In many technology areas, it is difficult to achieve significant energy savings; when they are, gains of a couple of percentage points are then celebrated as breakthroughs. There are technologies, however, that can deliver very significant reductions in energy use.
Foremost among these is the variable speed drive. It doesn’t make much noise or develop extreme temperatures or go through complex motions. In fact it sits in a cabinet and usually doesn’t even get a mention when the overall process is explained.
However, it can deliver great cuts in energy consumption, frequently nearly halving the consumption, and if applied in all relevant plants worldwide, it can deliver energy savings that equate to the electrical consumption of a country such as Spain.
The principle is simple: In the past, the motors that powered pumps were usually run at full power all the time, with the output controlled by valves. A drive regulates flow through direct control of the electrical power fed to the motor, so eliminating friction-based controls and associated losses. However, a lack of system standards for energy efficiency may result in up to 90 per cent of pump installations being incorrectly sized, leading to energy waste.
There may appear to be standards for everything, but in the area of energy efficiency there are still gaps. While there are standards for pump designs and many for the hydraulic data such as developed head, efficiency and net positive suction head, a search for standards providing guidance in system design is less likely to produce a result.
To use an analogy, if somebody were to buy a three-ton truck for use on shopping trips, it would do the job, but would not be a demonstration of energy efficiency - even if the truck selected boasted the best efficiency figures for three-ton trucks.
Figure 1 illustrates the problem faced by system design engineers. When planning a system, there is a degree of uncertainty as to the shape of the system curves (friction, pipe cross section changes and the number of bends in the final pipe layout all take their toll). These factors all add to the risk that the expected operating conditions will not be met.
There are three basic ways to address the changed operating conditions. If the changed condition is permanent, then the pump or fan size should be changed to match the load (adding to the installation cost). The pump or the fan speed can be changed, or the impeller can be modified, or a throttling device can be added, such as a valve, damper or guide vane. However, both of these options waste energy.
How systems get over-dimensioned
Despite careful analysis and design, many systems do not operate optimally. One reason is that many systems are simply sized too large to start with, resulting in higher operating and investment cost. To illustrate this, take the case of a system with a fan in a process plant. In this example, it is assumed that the ‘true’ nominal condition for the application is 100 units of flow, requiring 4,000 units of pressure (point ‘a’, Figure 3).
In order to be on the safe side concerning the maximum flow, the figure for the fan communicated to the engineer is 110 units of flow (point ‘b’ in Figure 4). With the assumed system curve, this would require a fan with a higher capacity (Figure 4, dashed yellow line) that can deliver 110 flow units and 5,000 units of pressure.
When establishing the fan capacity, the engineer estimates the overall pressure drop that these 110 flow units will cause (point ‘c’ in Figure 5). The pressure drop value that is calculated is increased by a ten per cent margin (point ‘d’ in Figure 5) because it is difficult to foresee whether the assumed number of bends in the duct will conform to that estimated. There is a possibly that the installation contractor will have to add bends to avoid other equipment, for example. Also, the cross-section of the duct may be uncertain. A smaller cross section would lead to a higher pressure drop. This therefore means that a ten per cent margin is not unreasonable.
So, what data are finally sent out in the requests for tender? Flow: 110 units at a pressure of 5,500 units (point ‘e’ in Figure 6). Even if the original assumptions were correct, the fan is now grossly oversized. At 100 units flow the necessary additional pressure drops over the damper must be about 2,800 units (pressure at ‘f’ minus pressure at ‘g’ in Figure 6). This corresponds to 70 per cent of the assumed correct total pressure. However, it is rare that 100 per cent of the design flow will be needed other than for very short bursts. Assuming that most of the time, 80 per cent of the flow rate will be required; the additional throttling needed across the damper will be about 6,000 units (pressure at ‘h’ minus pressure at ‘I’ in Figure 6). This corresponds to 150 per cent of the assumed correct total pressure.
The steps illustrated in this example are more common than they may seem. An additional factor is that, when it comes to the selection of a fan, this choice must be made from a standard range of fixed sizes. The next larger one will usually be chosen, with a motor sized to suit.
The correctly sized fan for this example at point ‘g’ should be 100 × 4,000 = 400,000 power units, and the normal running at point ‘I’ will require 80 x 2,500 = 200,000. The case above produces a requirement for a fan of at least 110 x 5,500 = 605,000 power units (150 per cent of the optimum). Correcting this with damper control leads to high levels of wasted energy. The additional losses at the 80 per cent flow point amount to 80 x (8,100 – 2,500) = 448,000 power units (120 per cent of the full power of a correctly sized fan). These figures will, in practice, become worse, because the fan will not be working at its optimum efficiency throughout the operating range. With a speed controlled fan instead of damper control, nearly all of this energy can be saved.
Hydraulic system gets an energy boost
The Corus site at Deeside develops and manufactures pre-finished steels. A hydraulic system is used on the production line for re-treating and inspecting strip material, driving actuators and web guiding systems in a 24-hour process. Typically, hydraulic systems waste much of their energy because a constant amount of fluid circulates at all times, although the work is only carried out in short bursts.
Corus Colours achieved significant energy savings by retrofitting the existing system with a variable speed drive. The pump speed was greatly reduced both when the system was in neutral and during cylinder actuation. When in neutral, power consumption was initially around 9kW. Under drive control, power consumption was reduced to 2kW, a reduction of 77 per cent. With the system under load, power consumption was reduced from 22kW to 12kW, a saving of 48 per cent.
With a 16 per cent duty on-load time for the system, the average energy saving over time was 70 per cent. The reduced energy consumption will allow a payback time of just 18 months and reduce the company’s carbon footprint by 33 tonnes of CO2 annually.
The reduction in energy consumption under load initially surprised the Corus engineers, as it should take the same amount of energy to move a hydraulic cylinder a given distance, regardless of whether a drive or direct-on-line operation is used. However, further tests showed that the drive used a lower motor speed to achieve the required pressure.
The drive automatically adjusts the pump speed to maintain system pressure, pressure feedback being returned to the drive from a transducer. The ABB industrial drive has built-in PID control that helps keep external values, like pressure, under control.
This application demonstrates that in any process where a restriction is used to control flow, energy can be saved, and in any process where volume can vary, energy can again be saved.
PLEASE REFER TO DIGITAL ISSUE FOR FIGURES