The two primary components in any water distribution system are the pumps that cause the fluid to flow and the valves that regulate this flow. And while such mechanisms represent mature technology that has been around since the start of the industrial age over 200 years ago, technological advancement has in no way plateaued. There remains virtually unlimited opportunity to improve the performance and efficiency of these basic mechanisms. These advances exist in the efficient use of materials for manufacturing, superior controls utilizing artificial intelligence algorithms, ever-higher levels of energy efficiency, and reducing operational costs that affect the bottom line.
Efficiency is of vital importance in determining the pump system’s economics. For a water supply plant, pump operations can represent 30% to 50% of the facility’s overall operating costs. For most communities, water supply and wastewater treatment plants are the single largest consumers of electricity. A small, 50 horsepower (hp) pump needs about 37 kilowatts (kW) to operate. At a national average cost of electricity of $0.065 per kilowatt-hour (kWh), this hypothetical pump would cost over $14 per day to operate (assuming a typical 6-hour running time per day), equivalent to about $5,270 annually. And this cost can vary greatly with location, season, and time of day—with peak demand time costs increasing by up to 80%. So, for water supply and wastewater treatment facilities (as well as any other user of pumps large or small), there is a strong incentive to maximize operating efficiency whenever possible.
PUMP TYPES AND OPERATIONS
There are two wide categories of pump types: positive displacement pumps and centrifugal pumps. The latter can be configured to operate either as a submersible pump operating underwater or outside the water as an extraction pump. Each type has an inherent advantage in managing different kinds of fluids: from clear water from water supply systems to water with high total suspended solids content taken from excavation pits; groundwater extracted from well points; fluids with large suspended particles; raw sewage containing large amounts of organics and man-made waste materials; and polluted water from sanitary systems, leachate removed from solid waste landfills, and highly viscous fluids (such as slurries and sludges, oils, hydrocarbons, etc.).
Specially designed pumps are used to handle other fluids besides water, or fluids with large suspended particles as well as floating waste and debris, and highly contaminated industrial waste flows or urban sewage. These include effluent pumps, grinder pumps, and sewage pumps. Each type is classified by the size or diameter range of particle sizes carried by the fluid and can manage these particles without clogging. Fluids with particles ranging up to 3⁄8 inch in size (equal to fine gravel) are managed by standard sump pumps. Fluids carrying particles up to ½ inch in size (approximately that of medium gravel) are handled by effluent pumps. Fluids containing very large particles (up to 2 inches, equivalent to cobbles) are managed by sewage pumps. Increased particle size requires higher flow rates, typically at the expense of increased operating head (or pressure). At the very largest particle sizes, over 2 inches, grinder pumps are used. These differ from other pumps in that they mechanically grind the particles into finer sizes and smaller diameters to make a slurry that is subsequently pumped.
The above specialty pumps are typically of the positive displacement type. This first pump category relies on the cyclical motion of a rigid or impermeable mechanical device that literally pushes a fixed amount of liquid with each movement cycle. The movement of the piston is regulated by a rotating cam assembly. The device can be a reciprocating piston that pushes the fluid out of its cylinder with each forward movement and allows more water to enter the cylinder as it moves back to its cycle’s starting position. The fixed amount of liquid in the cam and the known number of pumping cycles allow for very accurate measurements of pump flow without the need for direct metering. In this operation, check valves are essential to prevent the backflow of water from the discharge pipe back into the piston cylinder.
Though the reciprocating displacement motion is generic to this type of pump, the piston configuration is not. There are many types of displacement pumps. Each uses a unique mechanical configuration to move the fluid forward. For example, bellows pumps use a bellows cavity, whose interior space cyclically expands and contracts to propel the fluid, taking in fluid with each expansion and expelling it with each contraction. Similarly, diaphragm pumps utilize a vibrating flexible membrane to propel liquid to the discharge pipe. Some types of pumps literally pinch or squeeze liquid forward. These include flexible liner pumps, which use the inert surface of a liner and a rotating body block and peristaltic tubing that uses a pair of rollers to squeeze liquid through a series of flexible tubes. These pumps have the advantage of being able to operate dry without burning out and avoiding physical contact with the liquid to prevent cross-contamination.
Centrifugal pumps differ from displacement pumps in that they utilize (as their name suggests) centrifugal force to fling fluid towards the discharge. The centrifugal force is provided by the high-speed rotational spin of the pump’s impellor. Encased in a watertight housing, the impellor rotates around a central drive shaft. It is constructed of vanes or blades arranged so that the liquid (while subject to the centrifugal force) enters the pumps from a central inlet port and travels up the surface of the vane to the exterior where it is expelled. These pumps tend to operate at high flow rates but lower operating heads.
As with displacement pumps, there are several varieties of centrifugal pumps, each with its own unique operating characteristics and applications. In some models, the fixed impellor blades are replaced with flexible impellors. These will deform slightly while spinning to more efficiently trap and sweep up fluid being sent to the discharge pipe. Their efficiency is also matched by their corresponding lower costs. A similar design utilizes rigid vanes so as to operate at higher pressures. And some blades have a lobe shape whose curvature allows the pump to more effectively handle liquids with high viscosity.
Some centrifugal pumps utilize a completely different configuration than the standard impellor blades. Harkening back the design of the ancient Archimedes screw, progressive cavity pumps are used to handle high-viscosity fluids that may contain large particles and objects. A spinning screw raises the fluid through the pump until it is discharged into the system. Gear pumps use a rotational movement of two enmeshed gears to squeeze water out of the pump. These pumps have the advantage of operating with zero vibration at high pressures.
HOW TO MEASURE PUMP OPERATIONAL EFFICIENCY
Pump operation and efficiency are defined by a family of performance curves that compare various factors to either flow rate or operating head. The pump curve compares the pump’s inherent operating capacity in terms of its flow rate and operating head, with “head” being a measurement of fluid pressure defined in equivalent feet of water column. The unit weight of water is 62.4 pounds per cubic foot (pcf). One foot of head is, therefore, 62.4 pounds per square foot (psf) or 0.433 pounds per square inch (psi). This curve starts at a maximum resisting head (referred to as the “shut-off head point”) that represents the pump’s operating head at its start-up, and curves downward as the pump operates at increased flow rates.
Similarly, the fluid distribution system generates resistance to flow, which can also be measured in terms of head per flow rate. This second curve arches upward from a value of zero head for zero flow rate (when the pump is not operating). As the pump’s flow rate through the distribution system’s pipe network increases, so does the resisting head. The system resistance curve has a minimum resistance defined by its static head, determined by the elevation difference between the pump’s intake port and the distribution system’s highest discharge point. The system resistance curve increases from this minimum head value according to its velocity head and friction head.
Velocity head is calculated by dividing the square of the velocity by twice the acceleration due to gravity (g = 32.2 feet per second squared). Friction head is derived from the pipe's friction factor, which in turn is derived from the Moody diagram which compares the flow’s Reynold’s number with the curve associated with a smooth-walled pipe. Lastly, the Reynold’s number is derived from the flow velocity and the pipe diameter. Once the pipe flow’s friction factor is determined, Darcy’s equation can be used to determine its friction head losses.
Adding the static, velocity, and friction heads together gives the total system resistance head—which increases with flow rate and increased velocity. The pump’s operating point (defined in terms of matching head and flow rate) occurs where the downward pump performance curve crosses the upward arching system resistance curve.
Other operational curves are used to determine additional pump performance characteristics. The pump’s Net Positive Suction Head (NPSH) is defined as either the available head (NPSHa) or what is required (NPSHr) by the pump without being subject to potentially damaging cavitation.
The Brake Horsepower (BHP) curves are superimposed over the pump’s performance curve and show the total horsepower required to operate the pump at that particular combination of head and flow. BHP is typically measured in watts and is a measurement of the mechanical horsepower available at the shaft at specified revolutions per minute. It is the useful power that is actually produced by the pump. Information from this curve is used to determine energy requirements and power costs to operate the pump. The horsepower generated by a pump is calculated as follows:
HP = (H x Q x 8.33 x SG) / (33,000 foot pounds per minute)
Where:
HP = the pump’s power output (horsepower)
H = head (feet)
Q = flow rate (gallons per minute)
8.33 = conversion factor, weight of water (pounds per gallon)
SG = the specific gravity of water at a given temperature (equal to 1 at 68°F)
Lastly, there is the efficiency curve, which is measured in percent. Efficiency is the ratio between the amount of work actually performed by the pump and the amount of energy put into the pump from a power source. For example, a pump performing work equivalent to 90 watts, while being powered by a 100-watt power source, would have an operating efficiency of 90%.
The efficiency curve is parabolic. At the point of zero flow (the system is shut off), efficiency is 0% as no work is being performed. The curve arches upward with increased flow rates until it achieves an operating state referred to as the “Best Efficiency Point”—the flow rate where the pump is operating at maximum efficiency according to its size, design, impellor dimensions, etc. Then the curve arches back down to a low point associated with the termination of the pump’s performance curve at its maximum flow rate and zero performance head.
VALVE TYPES AND OPERATIONS
Pumps generate liquid flows at applied pressures. Valves regulate these flows and their operating pressures. As such, valves serve as a moderator to adjust the overall performance of the pumping system. One is necessary for the proper functioning of the other. Valves are one of the appurtenances and fittings that are part of the water distribution pipe system.
Valves perform various functions to regulate the operation of a water distribution system. This includes both shutting flow off completely and throttling it down to a lower flow rate. In the former, the pump may still be operating but unable to propel liquid forward, leading to possible damage to the pump if this is prolonged. Reducing allowable system flows increases the pump’s operating head and could reduce its efficiency.
Multiport valves can also be used to divert flows into branch lines of the pipe distribution system. This could result in a significant change to the pipe system’s resistance curve (if water is diverted to a section of the pipe network with a high static head and discharge point elevations, for example) and with it, a change the pump’s operating efficiency and operating head.
Valves can be used to regulate the pressure within the pipeline distribution system. This is primarily done by pressure relief valves, which allow built up water or steam to escape, thus reducing pressure within the pipes and preventing it from getting too high. This can be a problem not just for pump performance, but for the structural integrity of the piping system itself. Pressure management involves the installation of add-on pressure-reducing valves (PRVs) that reduce high-pressure locations in the system and minimize the potential for leakage. But the use of too many PRVs is costly and can affect necessary pressures and resultant flow rates within the system.
The types of valves are as varied as the functions they perform. Isolation and flow regulation valves include globe valves, gate valves, ball valves, plug valves, piston valves, pinch, and butterfly valves. Each uses a different mechanical configuration to perform the main task of shutting off flows. Safety pressure regulation and relief are performed by pressure relief valves and vacuum relief valves. Diversion and non-return valves include swing check valves and lift check valves.
The most common type of valve is the gate valve. This can be operated either manually or by automated machinery. A linear motion valve is either fully opened for full flow or fully closed for shut off and is commonly used in non-water applications such as fuel, lubricants, and hydrocarbons. Globe valves are used for flow regulation. Though more flexible than a gate valve and providing a better seal for shutoff, globe valves are more expensive. Check valves ensure that flow remains in one direction with any reversal of flow causing the valve to automatically shut off. Ball valves utilize a ball-shaped disc to shut off flow with a 90-degree rotary turn. Butterfly valves are typically used for large pipe diameter applications since they are relatively compact and lightweight. Pinch valves are specialty valves used with rubber piping to pinch off the pipe like a clamp and shut off flows.
As noted by Tim Fallon, product line leader at Mueller Water Products, “Choosing the right valve for the right application can significantly impact performance and energy conservation. Efficient valve utilization can be achieved by selecting pipes and valves for minimal head loss or high coefficient of flow (Cv).”
He also explains the effects the right valve can have on pump performance and overall system efficiency:
“There are many different types of valves that have been designed for the same flow control application. While they may provide the same result, some are more expensive to operate than others…. When the pumping process is complete, the pump turns off and a valve will close to keep the pumped water from flowing (in reverse) to the source. Many different valve types can be chosen to perform this task. A check valve is typically considered for this type of application. However, depending on the specific application, a butterfly, ball, cone, plug, or diaphragm valve can also be used to perform the same task of checking flow and can sometimes be more efficient during flow operation.
It’s important to consider which valve would provide the most efficient flow (i.e. high Cv) while the pump is in operation, not just the shut-off performance. A valve engineer should suggest the best solution based upon the application conditions and the type of pumping process being used. This engineer will also take into consideration the overall physical environment such as power availability or space constraints. Finally, it’s important to understand the client’s budget. Each of the previously listed valve types can be used to provide the same result. If efficiency is not taken into consideration, however, the client could spend more in electricity consumption over the life of the system than what was initially estimated due to a valve with a lower Cv.”
Mark Magda, a senior technical sales and training manager at Mueller Water Products, also provides insight on recent valve technology advances. “Utilities are taking a flow-based pressure management approach. Controlling pressures based on demand in the distribution system is resulting in significant reduction in pipe breaks, consequential damages, reducing non-revenue water loss. Efficiencies are gained simply by reducing water pressure during low demand—10 psi results in a 6% reduction in NRW; a 20 psi pressure reduction yields a 14% reduction in NRW; and a 30 psi reduction results in a 23% reduction of NRW.”
HOW TO MEASURE AND IMPROVE VALVE EFFICIENCY
Like the pumps they regulate, valves have their efficiency determined by comparing the energy put into the valve to make it operate and the actual work performed by the valve. Valve efficiency also determines overall cost-effectiveness. An indirect improvement of valve operational efficiency is provided by automated controls.
Traditionally, automatic shut off has been regulated by a smart electronic positioner or a solenoid valve. Digital positioners are also used for emergency shutdown applications. The positioner is controlled by a 4 to 20 milliampere (mA) or fieldbus signal. The overall control loop is run by a higher-level process control system that regulates the shut off in coordination with the operation of the rest of the piping system.
The energy source used to operate valves can include electrical power to operate the valve (measured in kilowatts, used to operate solenoids, actuators, and positioners) and energy in the form of pressure (measured in pounds per square foot, whether internal flow pressures or pressure released by a relief valve). Of the two, the one that costs the operator is electric power. The second is the power actually generated by the closing valve as it throttles or shuts off flow.
The following improvements show the greatest promise for increasing valve efficiency. First is the development of smaller, more compact actuators that still generate the required force needed to perform throttling or valve shut off. Similarly, right sizing the valve for the job is essential. No advantage is gained from a valve with an oversized flow coefficient. In keeping with right sizing, the operator should select valve accessories that match demand and function.
Small things also have big consequences. Valves and actuators should be chosen that both minimize leaks and the additional friction head they add to the overall pipe system. As noted above, each valve or appurtenance adds the equivalent of feet of system resistance head.
In addition to an overall “smart” control system and process control loops, individual smart positioners can also make a big difference. These can detect leakages and prevent unstable controls that are poorly matched to the control task. They can also monitor and analyze energy usage by their control valves. But most importantly they can gather and assess online data concerning a valves status and performance. This requires that the design of the control valve and its components be stored in the system’s memory along with its operating data such as pressure and flow rate.
PRESSURE MANAGEMENT CASE STUDY
The City of Farmington Hills is the second largest city in Oakland County, MI, with a population of over 80,000. Most of the city has excellent water pressure and adequate fire protection but, as the name suggests, hilly terrain presents pressure challenges in certain areas.
With roughly 468 miles of pipe in their water distribution system, the city uses an average of 8.6 million gallons per day (mgd) from the Detroit Water and Sewerage Department (DWSD). DWSD offers lower rates for communities that reduce their maximum day and/or peak hour demands. While the city’s average daily consumption is good, their peak hour factor of 4.4 and maximum daily consumption of 21 mgd comes at a high cost.
The desired operating pressure for a water system is 50–80 psi, with a state required minimum normal working pressure of 35 psi. To ensure adequate pressure for fire protection, the required minimum is 20 psi during emergencies. When pressure is too high, leakage rates go up. When it is too low, fire protection is compromised. Nine areas were identified with pressure concerns that needed to be addressed. To reach the desired operating pressure, two styles of PRV needed to be installed at five different points in the distribution system, ranging in size from 6 to 24 inches.
The first type of PRV installed was the Singer S106-2SC-PCO-PR-SC-SPI-MV. This model provides for remote SCADA control of flow and pressure. In the event that there is a loss of power or SCADA control, a pressure reducing override takes over, ensuring pressure is maintained through the system. The city also needed the valve fully open in the event of a low-pressure differential across the valve. The valve-opening pilot on the low inlet pressure setting ensures that the valve can go to a full open position on low inlet pressure to maintain fire flow and operating pressures in the system. The valve was also fitted with upstream and downstream pressure transducers tied into the SCADA to provide for this functionality.
The second type of PRV implemented in the system was the Singer S106-2PR-SC-SPI-MV. This valve is equipped with two pressure-reducing Model 160 pilot controls with different set points. Through SCADA control, the city now remotely sets the system maintenance pressure. In addition, non-revenue water loss and pipe breakage have been reduced because of the utility’s ability to lower pressure in off-peak usage times, typically during the night and non-summer seasons.
All valves have the SPI-MV, a single point electromagnetic flow meter that is built into the valve and ensures an accuracy +/- 2% throughout the specified velocity range. This enables Oakland County to measure flow into each of the corresponding pressure zones, track usage, and detect pipe breakage. It also allows the county to control and regulate the flow from their supplier, DWSD, during the peak flow hours. Furthermore, the SPI-MV allows the city to fill their tank at off-peak hours and control the total draw by reducing flow to districts as needed.
“The valves have exceeded the County’s expectations by allowing them to smoothly and effortlessly control flow rates within a 20 GPM dead band,” says Carrie Cox, assistant chief engineer for Oakland County. With the construction of the 3-million-gallon storage tank and accompanying controls, peak demands could be reduced, resulting in an annual savings of $3,300,000.
MAJOR SUPPLIER
Mueller Water Products is a leader in the advancement of valve monitoring, especially control valves. Mueller Water Products is on the cutting edge. As noted by Harold Mosley, Hydro-Guard product manager at Mueller Water Products, their systems combine Hydro-Guard pressure monitoring solutions with control valves to provide a monitored solution. By monitoring pressure changes at the control valve, they can provide near real-time data that allows a utility’s operations and management teams to perform a multitude of “smart” tasks. They have developed a dynamic hydraulic model of the distribution network, as opposed to a static model, to help utilities with critical decision-making. This allows the operator to gather intelligence from field installations to improve pump operations and reduce energy costs. The end result is the operator’s ability to enhance his knowledge of pressures, including transient activity in the transmission and distribution network, to extend the life of piping, valves, and equipment.
Their Singer automatic control valves also offer advanced technology with real-time data reporting on system upstream and downstream pressures, flow rates, and position of valve open. By installing an SPI (Single Point Insertion) Magnetic Flow Meter, the control valve becomes the system meter. This eliminates the need for separate meter pits, resulting in significant cost savings for the utility. This solution reduces space required for a separate meter vault, additional permitting, excavation costs, and labor.
