How to optimize industrial pumping systems

U.S. Department of Energy's Industrial Technologies Program
Tags: energy management

Don Casada is an Energy Expert for the U.S. Department of Energy, an authority on industrial pumping systems and a long-time contributing author to the DOE’s Energy Matters newsletter. In this question-and-answer article, Don outlines some main causes of excess pumping system energy use and describes methods of quantifying opportunities for reducing energy use, including those applicable to municipal water and wastewater applications.

Question: What are some principal causes of excess pumping energy use?
Answer: To provide some perspective on the answer to that question, it is useful to resort to the basic equation that expresses electrical energy as a function of system and component parameters:
This figure is an equation that reads as follows:  E equals Q times H times T times sg divided by 5,308 times ήpump times ήmotor times ήdrive., where
E = energy, kWh
Q = flow rate, gpm
H = head, ft
T = time, hours
sg =  specific gravity, dimensionless
5308 = conversion constant
ηpump = pump efficiency, fraction
ηmotor = motor efficiency, fraction
ηdrive = drive efficiency, fraction

Anything that causes the numerator to be bigger than it needs to be in order to meet the system functional requirements causes excess energy use. Similarly, if less than optimally selected components (pump, motor and drive) are used, the denominator shrinks and energy use is greater than it would ideally be.

Every type of opportunity to reduce pumping energy can be categorized in the terms of this relation. Ideally, of course, you would have equipment sized to meet the system flow and head needs (without excess of either) while operating at near best efficiency conditions – and you would have controls in place to only run the equipment when it was truly needed.

But, this is often not the case. Why not? There are many possible reasons, but let’s just list some of the more common factors, which are not altogether unrelated.

  1. Initial equipment specification and selection occurs before the system is actually operated, of course. Uncertainties in component and system modeling, conservative assumptions to accommodate those uncertainties, addition of margin in the equipment specification to cover future contingencies, and further margin increase “just to be sure” are examples of factors that often lead to installed equipment that is capable of providing considerably more flow rate or head than the system really requires.

  2. When the actual system requirement is considerably different than the pump capability, the pumping system efficiency will inherently suffer (and, it might be noted, so will reliability).

  3. Things do change with time, which is one of the reasons that margin is sometimes added in the initial specification/selection process. But when process requirements do change, it is often in unanticipated ways, and the existing equipment may be poorly suited for the new requirements.

  4. Energy costs frequently account for over three-fourths of the life-cycle ownership cost, but this is often unrecognized and therefore not factored into purchasing or other decision-making processes.

Question: How should one go about identifying and quantifying pumping-related energy opportunities?
Answer: Each individual plant, of course, has its own unique circumstances, so there is no single pattern that is best for all. That said, the approach endorsed by DOE has proven to be broadly effective, particularly for process industry applications. For those who have attended either a Pumping System Assessment Tool (PSAT) training or listened in on the free DOE-sponsored PSAT Webcasts which are held on roughly a monthly basis, this will be familiar territory.

The endorsed method implicitly recognizes that not all pumping systems are equal – and neither are the opportunities for improvement. The overall energy consumption is even more skewed toward the larger equipment than common sense would suggest. As an example, see Figure 1, which shows the population and installed power as a function of pump motor size at a large paper plant in Maine [1]. This profile is very similar to one from a DOE-sponsored study of the nation’s industrial motor population [2]. The point to be drawn from this graph is that the larger equipment uses a disproportionate fraction of the overall plant energy (e.g., equipment that is 100 horsepower and greater represents 20 percent of the population, but accounts for 85 percent of the connected load). So, our common sense intuition to focus on larger equipment is actually understated – it is really important to look at the big stuff – which generally runs a lot.

This figure is titled “Pump motor population and installed power distribution at a large paper plant.” The y axis is labeled “cumulative % of population, installed power,” and the numbers range from 0 to 100, in 10% increments. The x axis is labeled “Motor size range, hp” and the numbers range from ≥1000 to ≥10 running left to right. The coordinates of the population line are marked at: 0/≥1000; 5/≥500; 12/≥200; 20/≥100; 42/≥50; 100/≥10. The coordinates of the power line are marked at: 30/≥1000; 58/≥500; 75/≥200; 85/≥100; 95/≥50; 100/≥10.
Figure 1. Pump motor population and installed power distribution at a large paper plant

A second element in the DOE-endorsed method is to focus on centrifugal pump applications where the pump is driven by fixed speed motors (as opposed to adjustable speed drives), which exhibit one or more of the following symptoms:

  • Throttle valve-controlled systems

  • Pump bypass (recirculation) line normally open

  • Multiple parallel pump system with same number of pumps always operating

  • Constant pump operation in a batch environment or frequent cycle batch operation in a continuous process

  • Cavitation noise (at pump or elsewhere in the system)

  • High system maintenance

  • Systems that have undergone change in function or extent of demand.

The third and final element, which allows quantification of the opportunity, involves measurements of fluid and electrical parameters, including flow rate, pressures and motor power (or current as a proxy). These are not always readily available, and use of temporary test equipment may be required. And getting the data may be easier said than done – it can be a challenging job even for experienced field hands. The PSAT workshop currently devotes time to practical issues involving field measurements, but this is a subject that merits considerably more attention in the training realm.

PSAT is certainly not the only means to help one quantify opportunities, but that is exactly what it is geared toward doing, and in the author’s quite biased judgment, does it well, thanks to excellent algorithms developed by pump manufacturer representatives involved in the Hydraulic Institute standards program. PSAT does not, however, identify solutions. Its goal is to help the user understand if the potential savings merit the time to investigate solution alternatives.

It should be noted here that an essential element of using PSAT, or any other method to quantify opportunities, is to be sure that the true system functional requirements are always kept in mind. Simply assuming that the current operation is, by definition, equal to what is truly required, can result in the dominant savings potential being missed.

Question: Are there important opportunities that don’t fit in the standard symptoms list?
Answer: There definitely are energy and/or energy cost-reduction opportunities that are not flagged by the standard prescreening process, which is primarily geared toward process industry. One specific, very important pump-using community that merits a bit different take is the municipal water and wastewater industry.

Without going into the reasons why (lack of space), here are some specific things that the author and colleagues have found to be useful to consider in municipal applications:

  1. If the station pays a demand charge (and most do), does it have a poor load factor? If so, look for ways to better match the equipment to normal needs and to avoid situations where equipment that runs for relatively short periods – such as filter backwash pumps – operate concurrently with major loads, thereby setting the demand peak.

  2. On the flip side, does the municipality have extensive storage capacity that could potentially allow it to shift loads to times when electrical rates are lower? If so, and if the rate differential is sufficient, changes in operational patterns can yield good energy cost savings (note: this will not reduce energy consumption; it is more likely to actually increase it).

  3. In wastewater plants, are return activated sludge and aeration flows being adjusted to match the true system needs? If not, control schemes with appropriate strategies, including adjustable speed drives and/or on-off operation can be very helpful.

Some of the following documents are available as Adobe Acrobat PDFs. Download Adobe Reader.

  1. Data used in Figure 1 is courtesy of Tim Crocker, Powerhouse Superintendent, Domtar Woodland plant.

  2. United States Industrial Electric Motor Systems Market Opportunities Assessment, by Xenergy, Inc., for the U.S. Department of Energy and Oak Ridge National Laboratory, available for free download (PDF 6.6 MB).

Additional Resources

  1. Improving Pumping System Performance: A Sourcebook for Industry (PDF 3.1 MB).

  2. Pumping system tip sheets.

About Don Casada:

Don Casada is a consulting engineer with Diagnostic Solutions LLC in Knoxville, Tenn., and a DOE Energy Expert. He specializes in measurement and evaluation of pumping systems for energy, productivity and reliability improvements, and has led pumping system assessments at more than 125 industrial facilities. Don is the author and programmer of DOE’s Pumping System Assessment Tool (PSAT). He is also the developer of the associated PSAT end-user and PSAT Qualified Specialist (QS) training curricula, and serves as the lead QS instructor for DOE. He holds six patents involving valves and rotating machinery. In his spare time, Don enjoys spending time in the backcountry of his native Great Smoky Mountains.

This article is courtesy of the U.S. Department of Energy’s Industrial Technologies Program and appeared in its Spring 2009 issue of Energy Matters. To learn more, visit