Case Study from Rocky Mountain Institute - Efficient Pump Systems
Here's a story from Natural Capitalism: Creating the Next Industrial Revolution that illustrates the huge opportunities for improving energy efficiency in industrial pumping systems: In 1997, leading American carpet maker Interface was building a factory in Shanghai. One of its industrial processes required 14 pumps. In optimizing the design, the top Western specialist firm sized those pumps to total 95 horsepower. But a fresh look by Interface/Holland's engineer Jan Schilham, applying methods learned from Singaporean efficiency expert Eng Lock Lee, cut the design's pumping power to only 7 horsepower—a 92 percent or 12-fold energy saving—while reducing its capital cost and improving its performance in every respect.
The new specifications required two changes in design. First, Schilham chose to deploy big pipes and small pumps instead of the original design's small pipes and big pumps. Friction falls as nearly the fifth power of pipe diameter, so making the pipes 50 percent fatter reduces their friction by 86 percent. The system then needs less pumping energy—and smaller pumps and motors to push against the friction. If the solution is this easy, why weren't the pipes originally specified to be big enough? Because of a small but important blind spot: Traditional optimization compares the cost of fatter pipe with only the value of the saved pumping energy. This comparison ignores the size, and hence the capital cost, of the equipment—pump, motor, motor-drive circuits, and electrical supply components—needed to combat the pipe friction. Schilham found he needn't calculate how quickly the savings could repay the extra up-front cost of the fatter pipe, because capital cost would fall more for the pumping and drive equipment than it would rise for the pipe, making the efficient system as a whole cheaper to construct.
Second, Schilham laid out the pipes first and then installed the equipment, in reverse order from how pumping systems are conventionally installed. Normally, equipment is put in some convenient and arbitrary spot, and the pipe fitter is then instructed to connect point A to point B. The pipe often has to go through all sorts of twists and turns to hook up equipment that's too far apart, turned the wrong way, mounted at the wrong height, and separated by other devices installed in between. The extra bends and the extra length make friction in the system about three- to sixfold higher than it should be. The pipe fitters don't mind the extra work: They're paid by the hour, they mark up the pipe and fittings, and they won't have to pay the pumps' capital or operating costs.
By laying out the pipes before placing the equipment that the pipes connect, Schilham was able to make the pipes short and straight rather than long and crooked. That enabled him to exploit their lower friction by making the pumps, motors, inverters, and electricals even smaller and cheaper.
The fatter pipes and cleaner layout yielded not only 92 percent lower pumping energy at a lower total capital cost but also simpler and faster construction, less use of floor space, more reliable operation, easier maintenance, and better performance. As an added bonus, easier thermal insulation of the straighter pipes saved an additional 70 kilowatts of heat loss, enough to avoid burning about a pound of coal every two minutes, with a three-month payback.
Schilham marveled at how he and his colleagues could have over-looked such simple opportunities for decades. His redesign required, as inventor Edwin Land used to say, "not so much having a new idea as stopping having an old idea." The old idea was to "optimize" only part of the system—the pipes—against only one parameter—pumping energy. Schilham, in contrast, optimized the whole system for multiple benefits—pumping energy expended plus capital cost saved. (He didn't bother to value explicitly the indirect benefits mentioned, but he could have.)
Back