The MilliporeSigma Environmental Testing Partnership Program (ETPP) is a comprehensive offering designed to offer customers access to the...
Several types of smart products are available in both the commercial and industrial marketplace. The pump industry has also begun to incorporate computer technology to operate, control and protect pumps and their systems. A smart pumping system consists of a pump, variable speed drive or controller, instrumentation (when required), microprocessor and special software. The pump can be any standard centrifugal pump fitted with instrumentation to measure a process variable.
The value to the customer in using smart pumping systems is reduced life cycle costs, the major components of which include operating, maintenance, initial and installation costs.
System curve. A system curve comprises a static component and a dynamic component. The static component of the system curve does not change with flow rate. The dynamic component is essentially proportional to the square of the rate of flow. It is also a function of other variables such as pipe configuration/size, surface roughness, quantity and type of fittings/valves and fluid viscosity. These can be represented by a single system constant and the dynamic or frictional head can be expressed as: Hf = K Q2. The dynamic head constant K is a constant for a given system; however, if a control valve position changes in the system, the constant K will also change.
Conventional system. Figure 1 shows a typical control scheme for a conventional pumping system. Note the system curve includes a static component of 6.1 meters. This is the change in elevation between the suction and discharge source. In this type of system the pump operates at a fixed speed and the pump performance curve is based on an impeller diameter pre-selected to match the system requirements as closely as possible.
It is common practice to add a safety margin to the design point where it is difficult to accurately define system losses. This can result in an oversized pump that runs out too far on the curve and absorbs too much power. The total system head curve intersects the pump head-capacity curve at Point A with the control valve wide open. The design flow for this system is 638 m3/hr, so the friction head (Point B) must be increased to 37 meters in order to operate at this point.
The most common method of varying the capacity in a conventional system is to introduce a variable resistance device that will alter the system friction curve; this is the main function of a control valve.
Variable speed system. In a variable speed system the controller will match the pump output to system head requirements without the need for throttling a control valve. Safety factors and pressure margins typically built into dynamic head systems can be eliminated and in some cases result in a lower cost pump selection. Because the smart controller can adjust the pump speed to suit the required system conditions, only one impeller diameter need be stocked. This offers the benefit of lower inventory cost. In variable speed systems the design point no longer needs to be based on a fixed speed. Variable speed control is most effective in all friction head systems. The effectiveness diminishes somewhat for applications having high static head and low dynamic head curves, because the intersection of the pump and system curve moves further to the left of the best efficiency flow.
Systems with flat head curves can result in unstable flow, making control very difficult. Advanced products control the process variable by adjusting pump torque rather than speed. This provides a steeper, easy-to-control performance curve. Careful review of the application and product selection is required for these systems.
Figure 2 shows a variable speed system with system Curve A identical to that shown in Figure 1. Head-Capacity curves are shown at various speeds. If the desired operating flow is 638 m3/hr it is shown that the pump can operate at a substantially lower speed and head. In this example, the savings of the variable speed system over a conventional system are represented by the difference in head between points B (Figure 1) and C (Figure 2).
Life cycle operating cost. This difference in system head requirements can often translate to thousands of dollars in energy savings over the life of a pump. Table 1 exemplifies these savings and is based on a cooling tower installation represented by figures 1 and 2. Note that the total pump head has been reduced from 37 to 23 meters. As a result, the pump power has dropped nearly 30 kW and operating speed has been lowered by over 300 rpm.
The difference in energy costs between a conventional system and a variable speed system (see Table 1) represents a 35.5% decrease in operating costs over the life of the pump.
The primary components in pump failures are bearings and mechanical seals. Excessive vibration, excessive loads and/or poor lubrication are the primary causes of failure for these parts. Designing a pump with a larger shaft and bearings does not guarantee longer life. Many failures can be attributed to operator error and application factors.
The smart pump controller will match pump output exactly to system head requirements. This eliminates the need to add safety margins that will oversize the pump. With variable speed systems there is a better chance of finding a selection that will operate at or near the best efficiency flow; this usually occurs at a speed less than the pump maximum design speed.
Monitoring system conditions. Some of the more common causes of failures are attributed to the following upset conditions: dry running caused primarily by closed suction valves; deadheading due to a closed discharge valve; continuous operation below minimum flow; and cavitation due to runout conditions. Advanced smart pumping systems are capable of detecting all of these conditions by sensorless protection algorithms.
Protecting against transients. A sensorless flow monitor will detect and distinguish between a dry running condition, operation below minimum flow or closed discharge valve condition and a runout condition (too much flow). If flow falls outside specified user settings the response can be set to warn only, alarm and control or fault. Magnetic drive pumps can especially benefit from dry run protection. The alarm and control mode reduces the pump speed to a safe minimum speed until the transient has cleared and the pump can resume normal pumping operation. A protection delay can be set to distinguish between brief nuisance transients and prolonged process conditions. Other condition monitoring safeguards can warn and protect against various process variables, and setpoints can be selected to restrict operation to user specified ranges.
Smart pumping systems can also incorporate self-diagnostic features to compare current pump performance to the as new factory performance. One method of quantitatively predicting life cycle cost savings wherein reliability factors for operating speed, operating point and impeller diameter are assigned values between 0 and 1—higher values indicate more reliable selections. A reliability index, which is the product of the three reliability factors, can then be compared to pumps of similar design to give an indication of overall reliability. Table 2 shows the effect on life cycle maintenance savings for the cooling tower application using this method. It assumes an average MTBF for a conventional sealed pump of 18 months at an average cost per repair of $2,500.
The difference in pump life cycle maintenance savings for this smart pumping system represents a 33.5% decrease when compared to a conventional system, and MTBF is extended from 18 to 27 months.
When compared to total life cycle cost, smart systems can have an overwhelming advantage. The smart controller continuously monitors both pump and system conditions and matches pump output to system requirements exactly. Because the smart controller is a variable speed device, there is no need for an automatic control valve in most systems.
Advanced products are available that calculate pump flow by sensorless algorithms within ±5% of rated flow. This is sufficiently accurate for most applications. The sensorless flow value can also be used to provide pump protection, eliminating external flowmeters and/or auxiliary protective devices. Additionally, smart controllers have integral starters, and there is no need for a separate starter. One of the many safeguards built in to smart controllers is to protect against operation below minimum flow. Operation can also be restricted to user specified ranges. Depending on system design, recirculation lines and valves can also be eliminated. Because added safety margins are not required when controlling with a variable speed system, in some cases a smaller pump can be used. Table 3 shows that the life cycle initial cost savings for a smart pumping system compares favorably with a conventional system.
By reducing the amount of equipment in a system, both installation and maintenance costs are decreased. The installation costs associated with piping, air lines, wiring and communication lines can all be decreased by eliminating a control valve, flowmeter, separate starter, and recirculation line valve and piping (in some applications).
Table 4 shows the effect on life cycle installation savings for the application shown in Table 1. In this example, pump installation costs are based on 5x initial cost. Installation costs for the control valve, flowmeter and starter are based on 3x initial cost of these components. It is recognized that although a control valve is not required, some users continue to include them in their systems due to past practice or for auxiliary process use.
The difference in pump installation savings for this smart pumping system represents a 35% decrease as compared to a conventional system. Because the control valve and external flowmeter have been removed from the smart pumping system, maintenance for these items can also be eliminated, as shown in Table 5.
Total Life Cycle Cost
Taking into account these major life cycle cost components, the total life cycle savings for the smart pumping system is $268,650 for this installation. This represents a 35% saving over a 15-year equipment life. The present value of these savings, assuming a 10% interest rate, is $148,300, and the difference in initial capital investment is negligible. When comparing initial costs of conventional and smart pumping systems, smart systems can be very competitive, and when comparing total life cycle costs, smart pumping systems can have an overwhelming advantage.