SPE 52157

Selection of Artificial Lift

James F. Lea and Henry V. Nickens--Amoco EPTG/RPM

Copyright 1999, Society of Petroleum Engineers, Inc.

This paper was prepared for presentation at the 1999 SPE Mid-Continent Operations
Symposium held in Oklahoma City, Oklahoma, March 28-31, 1999.
This paper was selected for presentation by an SPE Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Society of Petroleum Engineers, its officers, or members. Papers presented at
SPE meetings are subject to publication review by Editorial Committees of the Society of
Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers is
prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300
words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Abstract
Selection of the most economical artificial lift method is
necessary for the operator to realize the maximum potential
from developing any oil or gas field. Historically the methods
used to select the method of lift for a particular field have
varied broadly across the industry, including
Determining what methods will lift at the desired rates
and from the required depths.
Evaluating lists of advantages and disadvantages.
Use of expert systems to both eliminate and select
systems.
Evaluation of initial costs, operating costs, production
capabilities, etc. using economics as a tool of selection.
This paper will highlight some of the methods commonly
used for selection and also include some examples of costs
and profits over time calculated to the present time as a tool of
selection. The operator should consider all of these methods
when selecting a method of artificial lift, especially for a
large, long-term project.
Introduction
In artificial lift design the engineer is faced with matching
facility constraints, artificial lift capabilities and the well
productivity so that an efficient lift installation results. Energy
efficiency will partially determine the cost of operation, but
this is only one of many factors to be considered.
In the typical artificial lift problem, the type of lift has
already been determined and the engineer has the problem of
applying that system to the particular well. The more basic
question, however, is how to determine what is the proper
type of artificial lift to apply in a given field.
Each of the four major types of artificial lift will be

discussed before examining some of the selection techniques.

Some additional methods of lift will also be discussed.
Preliminary comments related to reservoir and well factors
that should be taken into consideration are presented.
There are certain environmental and geographical
considerations that may be overriding issues. For example,
sucker rod pumping is by far the most widely used artificial
lift method in the United States. However, if we are in the
middle of a densely populated city or on an offshore platform
with forty wells contained in a very small deck area, sucker
rod pumping may be eliminated. Deep wells producing several
thousands of barrels per day cannot be lifted by beam lift and
other methods must be considered. These geographic and
environmental considerations may simply make our decision
for us; however, there are many considerations that need to be
taken into account when these conditions are not
predetermining factors.
Reservoir Pressure and Well Productivity. Among the most
important factors to consider are reservoir pressure and well
productivity. If producing rate vs. producing bottom-hole
pressure is plotted, one of two inflow performance
relationships (IPR) will usually occur. Above bubble point
pressure, it will be a straight line. Below bubble point
pressure, a curve as described by Vogel will occur. These two
curve types are shown in Figure 1 as a single IPR with a
bubble point at about 750 psi.
Some types of artificial lift are able to reduce the
producing sand face pressure to a lower level than others. The
reward for achieving a lower producing pressure will depend
on the reservoir IPR. For example, the well of Figure 1 would
have an absolute open flow (AOF) of about 670 bopd if no
gas was being produced. However due to the gas, the AOF is
reduced to about 580 bopd. If you are using a pumping system
on this well, there may be good reason for not lowering the
sand face pressure below about 500 psi as the increasing
amount of free gas may cause gas interference and
diminishing returns on production with lowered pressure. It
also would be difficult to lower the pressure as much
compared to some other lift methods using gas lift although a
gassy well would in general be beneficial for gas lift
applications.
In addition to the older conventional IPR expressions for
vertical wells, there are now available a number of IPR

J.F. LEA, H.V. NICKENS

models for horizontal wells. An input panel for one such

model is shown Figure 2. This allows calculation of inflow
from a horizontal well, which typically produces several
multiples of what a vertical well would produce in the same
formation.
Use of horizontal well inflow models for present or for
future depleted horizontal wells can be used to determine if
the flowing production rates can be economically increased
through the use of some method of artificial lift. If the
horizontal well is low pressure and ceases to flow, the inflow
model can estimate what the well could produce if supplied
with a form of artificial lift.
For a large project, reservoir models may be used to
predict expected inflow conditions of the expected life of the
project.
Reservoir Fluid. The characteristics of the reservoir fluid
must also be considered. Paraffin is a much more difficult
problem for some kinds of lift than for others. Sand
production can be very detrimental to some types of lift. The
producing gas-liquid ratio (GLR) is very important to the lift
designer. Free gas at pump intake is a significant problem to
all of the pumping lift methods but is beneficial for gas lift,
which simply supplements the lift energy already contained in
the producing gas.
Long Term Reservoir Performance and Facility
Constraints. Two approaches have frequently been taken in
the past to account for long term reservoir performance. In our
opinion, both of these approaches are extreme and wrong.
In some cases, we predict long term reservoir performance
and install artificial lift equipment that can handle the well
over its entire life. This frequently lead to the installation of
oversized equipment in the anticipation of ultimately
producing large quantities of water. As a result, the equipment
may operate at poor efficiency due to under-loading over a
significant portion of its total life.
The other extreme is to design for what the well is
producing today and not worry about tomorrow. This can lead
to change after change after change in the type of lift
equipment installed in the hole. We may operate efficiently
short term but spend large amounts of capital dollars in
changing equipment. For instance, the changing reservoir
conditions with time shown in Figure 3 would have to be
carefully considered in sizing artificial lift equipment for
current conditions and for some selected future period of time.
Reference 14 is concerned in detail with timing of artificial lift
methods.
In a new field development, the fluid handling requirement
can significantly increase the size and cost of the facilities
required to produce the field. With beam or electric
submersible pumps, only the produced fluid is handled
through the facilities. With gas lift, the injection gas
compression and distribution facilities and additional gas in
the production adds to the facilities. With hydraulics, the
power fluid pumps, power fluid injection lines and additional

SPE 52157

power fluid, many times combined with the production fluid,

adds to the fluid handling costs.
The design engineer must consider both long term and
short term aspects. Our aim is to maximize the present value
profit of the operation. The highest present value profit may or
may not result from greatest production rate available from the
well and may or may not anticipate a lift system change in the
future. Many of the introductory comments and observations
in the proceeding discussion will be included in lists of
advantages/disadvantages, expert systems and other types of
selection analyses discussed in following sections.
Types of Artificial Lift. The various major forms of artificial
lift are shown schematically in Figure 4. There are other
methods as well which will be mentioned as appropriate in the
following discussions, such as the ESPCP for pumping solids
and viscous oils. This system has a PCP pump with the motor
and other components similar to an ESP. Other methods
include long stroke modifications of beam pump systems.
The selection of the lift method should be a part of the
overall well design. Once the lift method is selected,
consideration should be given to the size of the well bore
required to obtain the desired production rate. More than once,
a casing program has been designed to minimize well cost and
then find that the desired production could not be obtained
because of the size limitation on the artificial lift equipment.
Even if production rates can be achieved, smaller casing sizes
can lead to higher long term production costs such as well
servicing and gas separation problems. If oil prices are low, it
is tempting to select a small casing size to help with current
economics. On rare occasions wells are drilled with the future
lift methods in mind.
The following sections will further detail each method of
lift with major advantages and disadvantages and how they
may be expected to perform in various well environments.
Selection by Consideration of Depth/Rate System
Capabilities. One selection criteria is the range of depth and
rate where particular lift types can function (Figure 5). This
chart is a slightly modified version of the original chart
published by R. Blais, Pennwell. The depth-rate ranges in
Figure 5 are approximate and there are many exceptions to
them, but they provide a quick idea of what systems are
available to lift with certain rates and from certain depths.
Particular well conditions can lead to wide divergences from
the initial selection from using these charts alone. Specific
designs are recommended for specific well conditions to
determine the rates possible from given depths.
Note how Figure 5 shows hydraulics systems can pump
from the greatest depths due to the U tube balancing of
produced fluid pressures with the hydraulic fluid pressure.
Gas lift has a wide range of production capacity. Beam pump
produces more from shallower depths and less from deeper
depths due to increasing rod weight and stretch as depth
increases. ESPs are depth limited due to burst limitations on
housings and energy considerations for long cables, but can

SPE 52157

SELECTION OF ARTIFICIAL LIFT

produce large production rates. Plunger lift is for low liquid

rates to primarily clear liquids from gas wells. Plunger is not
particularly depth limited, due to increased energy storage in
the casing annulus as depth increases.

Details of Major Systems.

Sucker Rod Pumping. Sucker rod pumping systems are
the oldest and most widely used type of artificial lift for oil
wells. Figure 6 shows a schematic of a typical rod pumping
system.
About 85 percent of all artificially lifted wells in the USA
are produced by rod pumps. This is also true in some areas of
S. America and Canada. About 80 percent of all oil wells are
stripper wells, making less than 10 bopd. A vast majority of
these stripper wells are lifted with sucker rod pumps. Of the
remaining 20 percent, about 27 percent are rod pumped, 52
percent gas lifted and the remainder lifted with ESPs,
hydraulic pumps and other methods of lift.
Although these statistics are ca. 1980, and are no doubt
somewhat different today, they indicate the dominance of rod
pumping for onshore operations. For offshore and higher rate
wells, the use of ESPs and especially gas lift increases
dramatically.
Sucker rod pumping systems should be considered for
new, low volume stripper wells because operating
personnel are usually familiar with these mechanically
simple systems and can operate them more efficiently.
Inexperienced personnel also can often operate rod pumps
more effectively than other types of artificial lift. Sucker
rod pumping systems can operate efficiently over a wide
range of well producing characteristics. Most of these
systems have a high salvage value.
Sucker rod systems also should be considered for lifting
moderate volumes from shallow depths and small
volumes from intermediate depths. It is possible to lift
1,000 barrels from about 7,000 feet and 200 barrels from
approximately 14,000 feet (special rods may be required).
If the well fluids contain hydrogen sulfide, sucker rod
pumping systems can lift 1,000 barrels of liquid per day
from 4,000 feet and 200 barrels per day from 10,000 feet
(exclusive of other mitigating conditions).
Most of the parts of the sucker rod pumping system are
manufactured to meet existing standards, which have
been established by the American Petroleum Institute.
Numerous manufacturers can supply each part, and all
interconnecting parts are compatible.
The sucker rod string, parts of the pump and unanchored
tubing are continuously subjected to fatigue. Therefore,
the system must be more effectively protected against
corrosion more than any other lift system to insure long
equipment life.
Sucker rod pumping systems and well dog-leg severity
are often incompatible. Deviated wells with smooth
profiles may allow satisfactory sucker rod pumping.

The ability of sucker rod pumping systems to lift sand is

limited.
Paraffin and scale can interfere with the efficient
operation of sucker rod pumping systems.
If the gas-liquid separation capacity of the tubing-casing
annulus is too low, or if the annulus is not used
efficiently, and the pump is not designed and operated
properly, the pump will operate inefficiently and tend to
gas lock.
One of the disadvantages of a beam pumping system is
that the polished rod stuffing box can leak. However, if
proper design and operating criteria are considered and
followed, that disadvantage can be minimized.
If the system is not sized to the well productivity and is
over-pumped without POC (pump-off control),
mechanical damage and inefficient pump operation will
occur.

Electrical Submersible Pumping (ESP). As an example

area where ESPs are applied extensively, THUMS Long
Beach Company was formed in April 1965 to drill, develop,
and produce the 6479 acre Long Beach Unit in Wilmington
Field, Long Beach, California. It was necessary to choose the
best method of lift for approximately 1100 deviated wells over
a 35 year contract period from four (4) man-made offshore
islands and one (1) onshore site. A schematic of a typical ESP
system is shown in Figure 7.
Advantages.
Adaptable to highly deviated wells - up to 80.
Adaptable to required subsurface wellheads 6' apart
for maximum surface location density.
Permit use of minimum space for subsurface controls
and associated production facilities.
Quiet, safe and sanitary for acceptable operations in
an offshore and environmentally conscious area.
Generally considered a high volume pump - provides
for increased volumes and water cuts brought on by
pressure maintenance and secondary recovery
operations.
Permits placing well production even while drilling
and working over wells in immediate vicinity.
Disadvantages.
Will tolerate only minimal percents of solids (sand)
production.
Costly pulling operations to correct downhole
failures (DHFs).
While on a DHF there is a loss of production during
the time well is covered by drilling operations in
immediate vicinity.
Not particularly adaptable to low volumes - less than
150 B/D gross.
Long life of ESP equipment is required to keep production
economical with high water cuts, approximately greater than

J.F. LEA, H.V. NICKENS

90%. Required improvements and recommendations based

upon experience are as follows:
Specify new stages and shaft in a rebuilt pump. Do not
reuse pumps except in a test case.
Pump designs are normally floating but use
compression to handle abrasives or to provide down
thrust resistance if cycling or gas slugging is expected or
if uncertain about rate - improves rate flexibility.
Low amperage, high voltage motors are preferred.
Motors run at 60F above ambient. Above 200F use
high temperature equipment.
Reuse motors in cool wells if cumulative run life < 1,200
days and passes QC inspection.
Use a modular (3 chamber -BSBSL) or a tandem (4
chamber) seal configuration for redundancy and
deviation angle resistance. Never reuse a seal chamber.
Use high temperature elastomers and oil where
warranted.
Cable best success is with 5KV, #4 solid conductor
(solid preferred to stranded) with barrier and braid and
heavy armor. Use cable with lead sheathing in high H2S
conditions.
Taped cable splices are preferred to molded.
Good cable handling practices are imperative in reducing
cable failures. Pull slow to prevent decompression
problems. Use 2 pre -formed super-bands especially
in deviated wells. Use no more than 7 splices per string
including motor lead extension and lower mandrel
connection. Try not to place splice near fluid level. Place
motor lead extension/round splice well above pump
outlet (~200) to keep cool. Use new cable with no
splices in hot wells.
The latest generation of motor controllers can collect and
store operational and forensic data (amps, kWh usage,
phase leakage, restart records, backspin, rotation, etc.)
and can provide restart lockout during backspin.
The Electrical Submersible Progressive Cavity Pump
(ESPCP). A schematic of a progressive cavity pump (PCP) is
shown in Figure 8. The PCP has a rotating metal rotor and a
flexible rubber molded stator. The rotating stator forms a
cavity that moves up as the rotor turns. The pump is well
suited for handling solids and viscous fluids as the solids that
move though the pump may deflect the rubber stator but do
not abrade or wear the stator or rotor to any appreciable
degree. In the past, most PCPs were powered by rotating rods
driven from the surface with a hydraulic or electrical motor.
Introduced in 1936, the PCP is of simple design and
rugged construction, and its low operating speeds enable the
pump to maintain long periods of downhole operation if it is
not subjected to chemical attack, excessive wear, or installed
at depths greater than about 4000 feet. The pump has only one
moving part downhole, with no valves to stick, clog or wear
out. The pump will not gas lock, can easily handle sandy and
abrasive formation fluids and it not plugged by paraffin,

SPE 52157

gypsum or scale.
With this system, the rotating rods would wear and also
wear the casing. The rotating rods would wind up on start
and unwind on the shut-down. Rotating rods must be sealed
at the surface and many installations would have oil leaks at
the surface.
To alleviate the problems with the conventional rotating
rod PCP systems, the ESPCP system is being made available.
This is not a new system. It has been run in Russia for a
number of years and also was available from ODI (ESP
vendor) a number of years ago. The new ESPCP system
(Figures 9-10) has a number of advantages over the rotating
sucker rod systems.
As shown in Figures 9 and 10, the PCP pump is located on
top of the assembly. There is problem of rotating the eccentric
rotor with the motor shaft because of possible vibration hence
a flexible connection is used. There is a seal section as in an
ESP assembly to protect the underlying motor from well-bore
fluids and also to accommodate and thrust in the internal
thrust bearing. Because the PCP usually turns around 3-600
rpm and the ESP motor turns around 3500 rpm under load,
there must be a way of reducing speed before the shaft
connects to the PCP.
Methods in use from the various manufacturers include
using a gear box to reduce the 3500 rpm to acceptable speeds
or using higher pole motors with lower synchronous speeds to
allow the PCP to turn at operational speeds. The motor is
located on the bottom of the assembly so fluids can pass the
motor and provide cooling as in the conventional ESP. Since
the ESPCP is not rod connected, it can be run into deviated or
horizontal wells. Some manufacturers refer to this system as
the PCSPS (Progressive Cavity Submersible Pump System) or
the ESPCP (Electrical Submersible Progressive Cavity Pump).
Advantages.
The pumping system can be run into deviated and
horizontal wells.
The pump handles solids in production well.
The pump handles viscous production well.
Several of the components are off the shelf ESP
components.
The production rates can be varied with use of a
variable speed controller (VSC).
There is one modification of this system whereby the
pump can be wire-lined out of the hole leaving the motor and
the rest of the system behind. This is desirable because the
pump is likely to have the shortest run life.
Disadvantages.
The unit does not tolerate heat due to the softening
of the stator material.
Gas must be separated to increase efficiency. It will
not gas lock but if ingesting large amounts of gas
continuously, or if pumped off, it will overheat and
damage will occur to the stator.

SPE 52157

SELECTION OF ARTIFICIAL LIFT

If the unit pumps off the well, the stator will likely
be permanently damaged.
The gearbox is another source of failure if well-bore
fluids or solids leak inside.
This pump is suited for deviated wells and can be run in
most locations of a horizontal well.
Summary. If you have a low pressure well with solids
and/or heavy oil, and the well temperature is not high, then
you could consider an ESPCP. If this is offshore or where
pulling the well is very expensive, you could consider the
option of the ESPCP that allows wire-lining out a failed pump
from the well while leaving the seal section, gear box, motor,
and cable still installed for additional usage. This modification
is in use in THUMS in Long Beach, CA.
Hydraulic Pumping. There are two kinds of hydraulic
pumps currently on the market; (1) positive displacement
pumps and (2) jet pumps. The positive displacement pump
consists of a reciprocating hydraulic engine directly coupled
to a pump piston or pump plunger (Figure 11). Power fluid
(oil or water) is directed down the tubing string to operate the
engine. The pump piston or plunger draws fluid from the well
bore through a standing valve. Exhausted power fluid and
production can be returned up a separate tubing string or up
the casing.
The jet pump is also shown in Figure 11. High pressure
power fluid is directed down the tubing to the nozzle where
the pressure energy is converted to velocity head. The high
velocity-low pressure power fluid entrains the production
fluid in the throat of the pump. A diffuser then reduces the
velocity and increases the pressure to allow the commingled
fluids to flow to the surface.
Combining the power fluid and production is called an
Open Power Fluid system. If production and power fluid are
returned up separate tubing, then this is a Parallel installation
with gas vented through the casing annulus to the surface. A
Casing installation requires the pump to handle the gas. Both
types are used with positive displacement pumps and with jet
pumps. In fact, most bottom hole assemblies can
accommodate interchangeability of jet pumps and positive
displacement pumps.
A Closed Power Fluid arrangement is where power fluid is
returned to the surface separately from the production.
Because the jet pump must commingle the power fluid and
production, it cannot operate as a Closed Power Fluid pump.
The most outstanding feature of hydraulic pumps is the
free pump (Figure 12). The drawing on the left of Figure 12
shows a standing valve (inserted by wireline) at the bottom of
the tubing and the tubing filled with fluid. In the second
drawing, a pump has been inserted in the tubing and is being
circulated to the bottom. In the third drawing the pump is on
bottom and pumping. When the pump is in need of repair, it is
circulated to the surface as shown in the drawing on the right.
The positive displacement pump, the jet pump and the closed

power fluid system previously shown above are all free

pumps.
Surface facilities required are a power fluid cleaning
system and a pump. The most common cleaning systems are
settling tanks located at the tank battery. Sometimes cyclone
de-sanders are used in addition to settling tanks. In the last
few years well site power plants have been very popular.
These are separators located at the well with cyclone desanders to remove solids from the power fluid.
Surface pumps are most commonly triplex plunger pumps.
Other types are quintiplex plunger pumps, multistage
centrifugal pumps and canned electric submersible pumps.
Surface pressure required is usually in the 1500-4000 PSI
range. It is important to specify 100% continuous duty for the
power fluid pump at the required rate and pressure. Low
volume (<10000 BPD), high pressure installations (>2500 psi)
use plunger type pumps.
Approximate maximum capacities and lift capabilities for
positive displacement pumps are shown in Table 3. In some
cases, two pumps have been installed in one tubing string.
Seal collars in the bottom hole assembly connect the pumps in
parallel hydraulically. Thus, the maximum displacement
values shown above are doubled.
A tabulation of capacity vs. lift is not practical for jet
pumps because of the many variables and their complex
relationships. To keep fluid velocities below 50 ft/sec. in
suction and discharge passages, the maximum production
rates vs. tubing size for Jet FREE PUMPS are approximately
as shown in Table 4.
Fixed type jet pumps (those too large to fit inside the
tubing) have been made with capacities to 17,000 B/D. Even
larger pumps can be made. Maximum lifting depth for jet
pumps is around 8000-9000 feet if surface power fluid
pressure is limited to 3500 PSI. The maximum capacities
listed above can be obtained only to about 5000-6000 feet.
These jet pump figures are only guidelines because well
conditions and fluid properties can have significant influences
on them. It should also be noted that the maximum capacities
listed above are for high volume jet pumps that require bottom
hole assemblies that are not capable of also accommodating
piston pumps.
Advantages.
FREE PUMP - Being able to circulate the pump in
and out of the well is the most obvious and
significant feature of hydraulic pumps. It is
especially attractive on offshore platforms, remote
locations, populated, and agricultural areas.
Deep Wells - Positive displacement pumps are
capable of pumping depths to 17,000 feet, and
deeper. Jet pumps can be run to 20,000 feet with
design target of 25% submergence at intake.
Speed Control - By changing the power fluid rate to
pumps, production can be varied from 10 percent to

J.F. LEA, H.V. NICKENS

100 percent of pump capacity. The optimum speed

range is 20 to 85 percent of rated speed.
Crooked Wells - Deviated wells typically present no
problem to hydraulic free pumps. Jet pumps can
even be used in TFL installations.
Sand Production - jet pumps, because they have no
moving parts, can handle sand and other solids with
hardened nozzle throats.
Viscous Oils - Positive displacement pumps can
handle viscous oils very well. The power fluid can
be heated or it can have diluent added to further aid
getting the oil to the surface.
Corrosion - Corrosion inhibitors can be injected into
the power fluid for corrosion control.

Disadvantages.
Power Fluid Cleaning - Removing solids from the
power fluid is very important for positive
displacement pumps. Surface plunger pumps are
also affected by solids in the power fluid. Jet pumps,
on the other hand, are very tolerant of poor power
fluid quality.
Pump Life - Positive displacement pumps, on
average, have shorter life between repairs than Jet,
sucker rod and electric submersible pumps. Mostly,
this is a function of the quality of power fluid, but
also, on average, they are pumping from greater
depths which is also a factor. Jet pumps, on the other
hand, have very long pump life between repairs
without solids or if not being subjected to cavitation.
Bottom Hole Pressure - Positive displacement
pumps can pump to practically zero bottom hole
pressure in the absence of gas interference and other
problems (lowest bottom hole pressure is a gas
gradient to the pump depth plus casing pressure) Jet
pumps cannot pump to low intake pressures. Jet
pumps require approximately 1000 PSI bottom hole
pressure when set at 10,000 feet and approximately
500 PSI when set at 5000 feet.
Skilled Personnel - Positive displacement pumps
generally require more highly skilled operating
personnel, or perhaps, just more attention, than jet
pumps and other types of artificial lift. There are
two reasons for this. First, pump speed needs to be
monitored daily and not allowed to become
excessive. Secondly, power fluid cleaning systems
need frequent checking to keep them operating at
their optimum effectiveness.
To answer the question, when do you use jet pumps and
when do you use positive displacement pumps?, our answer
is: Use jet pumps if the flowing (pumping) bottom hole
pressure is adequate (see disadvantage No. 3 above).
Gas Lift. Gas lift dominates the USA Gulf Coast and is
used extensively around the world. Most of these wells are on

SPE 52157

constant flow gas lift. Thus, the questions: Why choose gas
lift?, Where do you use constant flow? and When do you
select intermittent lift?
Constant Flow Gas Lift. A schematic of a gas lift system
is shown in Figure 13. Constant flow gas lift is recommended
for high volume and high static bottom hole pressure wells
where major pumping problems will occur. It is an excellent
application for offshore classic-type formations with water
drive, or waterflood reservoirs with good productivity indices
(PIs) and high gas-oil ratios (GORs). When high pressure
gas is available without compression or where gas is low in
cost, gas lift is especially attractive. Constant flow gas lift
supplements the produced gas with additional gas injection to
lower the intake pressure to the tubing, including lowering
formation pressure.
A reliable, adequate supply of good quality high-pressure
lift gas is mandatory. This supply is necessary throughout the
producing life if gas lift is to be effectively maintained. In
many fields the produced gas declines as water cut increases,
requiring some outside source of gas. The gas lift pressure is
typically fixed during the initial phase of the facility design
and as the water cut increases the depth of lift is decreased not
allowing the gas lift system to obtain the desired flowing
bottom hole pressure. Also the wells will produce erratically
or not at all when the lift supply stops or pressure fluctuates
radically. Poor quality gas will impair or even stop
production. Thus, the basic requirement for gas must be met
or other artificial lift means should be installed.
Constant flow gas lift imposes a relatively high back
pressure on the reservoir compared to pumping methods and
is at best only moderately efficient. The high back pressure
may significantly reduce production as compared with some
pumping methods and poor efficiency significantly increases
both capital cost and operating energy costs.
Advantages.
Gas lift is the best artificial lift method for handling
sand or solid materials. Many wells make some sand
even if sand control is installed. The produced sand
causes almost no mechanical problem to the gas lift
valve; whereas, only a little sand plays havoc with
most pumping methods.
Deviated or crooked holes can be gas lifted with
only minor lift problems. This is especially
important for offshore platform wells which are
directionally drilled.
Gas lift permits the use of wireline equipment and
such equipment is easily and economically serviced.
This feature allows for routine repairs through the
tubing.
The normal design leaves the tubing full opening.
This permits use of BHP surveys, sand sounding and
bailing, production logging, cutting, paraffin, etc.

SPE 52157

SELECTION OF ARTIFICIAL LIFT

High formation GORs are helpful rather than being

a hindrance. Thus in gas lift, less injection gas is
required; whereas, in all pumping methods, pumped
gas reduces efficiency drastically.
Gas lift is flexible. A wide range of volumes and lift
depths can be achieved with essentially the same
well equipment. In some cases, switching to annular
flow can also be easily accomplished to handle
exceedingly high volumes.
A central gas lift system can be easily used to
service many wells or operate an entire field.
Centralization usually lowers total capital cost and
permits easier well control and testing.
Gas lift has a low profile. The surface well
equipment is the same as for flowing wells except
for injection gas metering. The low profile is usually
an advantage in urban environments.
Well subsurface equipment is relatively inexpensive
and repair and maintenance of this subsurface
equipment is normally low. The equipment is easily
pulled and repaired or replaced. Also major well
workovers occur infrequently.
Installation of gas lift is compatible with subsurface
safety valves and other surface equipment. Use of
the surface controlled subsurface safety valve with
the 1/4-inch control line allows easy shut-in of the
well.
Gas lift will tolerate some bad design assumptions
and still work. This is fortunate since the spacing
design must usually be made before the well is
completed and tested.

Disadvantages.
Relatively high back pressure may seriously restrict
production in continuous gas lift. This problem
becomes more significant with increasing depths
and declining static BHPs. Thus a 10,000 foot well
with a static BHP of 1000 psi and a PI of 1.0 would
be difficult to lift with the standard constant flow
gas lift system. However, there are some special
schemes that could be tried for such wells.
Gas lift is relatively inefficient, often resulting in
large capital investments and high energy operating
costs. The cost of compressors is relatively high and
are often long delivery items. Costs in 1981 were
found to be $500 to $600 per horsepower for typical
land locations and $1000 to $1400 per horsepower
for offshore packages. The compressor presents
space and weight design problems when used on
offshore platforms. Also, the cost of the distribution
systems onshore may be significant. Increased gas
usage also may increase the size of flow line and
separators needed.
Adequate gas supply is needed throughout life of
project. If the field runs out of gas or if gas becomes

too expensive, one may have to switch to another lift

method. In addition, there must be enough gas for
easy start-ups.
Increasing water cut increases the flowing bottom
hole pressure with a fixed gas lift pressure. At some
water cut, another form of lift, such as ESPs,
should be evaluated to increase production be
reducing the flowing bottom hole pressure,
especially if the produced gas is low.
Operation and maintenance of compressors can be
expensive. Skilled operators and good compressor
mechanics are required for successful and reliable
operation.
There is increased difficulty when lifting low
gravity (less than 15 API) crude due to greater
friction. The cooling effect of gas expansion further
aggravates this problem. Also the cooling effect will
compound any paraffin problem.
Low fluid volumes in conjunction with high water
cuts (less than 200 BPD in 2-3/8" OD tubing)
become less efficient to lift and frequently severe
heading is experienced.
Good data are required to make a good design. Such data
may not be available and you may have to continue
operations with an inefficient design that does not
produce the well to capacity.
Potential gas lift problems that must be resolved.
Freezing and hydrate problems in injection gas lines.
Corrosive injection gas.
Severe paraffin problems.
Fluctuating suction and discharge pressures.
Wireline problems.
Dual artificial lift frequently results in poor lift
efficiency.
Changing well conditions, especially decline in BHP
and PI.
Deep high volume lift.
Valve interference multi-pointing.
Emulsions forming in the tubing

Intermittent Gas Lift. Intermittent gas lift method is

generally used on wells that produce low volumes of fluid
(~<200 bpd). Wells where intermittent lift is recommended
normally have the characteristic of (1) high PI and low bottom
hole pressure (BHP ) or (2) low PI with high BHP. Its use
stems from known major pumping problems or where
continuous gas lift is already installed or low cost high
pressure gas is available.
If an adequate, good quality, low cost gas supply is
available and plans are to lift a relatively shallow, high GOR,
low PI or low BHP well with a bad dog-leg that produces
some sand, then intermittent gas lift would be an excellent
choice. Intermittent gas lift has many of the same
advantages/disadvantages as constant flow gas lift, and the

J.F. LEA, H.V. NICKENS

major factors to be considered are similar. Only the

differences will be highlighted in the ensuing discussion.
Advantages.
Intermittent gas lift has a significantly lower
producing BHP than the constant flow methods.
It has the ability to handle low volumes of fluid with
relatively low production BHPs.
Disadvantages.
Intermittent gas lift is limited to low volume wells.
For example an 8,000 foot well with 2" nominal
tubing can seldom be produced at rates of over 200
BPD with an average producing pressure much
below 250 psig. Smaller sizes of tubing have even a
lower maximum rate.
The average producing pressure of a conventional
intermittent lift system is still relatively high when
compared to rod pumping. However, the producing
BHP can be reduced by use of chambers. Chambers
are particularly suited to high PI, low BHP wells.
The output to input horsepower efficiency is low.
More gas is used per barrel of produced fluid than
with constant flow gas lift. Also the slippage
increases with depth and water cut making the lift
system even more inefficient. However, slippage
can be reduced by use of plungers. In general if the
cycle time allows time for the plunger to fall, then
plunger should be used with intermittent lift if not
solids are present.
The fluctuation in rate and BHP can be detrimental
to wells with sand control. The produced sand may
plug the tubing or standing valve. Also surface
fluctuations cause gas and fluid handling problems.
Intermittent gas lift requires frequent adjustments.
The lease operator must alter the injection rate and
time period routinely to increase the production and
keep the lift gas requirement relatively low.
Conclusion. Gas lift has numerous strengths that in many
fields make it the best choice of artificial lift. However, there
are limitations and potential problems to be dealt with. One
has a choice of using either constant flow for high volume
wells or intermittent for low volume wells and there is little
difficulty in switching from one to the other. In addition, gas
lift can be used to kick off wells, unload water from gas wells,
or back flow injection wells. Gas lift deserves serious
consideration as a means of artificial lift.
Other Methods of Lift. There are other methods of lift that
will not be discussed in a paper of this length. Plunger lift is
commonly used to remove liquids from gas wells to maintain
a low gradient in the tubing. There is a chamber pump that
relies on gas pressure to periodically empty the chamber and
force the fluids to the surface with no gas mixing.

SPE 52157

Rather than try to examine all possible forms of lift, the

rest of the discussion below will be relegated to methods that
can be used to select the best form of lift for a particular
application. The methods should be applicable to any form of
lift under consideration.
Selection by Advantages and Disadvantages. Although the
previous sections detailed the major systems of artificial lift
and listed some advantages/disadvantages, there are other
more detailed listings of advantages and disadvantages.
Reference 1 by Clegg, Bucaram & Hein is the most
extensive and useful listing of the various advantages and
disadvantages of various lift systems under a broad range of
categories. Some of their information is open to discussion
such as their low limit on gas lift with viscous fluids, but in
general it is the best available list of artificial lift advantages
and disadvantages.
Tables 1 and 2 after Brown (Ref. 2) do provide a useful
summary of the various advantages and disadvantages of the
various lift systems that were described briefly in the
proceeding sections. It is probable that many artificial lift
systems have been and can be selected by using tables similar
to those generated by Clegg, Bucaram & Hein (Ref. 1) or the
Tables 1 & 2 repeated here from Brown (Ref. 2). However
there are other important considerations beyond a list of
advantages/disadvantages. Reference 3 is a brief summary of
advantages/disadvantages and selection criteria for various lift
systems separately presented by experts in a forum discussion.
Selection by Expert Programs. An advancement that should
be a step above a simple list of advantages/disadvantages is
the advent of expert programs or computerized lift selection
programs. These programs include rules and logic so they will
branch to select the best system of lift as a function of user
input of well and operating conditions. References 4-6 are
publications dealing with expert systems for the selection of
artificial lift systems.
Reference 4 describes an expert system with selection
criteria on:
1. Sucker rod
2. Hydraulic pump
3. ESP
4. Progressive pump
5. Continuous gas lift
6. Intermittent gas lift
7. Intermittent gas lift with plunger
8. Constant slug injection gas lift
9. Chamber gas lift
10. Conventional plunger lift
The program contains; an (1) Expert Module, a (2) Design
Module, and a (3) Economic Module. Module 1 is an expert
module that includes a knowledge base structured from human
expertise, theoretical written knowledge available and known
rule of thumb type calculations. Module 2 incorporates
simulation design and facility component specification
programs for all lift methods considered. Module 3 is an

SPE 52157

SELECTION OF ARTIFICIAL LIFT

economics evaluation module that includes a costs data base,

and cost analysis programs to calculate lift profitability.
Module 1 ranks the methods and also issues some
warnings, some of which may rule out high ranked methods.
Module 2 contains a suite of design methods with advice to
follow from Module 1. Module 3 utilizes the designs and
expected production rate to calculate profitability using
evaluation parameters such as net present value and rate of
return. It also includes investment costs and repair and
maintenance costs.
Reference 5 describes the program AL which decides,
from user input, what system among gas lift, hydraulic, sucker
rod or ESP pumping systems, is best for particular conditions.
Problems such as sand, paraffin, crooked hole, corrosion,
small casing, flexibility, and scale are used, with the stored
knowledge base and user input, to allow the program to rank
by score, the most appropriate method of lift for particular
conditions. Details of the programs input, structure, and
output are contained in the reference.
Reference 6 describes another expert system, which is very
encompassing in scope. The reference describes the OPUS
(Optimum Pumping Unit Search) program, later described
commercially as the Artilip program. The program consists of;
(a) a knowledge base containing the complete set of specific
information on the domain of expertise; (b) an inference
engine using the data and heuristics of the knowledge base to
solve the problem and (c) interactive modules enabling very
simple use of the expert system.
The structure of the rules in Reference 5 is of the form,
if (condition) then ( type of process)
For each process (i.e., lift method) , a suitability
coefficient (SC) from 1 to + 1 for the given condition is
defined, where SC = 1 eliminates the process from further
consideration and SC = +1 indicates a process well suited to
the given condition.
For example, the simple expression
if (Pump Temperature > 275F) then (ESP), -1
defines a rule that eliminates ESPs if the pump temperature
exceeds 275F (although this rule would have to be updated to
include use of hot line ESPs).
Intermediate values can be used to refine the system and
methods are presented for combining the coefficients into a
single coefficient. The program can combine the suitability
coefficients into one value for over-all evaluation. Other
details for the knowledge representation and the technical
and economic evaluation are given in the paper.
Another interesting feature (Ref. 6) is the presentation of
economical data for annual costs to be incurred by various lift
systems. The costs are presented in bar graphs showing how
the component costs would occur above the well head or
subsurface. For instance, much of the possible re-occurring
costs for ESPs can be subsurface whereas for gas lift, other

than wireline work, larger repair and servicing costs

associated with compressors would be above ground.
Selection by Net Present Value (NPV) Comparison. A
more complete selection technique will depend upon the lifetime economics of the available lift methods. The economics
in turn depend, for example, upon the failure rates of the
system components, fuel costs, maintenance costs, inflation
rates, anticipated revenue from produced oil and gas and other
factors that may vary from system to system. Reference 7-9 by
Etherton, et. al., by Smith and by Kol, et. al. are example
studies in selection that follow economically guided selection
techniques.
References 10-16 provide various references on artificial
lift in general, efficiency of lift methods, selection techniques
and limitations on various lift systems.
This section will illustrate the economic concepts for a
low-rate and high rate example. Methods considered are ESP,
gas lift, hydraulic pump and rod pump. It will use an enhanced
method of analysis similar to calculation methods available in
Ref. 17. The equations used in the following example analysis
are given in detail in the Appendix.
In order to use the NPV comparison method, the user must
have a good idea of the associated costs for each system. This
requires that the user evaluate each system carefully for his
particular well and be aware of the advantages/disadvantages
of each system and additional equipment (i.e., additional
costs) that may be required. Since energy costs are included in
the NPV analysis, an optimal design for each feasible method
must be determined before running the NPV analysis.
These factors force the user to consider all the selection
methods discussed previously to generate the necessary
information for the NPV analysis.
Low-Rate Example. Consider an example well with the
following characteristics.
Vertical Depth to Perforations
6000 ft
Separator Pressure
100 psig
Surface Temperature
100 F
Casing Size
7 inch
Tubing Size
3.500 inch, Gas Lift
2.875 inch, Other
Methods
Water Cut
50 %
Oil Gravity
30 API
Gas Gravity
.7
Water Gravity
1.03
Produced GOR
400 scf/bbl
Bubble Point
2227 psig
Static Reservoir Pressure
2000 psig
Productivity Index
1 STB/psi
There are common costs and economic variables that are
the same for all the different methods.
Fixed Costs
300 $/month
Fluid Disposal
.35 $/bbl water
Electricity
.05 $/kW-hr
Oil Revenue
12 $/bbl

10

J.F. LEA, H.V. NICKENS

Gas Revenue
1.25 $/Mscf
Inflation Rate
3 %/yr.
Discount Rate for Present Value
8 %/yr.
Oil Revenue Increase
1 %/yr.
To calculate the expected lifetime of the well, reasonable
reservoir production estimates must be supplied. For this
example, we assume that all lift methods (ESP, Gas Lift,
Beam Pump & Hydraulics) will produce initially at the same
rate, 1000 bbl/day with 50% water cut and 400 GOR. The
reservoir is assumed to decline immediately at 20 %/year
reduction in oil rate. Water cut is assumed to increase
maintaining the rate constant (water injection). At 90% water
cut, the simulation is stopped. The GOR is assumed to remain
constant for this example.
The actual possible initial production rate would differ for
each method, but for comparison purposes and to illustrate the
concepts, a rate of 1000 bbl/day for each method is assumed.
Different rates would possibly require different production
facilities and different initial costs. Thus each method should
be optimized and the associated required costs included in the
economic analysis.
Method specific costs must also be included as shown in
Tables 5-9.
Run Life Tables. The average pump run life for the pumping
systems, and the injected gas volume for gas lift, is required to
estimate the life-time costs for the different methods. The
values assumed for this example are listed in Table 10. The
last value in each table is used for subsequent years.
The different methods are compared by calculating the net
present value (NPV) income as a function of time until the
production rate decreases to the abandonment rate. This gives
a direct comparison of the different methods in terms of the
net revenue the well would be expected to produce.
Figure 14 shows results for the assumptions of this
example. Rod pumping would be the best method, showing to
be slightly more profitable than ESP, over the ~7 year life of
the well. The analysis is stopped at ~7 years when 90% water
cut is reached. These analyses also can be run for depleting
rates.
The NPV and total lifetime costs for each method are
summarized in Table 11. The operating costs are significant,
ranging from 14-26% of the NPV for this low rate example.
Reduction in operating costs could therefore be a significant
factor in selecting the optimum lift method.
To re-emphasize, the results will depend upon the
particular cost related data for each method. For this case,
however, it is likely that rod pump or ESP would be the most
economical method depending on the detailed cost data, and
gas lift and hydraulic pump would not be recommended.
High Rate Example. A well with productivity index PI =
24 bbl/d/psi is considered. Rod pumping cannot deliver the
rates required for this design and is eliminated from
consideration. Artificial lift designs with jet pump, ESP and
gas lift yield only are considered.
Production GOR and water cut is constant at 100 scf/bbl
and 1% respectively. Abandonment rate is 500 bbl/d, oil +

SPE 52157

water. The reservoir oil production declines at 20% year.

Operation and equipment specific costs are provided in
Tables 12-15.
Fixed Costs
1000 $/month
Water Disposal
.15 $/bbl
Electricity
.05 $/kW-hr
Oil Revenue
12 $/bbl
Gas Revenue
1.25 $/Mscf
Inflation Rate
3 %/yr.
Discount Rate for Present Value 8 %/yr.
Oil Revenue Increase
1 %/yr.
Run Life Tables. The average pump run life for the
pumping systems, and the injected gas volume for gas lift, is
required to estimate the life-time costs for the different
methods. The values assumed for this example are listed in
Table 16. The last value in each table is used for subsequent
years.
Figure 15 shows results for the assumptions of this
example. jet pump would be the best method, producing about
$4,000,000 more NPV than the ESP and $11,000,000 more
than gas lift over the life of the well. Although the curves
show that the NPV is near the point of no further increase, the
curves are terminated at the point where the abandonment rate
of 500 bpd occurs.
The NPV and total lifetime costs for each method are
summarized in Table 17. The operating costs are relatively
insignificant for this high rate case, ranging from 1.6 2.0 %
of the NPV for this high rate example. Operating costs
therefore are not as significant a factor in selecting the
optimum lift method for high rate wells.
Run Life Information. As shown in the above examples of
selection of artificial lift, one of the main factors in many
cases is knowing what the failure rates of are for the various
possible systems or the individual components of the systems.
Some typical examples of failure rates and costs are discussed
here.
Figure 16 shows failure rates from a group of 500+ beam
pumped wells over a period of years. The costs for downhole
lift replacement and servicing are shown over the bars on the
figure.
Figure 17 shows a pie chart breakdown of the major
causes for failure of the beam pump systems that went into the
accumulation of the failure rate data plot. If a lift selection
study is needed, data such as this for a field of similar
conditions would be very helpful in evaluating beam pump as
a candidate and being able to compare beam pump to others
methods of lift.
The conditions for the data presented here are a mix of
deep and shallower wells. The deep wells average about
8,880'-9,200'. They make about 926 BOPD and 8,562 BWPD.
The unit size averages out to be close to a 640. The shallower
wells will average to be about 4,200' deep. They make 2,885
BOPD and 37,577 BWPD. For the pumping unit they average
out to be about a 320. Different field conditions and methods
of operations will result in different failure rates and a

SPE 52157

SELECTION OF ARTIFICIAL LIFT

different distribution of failed components.

Figure 18 shows run lives of ESPs in the THUMS
operations offshore Long Beach, CA. There have been a
number of improvements to the submersible system at
THUMS that have been implemented over a 16 year period
that are responsible for increasing the mean time between
failures from 320 days in 1983 to over 1100 days in 1997.
These statistics are based on any cause that required the
downhole equipment to be pulled. Table 18 gives a summary
of the conditions in this operation.
Figure 19 shows ESP run lives for various fields. This
data was collected and presented in Reference 6 below for a
study for a study of artificial lift feasibility and what method
to use in a Siberian location. Table 19 shows one panel of
information collected for ESPs and details the costs and
equipment for a 900 bpd installation based on data collected
from other similar fields. This application detailed in
Reference 6 was for deep deviated wells drilled from onshore
locations in a marshy environment. Component service lives
are shown in Table 20. Targets and downside potentials were
established for this study as shown in Figure 19.
Reference 18 also includes various run life information
and selection criteria. Swan Hills (Alberta), Milne Point
(Alaska), the Amoco Congo field, the THUMS E. Wilmington
field, and the Amoco N. Sea field, the Montrose field were
used to help predict run lives for the Priobskoye field in
Siberia. More information on the conditions present in these
fields can be found in Reference 6. The learning curve
aspect of these fields is costly showing the time required to
come up to reasonable operational lives for the ESP
installations.
Table 21 shows some downhole hydraulic pump lives for a
collection of fields. The conditions for these fields are
presented in Reference 6 as well. The average life of the
pumps is about 114 days. Target, downside and industry data
is summarized in Figure 20 for the downhole hydraulic
pumps. Table 22 summarizes costs for operating with
Hydraulics in the study of Reference 6 for a 1000 bpd rate.
These type of costs would have to be gathered for a number of
rates and conditions for the data to be available for general
application to an artificial lift study.
No data is presented for gas lift system costs and failures
expected. Initial compressor costs are high but after
installation, most of the expenses are wireline work and new
or repaired valves, unless a major compressor fix or addition
is needed. Cost examples for other systems are not shown
here.
The data that is shown is, again, for particular fields and
may/may not be indicative of what you would expect for a
study that might be undertaken with other conditions present.
Conclusions
Discussion has been presented on various methods available
for the selection of the best artificial lift system for given
conditions. The discussion reviews methods from a depth-rate
feasibility map, tables of advantages and disadvantages,

11

through expert system programs containing feasibility,

technical, and economic programs.
Examples are given for calculating the net present value of
artificial lift methods as one example of how to economically
select the best method of lift. The examples show what data is
needed to allow the engineer to be able to make a choice that
should maximize profits over the life or a portion of the life of
the field. The examples presented assume that the user of the
NPV method has used previously mentioned methods of
reviewing advantages and disadvantages, and other feasibility
methods to be sure that the systems economically analyzed
can be used for given conditions.
Since the NPV method reviewed requires designs to meet
target rates, then the user is somewhat forced to evaluate harsh
conditions, etc., during the course of the design. He has to
then add gas separators, sand trim, or whatever is necessary to
try to meet target rates before the NPV analysis is performed,
so by necessity, various feasibility criteria has to be
considered. If target rates are not possible, then the system is
eliminated from consideration.
The reader is left with a menu of various possibilities for
selection of a lift method varying from use of simple charts
and tables, to economic analysis calculations, to use of
existing or future expert system computer programs.
References
1. Clegg, J.D., Bucaram, S.M. and Hein, N.W., New
Recommendations and Comparisons for Artificial Lift Method
Selection, SPE 24834; and Journal of Petroleum Technology,
1128, December 1993.
2. Brown, K.E., Overview of Artificial Lift Systems, Journal of
Petroleum Technology, 2384, October 1982.
3. Panel Discussion, Neely, B. Moderator, Selection of Artificial
Lift Method, SPE 10337.
4. Espin, D.A., Gasbarri, S. and Chacin, J.E., Expert System for
Selection of Optimum Artificial Lift Method, SPE 26967.
5. Heinze, L.R., Thornsberry, K. and Wit, L.D., AL: An Expert
System for Selecting the Optimal Pumping Method, SPE 18872.
6. Valentin, E.P. and Hoffman, F.C., OPUS: An Expert Advisor for
Artificial Lift, SPE 18184.
7. Etherton, J.H. and Thornton, P., A Case Study of the Selection
Procedure for Artificial Lift in a High Capacity Reservoir, SW
Petroleum Short Course 88.
8. Smith, G.L., Lease Operational Study Gas Lift vs. Submersible
Pump Lift G.H. Arledge C Lease, Scurry County, Texas,
SPE 6852.
9. Kol, H. and Lea, J.F., Selection of the Most Effective Artificial
List System for the Priobskoye Field, SPE ESP Workshop, April
26-28, Houston, TX.
10. Clegg, J.D., Artificial Lift Efficiency Depends on Design, The
American Oil & Gas Reporter, 48, June 1991.
11. Lea, J.F., Artificial LiftOperating at Lower Cost, SPE
Distinguished Lecturer Presentation.
12. Clegg, J.D., Artificial Lift: Producing at High Rates, SPE
Distinguished Lecturer Presentation.
13. Johnson, L.D., Selection of Artificial Lift for a Permian Basin
Waterflood, SW Petroleum Short Course, Lubbock, 1968.
14. Bennett, P., Artificial Lift Concepts and Timing, Petroleum
Engineer, 144, May 1980.

12

J.F. LEA, H.V. NICKENS

15. Weighhill, G.T., ESP Selection and Operating Strategy at Wytch

Farm.
16. Powers, M.L., The Depth Constraint of Electric Submersible
Pumps, SPE 24835.
17. Toolkit of computer programs by Integrity Consulting, Parker,
CO.
18. Lea, J.F., Patterson, J., Selection Considerations for Artificial
Lift, Artificial Lift Equipment Forum, Dubai, 1997.

Appendix
The economic equations used for the selection of lift methods
by economic analysis are summarized in this Appendix. The
equations are presented as pseudo code for readability. Values
not explicitly calculated are assumed to be user input values.
Initial Oil Rate (BBL/YR)
=
365.25 x Initial
Production Rate x
Initial Water Cut
Abandonment Oil Rate
=
365.25 x Total
Abandonment Rate
x Abandonment
Water Cut
Rdecl
=
Oil Production
Decline Rate/100
Years to Abandonment
=
- (Ln(Initial Oil
Rate) Ln(Aband.
Oil Rate))/Ln(1
Rdecl)
Initialize at Year 0
BOPD(0)
Water Cut(0)
GOR(0)
Cumulative NPV (0)
Cumulative Oil (0)

=
=
=
=
=

Initial Oil Rate / 365.25

Initial Water Cut
Initial GOR
0
0

Calculate production decline factor

R = Ln(1 Rdecl)
Begin loop to calculate costs and present value up to the
abandonment year.
FOR YEARS I = 1 TO YEARS TO ABANDONMENT
Calculate daily production rate at end of year 1
BOPD (I)
=
BOPD(0) x Exp(R* I)
Calculate maximum production for year I for full
production
Qmax (I)
=
365.25 * (BOPD (I)
BOPD (I - 1)) / R
Adjust for lost production
Qoil (I)
=
Qmax (I) - (Qmax (I) /
365.25) x
Days/Workover x
Workovers/Year
Calculate cumulative oil produced to end of year I
Cumulative Oil (I)
= Cumulative Oil (I-1) +
Qoil (I)
Straight line GOR and Water Cut

SPE 52157

WC (I)

WC (0) + I x
(Abandonment WC Initial WC)/ Years to
Abandonment
GOR (I)
= GOR (0) + I x
(Abandonment GOR Initial GOR)/ Years to
Abandonment
Calculate Water and Gas Rates for Year I
Qwat (I) = Qoil (I) x WC (I) / (1 WC (I))
Qgas (I) = .001 x Qoil (I) x GOR (I)
Calculate Required Cost and Revenue Factors
Rinflation
= (1 + Inflation Rate / 100) ^
(I - 0.5)
Rdiscount
= (1 + Discount Rate / 100)
^ (I - 0.5)
Roil
= (1 + Oil Price Increase
Rate/ 100) ^ (I - 0.5)
Requip
= (1 + Equipment Cost
Increase Rate / 100) ^ (I 0.5)
Relec
= Electricity cost / bbl
liquid produced = 24 x
kW/blpd x $/kW
Calculate Fluid Costs
Fluid Cost (I)
= Rinflation x Fluid
Disposal Cost/BBL x
(Qoil (I) + Qwat (I) )
Calculated Fixed Operating Cost
Fixed Cost (I)
= Rinflation x 12 x
(Common Fixed Cost +
Method Specific Fixed
Cost)
Calculate Workover Cost
Workover Cost (I)
= Rinflation x
Cost/Workover Day x
Days/Workover x
Workovers/Year
Calculate Equipment Cost. Equipment costs are specified
for each lift method and vary from method to method. For
each method, the type of equipment (pump, sucker rods,
tubing, ESP cable, etc), replacement cost for each type and the
anticipated frequency of replacement (either as a run life table
or a fixed replacement interval) are specified.
Total equipment cost is calculated by summing over all
identified method specific equipment, including the equipment
cost ONLY during those years where replacement is
scheduled from the run life table or specified fixed
replacement interval.
Equipment Cost (I)
= Equipment Cost (I-1)
FOR J
= 1 to Number of
Equipment Types
IF year I is a replacement year for Equipment (J) THEN

SPE 52157

Equipment Cost (I)

SELECTION OF ARTIFICIAL LIFT

ENDIF
END FOR
Calculate Electricity Cost
Electricity Cost (I)
=

Equipment Cost (I) +

Requip x Cost of
Equipment (J)

Rinflation x Relec x
(Qoil (I) + Qwat (I))
Calculate Total Costs for Year 1
Yearly Cost (I)
= Fluid Cost (I) + Fixed
Cost (I) + Workover
Cost (I) + Equipment Cost
(I) + Electricity Cost (I)
Calculate Total Income for Year I
Yearly Income (I)
= Roil x (1 Royalty/100)
x (Qoil (I) x $/BBl Oil
+ Qgas (I) x $/Mscf)
Calculate Net Present Value from Year I
Net PV (I)
= (Yearly Income (I) Yearly Cost (I)) /
Rdiscount
Calculate Cumulative Net PV from Year 0 to Year I
Cumulative NPV (I)
= Cumulative NPV (I-1) +
Net PV (I)
END FOR

13

14

J.F. LEA, H.V. NICKENS

SPE 52157

Table 1: Relative Advantages of Artificial Lift Systems

(After K. E. Brown, JPT, Oct., 1982)

Rod Pumping
Relatively simple
system design
Units easily
changed to other
wells with
minimum cost

Hydraulic
Piston
Pumping

Electric
Submersible
Pumping

Not so depth
limited-can lift
large volumes
from great depths

Can lift
extremely high
volumes,
20,000 B/D
(19078 m 3/d),
in shallow wells
with large
casing.

Efficient, simple
and easy for field
people to operate.

500 B/D (79.49 m

3
/d) from 15,000
ft. (4572 m) have
been installed to
18,000 ft.
(5486.4 m)

Applicable to slim
holes and multiple
completions.

Crooked holes
present minimal
problems.

Can pump a well

down to very low
pressure (depth
and rate
dependent).

Unobtrusive in
urban locations.

System usually is
naturally vented
for gas separation
and fluid level
soundings.
Flexible-can
match
displacement rate
to well capability
as well declines.

Power source can

be remotely
located.
Analyzable.
Flexible-can
usually match
displacement to
wells capability as
well declines.
Can use gas or
electricity as
power source.

Analyzable.
Can lift hightemperature and
viscous oils.

Downhole pumps
can be circulated
out in free
systems.

Can use gas or

electricity as
power source.

Can pump a well

down to fairly low
pressure.

Corrosion and
scale treatments
easy to perform.

Applicable to
multiple
completions.

Applicable to
pump off control if
electrified.

Applicable
offshore.

Availability of
different sizes.
Hollow sucker
rods are available
for slim hole
completions and
ease of inhibitor
treatment.
Have pumps with
double valving
that pump on both
upstroke and
downstroke.

Closed system will

combat corrosion.
Easy to pump in
cycles by time
clock.
Adjustable gear
box for Triplex
offers more
flexibility.
Mixing power fluid
with waxy or
viscous crudes
can reduce
viscosity.

Currently lifting
120,000 B/D
(19068 m 3/d)
from water
supply wells in
Middle East
with 600-hp
(448-kW) units;
720-hp (537kW) available,
1,000-hp (746kW) under
development.
Unobtrusive in
urban locations.
Simple to
operate.
Easy to install
downhole
pressure
sensor for
telemetering
pressure to
surface via
cable.
Crooked hole
present no
problem.
Applicable
offshore.
Corrosion and
scale treatment
easy to
perform.
Availability in
different size.
Lifting cost for
high volumes
generally very
low.

Gas Lift
Can handle
large volume of
solids with
minor
problems.
Handles large
volume in highPl wells
(continuous
lift). 50,000
B/D (7949.37
m 3/d).
Fairly flexibleconvertible
from
continuous to
intermittent to
chamber or
plunger lift as
well declines.
Unobtrusive in
urban
locations.
Power source
can be
remotely
located.
Easy to obtain
downhole
pressures and
gradients.
Lifting gassy
wells is no
problem.
Sometimes
serviceable
with wireline
unit.
Crooked holes
present no
problem.
Corrosion is
not usually as
adverse.
Applicable
offshore.

Hydraulic
Jet Pump

Plunger lift

Retrievable without
pulling tubing.

Retrievable without
pulling tubing.

Has no moving parts.

Very inexpensive
installation.

No problems in
deviated or crooked
holes.
Unobtrusive in urban
locations.

Automatically
keeps tubing clean
of paraffin, scale.
Applicable for high
gas oil ratio wells.

Applicable offshore.
Can use water as a
power source.
Power fluid does not
have to be so clean
as for hydraulic piston
pumping.
Corrosion scale
emulsion treatment
easy to perform.
Power source can be
remotely located and
can handle high
volumes to
30,000 B/D (4769.62
m 3/d).

Can be used in
conjunction with
intermittent gas lift.
Can be used to
unload liquid from
gas wells.

Progressive
Cavity
Pumps
Some types
are retrievable
with rods
Moderate Cost
Low Profile
Can use
downhole
electric motors
that handle
sand and
viscous fluid
well
High electrical
efficiency

SPE 52157

SELECTION OF ARTIFICIAL LIFT

15

Table 2: Relative Disadvantages of Artificial Lift Systems

Rod
Pumping
Crooked
holes present
a friction
problem.
High solids
production is
troublesome.
Gassy wells
usually lower
volumetric
efficiency.
Is depth
limited,
primarily due
to rod
capability.
Obtrusive in
urban
locations.
Heavy and
bulky in
offshore
operations.
Susceptible to
paraffin
problems.
Tubing cannot
be internally
coated for
corrosion.
H2S limits
depth at
which a large
volume pump
can be set.
Limitation of
downhole
pump design
in small
diameter
casing.

Hydraulic
Piston
Pumping
Power oil
systems are a
fire hazard.
Large oil
inventory
required in
power oil
system which
detracts from
profitability.
High solids
production is
troublesome.
Operating costs
are sometimes
higher.
Usually
susceptible to
gas
interferenceusually not
vented.
Vented
installations are
more expensive
because of
extra tubing
required.
Treating for
scale below
packer is
difficult.
Not easy for
field personnel
to troubleshoot.
Difficult to
obtain valid well
tests in low
volume wells.
Requires two
strings of tubing
for some
installations.
Problems in
treating power
water where
used.
Safety problem
for high surface
pressure power
oil.
Lost of power
oil in surface
equipment
failure.

Electric
Submersible
Pumping

Gas Lift

Not applicable to
multiple compilations.

Lift gas is not always

available.

Only applicable with

electric power.

Not efficient in lifting

small fields or one
well leases.

High voltages (1,000

V) are necessary.
Impractical in shallow,
low-volume wells.
Expensive to change
equipment to match
declining well
capability.
Cable causes
problems in handling
tubulars.
Cables deteriorate in
high temperatures.
System is depth
limited, 10,000 ft.
(3048.0 m), due to
cable cost and inability
to install enough
power downhole
(depends on casing
size).
Gas and solids
production are
troublesome.
Not easily analyzable
unless good
engineering knowhow.
Lack of production
rate flexibility.
Casing size limitation.
Cannot be set below
fluid entry without a
shroud to route fluid
by the motor. Shroud
also allows corrosion
inhibitor to protect
outside of motor.
More downtime when
problems are
encountered due to
entire unit being
downhole.

Difficult to lift
emulsions and
viscous crudes.
Not efficient for small
fields or one-well
leases if compression
equipment is required.
Gas freezing and
hydrate problems.
Problems with dirty
surface lines.
Some difficulty in
analyzing properly
without engineering
supervision.
Cannot effectively
produce deep wells to
abandonment.
Requires makeup gas
in rotative systems.
Casing must
withstand lift
pressure.
Safety problem with
high pressure gas.

Hydraulic
Jet Pump
Relatively
inefficient lift
method.
Requires at least
20%
submergence to
approach best lift
efficiency.
Design of system
is more complex.
Pump may
cavitate under
certain
conditions.
Very sensitive to
any change in
back pressure.
The producing of
free gas through
the pump causes
reduction in ability
to handle liquids.
Power oil
systems are fire
hazard.
High surface
power fluid
pressures are
required.

Plunger Lift

Progressive
Cavity
Pumps

May not take

well to
depletion;
hence,
eventually
requiring
another lift
method.

Elastomers in
stator swell in
some well fluids

Good for lowrate wells only

normally less
than 200 B/D
(31.8 m/d).

Rotating rods
wear tubing;
windup and
after-spin of
rods increase
with depth

Requires more
engineering
supervision to
adjust properly.
Danger exists
in plunger
reaching too
high a velocity
and causing
surface
damage.
Communication
between tubing
and casing
required for
good operation
unless used in
conjunction with
gas lift.

POC is difficult
Lose efficiency
with depth

16

J.F. LEA, H.V. NICKENS

SPE 52157

Table 3: Capacities of Reciprocating Hydraulic Pumps

Tubing
Working Fluid
Maximum Pump
Size
Level, ft.
Displacement, B/D
2-3/8"
6000 to 17000
1311 to 381
2-7/8"
6000 to 17000
2500 to 744
3-1/2"
6000 to 15000
4015 to 1357
Table 4: Capacities of jet free pumps
Tubing
Production B/D
2-3/8"
3000
2-7/8"
6000
3-1/2"
10000

Target Rate (bbl/day)

Initial Installation ($)
Energy Efficiency (%)
Intake Pressure (psia)
Lift Energy (kw/bbl/day)
Workover Cost ($/day)
Wireline Cost ($/day)
Injection Gas ($/Mscf)
Other Costs($/month)

Table 5: Lift Methods Costs: Low Rate Case

Beam
Hydraulic
Gas Lift
1000
1000
1000
141000
173000
239000
58
16
15
900
900
900
.025
.096
.100
1000
1000
1000
1000
.24
200
2900
600
(Maintenance)
(Maintenance)
(Compressor
Maintenance)

ESP
1000
105000
48
900
.031
1000
225
(Inventory)

Table 6: Beam Pump Equipment Costs, Low Rate Case

Item
Cost ($)
Life (yrs)
Tubing
80000
15
Rods
20000
4
Pump
6000
Run Life Table
Table 7: ESP Equipment Costs, Low Rate Case
Item
Cost ($)
Life (yrs)
Tubing
80000
15
Pump
25000
Run Life Table
Protector
4000
Run Life Table
Separator
5000
Run Life Table
Motor
15000
*Run Life Table
Cable
50000
6
Cable Protector
20000
15
Transformer
12000
15
VSD
35000
15

Item
Tubing
Valve
Mandrel

Motor life assumed 2 x pump life

Table 8: Gas Lift Equipment Costs, Low Rate Case
Cost ($)
Life (yrs)
80000
15
2000
3
5000
10

Replace
2
6

SPE 52157

SELECTION OF ARTIFICIAL LIFT

17

Table 9: Hydraulic Pump Equipment Costs, Low Rate Case

Item
Cost ($)
Life (yrs)
Tubing
80000
15
Pump
20000
Run Life Table

Year
1
2
3
4
5

Table 10: System Run Life Data

Hyd. Pump
Beam Pump
ESP Pump
(days)
(days)
(days)
300
200
300
400
600
750

Inj. Gas Volume

(Mscf/d)
600
700
800
900
1000

Table 11: Summary of Low Rate NPV Calculations - Constant Rate

Method
NPV (MM$)
Costs (MM$) Cost/NPV
ESP
4.7
.82
.17
Gas Lift
3.80
.82
.22
Hydraulic Pump
4.28
1.12
.26
Rod Pump
4.70
.67
.142
Table 12: Equipment Operational Costs, High Rate Case
Jet
Gas Lift
ESP
Target Rate (bbl/day)
17000
17460
17020
Initial Installation ($)
200000
265000
150000
Energy Efficiency (%)
21
16
41
Lift Energy (kw/bbl/day)
.042
.056
.022
Workover Cost ($/day)
2000
2000
2000
Wireline Cost ($/day)
2000
Injection Gas ($/Mscf)
.24
Other Costs($/month)
2900
3000
225
(Maintenance)
(Compressor
(Inventory)
Maintenance)
Table 13: ESP Equipment Costs, High Rate Case
Item
Cost ($)
Life (yrs)
Tubing
80000
15
Pump
25000
Run Life Table
Protector
4000
Run Life Table
Separator
5000
Run Life Table
Motor
15000
*Run Life Table
Cable
50000
6
Cable Protector
20000
15
Transformer
12000
15
VSD
35000
15
* Motor life assumed 2 x pump life
Table 14: Gas Lift Equipment Costs, High Rate Case
Item
Cost ($)
Life (yrs)
Replace
Tubing
80000
15
Valve
2000
3
2
Mandrel
5000
10
6

18

J.F. LEA, H.V. NICKENS

SPE 52157

Table 15: Hydraulic Pump Equipment Costs, High Rate Case

Item
Cost ($)
Life (yrs)
Tubing
80000
15
Pump
20000
Run Life Table

Year
1
2
3
4

Method
ESP
Gas Lift
Jet Pump

Table 16: System Run Lives, High Rate Case

Hyd. Pump
Inj. Gas Volume
Beam Pump
ESP Pump
(days)
(Mscf/d)
(days)
(days)
300
200
300
300
400
600
750
Table 17: Summary of NPV Analysis, High Rate Case
NPV (MM$)
Costs (MM$) Cost/NPV
218.2
3.4
.016
211.1
4.0
.019
222.9
4.4
.020

Table 18: Field conditions for THUMS field where MTBFs are illustrated in Figure 18
Zone
On/Offshore
Active Producers
General Description

Well Production, BFPD

Pump Intake
Pressure, psi
Vertical Depth, ft
Oil Gravity, oAPI
GLR, scf/bbl
Avg. Resv. Press., psi
Avg. Temp. oF
Avg. W ater Cut, (%)
Avg. Viscosity, cp
Scale
Abrasives
CO2, ppm,
H2S, ppm
Emulsion Problems
Casing Size
Liner Size
Completion
Tubing Size

Ranger
Terminal
Offshore
Offshore
439
128
Unconsolidated
Poorly
Sandstone
Consolidated
Sandstone
200 - 5,500
120 - 3,500

UP Ford
Offshore
44
Moderately
Consolidated
Sandstone
40 - 1,500

100 - 850
100 - 850
100 - 600
2,100-3,200
2,800 - 4,200 4,100 - 7,100
15
20
28
11
35
80
1,000
1,100
1,400
130
160
210
94
82.5
80
80
15
5
CaCO, BaSO4 CaCO, BaSO4
CaCO
light - heavy
light - heavy
light - heavy
0-5%
0-5%
0-1%
0-4000
0-4000
0-2000
0-4000
0-4000
0-2000
Y
Y
Y
8-5/8", 32#
8-5/8", 32#
9-5/8", 40#
6-5/8", 28#
6-5/8", 28#
7", 26#
Gravel Pack
Gravel Pack Slotted Liner
2-7/8", 6.4#
2-7/8", 6.4#
2-7/8", 6.4#

SPE 52157

SELECTION OF ARTIFICIAL LIFT

19

Table 19: Summary of lives and costs of various components of an ESP system (Ref. 6)
900 BPD TARGET CASE

ite m
c o st

PV

/ ye a rly c o st
1
2

Wo rko ve r fre q ue nc y
investment:
p ump / p rotec to r
se p a ra tor
c a b le
mo to r
d o w nho le se nso r
tub ing
ste p -up tra nsfo rme r
VSD/ sw itc h b o a rd
c a b le p ro te c tors
opera ting costs:
e le c tric ity
w o rko ve r
inve nto ry
o ve rhe a d
TOTAL ESC . EXPENSE
ca shflow

24.3
3.9
50.3
20.4
5.6
83.0
10.2
31.2
14.0

15.0

(977) 257.9

equipment:
389 sta g e p ump
120 HP mo to r
No 2 c a b le

2.8

1.1

0.6

0.5

0.5

0.5

68.5
11.1
50.3
28.7
15.8
83.0
10.2
31.2
14.0

26.5
4.3

14.6
2.4

12.1
2.0

12.1

20.4
42.3
2.7
22.7
400.8
-401

10

11

12

0.5

0.5

0.5

0.5

0.5

0.5

12.1

12.1

12.1

12.1

12.1

12.1

12.1

5.1

5.1

5.1

5.1

5.1

22.1
7.5
2.7
3.0
69.2
-69

22.4
7.5
2.7
3.0
72.3
-72

22.7
7.5
2.7
3.0
75.6
-76

22.9
7.5
2.7
3.0
79.1
-79

23.2
7.5
2.7
3.0
82.7
-83

11.1
5.6

6.1
5.6

5.1
5.6

5.1

5.1

50.3
5.1

20.6
16.4
2.7
5.2
96.0
-96

20.9
9.0
2.7
3.7
70.2
-70

21.1
7.5
2.7
3.4
66.9
-67

21.4
7.5
2.7
2.9
60.5
-61

21.6
7.5
2.7
2.9
63.3
-63

21.9
7.5
2.7
6.0
133.7
-134

a ll c ost x $1000

Table 20: Summary of some lives of ESP equipment derived for the study in Reference 6.
Estimated cumulative service life for ESP components1.
Component/case
Pump/intake
Separator2
Motor
Cable
d.h. sensor
Transformer
VSD3
Tubing
1

150 BFPB
target

300 BFPD
target

>450 BFPD
target

150 BFPD
downside

target curve
target curve
2x target
6 years
target curve
15 years
5 year
15 years

target curve
target curve
2x target
6 years
target curve
15 years
5 year
15 years

target curve
target curve
2x target
6 years
target curve
15 years
5 year
15 years

<250 curve
<250 curve
2x <250
6 years
<250 curve
15 years
5 year
15 years

Cumulative service life in this table is related to the estimated run life depicted in Figure 19.
Rotary separators will be used for the first 5 years only
VSDs will only be used for the first 5 years.

2
3

300 BFPD
downside
interpolated
interpolated
2x interp.
6 years
interpolated
15 years
5 year
15 years

>450 BFPD
downside
>500 curve
>500 curve
2x >500
6 years
>500 curve
15 years
5 year
15 years

20

J.F. LEA, H.V. NICKENS

SPE 52157

Table 21: Summary of hydraulic pump lives for various fields.

Operator
Citronell Unit
Operator,

Depth

Production

(ft)

(bpd)

10-11000

300-400

N. Of Mobile, Al.
Texaco
Barre Field,

Triplex, oil system

3000 psi inj
Several wells on
one pump
Triplex, oil system
1800-2400 bpd oil

49

Use Kevlar spring loaded plungers & liners

75

Like soft pack Triplex

100-450

10,500

15-50

Triplex, oil
Vortex to clean oil

165

10,000

75

Triplex, 3000-3500
psi

90

at 3900 psi

Woodland Unit
MWJ
Baum/Sanders

Comments

(days)

15,000

S. Al.
Unocal
Wyoming

RUN LIFE DATA FOR HYDRAULICS

JFL / TRC
Power System
Pump Run Life

Single string systems

Pressure annulus to bring pump up
Takes about 1000 psi to move pmp up

in New Mexico

no trouble

Recip pumps, corrosion treat BHA's

1 inch vent string
Production up casing

J. Schlagel

Injection down tubing (2 3/8)

Recommends individual pwr supply

Marathon
Cody Unit, WY

7,500

475

Triplex,uses
water and oil

180 (min)

Slower pumps (lt. 45 spm) may

run 3 years..

Andy Franklin

for pwr fluid

Pump repair: $1500-2000

(likes now with

Vortex cleanup

For frac cleanup the fill

experience)

4 spd trans

well with liquids, circulate and filter with

vortex unit at surface
Single string- monitor tub press for
pump off..trying VSD
200-300 psi on casing brings up pmp

UNOCAL
Huntington Beach

4,000

295

Joe Gonzales

1st Triplex with oil

now ESP with
water
some 3000 psi

Use 3 string, main& 2 side strings

Free pump installations

some 2000 psi

BHA's 3 years

Power water not mixed with production

side strings leak, pull only one string if side string
few hrs to pull side string
4 hrs to round trip new pump
ESP's less maintenance, more energy to run
Triplex's more maintenance, less energy to run

Cook Inlet
UNICAL
Dean Geisert
likes hydraulic
better
than ESP's

7,500

105 avg

Triplex & ESP

Oil system
tank only
separation
no vortex cleanup
3550 psi
generated
60 hz esps

180 plus

Use recip and jet pumps

2 1/2 hrs to replace pumps
Tubing tripped not more than 3-4 years
two strings, open annulus
check fluid level with echometer
runs two strings simultanously

gears on triplex's
Average Pump Run Life, days
114.5

SPE 52157

SELECTION OF ARTIFICIAL LIFT

21

Table 22: Summary of run lives and costs assumed for one rate for study of Reference 6.
1000 B PD

PV

it e m
cost

/ y e a r ly c o s t
1
2

p u m p r e p a ir f r e q .

2 .0 0

in v e s t m e n t :
d o w n h o le p u m p
a s s o c ia t e d e q u ip m e n t
p o w e r f lu id p u m p *
p o w e r f lu id s y s t e m
a u t o m a t io n e q u ip m e n t
t u b in g

1 0 .0
3 0 .0
5 0 .0
1 1 0 .0
1 0 .0
1 5 5 .4

o p e r a t in g c o s t s :
e le c t r ic it y
w o rk o ve r
w e ll a t t e n d e n c e
s u r f a c e e q p m . m a in t .
d o w n h o le p u m p r e p a ir *
o ve rh e a d
TO TA L E S C . E X PE N S E
c a s h f lo w

H y d r a u lic P u m p S y s t e m s

TA R G E T C A S E

2 .0 0

2 .0 0

2 .0 0

2 .0 0

2 .0 0

1 0 .0
3 0 .0
5 0 .0
1 1 0 .0
1 0 .0
1 5 5 .4

3 8 .4
6 0 .0
2 .5
1 1 .2
8 .0
2 9 .1
5 1 4 .6
-5 1 5

3 0 .0

4 .0

(1,222) 3 9 9 .4

2 .0 0

10

11

12

2 .0 0

2 .0 0

2 .0 0

2 .0 0

2 .0 0

4 1 .7

4 2 .2

4 2 .7

4 3 .2

4 3 .7

2 .5
1 1 .2
8 .0
3 .8
8 8 .5
-8 8

2 .5
1 1 .2
8 .0
3 .8
9 2 .7
-9 3

2 .5
1 1 .2
8 .0
3 .9
9 7 .2
-9 7

2 .5
1 1 .2
8 .0
3 .9
1 0 1 .9
-1 0 2

2 .5
1 1 .2
8 .0
3 .9
1 0 6 .8
-1 0 7

1 0 .0
1 0 .0
5 0 .0

3 8 .8

3 9 .3

3 9 .8

4 0 .2

4 0 .7

2 .5
1 1 .2
8 .0
3 .6
6 6 .7
-6 7

2 .5
1 1 .2
8 .0
3 .7
6 9 .9
-7 0

2 .5
1 1 .2
8 .0
3 .7
7 3 .3
-7 3

2 .5
1 1 .2
8 .0
3 .7
7 6 .8
-7 7

2 .5
1 1 .2
8 .0
3 .7
8 0 .5
-8 1

4 1 .2
3 0 .0
2 .5
1 1 .2
8 .0
9 .8
2 1 8 .5
-2 1 9

a ll c o s t x $ 1 0 0 0

2000

Pressure, psig

1500

1000

500

1
0

Inflow @ Sandface (1)

Inflow (1)
Not Used
Not Used
Not Used
Not Used
Not Used
Not Used

100
Not Used
Outflow (A)
Not Used
Not Used
Not Used
Not Used
Not Used

200

300

400

Liquid Rate, Bbl/D

Reg: James F. Lea - Amoco

Figure 1: IPR with bubble point below static reservoir pressure.

Figure 2: Schmetic of geometry of horizontal well inflow model.

500

600

22

J.F. LEA, H.V. NICKENS

SPE 52157

2000

Pressure, psig

1500

1000

500

5
0

Inflow @ Sandface (1)

Inflow (1)
Case 2 (2)
Case 3 (3)
Case 4 (4)
Case 5 (5)
Not Used
Not Used

100

200

Not Used
Outflow (A)
Case 2 (B)
Case 3 (C)
Case 4 (D)
Case 5 (E)
Not Used

Figure 3: IPRs decreasing with time.

Figure 4: Major Artificial Lift Systems (from Trico)

3
300

Liquid Rate, Bbl/D

Inflow
Declining reservoir press, psia

Reg: James F. Lea - Amoco

2
400

1
500

Inflow
(1) 2000.0
(2) 1800.0
(3) 1600.0
(4) 1400.0
(5) 1200.0

600

SPE 52157

SELECTION OF ARTIFICIAL LIFT

A r t if ic ia l L if t :

R a t e

v s . D e p t h

10

100

v s . M

e t h o d

Ref: Pennwell AI Methods

Chart, 1986

BPD
1

23

1000

10000

100000

0
2000

Plunger

4000

DEPTH, FT

6000

Gaslift

8000
10000

Beam

12000

ESP
14000
16000
18000

10

100

0
2000

Hyd. Jet

4000

DEPTH, FT

6000

PCP

8000
10000
12000
14000
16000

Hyd.
Recip.

18000
Figure 5: Depth/Rate Selection Chart after Blais

1000

10000

100000

24

J.F. LEA, H.V. NICKENS

Figure 6: Schematic of Beam Pumping System

Figure 7: Schematic of Typical ESP system

SPE 52157

SPE 52157

SELECTION OF ARTIFICIAL LIFT

25

Figure 8: Schematic of PC pump.

The ESPCP System

y Progressing cavity pump driven
by submersible motor
y Replaces Rod-Driven PC Pump
Units:

PCP
Flex Shaft
Assembly

Deviated wells
Viscous production

Cable

Seal Section

Gear
Reducer
Electric Motor

Figure 9: Schematic of ESPPC system

26

J.F. LEA, H.V. NICKENS

SPE 52157

The ESPCP System

y Standard components include:
Seal section
Motor

PCP

Cable

Flex Shaft
Assembly

PC pump
y New components include:
Intake with flex shaft
Gear reducer

Cable

Seal Section

Gear
Reducer
Electric Motor

Figure 10: Components of an ESPPC System

Figure 11: Reciprocating hydraulic pump and Jet Hydraulic Pump

SPE 52157

SELECTION OF ARTIFICIAL LIFT

Figure 12: Showing operation of free hydraulic pump installation.

Figure 13. Schematic of gaslift system:

27

28

J.F. LEA, H.V. NICKENS

Present Value
MM $

SPE 52157

ESP

Gas Lift

Hydraulic Pump

Rod Pump

5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
0

Years
D:\LANG\VB4\ECON\LORATE1.PEP

Figure 14: Summary of Low Rate NPV Analysis

Present Value
MM $
225
200
175
150
125
100
75
50
25
0

Years
Figure 15: Summary of High Rate NPV Analysis

10

11

12

13

14

15

16

17

18

SPE 52157

SELECTION OF ARTIFICIAL LIFT

29

Ty p ica l B ea m S y s t em Fa ilu r es P er Yea r - 532 w ell s

35 0

1.2 0

1.14

30 0
25 0

0.74

20 0

0.81
$9 7 2 .4 K

P u lls

1.0 0

F PW PY

0.8 0

$8 6 3 .4 K

0.60

15 0

$6 4 1 .6

0.45

10 0

0.6 0

$567.5K

$8 6 3 .7 K

0.51
$4 6 1 .1 K

0.41

0.39

0.4 0

0.29

50

0.2 0

0.0 0
1989

1 990

19 91

199 2

1993

1 994

1 995

19 96

199 7

Figure 16: History of typical beam pump opeation: failures per year with approximate associated costs.

Total P um p Failures = 418

Typical D istribution of B eam P ump Failures

Total Tubing Failures = 224

Total B eam Failures = 1110

Total P in & Coupling Failures = 206

P olis h Rods
5%

Other
3%

Rod B odies
15%

Total Rod B ody Failures = 171

Total P olis h Rod Failures = 58

Total "Other" B eam Failures = 33

P um ps
38%

P in & Couplings
19%

Tubing
20%

Figure 17: Typical distribution of failures among the beam pump system components

30

J.F. LEA, H.V. NICKENS

SPE 52157

1,4 00
1,3 00
1,2 00

A c tiv e E S P s

1,1 00

M TB F , D ay s

1,0 00
9 00
8 00
7 00
6 00
5 00
4 00
3 00
2 00
1 00
0
1983

1984

1985

1986

1987

1988

1989

1990

1991 1992

1993 1994

1995

1996

1997

Figure 18: History of increasingly better run lives (MTBF) in the THUMS field.

PRIOBSKOYE average ESP run life prediction

900

Sw an Hills (Im
Target

800

Runlife (days)

700
Sw an Hills

600
Congo

500

Dow n side > 500BPD

400
300

Dow n side < 250BPD

Montrose

200

Milne Point

100
0
0

10

Years of deployment

Figure 19: Failure data from a number of field locations and also target values for the
study in Reference 6.
400
350

Days

300
250

industry data

200

target

150

downside

100
50

BFPD
NOTE : several values reported as 180 days plus

Figure 20: Run life of downhole hydraulic pumps

1000

900

800

700

600

500

400

300

200

100