arise from the
combination of separate static and dynamic effects, i.e., fluid density and wall friction components. This
relationship can be generalized for two-phase flow. The volume fraction of gas at local conditions within a
two-phase mixture is defined as e. The flowing density of the mixture is then defined by equation (5) as
follows:
p mixture = e pg + (1 - e) p1
The second term on the right is assumed to account for all internal irreversibilities, particularly those related
to wall friction and relative motion of the gas and liquid. Integration requires separate correlations for e and f,
together with equations of state and property relationships for the gas and liquid phases. Several variations
and modifications have been made to this basic relationship to improve upon the method to obtain better
correlations between predicted and measured values. Fortunately, it is not necessary to manually calculate
tubing pressure gradients. Acceptable results can be obtained using graphical techniques (provided the
graph covers the required conditions) or computer methods, some of which give accurate results provided
that predictions do not exceed the bounds upon which the correlations are based.
Graphical methods generally are presented as shown on Figure 6, with a depth scale on the ordinate and a
pressure scale on the abscissa. Pressure traverses are provided for a given tubing size and liquid rate at
various gas/liquid ratios. Knowing the well depth, wellhead (or bottom hole) pressure, and the GLR, the
pressure traverse for the tubing string can be determined. For the example on Figure 6, given a bottom hole
pressure of 2040 psia, a flowing wellhead pressure of 410 psia would be expected if the well depth is 600
feet.
If computer Timeshare is available, the tubing pressure gradients can be readily evaluated for many
conditions. Appropriate examples and instructions are provided in the WELLFLOW manual, which can be
easily referred to.
One frequently used method to determine the best tubing size is to prepare a plot as shown on Figure 7. In
this example, the goal is to select a tubing size which will allow maximum well fluid production with a
minimum wellhead pressure or 170 psig. In this example, the optimum tubing size could be either the 1.9
inch or 2-3/8-inch tubing. Tubing cost, availability, and compatibility with other well equipment
considerations would determine which size was best.
Position 1 represents stable flow conditions and position 2 represents unstable flow conditions. As shown
on 8B, position 1, if for some reason the choke pressure increases, or decreases, the well inflow conditions
will tend to be returned to the original intersection position, the criteria for stability. However, as shown on
Figure 8C, if the rate decreases slightly for some reason, the net effect is to impose a higher back pressure
on the well. This decreases the well rate and can lead to the well “dying”. Thus, flow conditions are
unstable.
A summary of methods for predicting flow through positive chokes is provided on Tables 2-4. A widely
accepted method for calculating noncritical multiphase flow through wellhead chokes is not available.
The most common method for gas lifting is to install the valves in the tubing such that gas from the annulus
can enter the fluid in the tubing string. Occasionally the reverse (gas down tubing with fluid up annulus) is
applied for exceptionally high volume wells. It is intended that only one gas lift valve at a time be opened.
For a given tubing size, the gas-liquid ratio that will give the lowest tubing intake pressure becomes
increasingly less as the producing liquid rate increases. The gas-liquid ratio to achieve minimum pressure
becomes very high for low rates of production. When this occurs, intermittent gas lift may be more efficient.
Differential pressure valves operate based on the pressure differential between the tubing and casing
pressures. They are normally open, with the open force provided by a spring. Casing pressure tends to
close the valve and tubing pressure tends to open the valve. When the differential reaches a predetermined
amount, the valves will open.
As shown in Figure 13, an iterative process is used to arrive at the setting depths of the gas lift valves. At
start-up conditions, the upper-most valve is closed, and deeper set valves are open due to fluid gradient in
the annulus. As gas pressure in the annulus is increased, the gas enters at the uppermost valve and
decreases the tubing gradient above that point. Annulus fluid is lowered and the gas begins to U-tube
around the next lower gas lift valve. As this occurs, the upper gas lift valve closes due to the decreased
tubing pressure. The process is repeated until gas is being injected at the lowermost valve with all valves
above it closed.
Figure 14 shows conditions for a well which has been shut-in for a period of time. The tubing is filled with
“dead” oil (gas free) that has accumulated. This well would normally be returned to production by initiating
flow through the well control choke. Figure 15 shows typical flowing pressure as the well comes on
production. Wellhead pressure initially begins to drop as the long column of “dead” fluid moves up the
tubing string. As fresh wellbore fluids enter the tubing string, gas cutting of the column occurs and the
wellhead pressure begins to increase due to the lighter column in the tubing string (and relatively constant
bottomhole pressure). Generally, the wellhead pressure will again decrease somewhat until stabilized flow
conditions exist. A number of variations can occur, particularly with weak flowing wells.
Sometimes, weighted fluids are placed in the tubing string to insure that the well cannot flow to the surface.
This can occur during well completion, workover or killing operations. A brief discussion of typical weighted
fluids follows.
For fluid densities from 9.0 to 11.6 pounds per gallon, clear solutions of calcium chloride in water are most
often used. These fluids are economical and will not precipitate over reasonable periods of time.
Precipitation can cause tubing or perforation plugging, or possible formation impairment. When using these
fluids to balance formation pressure, the effect of wellbore temperature on the fluid density must be
considered. This can be done using the relationship shown on Table 5. Sometimes, higher density fluids
are necessary. These needs can be met with solutions of calcium bromide and calcium chloride in water as
shown on Figure 15. Occasionally, even higher density fluids are required, and these needs can be met
with solutions of zinc bromide, calcium bromide, and calcium chloride in water, as is also shown on Figure
16. The latter solutions are exceptionally expensive.
These solutions are normally used rather than drilling mud because of plugging damage that mud can cause
over an extended period of time. Loss control agents are sometimes added to the weighted fluid solutions
to minimize their loss to the formation. These can be nondamaging filter cake forming materials or viscosity
increasing agents. Whatever is used, it is acid soluble so that it can be dissolved if necessary.
The same wellbore conditions shown above on Figure 14 are shown on Figure 17, but the formation pressure
is now balanced with a weighted solution (neglecting wellbore temperature effects). Wellhead pressure is
zero and the well will not flow, and it is commonly referred to as “dead”.
Several methods can be used to place a dead well onto production. Each requires patience, and some
wells will require more patience than others. If the well is equipped for gas lifting, this is the best method to
use. Gas is injected, displacing the kill fluid slowly to the surface and allowing the lighter wellbore fluid to
enter the tubing. As the heavier fluid is displaced, the well will begin to flow using natural reservoir energy
and gas injection can be stopped. The next method involves slowly displacing the heavier fluid into the
formation with a lighter fluid from the surface. This method is frequently referred to as “bullheading” fluid into
the formation. Care must be taken to avoid overpressuring the wellbore equipment, fracturing the formation,
or otherwise damaging the completion. Pump-in is stopped, and the heavier fluid allowed to settle due to
gravity around the wellbore. The well is slowly opened to production, allowing lighter fluid to enter the tubing.
The process must sometimes be repeated many times, or “saw-toothing” before sufficient lighter fluid enters
the tubing and allowing the well to flow naturally.
If gas lifting is not possible and pump-in methods are not feasible due to a deep set standing valve or high
injection pressures, the well can be entered vertically through the wellhead tubing swab valve. Wireline
methods can be used to bail the heavier fluids out of the tubing string. Also, small diameter “spaghetti”
strings can be inserted down the tubing and gas injected down this string to displace the heavier fluid,
allowing the lighter wellbore fluid to enter the tubing.
If the well is equipped for TFL, lighter fluid can be pumped down one string, through the downhole circulation
member, and back up the other tubing string. This is usually done at a relatively high pump rate to displace
the heavier fluid. The well is then allowed to start flowing as described earlier.
Well flowing pressure can provide an early indication of possible downhole changes or problems. Also, if a
more rapid decrease in pressure is observed than would be expected by normal gradual depletion of
reservoir pressure, choke cutting or water entry may be occurring. A pressure increase can indicate a GOR
increase or plugged choke. If a substantial pressure drop exists across the choke, minimum temperature
change will be observed with low GOR production. For high GOR production, a significant temperature
change will be apparent and it may be necessary to inject hydrate inhibitor upstream of the choke to prevent
hydrate blockage at the choke.
Well tests are performed to determine if the well is producing as it should. This information provides
valuable reservoir management information. A typical well test separator is shown on Figure 18. Production
is routed from the well to the test separator where oil, gas, and water are separated and measured.
Conditions are allowed to stabilize, usually for a few hours. During the test period, which is normally not
less than four hours or more than twenty-four hours, the volume of oil, gas, and water is determined. These
results are then expressed in barrels per day for the oil and water, and in thousand standard cubic feet per
day for the gas. Sometimes, the separated oil still has water in it. If so, it is sampled and the dispersed
water content determined. This volume of water is added to the separated and measured water volume and
subtracted from the measured oil volume. The term percent B.W.&W. is frequently used in conjunction with
well testing. It stands for Basic Sediment and Water. Field personnel more often describe it as B. S.
water, which is not the correct terminology. This measurement is made to accurately determine the net
volume of produced oil. Also, results can indicate whether unusual amounts of solids (normally sand) are
being produced.
Frequently, a wellbore penetrates several producing zones. Only one, or a few, of these zones are opened
to the wellbore at one time in most modern-day completions. This is done to assure maximum reservoir
depletion and recovery efficiency. Sometimes, zones behind casing pipe contain producible reserves after
the open zone(s) have been depleted. When this condition exists, a workover will be performed to plug off
the depleted zone(s) and open up zones that are still behind pipe. Also, existing producing zones may have
developed some type of difficulty that can be corrected or improved by workover operations. This can
include early gas breakdown on top of an oil zone, or early water breakthrough (frequently called water
conning) on bottom of an oil zone. Sometimes, these conditions can be corrected by plugging off, or
squeeze cementing, the appropriate section of the producing zone.
Several types of downhole mechanical failures can occur which require workover operations to correct.
These can include wellhead seal failures, downhole safety valve failures, tubing leaks above or below the
downhole safety valve, packer leaks, casing leaks, and completion component (such as gravel pack)
failures.
Changing producing circumstances can also require workovers to best meet current producing needs. For
example, a tubing size which was necessary to meet early high producing rates, may prevent a well from
flowing during later life lower producing rates. Rather than installing gas lift equipment to allow producing the
well, it may be more economical to install a smaller tubing which could allow producing the well to depletion.
For safety reasons, most workovers are performed through BOP equipment. The tree is removed from the
wellhead (normally the well is killed and tubing plugs set first) and the BOP installed. Workover activities
performed, BOP removed, and the tree reinstalled.
Typical wellbore work needs can include paraffin cutting, downhole safety valve servicing, setting or
recovering plugs, standing valves, circulation control valves, gas lift valves or dummies, and shifting sliding
sleeves. Also, bottomhole pressure and temperature measurements can be obtained. Less often, logging
and perforating activities are performed. Also, sand bailing or washing may be performed.
A representative wireline rig-up is shown on Figure 19. The wireline lubricator assembly is mounted on the
tree above the swab valve. If the well is abnormally pressured, or if unusually difficult operations are
anticipated, a small BOP assembly is sometimes mounted between the tree and the wireline lubricator.
The types of wireline components are discussed thoroughly in the various wireline equipment manufacturers’
catalogs. These should be referenced if further information is necessary.
PDT or TFL methods can also be used to perform the well operations provided by wireline. A typical TFL
system is shown on Figure 20. TFL components were described above under the “Well Completion”
seminar. A brief description of TFL equipped well operations follows.
For the TFL system shown on Figure 20, the downhole safety valve requirement is met with a velocity type
subsurface controlled subsurface safety valve (SSSV). Regulations require either that the SSSV be pulled
and inspected annually, or that it be demonstrated that it can be flow closed. This discussion will describe
the TFL operations which would be typically performed to pull the SSSV. It is assumed that this attempt
results in stuck tools which must be recovered and that the SSSV cannot be pulled. Thus, it is tested by
flowing it closed. The SSSV is located below the H-member in the short string (SS) and the long string (LS)
is plugged. Thus, normal production is through the SS standing valve, the SSSV, and to the surface via the
SS, LS, or both.
With a frequently used closed system, it is sometimes useful to “fingerprint” the hydraulic system. As
shown on Figure 21, the pressure drop under steady state pump conditions is very predictable. If this
information is developed before any paraffin accumulation existed, the measured pressure drop at any later
point in time can indicate whether substantial paraffin accumulations exist. In this particular instance,
paraffin was present in the flowline between the surface and the wellhead. Thus, a TFL tool string is placed
into the SS lubricator at 1. Tree valves would be closed while this operation is in progress so that it is not
necessary to operate above the well shut-in pressure. The tool string would be pumped down the SS until
paraffin build-up is encountered, as indicated by an increase in the pump pressure. The tool string would be
worked back and forth in the SS until the tool string no longer encounters paraffin build-up. In this case,
within a few hundred feet of the wellhead. Tool string location is determined by monitoring pumped volumes
as shown on Figure 22. If the pump rate is not constant, as in this example, it is best just to note the net
pump in volume behind the tool string. Tool string location monitoring methods are not very accurate and
care must be taken not to pump them into closed valves or obstructions at high velocity.
Careful considerations are necessary to prevent getting the fishing string locked onto the stuck tools with no
means of release. Thus, a brief discussion of hydraulic system dynamics, accelerator/oil jar relationships,
and pin shearing follows.
First, pressure difference across the tool string, which is several thousand feet away, does not happen
instantaneously. As shown on Figure 24, two elements are involved. First, with the system equalized and
pressured up on both sides, pressure is maintained on the LS side and the SS side opened fully. The
pressure wave travels down the SS from the surface to the tool string (SS side only). As shown on Figure
25, by holding 4000 psi on the LS, and allowing the SS pressure to fall off as long as possible, the
maximum pulling force (Fp) is obtained. Thus, the oil jar release time should be designed to allow the
maximum pressure difference possible to occur. This effect is represented on Figure 26. Recalling that the
need is to up shear the steel pin in the original pulling tool, Figure 27 shows the amount of shear energy
required per pin shear plane. This device has four shear planes, thus the total impact shear energy would
be 152 ft-lb.
Second, the accelerator is used so that the oil jars do not release into the very long fluid column, which
greatly dampens the magnitude of the resulting “hit”. The operating principle of an oil jar/accelerator
assembly is shown on Figure 28. It is clear that the tool string weight, oil jar stroke, and accelerator spring
constant must be properly designed to provide the required energy to release the pulling tool by shearing the
pin. These relationships are shown on Figure 29.
Thus, the fishing tool assembly is equipped with the appropriate oil jar, accelerator, a high strength shear up
pin (so that the fishing tool string does not release before the stuck tool string) and a shear down release pin
(so that the fishing string can be released if the stuck tool string cannot be recovered). The fishing tool is
pumped down the SS and engages the stuck tools. High pressure is equalized across the tool strings, and
SS pressure is released. The stuck tool string comes loose when the oil jar “hits,” and both strings are
pumped back to the surface and recovered. Unfortunately, the SSSV and lock assembly are not recovered.
Thus, it is necessary to flow test the SSSV closed, shut-in the well, or perform a costly workover. The
former is clearly the most acceptable choice.
The tubing strings can be pressured up with oil to above the well shut-in pressure. Then, by quickly
releasing the pressure at the surface, the SSSV will close with only momentary flow—provided the SSSV is
functioning. This operation, including reopening the SSSV by pressuring up on top of it, is shown on Figure
30. Thus, the well can be safely produced and shut-in or costly workover is avoided or deferred.
TABLE OF CONTENTS
LIST OF FIGURES
FIGURE 1. CALM System
FIGURE 2. Exposed Location Single Buoy Mooring (ELSBM) System
FIGURE 3. Single Buoy Storage (SBS) System
FIGURE 4. Shell-Castellon Field Development
FIGURE 5. Spread Mooring Arrangement
FIGURE 6. Turret Mooring Arrangement
FIGURE 7. CALM Rigid Arm System
FIGURE 8. Buoyant Tower SPM Terminal
FIGURE 9. Single Anchor Leg Mooring (SALM) System
FIGURE 10. Buoyant Tower for Fulmar Field
FIGURE 11. URI Mooring System
FIGURE 12. Restoring Force In Gravity System / Restoring Force In Buoyancy System
FIGURE 13. Force/Excursion Diagram for Different Systems
FIGURE 14. Force Generation
FIGURE 15. Mooring Force is Function of System’s Energy Potential
The first SPM used for this purpose was the catenary anchor leg mooring (CALM) shown in Figure 1.
Later, the system moved to offshore oilfields where it was used to provide storage for processed crude. A
tanker or barge was permanently moored to the SPM in the vicinity of the platform. The crude was collected
from the platform via a small diameter subsea pipeline and a flexible riser via the mooring system into the
barge. In order to export the crude, a second offloading terminal was placed in the field. Large-diameter
pipelines connected the offloading terminal with the storage barge, allowing unloading of the barge within a
24-hour time span.
The early concepts were used in fairly mild areas where mooring a vessel permanently with simple means
posed no problems.
When operators moved into more hostile waters, such as the North Sea, the storage concept had to be
modified since technology at that time did not permit mooring a vessel permanently in such an environment.
Instead, a combination storage and loading terminal was conceived. The vessel would hook up to the
terminal and receive crude from the platform at the rate of the field production. When the tanker was full, it
would sail to port and another vessel would take its place. When weather conditions became too severe,
the tanker would disconnect and the field would be temporarily shut-in.
The first operator who applied such a scheme was Phillips in its North Sea Ekofisk Field.
Buoy manufacturers started to develop alternatives to the simple CALM which were more suitable for hostile
environments. One development was the Exposed Location SPM (ELSBM), used by Shell/Esso in Aukfield,
shown in Figure 2 which eliminated the floating hose.
Until then, all of these systems were based on the old CALM buoy principle which functioned fine in
conditions for which it was originally intended: Export terminal in mild conditions. Soon it became clear
that the old CALM buoy was not all that great for storage and rough environments.
The main drawbacks appeared to be the use of nylon mooring hawsers and floating hoses. The continuous
wave forces causes fatigue at the hose nipple which could lead to premature hose failure. The nylon hawser
is internally abraded by the cyclic stretch and loses its strength over a period of time. In export terminals,
there is ample time to maintain and replace these items between loadings. A permanent storage terminal,
on the other hand, is moored 100% of the time, which speeds up wear and furthermore complicates
replacement. Also, replacement of hawsers and hoses in the North Sea could be impossible for a long
period of time due to adverse weather conditions.
For this purpose, the rigid yoke moorings were designed. The first rigid yoke mooring was the Single Buoy
Storage system (SBS) installed in 1974 offshore Tunisia, and shown in Figure 3.
The advantages are clear: The nylon hawser is replaced by a steel structure with predictable strength
qualities, the floating hoses are eliminated and replaced by hard piping, and the overall motion characteristic
of the tanker is improved. This last point relates the slow oscillatory “fishtailing” of the vessel, which is
normally encountered with hawser moorings.
The stable position of the storage tanker also allows the offloading tanker to moor alongside the barge
instead of using a separate terminal. The crude is transferred directly between the ship’s manifolds.
A dozen or so rigid arm systems of various designs have been installed since then. The desire in later
years to develop marginal fields where a platform was too costly or simply not feasible has led to the latest
development in mooring systems, i.e., the application in floating production systems. Shell was the first
operator to exploit a tanker-based production facility in its Castellon Field offshore Spain in 1977, as shown
in Figure 4. A subsea well produces directly into a permanently moored tanker, where on-board process
facilities separate the oil from gas and water. The oil was stored in the tanker holds while the gas and water
were disposed. Shuttle tankers came alongside the barge periodically to offload the crude.
The major disadvantage of the bow turret is the limitation in water depth. Heave response of the nose of a
tanker can be 1.5 times the wave height. This vertical motion will cause high dynamic loads in the anchor
chains which becomes more predominant when the ratio of water depth to vertical motion is small. On a
tanker, the turret must extend downward to near the level of the keel of the vessel to prevent the mooring
chains from hitting the bulb on the bow. This further decreases the actual catenary depth, which reduces
the flexibility. The catenary mooring lines can be made longer and thus more flexible for a flat-nosed barge
or by removing the bulb from the tanker. However, in general, the turret mooring cannot be used in water
depths less than 100 feet except in very calm weather.
by floating hoses. In rough weather the motions of the floating hose relative to the buoy and vessel are
large, causing high bending stresses in the connection at the buoy and where the hose changes from
horizontal to vertical at the vessel side. Any maintenance work or operation on the buoy requires transport
of people from the vessel by boat. In calm weather the vessel tends to ride up to the buoy, which can cause
damage to the buoy and the floating hose.
A consequence of eliminating the hawser is a reduction in mooring elasticity. In deep water this will not
affect the systems limitations since the chain system alone provides sufficient elasticity.
In shallow water, however, the catenary system becomes less elastic and the hawser will contribute a
greater part to the overall system elasticity. Exchanging the hawser for a rigid arm in that case will reduce
the operational limits of the system.
With increasing water depth, the bending moments in the tower, due to vessel offsets and waveloading,
increase progressively and the structural needs to cope with the bending stresses make this concept
economically prohibitive in water depths greater than 600 feet.
In shallower water, the riser can be eliminated using the buoy as a tower, as shown in Figure 10.
For a production riser application, multiple lines are brought around the universal joint below the buoy
through hoses or swivels to the top of the buoy. A multiproduct swivel carries the fluid from the buoy to the
rotating turntable. A rigid arm must be used in this case, since the rocking motion of the buoy and the
relative vertical motion of the water surface would pose severe stress conditions on surface or subsurface
hoses.
The SALM concept can be used in virtually any water depth. Unfortunately, the cost of the system is high
because the buoy will be large and must have a deep draft in order to be stable in the wave troughs.
In deep water this problem becomes more pronounced since the riser tension has to increase in order to
limit the riser angles. Large riser angles would cause the yoke to move down a considerable amount,
bringing the fluid swivel under water. The amount of buoyancy required to provide sufficient riser tension in
deeper water would therefore be enormous, and the economic use of the SALS is probably limited to water
depths of approximately 1000 feet.
Another gravity system is the URI Mooring. A weight is suspended from the riser creating a tension force.
When the ship is pushed from its neutral position, the riser inclines and causes a horizontal component of
the riser tension force, which tends to pull the ship back. A new equilibrium position is found when the
restoring force equals the environmental force acting on the vessel.
Restoring force indicates the capacity of the system to generate a reaction. The performance of a mooring
system, however, depends also on the excursion.
The force/excursion diagram, given in Figure 13, shows the system performance and is one of the criteria to
judge whether that mooring system is suitable for a particular environment.
vessel size
wave height and period
current speed and direction
wind speed and direction
characteristic of mooring system
Before we address these parameters, the mechanisms of a mooring system should be clarified. The main
purpose of a mooring is to restrain the horizontal motion of the vessel. Absolute restraint is impossible,
however. To limit the wave induced short period motion of the tanker would require forces beyond what is
physically possible. Hence the six degrees of short period vessel motion cannot and should not be
restrained by the mooring system. What can be limited is the long period surge, sway and yaw of the
vessel.
Since the short-period vessel motions dictate an elasticity in the system, the ship will respond to force
variations in the long natural period of the spring-mass system, formed by ship and mooring.
This natural period will affect surge, sway and yaw. The exciting force to this phenomena is caused by wave
drift forces, wave grouping, wind speed variation and direction changes, and current fluctuations.
This force system on the vessel is complex. A wind gust can initiate a yaw motion which will change
relative current heading and wave drift force in its turn. A continuous change in force and moment balance
takes place.
The result of this slow oscillatory motion is that the mooring force will be larger than a static equilibrium
check would indicate. The magnitude of this amplification depends on the characteristic of the mooring
system. A very stiff system will in general induce higher mooring loads; on the other hand, a sloppy system
is equally unsatisfactory since the vessel can built up momentum first, which then causes high inertia loads
when enough restoring force can be generated.
The general tendency is that the mooring force increases with vessel size, wave height, current speed and
wind speed. However, there are certain effects which can upset this generalization. A larger vessel size
can bring the natural period of the mooring system outside the range of predominant wave grouping;
moreover the short period forces become smaller with increasing vessel size. These two effects can result
in a larger mooring force with the smaller vessel.
A short-crested wave, encountered in typhoon areas without sufficient fetch for the sea to fully develop,
could cause higher mooring loads than a larger wave with a longer period. When current is running in the
direction of wave propagation and vessel hading, the static mooring force will increase with current speed.
The current will introduce, however, a damping effect which could cause a reduction in long period surge.
When current comes in perpendicular to the waves, the vessel will take a heading somewhere between
waves and current. The current will then generate a lift effect which will push the vessel forward into the
waves and reduce the mooring load.
The characteristic of the mooring system is the keystone in the mooring force concept. Firstly the system
should have enough strength to withstand the mooring force; secondly, the system should yield enough
elasticity to limit the mooring force. This can be clarified with the following picture. Figure 14 shows the
horizontal motion of the system over a period of time. The system makes long period excursions,
superimposed on which are wave-induced short-period motions. Projecting this picture on a typical
force/excursion diagram of a catenary chain system shows some interesting phenomena. Again we find the
quasistatic equilibrium position A with the long-period mooring force, FA. The dynamic motions create a
force variation around this point, the magnitude of which depends on the gradient of the force/excursion
diagram. A small force gradient obviously generates small dynamic forces. A high force gradient could
cause forces far in excess of the breaking strength of the chain.
A measure of effectiveness of the mooring system is the energy potential represented by the (usable) area
under the curve. Figure 15 shows a hypothetical energy curve. A concave curve will yield high quasistatic
and dynamic forces, while a convex curve with the same energy potential will induce smaller forces.
Only horizontal motions have been considered in these graphs; vertical motions, particularly in shallow
water, will h ave similar, and sometimes worse, effects as the horizontal dynamic motions.
MARINE SYSTEMS
TABLE OF CONTENTS
2.0 STABILITY........................................................................................................................... 3
2.1 Initial Stability ............................................................................................... 3
2.2 Complete Stability ......................................................................................... 3
2.3 Dynamic Stability .......................................................................................... 4
2.4 Stability Standards ........................................................................................ 4
2.5 Damage Stability ........................................................................................... 5
3.0 MOTION............................................................................................................................... 6
3.1 Environment .................................................................................................. 6
3.2 Idealized Environment .................................................................................... 6
3.3 Statistical Analysis........................................................................................ 6
3.4 Vessel Motion Definition................................................................................. 8
3.5 Ship - Environment Interaction ........................................................................ 8
3.6 Drillship - Semisubmersible Comparisons ........................................................ 9
LIST OF FIGURES
FIGURE 1. Ship Planes
FIGURE 2. Lines Plan
FIGURE 3. Table of Offsets
FIGURE 4. Displacement and Other Curves of Form
FIGURE 5. Simpson Integration for Area
FIGURE 6. Simpson Integration for Displacement
FIGURE 7. Cross Section of Typical Floating Body
FIGURE 8. Cross Section of Typical Floating Body
FIGURE 9. Cross Section of Typical Floating Body
FIGURE 10. Cross Section of Heeling Vessel
FIGURE 11. Cross Curves of Stability
FIGURE 12. Curve of Statical Stability
FIGURE 13. Righting & Heeling Moment Curves
FIGURE 14. Overturning Moment due to Wind
FIGURE 15. Limiting Downflooding Angle
FIGURE 16. Compartment Flooded for Damage Stability Calculations
FIGURE 17. Floodable Length Curve
FIGURE 18. Load Line Markings
FIGURE 19. Current Profile
FIGURE 20. Regular Wave
FIGURE 21. Linear Spectral Analysis
FIGURE 22. Six Degrees Vessel Motion
FIGURE 23. Heave Rao Curve
FIGURE 24. Tons per Inch
FIGURE 25. Heave Rao Curves Ship vs. Semi
In order to quantify the shape of the ship, a mapping technique is used. Mutually orthogonal plane sets are
“passed through” the ship and the intersections of the planes with the boundaries of the ship plotted as
shown in Figure 1.
The first set of planes are known as waterplanes. These are parallel to the water surface. The intersection
of one of these planes and the ship produces a waterline.
The set of planes parallel to the length and perpendicular to the waterplanes are called buttock planes and
the intersections are buttocks.
Transverse planes are perpendicular to both the previous sets and result in section lines.
The lines defined above are assembled on a drawing called a lines plan; the waterlines grouped on a half-
breadth plan, the buttocks on a profile plan and the sections on a body plan. An example is shown in
Figure 2.
From this plan, measurements are taken off. These are grouped in a table called a table of offsets as shown
in Figure 3. The table of offsets represents the numerical definition of the shape of the ship. See Figure 3.
The important characteristics of the underwater volume are grouped into one plot as shown in Figure 4:
? or displacement
CB or center of buoyancy
Awp or area of the waterplane
CF or center of flotation
I or waterplane inertia
BM or metacentric radius
The above relate to the behavior of the ship. In addition, characteristics may be calculated which indicate
the fatness or slimness of the volume. All of the characteristics are calculated for a specific assumed
waterplane. The most commonly encountered are defined by the design draft.
The characteristics of the hull plating which are determined are the weight of plating, its center of gravity and
its inertia. These properties are combined with a calculation of the weight, center of gravity and inertia of the
ship’s structure, outfitting, consumables and cargo to yield properties for the ship as a whole. This includes
? and CG.
Because most ships have lines which are not mathematical equations, numerical integration techniques
must be used to determine the volume and hull properties. Nowadays, this is done with sophisticated
computer programs and advanced formulas and up until about six or seven years ago, naval architects spent
a great deal of time doing this. The most widely used formula is known as Simpson’s Rule. Figure 5 is an
example of transverse sectional area. This rule assumes a curve is a second-degree parabola. An even
number of equal spaces, n, is imposed on the area to be integrated by (n + 1) ordinates. The area is
considered to consist of n/2 parts, each with a base equal to twice the common interval, S, between
ordinates as shown in Figure 6.
The characteristics of the underwater volume are calculated for various drafts and plotted as shown on
Figure 4.
2.0 STABILITY
The weight of the ship must be balanced by the displaced water. Thus the weight will determine at what
draft the ship floats. At this waterline, the volume of water will have some centroid, B, which is measured
vertically from the keel as shown in Figure 7. The ship has some center of gravity, G, also measured from
the keel. At even keel, the line of action of the weight, through G, lines up with the line of action of the
buoyancy, through B, and there is equilibrium.
Now assume an applied moment cause the ship to roll as shown in Figure 8. At the inclined waterline,
there is a new center of buoyancy. There is now a separation between the line of action of the weight and
the line of action of the buoyancy. This separation is the righting level, GZ. The couple created by the
separation of weight and buoyancy is called the righting moment and is equal to the displacement times the
lever or: ? GZ as shown in Figure 9.
The above discussion refers to a ship with positive stability. This is defined by the metacenter being above
the center of gravity. If this is true, an applied moment will result in a restoring moment resisting heel.
Negative stability occurs when the metacenter is below the center of gravity. An applied moment is now
assisted by the “restoring” moment. It becomes a heeling moment as shown in Figure 10.
The relation of the center of gravity to the metacenter, GM, is calculated from the known properties of the
ship.
GM = KB + BM - KG
There is an important correction to this equation which accounts for the effect of liquids in tanks on the ship.
The center of gravity of the ship as used above is assumed constant. If there are partially filled tanks on
board, the effect of heeling will raise the center of gravity of these tanks, thus raising the center of gravity of
the ship. Thus stability calculations include a “free surface” term to correct the KG term.
The righting arm for the know condition varies with angle of heel and is usually plotted as GZ (feet) versus
angle of heel (degrees). This is a curve of statical stability as shown in Figure 12.
The effect of an external heeling moment can now be compared to the restoring moment. If, for example a
wind is blowing against the ship’s beam then the wind and the ship’s hull in the water will produce a couple
tending to overturn the ship as shown in Figure 14. If the ship is initially at rest, the wind will cause heeling
to point A (See Figure 13). This value of inclination is the steady state value. Point B is an unsteady
position. If the angle of heel is increased, the ship will roll over. Point C represents the maximum moment
that can be generated by the ship. Points A and B are referred to as, respectively, the first and second
intercepts.
Another criteria involves the angle at which the ship begins to take on water through an opening, such as a
hatch. This downflooding angle, as illustrated in Figure 15, may restrict how far the vessel can heel.
In addition to wind as a heeling moment, several other conditions will cause unusual inclination. The
following conditions may also need consideration during design:
The discussion thus far has been about paper characteristics. Once the ship is in the water, a test may be
made to check the accuracy of the calculations. This test is called an inclining experiment. Weights are
moved around on the ship and the change in the heel of the ship is measured. This test ends up relying on
calculated characteristics, however, as the displacement is needed. It’s a very small ship that can be
actually weighed.
Thus, the stability and flotation of a ship is investigated by mathematically damaging the ship. This is
illustrated in Figure 16. For a given ship, compartments are flooded and the impact on heel, trim and
righting moment calculated. Interestingly, two methods used for flotation calculations are the “added weight”
method and the “lost buoyancy” method.
Another set of calculations is performed to derive a floodable length curve. This curve, shown in Figure 17,
represents the maximum length of compartment, centered at point of penetration, which could be flooded
without causing the ship to trim more than an acceptable amount. The trim is without list.
The calculations of damaged condition that a ship has some reserve buoyancy. This is buoyancy not
normally at work but which is needed in a damaged state. To ensure that ships keep enough reserve
buoyancy during operation, they are assigned a load line. This mark, as shown in Figure 18, is on the side
of the ship and indicates the maximum draft to which a ship may be loaded.
3.0 MOTION
3.1 Environment
The ship’s environment is a widely varying one. The effect of this environment can be studied by breaking it
down into three segments. Wind is the movement of the air portion. Current is the movement of the water
portion. These are generally defined by steady-state values. Waves are the changing air-water interface.
They are the most difficult part of the environment to define.
Current is also a widely varying force, though it does not usually vary as rapidly as wind. It, too, is usually
taken as a steady-state velocity and direction. For drilling vessels, it is important to know the variation of
current with depth. This “current profile” is usually presented as uni-directional as shown in Figure 19.
Often, two profiles are used. The average, or mean, values represent the most commonly expected current.
The maximum values represent either measured short-term extremes or the uncertainty of the measurer.
Waves are caused by a disturbance of the water, either by some physical displacing force or, more
commonly, by the passage of wind. An idealized wave train, as shown in Figure 20, has certain measurable
properties, such as height and length. While regular waves are useful as a means to understanding, they
rarely occur alone. Most sea conditions of interest consist of groups of regular waves superimposed on
each other.
“linear” quantities
“non-linear” quantities
The first group contains functions which have a linear response to wave height. Short-period system
motions fall in this category. Most of the mooring system parameters, however, fall in the second category.
All the system forces, for instance, are not linear-related to wave height.
In order to predict certain maximum design loads, statistical methods have to be applied. A different
approach for the two categories is necessary.
Determination of probability levels for the first kind is straight-forward and is done with the wellknown
spectral analysis technique. Figure 21 illustrates the following relationships.
2
1 - z
Normal distribution function : f(z) = e 2 mo
2π mo
2
z z2
Rayleigh distribution function : f(z) = _ e- 2 mo
mo 3
The probability of the exceedance of a certain value of the heave amplitude can be calculated as:
2
- z z2
P( Z a f a) = ∫∞ z a _e 2 mo dz a = e- 2mo
mo 4
The expected maximum heave amplitude can be calculated according to:
(
Za max = 2mo* ( ln N )1/2 + .29( ln N )-1/2 ) 5
where: N = number of oscillations
For instance, in a six-hour storm with an average wave period of 10 seconds, the number of oscillations
would be N = 6 x 60 X 60/10 = 2160 and the
For a ship in still water, a force producing surge, sway or yaw produces no restoring force. This is because
no change is made in the shape of the waterplane. A force producing roll or pitch alters the waterplane and
a restoring force is generated. Heave alters the displacement of a ship and force causing this is resisted.
A moored vessel, such as a drilling rig, must resist the offsetting force of wind. In this case, the wind is
causing deleterious surge or sway which must be resisted by a mooring system. The ability of a vessel to
remain on location is usually measured in terms of the maximum wind which can be resisted. This load is
in conjunction with other environmental loads.
A head-on current has the direct effect of reducing a vessel’s speed. At other angles, the effect of current is
nearly direct. For a stationary vessel, current causes an offsetting force just as wind does. The ability of a
ship to resist current is usually expressed as the maximum velocity in which headway can be maintained or
station may be kept.
The response of a vessel to waves is a very complex problem. One wave passing a ship can effect
movement in all six degrees of freedom. For this reason, response is most often investigated by isolating
the motions as a result of regular waves.
If a ship is moving through a regular wave train, it will have some heave response, for example. This heave
will be different for different waves and ship speeds. Taking a moored ship, or one with no forward speed, a
relation can be found between wave period and vessel heave. This is usually plotted as heave of the ship
divided by wave height versus wave period as illustrated in Figure 23. This non-dimensional response
amplitude operator (RAO) curve is characteristic of the ship and can be compared to curves of other ships.
Similar curves can be developed for other degrees of freedom, speed of advance and vessel orientation.
Response to irregular seas can also be predicted using wave energy spectra and the RAO curves. These
predictions are each good, then, for a specific:
Vessel speed
Vessel orientation
Vessel degree of freedom
Wave spectrum
The response of a ship to waves can be compared to many criteria. the maximum permitted response in
one of the degrees of motion may limit operations. On some vessels, acceleration must be kept below
certain levels. On the other hand, too slow a vessel response may cause prohibitive deck wetness.
Harmonic pitch response and flat bottom forward may combine to produce slamming. At the other end of
the ship, propeller emergence may be a problem. A speed reduction may be required due to one or all of
the above. The general term for the relation of a ship to its response in waves is seakeeping. There is good
seakeeping and bad seakeeping but no one number which can be calculated or measured to define it.
The semisubmersible first became popular as a drilling platform because of its motion characteristics.
Reduced roll, pitch and heave response to waves is due to the small waterplane area in relation to
displacement. The accelerating forces due to a wave passing are directly related to the waterplane area and
inversely proportional to displacement. Thus, a ship will be accelerated more in a small period wave train as
shown in Figure 25. The dip in the ship response curve indicates some harmonic tuning between the ship
length and the wave length at about 10 second period.
The term “scale factor” refers to the relationship of the dimension of the prototype compared to the
dimension of the model. For example, if the prototype dimension is 60 times the dimension of the model,
the scale factor, designated by ?, is 60.
Normally the principal effects to be modeled in SPM model testing are forces due to gravity waves.
Therefore, Froude’s-law scaling is used for these model tests.
By Froude’s-law scaling for dynamic similitude, the following condition must be satisfied:
V2 V2
=
gL p gL m 7
where V = velocity
g = acceleration of gravity
L = characteristic length
p represents prototype
m represents model
Since, the acceleration of gravity is the same on the model as it is on the prototype, it follows that:
2
Vm Lm
= = 1/λ
Vp Lp 8
Thus, the ratio of the scale velocities is ?-1/2. The ratio for time scale is also ?-1/2 since
Tm Lm/Vm Lm Vp
Tp
= = ( )
= 1/λ λ‰ = λ -‰
Lp/Vp Lp Vm 9
Further, the ratio for forces is given by ?-3 since:
3
Fm (s.g.)m xL m3 x am Lm
= = _ λ -3
Fp (s.g.)p _x Lp3 x ap Lp
=
10
Both specific gravity and acceleration are the same for model and prototype.
Testing at as large a scale as practical is desirable because the accuracy of the results will be better. The
accuracies of the modeled environments and the dimensions of the model will be proportionately better at a
larger scale. The ability to scale forces will also be better at a larger scale because certain phenomena
scale better by one scaling rule than by another, and the deviations are greater at smaller scales.
The effect of scale is more important in modeling small objects, such as cargo hoses, than it is in modeling
very large objects, such as tanker hulls. For example, above a critical Reynolds number, viscous-drag
coefficients are essentially independent of velocity, but small objects are more apt to have Reynolds
numbers below the critical range, and thus not scale properly.
Force and moment measurements are much more accurate at larger model scales. There is a practical
limit to how small and how accurate force transducers can be fabricated and calibrated. A one-Newton error
in a transducer produces a much larger error when scaled up from a small model. Furthermore, at small
scales the presence of a relatively large or heavy transducer can influence forces on the model or the
response of the model, and in turn produce errors in the results.
The model scale is limited, however, by facilities and cost. Limitations of available model test facilities and
equipment will govern more than cost. The length, width, and depth of the model basin, the capacity of the
wave generators, and the pumps and fans used to produce current and wind will limit the size of the model
which can be tested. There may also be a limiting relationship between the depth of water in the basin and
the height of wave or velocity of current which can be produced.
At a scale of 1/50, a model of 550,000 dwt tanker will be approximately 8.3m (27 ft) long and will weigh
approximately 50 kN (11,000 lbs) at loaded draft.
Several types of mechanisms are used to generate waves in the various model basins. The principal types
of wave generators are mechanical flaps or paddles, mechanical plungers, and pneumatic generators.
Wave generators generally are limited in their ability to produce short-period (high frequency) waves. For
example, if a wave generator cannot cycle faster than 1.5 hertz, it will not be able to produce waves with
periods shorter than about 5 seconds at a length scale factor of 60. This inability to produce the shorter
period components of wave spectra is common to all wave basins.
When testing in survival conditions with cargo hoses, high-frequency response characteristics may be of
major concern. In such cases it will be necessary to match the high-frequency portion of the spectrum
more precisely. this may necessitate modeling the buoy and cargo hose system (without a tanker moored)
at a larger scale.
The most generally accepted method of modeling wind during SPM model testing is to produce the forces
and moments on the vessel by means of a generated wind field. A bank of fans is arranged perpendicular to
the desired wind direction at some distance from the model, such that a uniform wind field is developed over
the area occupied by the vessel at the SPM.
At the beginning of a test series in which wind will be modeled, a series of constrained-vessel wind-force
tests should be conducted to verify that Froude’s-law-scaled forces and moments are produced. The
longitudinal wind force, lateral wind force, and wind yaw moment are measured on a restrained vessel for a
number of different wind directions.
During the constrained tanker wind-force tests, the forces and moments that are measured are compared to
those forces and moments that have been determined in wind-tunnel model tests. The velocity of the wind
field generated by the bank of fans is adjusted until a reasonable match is obtained.
Another method of modeling wind forces is by several fans mounted on the vessel itself. Their velocity may
then be regulated by a yaw sensor coupled to a computer, such that they produce the desired forces on the
vessel according to Froude’s-law scaling.
In SPM model tests, current may be modeled according to Froude’s-law scaling. The predominant effects of
current on the vessel are form and viscous drag, and, in shallow water, a difference in water level across the
vessel due to blockage. The latter phenomenon is a gravity effect and is therefore accurately modeled by
Froude’s-law scaling. Viscous drag effects are minimal except at angles near bow-on to the current. The
form or pressure drag is of viscous origin, but appears in general to be independent of Reynolds number in
the range of interest in SPM tests with ship-shaped bodies.
The most realistic method of modeling current in SPM model tests is to generate a flow of water in the
model basin, with a velocity proportional to the prototype current using Froude’s-law scaling. Such a current
flow is usually produced by pumping water through a bank of inlet ducts along the side of the basin, and
returning water to the pumps through ducts on the opposite side of the basin.
The model, both vessel and mooring system, must represent the prototype in geometry and mass
distribution. Center of gravity, center of buoyancy, and natural periods should be accurately modeled so
that vessel dynamics and stability are identical to the prototype.
Since the elasticity of the mooring system is a crucial parameter, extreme care should be taken that the
load-elongation curve of nylon bow hawsers, chain weights, and chain elasticity are properly modeled.
The quality of the instrumentation and data recording system is very important. The care taken in
accurately modeling the environment and the system is lost if the recorded forces and motions are distorted
by the instrumentation system.
5.1 Purpose
Classification societies are independent, objective, non-profit agencies which provide guidelines for the
design, construction and maintenance of sound vessels. A ship which is built to the requirements of the
rules of a classification society, under the inspection of the society’s surveyors, and with materials and
other components tested to society specifications is granted “classification” by formal action of the society’s
committee. This fact is then published in the society’s register and made public knowledge.
Classification establishes that an owner has used “due diligence”, it informs a shipper that he is using a
sound vessel, and it helps an underwriter determine the risks of insuring ship or cargo.
5.2 Societies
The following are recognized as major classification societies. While the specific rules of each differ, the
intent is the same.
Scantlings
Materials
Arrangements
Mooring equipment
Machinery
Controls
Workmanship
Stability