Part 16(b)—Understanding Pump Drivers, Part 2
By Ed Butts, PE, CPI
In the last installment of The Water Works, we began overviewing the various types of drivers for pumping equipment used for deep and shallow well and booster applications with a detailed look at electric motors. Now we will conclude the topic by outlining engines, generators, and geardrives used for this same purpose.
Introduction to Engines
Engines used to generate the power necessary to drive a pump, compressor, or generator follow the same basic principles of horsepower development and transfer that an electrically powered motor uses by converting one form of energy to another form of energy.
In this case, however, the output rotation and resultant power is derived from the chemical value of the fuel used (i.e., potential energy) by converting to an output of kinetic energy through a series of controlled implosions from an assembly of alternating pistons contained within a series of closed cylinders rather than the electromotive magnetic force developed from electricity.
The efficiency of an internal combustion engine depends on several factors. The two most important are the expansion (compression) ratio and the specific heat ratio of the fuel in use. For any heat engine, the work which can be extracted from it is proportional to the difference between the starting pressure and the ending pressure during the expansion phase, generally expressed in pounds per square inch-atmospheric, or psia. Therefore, increasing the starting pressure is an effective way to also increase the work extracted.
The generated horsepower rating and output of an engine also does not follow the same relationship of that within an electric motor. This is due to a massive difference in the energy conversion from the fuel to the engine along with a much lower inherent efficiency—around 25% to 30% for a well-tuned gasoline engine and up to 40% to 45% for a typical diesel engine. This basically means for every single brake horsepower (BHP) required by the load, a gasoline engine must generate nearly 3-4 HP from the fuel and a diesel engine approximately 2.5-3.5 HP from its fuel.
Engine performance curves are designed to account for these losses, as well as the parasitic losses associated with the engine’s water pump, cooling fan, alternator, and other loads. This generally results in curve values displaying the net HP output of the engine.
The principle behind the energy obtained from an engine uses what is known as a combustion power cycle. This cycle is similar in concept to a vapor power cycle where energy is extracted from the vapor produced from a fluid, such as vapor from water for steam generation or boiler feed. However, unlike the vapor cycle, the fluid to generate energy in a combustion cycle is capable of only being used once in the cycle and cannot be reused or returned to its original state for use in another cycle from makeup fluid.
The technical terms applied to a typical combustion power cycle is an air-standard Otto cycle for gasoline or similar fuels or an air-standard diesel cycle for diesel fuel.
In pure physics, an Otto cycle includes an initial isentropic air-fuel mixture development and compression in a cylinder, followed by ignition in the cylinder through a constant volume combustion explosion. This results in an isentropic expansion of the air-gas mixture.
In contrast, a diesel cycle has an isentropic compression of the fuel followed by a non-explosive combustion at a relatively constant pressure, resulting in the isentropic expansion.
The definition of an engine can be applied to fit many types of machines, but for the purposes of our discussion an engine will be limited to that found in an internal combustion engine. That is, an engine using fossil fuel (hydrocarbons such as potential energy sources, gasoline, natural gas, biogas blends), propane, diesel, or biodiesel blends.
All these types of engines convert the known and predictable value of the potential chemical energy contained in a unit volume of a specific type or blend of fuel by mixing it with an oxidizer to create the combustible mixture. Generally, the source of the oxidizer is oxygen derived from the open atmosphere from the surrounding ambient air, a readily available source of oxygen since it comprises approximately 21% of the volume of atmospheric air at sea level. In a gasoline engine, between 12 to 18 parts of air by weight is used per one part of gasoline to create the proper fuel-air mixture. A diesel mixture consists of approximately 15 to 23 parts of air by weight per part of fuel.
If there is not enough oxygen for proper combustion, the fuel will not burn completely and will produce less output energy. This is referred to as a lean condition.
An excessively “rich” air to fuel ratio will increase pollutants from the engine. If all the oxygen is consumed because there is too much fuel, the engine’s power is also reduced.
For a stationary engine, the fuel is either generally stored in bulk volume on-site or transmitted and delivered to the engine through a dedicated supply line. With gasoline and similar fuels, it is directed or pumped into a special chamber (often referred to as a carburetor) and mixed with sufficient air to create a combustible form of thermal energy. This mixture is then sent or injected to each engine cylinder, in turn, and combusted to create the resultant mechanical energy.
Depending on the type of engine and fuel, the fuel-air mixture or air alone is then routed to the upper region of a cylinder where the mixture is immediately compressed to a typical value between 150–175 psi and then:
- Spark-ignited using a high-voltage DC spark-producing device (spark plug) (with gasoline, natural gas, or LPG). See Figure 1.
- Compression ignited (with diesel) using an extreme value of combined pressure at roughly three to four times the value (450–700 psi) for gasoline, which also adds heat onto the air
and results in the thermodynamic explosion of the fuel-air mixture when the fuel is injected into the cylinder and contacts the hot and highly compressed air. See Figure 2.
Whatever they are called—drives, gearheads, gearboxes, driveheads, heads, or the preferred term of right angle geardrives—for well pumps, they are generally used to transfer the output power from a horizontal engine or electric motor to the input of a vertical lineshaft driving a deep well vertical turbine, mixed, or axial flow pump.
A typical right angle geardrive (Figure 3) uses an enclosed set of bevel gears on a horizontal orientation matched to another set of bevel gears oriented to drive on a vertical plane. The up or down ratio change in speed between the two axes depends on the size of the gears and the teeth of the respective gears.
Speed changes are usually expressed as a direct ratio with the input (driver) horizontal speed versus the output (driven) vertical speed proportionally indicated. For example, an input speed of 1800 RPM with a desired output speed of 3600 RPM would be expressed as a 1:2 speed increasing ratio, while the same input speed of 1800 RPM to a desired output speed of 600 RPM would be expressed as a 3:1 speed decreasing ratio.
A commonly used 1 to 1 or 1:1 ratio neither increases nor decreases the driver to driven speed, but simply accepts the input speed at one value and delivers the same speed to the vertical shaft. Appurtenant geardrive ratios along with the associated input and output speeds are shown in Table 1.
The horsepower rating of a normal right angle geardrive is dependent on several factors, including the size and metallurgy of horizontal and vertical gears, shafts, and bearings; specific gear design and number of teeth; rotational speed and ratio; gear and bearing lubrication efficiency; and thrust capacity.
Oil-bath lubrication is typically used to distribute adequate lubrication onto the internal meshing surfaces (teeth) and the input and output bearings from an oil reservoir and feed pump located in the bottom of the geardrive. Once delivered to and having lubricated the necessary gears and bearings, returning oil drops down into the oil reservoir for cooling and repumping back to the required areas.
A simple but continuous oil spray or drip lubrication followed by cooling of the oil in the reservoir is often conducted as the sole method of lubrication. But in many cases supplementary cooling of the oil by routing a steady flow of water through the geardrive is performed to stabilize and lower oil temperatures and provide better heat dissipation and lubrication, particularly in regions or enclosed areas with high ambient temperatures above 90°F or severe conditions.
The efficiency of a geardrive also depends on the degree of its full load rating and speed variation. However, most geardrives are slightly higher in efficiency at fully loaded conditions than at reduced or partial loads, but operate at a reasonably high efficiency nonetheless, usually doing so between 95% to 98% (Table 2).
Special Geardrive Configurations
There are numerous variations, combinations, and options available with right angle geardrives. These include speed ratios (up or down), shaft size, horsepower rating, self-release coupling or non-reverse ratchet, hollow or solid shaft, rotational direction, thrust capacity, and multiple drive combinations. However, there is one basic factor common to all right angle geardrives. They accept a horizontal input speed of one value and convert it vertically to another output speed through a ratio change in speed, up or down.
Beyond that factor, the remaining three fundamental options to a geardrive are:
- One or two horizontal inputs and one vertical output
- One or two horizontal inputs or one manual vertical input and one vertical output (combination head)
- One or two horizontal inputs or one automatic vertical input and one vertical output (clutch driven-combination head).
Right angle geardrives can be configured with an electric motor mounted above it that normally drives the well pump with the drive simply acting as a spacer and remaining stationary between the motor and discharge head or going along for the ride.
Although this application during normal electric motor operation of a combination head does not normally involve the geardrive itself, a steady bearing mounted in the geardrive is used to maintain smooth operation and prevent vibration. If failure of the motor or electric power occurs, a rapid manual conversion to the geardrive can be conducted to allow alternate operation from an engine.
Handling the pump thrust is performed by way of the electric motor’s thrust bearing (if remaining in place) or a separate thrust bearing in the geardrive if the motor is removed for repair and the pump load applied to the drive.
A final type of combination geardrive still uses an electric motor mounted above the geardrive, but in this case the system is configured using a special set of controls that can start and warm up the auxiliary engine (or other standby power source) upon power failure or an independent control signal.
In addition to starting the standby engine, the controller also relays a DC output voltage to an electric clutch mounted on the geardrive. This clutch automatically engages the standby engine to the driven load, usually a well pump, to replace the electric motor, but the motor usually carries the thrust load if or until it is removed for repair.
Upon restoration of electric power or satisfaction of pressure or another input signal, the electric clutch on the geardrive is disengaged, removing the load from the standby source and returning it to the electric motor. The proper application of lockouts and safety devices in the control system preclude the simultaneous operation of the electric motor and standby power source.
This type of system is available under various names depending on the manufacturer, although the most recognized name is likely the Redi-Torq system manufactured by Johnson Gear Co. This system is popular and widely used where continuous delivery or pumping of water is a necessity. For example, many closed-loop water systems without reserve water storage or stormwater/sewage pumping systems often use this method. An example of a 1500 GPM water pumping system built in 1985, using an electric motor with a diesel engine and right angle Redi-Torq style of geardrive, is illustrated in Figure 4.
Drive Couplings and Drivelines
Many end-suction or split case centrifugal pumps do not include a power conversion means, and are attached directly to the engine using an SAE fit (Society of Automotive Engineers), either through a clutch that is direct (frame) mounted or a power transmission means, such as a stub shaft, belt drive, or gearbox.
The need to either increase or decrease speed between the engine (driver) and pump (driven) is generally accomplished through either a belt-drive arrangement or a speed reducing or increasing gearbox. Many pump units require some method of coupling between the pump and driver. These devices are most commonly called drive couplings and they come in many different configurations and styles.
Most drive couplings, however, have one common variable: two solid halves used to connect the pump and motor, and a flexible element between these parts. The connection half-couplings are made from steel, cast iron, or aluminum while the flexible element (almost always placed between the two halves) is made from natural or synthetic rubber.
Problems with drive couplings are typically confined to three distinct situations: original misapplication and undersizing, misalignment, element fatigue.
Problems with the original application are most often the result of using too small of a flexible element. Since the horsepower required by the pump must be totally transferred across this element, proper selection is critical for proper life and performance.
Generally, selection of drive couplings is performed by considering three factors: rotative speed, horsepower, and expected life. Application using only one or two of the three will often result in premature failure and increased repair cycles.
Usually, oversizing of a coupling is not necessarily a bad economic decision, especially when considering the cost of one small component against the importance it may have in the operation of the plant.
When replacing an existing unit, I suggest double-checking the original selection to verify the application. This simple step could save future problems and downtime.
Misalignment is a very common problem and is more of a critical problem when using couplings with little built-in flexibility. All couplings are designed for some measure of deflection and every technician should verify and record the allowable deflection for their specific drive coupling.
Considering use of a coupling with a greater degree of deflection (such as a bellows or tire-type coupling) can often solve ongoing problems with a different type of coupling. Whichever coupling is used, alignment across both ends, driver and driven, are critical and must be maintained.
Newer methods of alignment, such as laser alignment, are an excellent tool to help with those stubborn couplings that just seem to wear out too soon. A facility that uses many units with flexible drive couplings should consider investing in a laser alignment tool, as the investment can pay itself back in a short time.
Although laser alignment, when properly performed, is a superior method of alignment, other time-honored methods, such as caliper and straightedge alignment, can also be used. The key to good alignment between a pump and driver is knowledge of the alignment method used.
Finally, element fatigue is a condition in which the rubber element breaks or starts developing fatigue cracking in a short period of time. This particular problem is often due to overloading of the element, jogging or repeated starting and stopping cycles, or excessive stress during start-up.
Verification of the allowable shock load of the coupling and possible stepping up to the next size will usually alleviate this problem. In some cases, installation of a reduced or soft starter will not only extend coupling life, but decrease wear and tear on other system components such as electrical equipment.
Generators are popular in water well and water works applications for primary, supplementary, or emergency backup service. Proper sizing of a generator is crucial to the long-term success of any generator installation and requires a good working knowledge of electricity and motor load characteristics, as well as the varying requirements of the other electrical equipment comprising the load.
When analyzing an electrical load, consult the nameplate data on each major piece of equipment like motors to determine the starting and running requirements in terms of watts, amps, and voltage.
When choosing a generator output for water applications, it is suggested the designer select a kilowatt rating that is approximately 20% to 25% higher than the peak load. For example, if the design load is 100 kilowatts, selecting a 125 kW generator set is recommended.
A higher-rated generator will operate comfortably at approximately 75% to 80% of full capacity and provide a margin of needed flexibility if the load increases in the future. This is particularly important on deep well applications where the motor horsepower may increase in the future to handle higher capacity and pumping head.
For safety reasons, a qualified electrician or installation technician who is familiar with applicable codes, standards, and regulations should perform all wiring and installation procedures. It is also essential to comply with all applicable regulations established by OSHA and maintain strict adherence to all local, state, and national codes.
Before selecting a generator, check for all municipal ordinances that may dictate requirements regarding placement of the unit (setback from building and lot lines), electrical wiring, gas piping, fuel storage (for liquid propane or diesel tanks), and sound and exhaust emissions.
All generators work on the principle of dynamically induced electromotive force (EMF). The change in flux associated with the conductor can exist only when there is a relative motion between the conductor and the flux. This relative motion can be achieved by either rotating the conductor with respect to the flux or by rotating the flux with respect to the conductor.
So, a voltage and current are generated in a conductor so long as there exists a relative motion between the conductor and the flux. Such an induced EMF is called a dynamically induced EMF.
Thus, a generating action requires the following basic components to exist:
- Conductor or a coil
- Relative motion between the conductor and the flux.
In order, that current can be obtained from an electric circuit and an electromotive force (voltage) that must be established and maintained between the two ends of the circuit. This electromotive force may be established in several ways, one of which is by means of an electromagnetic generator.
Michael Faraday discovered an electric potential can be established between the ends of a conductor in the following three ways:
- By a conductor moving or cutting across a stationary magnetic field (DC generator)
- By a moving magnetic field cutting across a stationary conductor (AC generator)
- By a change in the number of magnetic lines enclosed by a stationary loop or coil (transformer).
Faraday’s law states: “The EMF (electromotive force) induced between the ends of a loop or coil is proportional to the rate of change of magnetic flux enclosed by the coil; or the EMF induced between the ends of a bar conductor is proportional to the time rate at which magnetic flux is cut by the conductor.”
This law emphasizes rate of change or rate or flux cutting rather than density or extent of magnetic field.
Lenz’s law states, “A change in the magnetic flux passing through or linking with a loop or coil causes EMF to be induced in a direction to oppose any change in circuit conditions, this opposition being produced magnetically when current flows in response to the induced EMF.”
Whenever there is a change in current in a magnetizing coil, a voltage is induced which tends to prevent the change. Thus, if we attempt to diminish the current flowing in a magnetizing coil, a voltage will be developed that will tend to keep the current unchanged. Likewise, if we attempt to establish a current in a magnetizing coil, a voltage will be developed that will tend to keep the current from increasing.
To produce voltage, it is necessary to move a conductor through a magnetic field as stated above. Mechanical energy is required to provide motion to this conductor. With the field energy remaining constant, the conductor is changing mechanical energy into electrical energy.
There is a definite relationship between the direction of the magnetic flux, the direction of motion of the conductor, and the direction of the induced EMF. The voltage and current output are perpendicular to both the motion of the conductor and the magnetic field. The voltage produced at any instant of time is proportional to the number of turns in the coil multiplied by the rate of change of flux.
For example, the rotational field and sine wave begins at 0° as shown in Figure 5. The flux (EMF) then increases as the field moves toward 90° and becomes zero again at 180° since the plane of the coil is now parallel to the magnetic field.
The flux then increases in the opposite direction, reaching a negative maximum at 270° and diminishing again to zero at 360°. The flux reverses and increases again in the original direction to reach a maximum at 90°.
If the coil shown in Figure 5 were rotated at a constant speed in a uniform magnetic field, a sine-wave of single-phase AC voltage would be obtained where both the amount of flux enclosed and the induced voltage are plotted against time, as indicated in the bottom of Figure 5.
A three-phase generator will simply have three single phase windings. When starting motors, large voltage and frequency dips may occur if the generator isn’t sized properly. Other loads connected to the generator output may be more sensitive to voltage and frequency dips than the motor or motor starter, and this may cause problems.
For example, a rate of change greater than 1 Hz/sec in generator frequency may cause some static UPS units to malfunction. If the load on the generator set is a single large motor, particularly one requiring high starting torque, a number of problems can occur. They include sustained low-voltage operation that can cause overheating, extended load acceleration times, opening of circuit breakers or motor protective devices, and more.
When a generator is attached to a prime driver, such as an internal combustion engine, to create the rotational force and power needed to operate the generator in order to generate electric power, the combination is usually referred to as a generator set or genset for short. A generator set’s ability to start large motors without excessive voltage and frequency dip is a function of the complete system working in concert. This includes:
- Engine size and output HP
- Generator’s capacity
- Response of the generator’s excitation system
- Energy stored in the rotating inertia of the generator set
- Acceleration of the motor and its load.
All these factors must be considered for proper genset sizing. A simple rule of thumb for estimating the size of an engine-generator set for single motor starting is: 1 kW of generator set rating is required per each 3/4 to 1 HP of motor nameplate.
Most induction motors have typical starting characteristics. The curve of motor current versus speed shows that during full voltage starting, the motor draws approximately six to seven times its full load current. This current remains high until the motor accelerates and reaches about 80% of speed. This high inrush current causes a resultant dip in the generator output voltage.
The electric power initially required by the motor (with the motor at rest) is about 150% of its rated power. The power required by the motor peaks at about 300% of rated power and 80% of speed with full voltage applied. However, the genset supplies less than 300% power because the starting voltage is lower than the full voltage during acceleration and because the genset’s rotating inertia also transfers energy to the motor.
The motor must develop greater torque than required by the load. The motor’s torque curve at full voltage is above the load’s torque curve. The difference between the torque developed by the motor and the torque required by the load determines the rate of acceleration. Since torque is proportional to voltage, any reduction in voltage means a proportional reduction in torque.
A properly sized genset will support the high starting kVA requirements of the motor and maintain sufficient output voltage for the motor so it can develop adequate torque to accelerate the load to rated speed. All gensets use synchronous generators with exciters. They are primarily available with shunt (self-excited), excitation boost, auxiliary winding, or permanent magnet generator (PMG) (Figure 6) excitation systems.
All methods use some type of automatic voltage regulator (AVR) to supply DC output to the exciter stator. The exciter rotor AC output is rectified to a DC input for the main generator rotor. More advanced systems use an additional input to the AVR.
Although the shunt method is the most cost-effective method of excitation for linear loads, the PMG is the most commonly used method for AC standby and emergency service to operate motors and varying non-linear loads and provides excitation power independent of the generator terminal voltage. As such, it can maintain full excitation even during transient loading events such as the starting of a motor.
Full excitation power results in a less extensive voltage dip and improved recovery times. Though a voltage dip often causes various problems, a controlled reduction in voltage at the motor terminals can be beneficial, but only when a reduction in motor torque is acceptable.
Reducing motor starting kVA can reduce the required size of the genset, reduce the voltage drop, and provide a softer start for the motor loads.
When sizing gensets, you must first determine the acceptable level of motor torque required during starting, or the loads will accelerate slowly or even fail to reach full speed and ultimately cause motor damage.
Solid-state starters can adjust the starting torque, acceleration ramp time, and current limit for controlled acceleration of a motor when it starts.
For the purpose of sizing a genset, the current limit adjustment reduces the inrush current and may be used to reduce the starting kW and kVA requirement on the generator.
The range of available current limit settings is typically from 150% to 600% of full-load current. A 600% current limit setting on the solid-state starter results in a genset sizing that’s roughly the same as an across-the-line starting method. A 300% current limit setting reduces the starting kVA by up to 50%.
Use of the current limit setting also reduces motor torque available to the load. From a genset sizing perspective, an extended acceleration ramp time and low current limit setting (if appropriate for the motor and the driven mechanical load) would result in the least voltage and frequency variations.
One downside to using solid-state motor starters is their integral SCRs (silicon-controlled rectifiers) will cause voltage distortion. To compensate, it may be necessary to oversize the generator. The recommendation is two times the running kW load except when using an automatic bypass.
If the solid-state starter does have an automatic bypass, the SCRs are only in the circuit during starting. Once the motor is running, the bypass contactor closes and shunts the SCRs. In this case, the voltage distortion can be ignored during starting and additional generator capacity is not needed.
All versions of variable frequency drives (VFDs) are current limiting and reduce starting kW and kVA. The current drawn by these drives is nonlinear (i.e., possessing harmonics), which causes a distorted voltage drop across the reactance of the generator.
Since VFDs are nonlinear devices, an additional generator capacity sizing factor to keep voltage distortion to a reasonable level of approximately 15% total harmonic distortion (THD) or less must be included. The larger the generator, the greater the reduction in impedance of the power source (generator), which in turn reduces the effects caused by the harmonic current distortion.
For six-pulse VFDs, a typical generator sizing factor would be twice the running kW of the drive, which offsets any reduction in starting kW and kVA. If the drive is the pulse width modulated (PWM) type (or includes an input filter to limit current distortion to less than 10%), then the sizing factor can be reduced to 1.40 times the running kW of the drive.
Support Brackets and Bases
What can I say—this is where it all starts. Without adequate support bases under the pump and driver, the unit will vibrate and soon fail. Too many cases of inadequate support under the pump/driver have caused numerous problems with the unit itself. This includes excessive wear of running surfaces, excessive vibration, noise, premature failure, and breakage to name a few.
Generally, sizing of the support base is the job of the design engineer, but in many cases new installations or retrofits require the technician to become his own engineer.
In most cases, the answer is one word: mass. Simply put, mass is weight, and the more resisting weight you can provide for the pump to rest on, the less likelihood it will start to walk to China!
Obviously, when I discuss mass I also assume adequate methods of securing the pump to the platform have been provided. Supporting mass is useless if the anchor bolts are not securely held into the base.
In lieu of using a heavy and often expensive concrete base, a sturdy base fabricated from steel can be used. When using a steel base, however, remember steel loves to corrode—especially when in continuous contact with water and most definitely when in chlorinated water, as is often the case with water pump bases.
Good fabrication techniques, including full penetration welds and a good coating system, can significantly lengthen the life of a steel base. Support brackets for pipe should also be anchored to a resisting mass (most often a concrete floor) and should fully support the pipe within 1 or 2 feet of the suction and discharge nozzles on the pump.
This concludes this two-part series on the various kinds of drivers used for pumps. In the next installment we will discuss the many diverse piping and valving methods used for pumping installations and stations.
Until then, keep them pumping!
Ed Butts, PE, CPI, is the chief engineer at 4B Engineering & Consulting, Salem, Oregon. He has more than 40 years of experience in the water well business, specializing in engineering and business management. He can be reached at firstname.lastname@example.org.