EUR
en
Submersible pumps are pumps with connected motors that can be submerged into the sump, pit or well. In some cases the submerged, hermetically sealed motors are filled with oil and in other cases they are filled with air along with separate water cooling of the motor casing. The smaller size pumps are furnished with air cooled motors requiring no coolant flow through the pump motor casing. The larger pumps are cooled with pumpage most frequently, utilizing the impeller back vanes as the pumping means. Seal chambers filled with oil separate the motor from the pump on some units. A submersible slurry pump with liner is shown in Fig. 6.10, a dewatering pump in Fig. 6.11 and a submersible propeller pump installation in Fig. 6.12, at Modesto, California, USA. These four pumps have a combined flow rate of 78 _MGD_. Diffuser and volute pumps of many types have been furnished in submersible configurations. Submersible pumps can be portable and are commonly used as contractors pumps for utility cleanup. Fig. 6.23 shows a cutaway of a portable submersible and Fig. 6.24 shows the vortex pump version of that same pump used for slurry transport duties, such as tank emptying. Further coverage of these pumps is included in Chapters 24 and 25.
Submersible pumps should strictly be termed ‘submersible motor’ pumps or ‘submersible pumpsets’. The motor design (Plate 32(c)) is the main difference from more conventional designs. The pump, driven by a submersible motor, is very similar to a pump driven by a vertical spindle ‘dry’ motor, although some differences are given below. Submersible pumps gained in popularity because they usually result in a cheaper installation than one using dry motors. Motor reliability and its location out of the sight and hearing of any attendant were issues but these have been largely overcome by improvements in the motor design, particularly in the insulation and in the instrumentation used for monitoring pump performance. Properly chosen submersible pumps have proved reliable in service over many years; submersible designs are now available from specialist manufacturers for a very wide range of duties.
Many submersible pumps in water supply are installed in drilled boreholes. The high cost of drilling is affected by the borehole diameter; therefore the diameter of the submersible pump is of great importance. Designers must produce pumps and motors of small diameter. Mixed flow pumps produce more flow at a given casing diameter than radial flow pumps and are suitable for borehole pumps. However, they produce less head and more pump stages are needed. This results in pumps longer and narrower than more conventional designs. For the same reasons, submersible motors are longer than equivalent dry motors. They are nearly always two-pole designs (Section 19.21) to develop more power from a given size motor and to run at the highest available speed to maximize pump output, hence reducing overall cost. Naturally the mechanical design of the pump, especially its bearings, must be appropriate for the chosen speed. The disadvantages of a higher speed are increased wear, particularly if the pumped water contains abrasive solids, and reduced suction capability, so that deeper submergence may be required.
In a typical borehole installation, the pump is directly coupled to the submersible motor, which is underneath, and power is supplied to the motor through waterproof cables clipped to the outside of the riser pipe. The water inlet is between pump and motor with the outlet from the final pump stage leaving axially. The motor is normally a water-cooled fixed-speed caged induction motor, specially designed for underwater running. However, where varying output is needed, a variable frequency power supply may be used (Section 19.24), although at significant extra cost. If there is any risk that inflow to the borehole could be predominantly from a higher level than the pump inlet, or if the pump is installed in a large body of water so that the pumped flow does not pass over the motor, a motor shroud should be used to ensure a cooling flow passes over the motor.
Whilst all submersible motor windings will meet the required temperature class there can be an issue with the insulation breaking down due to the voltage withstand level that the motor can accept (Section 19.24).
Submersible pumps are relatively quick and easy to install. The rising main is free of the spindle and sleeving needed with the vertical spindle pump; a large thrust bearing to support the heavy rotating parts is not required. Submersible pumps need not be installed truly vertically, which may be a big advantage in very deep wells. They are even sometimes used horizontally as booster pumps in distribution mains (Plate 32(d)). Submersible-pump reliability in non-corrosive waters has been proven over the years; even in corrosive waters they can be withdrawn for attention or replacement more easily than pumps of the vertical spindle design. Some modern borehole installations are now designed without any surface housings, although provision still needs to be made for access for a mobile crane or sheerlegs for withdrawal of the pumpset and its riser pipe. This simplification can make substantial cost savings.
Submersible pumpsets may be less efficient than the vertical spindle design, partly because of the special design of the motor but also because of the higher number of stages needed to achieve a given duty. This can be important if the pumping duty is wrongly estimated, because of the pronounced peak in the efficiency curve with the multi-stage unit. However, submersibles gain by avoiding the transmission shaft losses of the vertical spindle design.
The heart of the ESP unit is the submersible pump and the design and analysis of the whole ESP system cannot be understood without a basic comprehension of the operation of the pump. Hence the description of the system's components must be started with a thorough analysis of the construction and operation of the centrifugal pumps.
The submersible pumps used in ESP installations are multistage centrifugal pumps operating in a vertical position. Although their constructional and operational features underwent a continuous evolution over the years, their basic operational principle remained the same. Produced liquids, after being subjected to great centrifugal forces caused by the high rotational speed of the impeller, lose their kinetic energy in the diffuser where a conversion of kinetic to pressure energy takes place. This is the main operational mechanism of radial and mixed flow pumps, as detailed in Chapter 2.
Fig.3.2 illustrates the main parts of an ESP containing mixed flow stages. The pump shaft is connected to the gas separator or the protector by a mechanical spline coupling at the bottom of the pump. Well fluids enter the pump through an intake screen and are lifted by the pump stages. The figure illustrates a standard intake with several large (minimum 1″) screened ports and bolted to the bottom of the pump housing with its shaft connected to the protector shaft by a spline coupling. Integral intakes are also available. Other parts include the radial bearings (bushings) distributed along the length of the shaft providing radial support to the pump shaft turning at high rotational speeds. An optional thrust bearing takes up part of the axial forces arising in the pump but most of those forces are absorbed by the protector's thrust bearing.
The liquid-producing capacity of an ESP depends on the following factors:
• the rotational speed provided by the electric motor,
• the diameter of the impeller,
• the design of the impeller (characterized by its specific speed),
• the actual head against which the pump is operating,
• the thermodynamic properties (density, viscosity, etc.) of the produced fluid.
The ESPs are designed to operate in the proper direction of rotation; as seen from the top, they can be driven in clockwise (CW) or counterclockwise (CCW) directions. The actual direction depends solely on the electric motor's rotation, which is easily set during installation in accordance with the pump's requirement. CW rotation, used by many manufacturers, involves the risk of unscrewing of threaded equipment parts (pumps, protectors, motors, etc.) during startup and running because of reactive torque. To prevent such failures “lock plates,” i.e., steel pieces are welded across the equipment heads/bases and housings against possible separation of threaded connections. In contrary, if the pump is rotated CCW, unscrewing does not occur.
The pump will operate with the wrong direction of rotation as well but its output will be much reduced. This fact may be used to detect the proper rotation of the equipment by observing the developed wellhead pressure and/or the produced liquid rate immediately after starting the ESP unit. If measured performance data do not match design values, the direction of motor rotation must be changed.
Conventional ESP installations run on AC power with a constant frequency of 60 or 50 Hz. ESP motors in 60 Hz electric systems rotate at a speed of about 3,500 RPM, whereas in case of a 50 Hz power supply the motor speed is about 2,900 RPM. For constant speed applications, the most important factor is impeller size, which, of course, is limited by the inside diameter of the well casing. Pumps of bigger sizes can produce greater rates, although impeller design also has a great impact on pump capacity.
Present-day ESPs come in different capacities from a few hundred to around 80,000 bpd of liquid production rate and in outside diameters (ODs) from around 3″ to 11″. Smaller units contain “pancake” type impellers with radial discharge and are used up to the rates of 1,500–3,500 bpd, above which mixed flow impellers are used.
The length of individual ESPs is limited to about 20–25 ft, for ensuring proper assembly and ease of handling. Tandem pumps are made up of several pump sections (up to three) and are used to achieve higher operational heads usually required in deeper wells. Thus several hundreds of stages can be run, the maximum number of stages being limited by one or more of the following factors:
• the mechanical strength of the pump shaft, usually represented by the shaft's horsepower rating,
• the maximum burst-pressure rating of the pump housing,
• the maximum allowed axial load on the unit's main thrust bearing (usually situated in the protector section).
Individual stages in ESPs, provided they are of the same impeller design, handle the same liquid volume and develop the same amount of head. The heads in subsequent stages are additive so a pump with a given number of stages develops a total head calculated as the product of the total number of stages and the head per stage. This rule allows one to find the number of stages required to develop the total head to overcome the total hydraulic losses, valid at the desired liquid production rate in a well.
As the size of well casing limits the OD of the ESP equipment that can be run, pump selection is heavily restricted by the actual casing size.
The performance of ESPs is characterized by the pump performance curves introduced in Section 2.2.2.3. These are plotted in the function of the pumping rate and represent
• the head developed by the pump
• the efficiency of the pump
• the mechanical power [brake horsepower (BHP)] required to drive the pump when pumping water.
These curves are experimentally obtained with freshwater under controlled conditions described in API RP 11S2[1] at an operating temperature of 60°F. Tests on submersible pumps are made by driving the pump at a constant rotational speed, usually 3,500 RPM for 60 Hz service. This is the speed generally assumed as the actual speed of a fully loaded ESP motor having a synchronous speed of 3,600 RPM. Experiments with other fluids or speeds must be corrected to these standard conditions using the affinity laws. All performance parameters must represent the operation of one or a specified number of pump stages clearly indicated on the chart.
Performance curves of pumps in 50 Hz service are typically published for a fixed rotational speed of 2,917 RPM. As described by Butlin[2], this number was chosen based on a misunderstanding of slip in electric motors, using the concept of the “percentage slip.” It was erroneously assumed that motor slip is found as a percentage of the synchronous speed. Slip (defined as the difference between the motor's synchronous and actual speeds), however, depends on motor voltage and torque loading only. This means that the same motor at the same pump load will have the same amount of slip, independent of the electric frequency. The synchronous speed of ESP motors at 50 Hz service being 3,000 RPM, the actual full-load motor speeds must be around 2,900 RPM or even less for small-diameter motors.
Because of manufacturing tolerances and other factors, a variance in performance as compared to published performance curves can exist from one pump to another of the same design and the same manufacturer. The accepted tolerances between published curves and actual pump performance are specified as follows:
• The pump's best efficiency point (BEP) at the nominal flow rate should be at least 90% of the published value.
• The head–flow rate performance curve in the recommended operating range should have a tolerance of ± 5%.
• The tolerance of pump power–flow rate curve in the recommended operating range should be ± 8%.
The sample curves shown in Fig.3.3 represent the performance of one stage of a given pump design. Pump performance curves include the head developed by the pump, the power required to drive the pump, and the efficiency of the pump, all in the function of the pumping rate. The performance parameters belonging to the BEP represent the criteria for an optimum utilization of the pump, around which the recommended range of operation is indicated.
The recommended range of pumping rates for any ESP is strictly related to the variation of axial forces occurring in the pump. As discussed in Section 2.3.2.5, static and dynamic axial forces occurring in pump stages are the results of different phenomena and may take the form of downthrust or upthrust. Fig.3.4 presents the relationship between the axial forces and the recommended operating range of the pump; their interaction is detailed in the following.
The lower part of Fig.3.4 schematically shows the change of axial forces in an ESP stage in the function of the liquid rate produced by the pump. Downthrust is basically determined by the head developed because its main component comes from the pump's discharge pressure acting on the top and bottom shrouds of the impeller. Its variation with pumping rate, therefore, follows the shape of the pump's head–rate performance curve, as shown in the figure. It is at a maximum at shut-in conditions (at a pumping rate of zero) and diminishes to zero where the pump's head decreases to zero. Upthrust forces arising in the pump stage, on the other hand, are the result of the change in inertial forces and are proportional to the kinetic energy of the liquid pumped. Thus their variation with pumping rate follows a second-order curve.
The sum of the up- and downthrust forces is shown in a bold line representing the net thrust arising in the pump stage. As seen, the operation of ESPs is dominated by downthrust forces because the net axial force points downward in the largest part of the operating pumping rates.
As discussed in Section 2.3.2.5, axial forces developed in ESPs must be compensated, otherwise the axial movement of the impellers and the pump shaft leads to mechanical damage of the stages. Elimination of such forces is accomplished differently in stages with fixed impellers from stages with floating impellers.
• In fixed impeller pumps all axial forces are transmitted to the pump shaft and must be balanced by the main thrust bearing, situated in the protector section of the ESP unit. This solution necessitates the use of thrust bearings of relatively large capacity.
• In floating impeller pumps, on the other hand, most of the axial forces are taken up by the forces arising in the up-, and downthrust washers installed on the impellers. The benefit of this impeller arrangement is that smaller-capacity thrust bearings are required in the protector.
In the following discussion, let us assume a floating impeller pump and determine the safe operating range of such ESPs. In these pumps, most of the net thrust illustrated in Fig.3.4 must be absorbed by the thrust washers in the pump stage. Typical arrangement of thrust washers is illustrated in Fig.3.5 from which it can be seen that the load-carrying area of the upthrust washer is much smaller than that of the downthrust washers; therefore, upthrust is more dangerous for the pump's operation than downthrust. To safeguard against the occurrence of upthrust, a minimum of downthrust is assumed and a safety zone in pumping rates is created, as shown in Fig.3.4. As illustrated in the figure, this defines the upper boundary of the pump's applicability in such a way that upthrust conditions are safely avoided.
As downthrust forces are absorbed by the combined effort of the washers and the unit's main thrust bearing situated in the protector, their combined capacity defines the maximum acceptable axial load that can be allowed. The pumping rate belonging to this load defines the minimum recommended pumping rate of the ESP.
Any given pump, therefore, must be operated in the recommended operating range defined by the two boundary rates described previously. Keeping the operation of an ESP inside the recommended pumping rate range poses the most severe restriction in the application of ESP equipment and should never be forgotten by users.
A floating impeller centrifugal pump's operating modes are illustrated in Fig.3.6, in which the ranges of upthrust and downthrust conditions are defined. In the recommended range the impeller is said to “float” between the two diffusers but is forced against the diffusers when pumping liquid rates outside the range. Pump stages are damaged quickly because of the resultant mechanical wear that occurs in upthrust or downthrust conditions.
Although ESP performance curves are normally established by experiments, analytical models are also available [3] to calculate the pump's head performance curve. The use of pump performance curves in computer programs is facilitated by the fact that head–rate and power–rate functions can be fitted very accurately by polynomial functions of the liquid rate. The use of the following mathematical formulas is universally accepted:
(3.1)H=∑i=0 N a i q i
(3.2)BHP=∑i=0 M b i q i
where
_H_=head developed by pump, ft
[BHP](https://www.sciencedirect.com/topics/engineering/brake-horsepower "Learn more about BHP from ScienceDirect's AI-generated Topic Pages")=required pump power when pumping water, HP
_q_=liquid rate, bpd
_a_ _i_, _b_ _i_=coefficients
_N_, _M_=maximum number of coefficients, –.
Because of the regular shape of the pump performance curves, they can be fitted very precisely by the above formulas, using a maximum of eight to nine coefficients: _N_≤9 and _M_≤9. Computer programs utilize this feature and can reliably describe the performance of different pumps with the help of a proper data base of pump coefficients.
In ordinary pumping situations the ESP, while pumping a given liquid rate, develops a positive head, i.e., both the liquid rate and the head are positive. Special cases, however, may necessitate the description of the pump's operation at negative rates and/or heads, e.g., when injecting liquid into the well through an inoperative pump. In such cases a four-quadrant representation of the pump's performance may be useful where the liquid rate (abscissa) and the head developed (ordinate) axes are extended into negative regions [4]. Out of the four possible combinations of positive and negative values of rates and heads, three versions are possible:
• Normal operations take place in the first quadrant with positive rates and heads.
• In the second quadrant with positive flow rates but with negative heads due to the pressure drop occurring in its stages the ESP acts as a turbine. This can happen when the well is produced by another artificial lift method (e.g., gas lifting), with the pump not energized. The pressure drop across the stages creates great upthrust forces and the pump can easily be damaged.
• If fluid is injected through a pump not driven by its motor, flow rates are negative but head is still positive because the pressure decreases downward across the pump. The pump operates in the fourth quadrant and all thrust forces are in the downward direction. These unbalanced forces will result in heavy downthrust wear of the pump stages.
Stages with floating impellers are the original and simplest constructions where the impellers are free to “float” axially (move up and down relative to the pump shaft) because they are not fixed to the pump shaft in the axial direction. As shown in Fig.3.7, impeller hubs are not stacked so there is a vertical distance between the successive hubs. Furthermore, each stage contains downthrust washers to absorb the axial forces occurring during operation as well as to seal and minimize recirculation of fluids within the stage. Although called “floater” impellers, its washers are normally in contact with the diffuser pad because of the axial load on them. Because of this condition, the unit's main thrust bearing, situated in the protector section, has to carry only the downward thrust acting on the pump shaft.
The downthrust on the shaft is due to its cross-sectional area being exposed to a large pressure differential that equals the difference between the pump's discharge and suction pressures. This load is directly transferred to the protector and has to be taken into account when selecting the right protector.
The benefits of floating impeller design include
• the elimination of having to fix the impellers axially, a time-consuming work requiring high precision,
• the building of pumps with several hundreds of stages is possible,
• smaller-capacity thrust bearings are needed in the protector section because most of the hydraulic thrust is absorbed inside the pump
• lower investment cost, as compared to fixed impeller pumps.
Limitations are related to the load-bearing capacity of available thrust bearings, which, in turn, are restricted by the annular space available in different casing sizes:
• such pumps are usually manufactured in smaller diameters, up to a size of about 6″,
• the recommended operating range is somewhat narrower than that for the same pump with fixed impellers.
Fixed impellers are locked on the pump shaft in the axial direction, their hubs being in contact, see Fig.3.8. As the stages are not equipped with downthrust washers, the axial thrust developed on them must be fully carried by the unit's main thrust bearing in the protector section. Because of this, these pumps may be operated outside their normal operating ranges without much damage to the stages. Pumps with such stages are often called “compression pumps” and are commonly used in larger-sized ESP units (greater than 6″ in diameter) capable of producing large volumes of liquids. As already mentioned, they may have a wider operating range than pumps of the same type with floating impellers. Generally, they tolerate pumping of fluids containing abrasives much better than floating impeller type pumps.
The thrust to be carried by the main thrust bearing of the ESP unit is much greater than that for floater pumps. The total axial load has two components: (1) the shaft load due to the differential pressure acting on the shaft cross-sectional area and (2) the sum of the axial loads occurring in the pump stages. Clearly, thrust bearings of much greater capacity than that for floater pumps are required for proper operation.
Limitations of fixed impeller “compression” pumps include the following:
• they are more difficult to manufacture because impellers must be fitted very precisely along the pump shaft,
• investment costs are higher because of manufacturing requirements,
• the maximum number of stages in one pump is limited to about 80–100,
• protectors with high-capacity thrust bearings must be used.
During its operation, the centrifugal pump converts mechanical energy provided by the submersible motor into pressure increase of the fluid pumped. As the efficiency of energy transformation is inevitably less than 100%, there are energy losses involved. These losses end up as heat generated in the pump that is absorbed by the fluid flowing up the tubing string. To find the temperature increase in the produced fluid, a balance of the heat generated and the heat absorbed at steady-state conditions must be written up.
The power exerted by the pump to lift a given amount of liquid against the operating head equals
(3.3)P hydr=7.368 10−6 H q γ
where
_P_ hydr=pump hydraulic power, HP
_H_=head generated by the pump, ft
_q_=pumping rate, bpd
_γ_=specific gravity of the fluid, –.
The mechanical power required to drive the pump is found by considering the pump's efficiency:
(3.4)BHP=P hydr η p
where
BHP=power required to drive the pump, HP
_P_ hydr=pump hydraulic power, HP
_η_ _p_=efficiency of the centrifugal pump, –.
The energy wasted in the pump is the difference between the required BHP and the hydraulic power. This is the energy converted to heat, _Q_, in the pump; if expressed in British thermal units per minute, we obtain the following formula:
(3.5)Q=42.41 P hydr(1 η p−1)
where
_Q_=wasted energy converted to heat, BTU
_P_ hydr=pump hydraulic power, HP
_η_ _p_=pump efficiency, –.
The heat absorbed by the fluid flowing through the pump can be calculated from the fluid rate, the temperature increase, and the fluid's thermodynamic properties as
(3.6)Q=350 1,440 c q γ Δ T f
where
_Q_=heat absorbed by the fluid, BTU
_c_=specific heat capacity of the fluid, BTU/lb/°F
_q_=pumping rate, bpd
_γ_=specific gravity of the fluid,–.
Δ _T_ _f_=temperature rise in the fluid, °F.
As the heat generated in the pump is equal to the heat taken by the fluid, simultaneous solution of Eqs. (3.5) and (3.6) and substitution of _P_ hydr (Eq. 3.3) into the resulting equation gives the temperature rise of the fluid pumped:
(3.7)Δ T f=H(1−η p)778 c η p
where
Δ _T_ _f_=temperature rise in the fluid, °F
_H_=head developed by the pump, ft
_c_=specific heat capacity of the fluid, BTU/lb/°F
_η_ _p_=pump efficiency, –.
As seen from the equation, the temperature rise of the fluid pumped does not depend on the pumping rate or the fluid gravity. The effects of the governing parameters are the following:
• Fluid temperature increases proportionally with the total head developed by the pump.
• Pump efficiency has a definite effect; at low efficiencies, more heat is generated and the fluid is heated up more.
• Fluids with high heat capacities (water or wellstreams with high water cut) provide more cooling of the pump.
The temperature rise found from Eq. (3.7) refers to the inflow temperature of the fluid entering the pump and the outflow temperature from the pump to the tubing string. In most cases the effect of this on the operation of the pump is negligible except when fluids with extremely high viscosity are pumped.
For most of the history of centrifugal pump applications, the design of the pump stages was more an art than science. Development of new stages involved an iterative design process heavily relying on laboratory measurements and the final product usually did not meet all the requirements for efficiency and reliability. This situation has completely changed since the introduction of computational fluid dynamics (CFD) program packages that permit one to design and test different stage geometries on the computer. This novel method allows a complete control on the hydraulic processes that occur inside the pump stage and helps designers build stages for different requirements: optimum pump efficiency, maximum head development, etc.
Nowadays, most major manufacturers take advantage of CFD methods and offer centrifugal pump stages with significantly improved characteristics than the previous designs allowed. One of the first computer-designed stages was the “Centurion” offered by Centrilift [5] (Fig.3.9). The objective of the designer was to develop the highest possible head and, at the same time, to increase the pump's resistance against sand erosion. The most significant change in the stage's geometry is the extra-large (more than 30% greater than in conventional pumps) vane opening, as shown in Fig.3.10. Sand problems are minimized by the swirl-suppression ribs that divert the sand-laden fluid into the main flow path of the stage, thus decreasing the erosional damage caused by sand particles on the diffuser wall [6]. The high head developed by the Centurion stage has many advantages: shorter pumps are needed, the necessity of tandem pumps may be eliminated, and equipment installation time is reduced.
The advantages of CFD designs are illustrated in Fig.3.11 that presents a comparison of two similar REDA stages, the DN2150 of the old design and the computer-designed D2400N [7]. As seen, the new stage attains an efficiency of 68% as compared to 56% of the old design; this proves the excellent hydraulic performance of the stage designed by CFD methods. The company offers several models with similarly high pump efficiencies for different production environments [8].
The materials and manufacturing processes have also improved over the years [9,10]. The first ESP stages were simple grey iron castings but stronger and harder materials soon followed to cope with enhanced requirements. Today the standard material of pump parts is Ni-Resist, an iron alloy with about 15% nickel content; it has increased hardness and improved resistance against corrosion and abrasion. The improved Type 4 Ni-Resist contains approximately 25% nickel that further increases the metal's corrosion resistance and allows its heat treatment, but it is rarely used because of its high cost.
The pursuit for better performance at lower costs resulted in the wide use of coatings applied to impellers and diffusers to increase their run life. Surface coatings applied include nickel, tungsten carbide, etc. All these have the disadvantage that sand erosion can remove them from the surface because sand is harder than the mostly used nickel. Diffusion coatings can eliminate the problems associated with simple coatings but require heating the parts during application and can thus distort the shape of impellers and diffusers.
The conventional manufacturing process associated with the traditional materials just discussed is sand molded casting in which molten metal is cast in sand molds. Sand molds are cheap and heat resistant and can provide the necessary surface smoothness for stage parts but the complexity of the stages that can be created is limited. Another disadvantage is that not every metal lends itself to the sand casting process; anyway, most pump stages are manufactured by sand casting everywhere today.
The latest innovation that revolutionized the industry is the introduction of powder metallurgy technology; the first application in pump stage manufacture was by the Russian company Novomet in 1991 [11]. Conventional powder technology, well known in other industries, involves mixing, molding, and sintering of metal powders into the required final shape. To reach the complex shapes of impellers and diffusers the process is divided into several steps:
• Powdered (finely ground) raw materials (iron, graphite, copper, etc.) are mixed using exact recipes.
• Properly selected parts of impellers and diffusers are injected into molds and pressed into the desired cold-formed shapes using high-pressure presses.
• Parts are assembled and joined into nonsintered blanks.
• The blanks are sintered (heat treated) at high temperature for extended periods while metal particles bond together to form the final solid object.
• Final machining finishes the manufacturing of impellers and diffusers.
The powdered metal manufacturing process produces pump stages with the following very important improvements as compared to stages made with the conventional sand casting method:
• More complex stage geometries can be manufactured especially in conjunction with CFD design.
• Geometrical tolerances and surface finishes are much improved.
• Optimum selection of metal combinations for specific pump parts is possible; pump stages for specific purposes (corrosion, erosion resistance, etc.) can be built.
• Pump stages are more balanced and produce lower amounts of vibrations thereby increasing the reliability of the pump. Thanks to this feature, pumps can be operated at much higher speeds (up to 6,000 RPM) than their conventional counterparts.
• The process lends itself to cost-effective mass production.
Due to the reduction of the driving force which lifts the reservoir from downhole naturally, pumps were commonly used to increase the backpressure for production. The electric submersible pump (ESP) is an effective and economical method of lifting large volume of the fluids from downhole under different well conditions. ESP system requires a large electricity supply, but it is less complex and more efficient than delivering gas to gas lift systems.
Different from the surface pump system, the ESP systems are particularly designed to be immersed in fluid. It can be either located in a well or on the seabed. The ESP motors are pressure balanced with the environment, whether that is downhole pressure or water pressure in subsea conditions.
Optional components of the ESP system may include tubing joints, check valve, drain valve, downhole pressure and temperature transmitters, etc. Figure 2-17 shows a typical ESP configuration in downhole.
The selection of ESP types mainly depends on the well fluid properties. Following are the three major types of ESP applications:
• High water-cut wells producing fresh water or brine;
• Multi-phase flow well with high GOR;
• Highly viscous fluid Well.
The pump rate is a function of the rotational speed, the number of stages, the dynamic head acting against the ESP and the pumped fluid viscosity. These factors dictate the differential pressure across a pump system, and therefore the flow rate. However, for a given pump, there is an optimal design flow rate that maximizes pump efficiency and run life. Figure 2-18 shows the operating range recommended by ESP manufacturers.
Sizing of ESP is based on predicted completion performance, or flow rate. This usually involves examination of the well inflow performance relationship (IPR), which describes the production response to changes in bottomhole pressure (BHP).
Data required for calculation and sizing of ESP includes well data, production data, well fluid conditions, power sources and possible problems etc. Calculations for designing an ESP system include:
• Determination of Pump Intake Pressure;
• Calculation of total dynamic head;
• Selection of pump type;
• Check of load limits;
• Selection of accessory and optional equipments.
Submersible pumps are used to enhance the production of hot water wells. The energy transfer process is treated assuming unsteady flow of inviscid fluid. Theoretical and performance head curves are studied. Cavitation and its measurement is analyzed.
This chapter discusses the electrical submersible pump (ESP) components and their operational features. The ESP system proved to be an efficient means of producing liquid from oil and water wells. The classical or conventional installation is illustrated where the ESP unit is run on the tubing string and is submerged in well fluids. The heart of the ESP unit is the submersible pump and the design and analysis of the whole ESP system cannot be understood without a basic comprehension of the operation of the pump. The submersible pumps used in ESP installations are multistage centrifugal pumps operating in a vertical position. Although their constructional and operational features underwent a continuous evolution over the years, their basic operational principle remained the same. Produced liquids, after being subjected to great centrifugal forces caused by the high rotational speed of the impeller, lose their kinetic energy in the diffuser where a conversion of kinetic to pressure energy takes place.
An electric submersible pump is a widely used artificial lift equipment which is used to produce oil where natural production is not possible due to various reasons, for instance, low bottomhole pressure, liquid loading, heavy oil presence, etc. ESPs have a long application history in the oil and gas industry,
Bookmark
Daniel Féau processes personal data in order to optimise communication with our sales leads, our future clients and our established clients.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.