Wind-Electric Pumping Systems for Communities

[1]Presented at the First International Symposium on Safe Drinking Water in Small Systems, May 10-13, 1998, Washington, D.C.,

Michael L.S. Bergey

Bergey Windpower Co.

Norman, OK, USA


Wind-electric pumping systems (WEPS) combine high reliability, low maintenance small wind turbines and “off-the-shelf” alternating current (AC) electric centrifugal pumps to provide a simple and robust remote water delivery system. In a WEPS system the 3-phase AC motor for a centrifugal electric water pump is directly powered by the variable voltage, variable frequency AC electrical output from the wind turbine’s alternator. Unlike mechanical wind pumps, these new systems require no scheduled maintenance and can operate autonomously for periods of years between inspections. For areas with even modest wind resources (4 m/s or greater), WEPS can provide a cost-effective alternative to small diesel pumps for both drinking water and small plot irrigation. WEPS technology was developed by the U.S. Dept. of Agriculture, the U.S. Dept. of Energy, and private industry in the 1980’s and commercial systems ranging from 1 – 10 kW (1.3 – 13 Hp) have now been installed at several hundred sites in more than 20 countries.

This paper provides an overview of the technology, its performance and costs, and provides case studies from community water supply projects in Morocco, Brazil, and Bolivia. Supporting information on modern small wind turbines and wind resources is provided. Finally, rules-of-thumb for determining applicability and feasibility, a roadmap for project implementation and sources for technical assistance are presented.



The best situation for developing a new community water supply system occurs when the water source is higher in elevation than the community (or the community storage tank).  This allows the work of delivering the water to the community to be done by gravity.  When the water source is lower in elevation than the community, on the other hand, some form of pumping, or artificial lift, is required to deliver the water.  When grid electricity is available this lift is readily provided with electrical pumps.  But, the electricity distribution grid does not extend into many rural areas and alternate power sources must be utilized.  The technical options available for off-grid pumping include manual labor, animal traction, hand pumps, windmills, internal combustion engines (diesel, gasoline/petrol, etc), solar, and wind-electric systems.  The latter two technologies must be considered emerging since they have not been widely implemented.

The most common technology for off-grid communities, diesel engine pump sets, is also the least sustainable.  Diesel engines are attractive to infrastructure developers and donors because they are inexpensive to purchase and they are widely available; but they are also heavy polluters, can be expensive to operate, and their reliability is sensitive to proper operations and regular maintenance.  Fuel availability is seldom a problem but the support infrastructure for diesel pumps commonly breaks down, leaving users without safe water for long periods.  It is amazing how often diesel pumps are automatically employed in development projects in spite of the overwhelming evidence that they have poor sustainability.  At the Naima project is Morocco, one of the “case studies” in this paper, for example, the pump building at Ain Tolba held one working diesel and its four broken predecessors when wind systems were added in 1990.


Wind power has been used to supply water to homes and communities for hundreds of years.  Locally manufactured windpumps can be found in most countries, though they tend to be small in size ( < 5 Hp).  These units have a wide range of sophistication and reliability and their reliability seems to be proportional to the unit’s weight and cost.  In general, the heavier the construction and the more costly the product, the more reliable they are.  Lighter weight and faster running designs, such as those promoted by the now defunct Dutch CWD, have not been able to achieve the level of reliability required for community water supply applications.  Mechanical windpumps must be optimized for a particular wind speed because the linear power requirements of their volumetric pumps do not match well with the cubic increases in power available from the wind rotor (blade assembly) with increasing wind speed.  Manufacturers optimize the performance at low wind speeds because delivering some water in the low wind speed periods is usually more important than maximizing total delivered water volume over the year.  As a result, most windpumps have excellent low wind speed performance.  They are able to begin pumping at wind speeds as low as 2.5 m/s (for low lifts) and they reach their peak efficiency in the range of 4-7 m/s.  Mechanical windpumps are probably the best choice for using wind energy when the annual average wind speed is less than 4 m/s.

Mechanical windpumps do, however, suffer from several important weaknesses.  As a system they require regular maintenance, not so much for the windmill itself, but for the seals on the pumps.  These seals generally require replacement every one to two years and this replacement requires that the riser pipe be pulled to access the pump.  Another problem is that the mechanical linkage to the pump requires that the windmill be placed directly over the water source.  In some situations this is physically difficult and in others it means that the windmill can not be placed so as to capture the most wind.  The products on the market also tend to be small and mostly aimed at pumping for livestock.  Few models are available in sizes suitable for small communities and only a handful of these larger models have been installed.  For these reasons mechanical windpumps do not offer a competitive alternative to the diesel pump for community water supply.


Wind-electric pumping systems is an emerging technology that combines modern high-reliability small wind turbines and standard electric centrifugal pumps to provide a reliable and cost-effective alternative to diesel pumping for community water supply.  The U.S. Dept. of Agriculture and Bergey Windpower Co. developed the technology in the 1980’s, under partial funding from the U.S. Dept. of Energy.  More recently Bergey Windpower has extended the technology to cover volumetric pumps, such as jack-pumps, for deep well applications and the national renewable Energy Laboratory (part of US-DOE) has developed a comprehensive analytical model of the electrical system.  The cornerstone of wind-electric water pumping technology, however, has been the advent of very high-reliability small wind turbines that can operate for years at a time without any maintenance.


With small wind turbines used in rural applications reliability is the critical benchmark of product maturity.  Many people involved in rural water system development formed their opinions of small wind turbines more than twenty years ago when the first wave of products came on the market in response to the energy crises of the early 1970’s.  These systems, as a general rule, were not reliable enough for rural projects and the companies behind them were not capitalized sufficiently to resolve the operational problems that arose in the field.  Pilot projects in the late 1970’s and early 1980’s were generally not successful and few development agencies have actively pursued the technology since that time.  But, the technology of small wind turbines has continued to evolve and within the last fifteen years much more advanced and sophisticated products have emerged and these products are considerably more reliable than the earlier products.

Modern small wind turbines, such as the 10 kW unit shown in Figure 1, utilize aerospace technologies, and refinements borne of millions of hours of field experience, to provide simple designs with a minimum of moving parts and the ruggedness to endure harsh weather events year after year.  These turbines which range in size from 50 Watts to 10 Kilowatts, and are available from approximately 15 manufacturers worldwide, typically have only three or four moving parts and require no scheduled maintenance.  They operate at variable speed, use passive controls for overspeed protection, and have purpose-built direct drive low-speed alternators, which eliminates the need for a speed-increasing gearbox.  They are able to operate automatically in all weather conditions and most are designed for operational lives of 20 or 30 years.  These high-reliability small wind turbines should be inspected about once a year, but it is not uncommon for them to operate for ten years or more without any attention whatsoever.  They have proven in over 150,000 installations and more than a billion operational hours that the technology is mature enough for critical rural power applications such as rural electrification and community water supply.  Figure 2 shows the most common applications for modern, high-reliability, small wind turbines.


The foundations of wind-electric pumping technology are the ability of standard 50 or 60 hertz induction motors to be operated at variable speed and a fortunate match between the power needs of a centrifugal pump and the power availability from a wind turbine.  A three-phase induction motor can be operated at variable speed by changing the frequency of its electrical supply, so long as the voltage is also changed correspondingly.  With a variable speed wind turbine the frequency output from the generator varies with the wind speed as the rotor speeds up or slows down in response to wind gusts.  The power available from a wind turbine varies as the cube of the wind speed.  Following the Affinity Laws, the power required by a centrifugal pump varies as cube of its speed.  Therefore, if a pump is optimally matched to a wind turbine at one speed or frequency, it is optimally matched at all speeds or frequencies.

As shown in Figure 3 the wind turbine in a wind-electric pumping systems is directly connected to the pump without need for batteries or power converters such as inverters.  The pump and system controller merely serves as a smart switch to turn the pump on or off in response to system conditions.  The pumps used are standard “off the shelf” 50 or 60 Hz AC submersible or surface pumps.  These pumps are operated over a frequency range of 25 – 100 Hz and at peak output may supply 250% of their nameplate flow rate.  This application, however, does require more stages on the pumps because of the need to develop adequate pressures at lower pump speeds.  The most common brands of pumps and motors used in these systems are Grundfos stainless steel submersible pumps with Franklin Electric motors, but many other brands have been used successfully.

From the users standpoint wind-electric pumping technology offers a more sustainable alternative to diesel pumps at costs well below that of solar pumping systems or grid-extension.  Economics are important and in some cases wind-electric pumping systems can today provide a least cost pumping solution on a life cycle basis.  In addition, with future reductions in costs as small wind turbine production volume increases, the economic competitiveness of the technology is expected to improve over time.  However, the very best feature of wind-electric pumping systems for community water supply is their sustainability.  These systems do not require fuel (unless a back-up generator is employed) or regular maintenance, so they are able to operate at 100% availability for years at a time without any attention.  This reduces the importance of maintaining a local support infrastructure.  Regional support capabilities will be sufficient in most cases.  The weakest part of these systems is their electronic controllers, which are used to connect and disconnect the pump motor from the wind turbine, depending on the operating mode of the system.

Since power is transmitted by electrical cables it is possible to separate the wind turbine from the pump with wire runs up to 700 meters.  In hilly terrain this means that the wind turbine can be installed on top of a hill where the best wind resource is located, while the pump can be placed in the valley where the water supply is located.  This can provide a large boost to system performance.  The electrical controls in the system allow the use of pressure switches and float switches for control purposes and some systems have even been configured to provide both water pumping and electrification.  It is possible to configure wind-electric pumping system to both deliver and treat water to provide a safe water supply even where the water source is not safe.


Commercial systems are available is sizes from 1 – 10 kW (1.3 – 13 Hp).  Both larger and smaller systems are under development.  Figure 4 shows the performance and costs for a 1.5 kW wind-electric water pumping system.  Figure 5 shows a typical community water supply application for this size system.  Figure 6 shows the performance and costs for a larger 10 kW wind-electric water pumping system.  Figures 8 and 8 show typical community systems with this size wind turbine.

Wind is an intermittent resource, so reliable water delivery requires storage and, sometimes, a back-up generator to power the pump during extended period of low wind.  Storage tanks are typically sized for 3 – 6 days of supply.


Wind-electric water pumping systems will sometimes provide the least-cost approach for artificial lift in community water supply systems.  But, determining the least-cost approach can be difficult and time-consuming.  A more efficient method is to screen potential sites based on “rules-of-thumb” that have been developed by industry based on experiences with hundreds of projects.  Wind-electric pumping systems should be considered:

  • When “m4” (daily volume required, in cubic meters, multiplied by pumping head, in meters) ranges from 200 – 10,000, and
  • When the average annual wind speed at 10 meters is above 4 m/s (9 mph), preferably above 5 m/s (11 mph), and
  • Where diesel pumps have not been sustainable, or
  • Where the utility grid is more than 2 kilometers (1 ¼ miles) away

One of the biggest obstacles to wider use of wind-electric pumping systems, and wind power in general, is the very poor wind and misleading wind resource data available in most countries.  This data, often in the form of national wind resource maps, is taken from meteorological stations that were never meant to provide accurate wind energy resource information.  The wind speed sensors (anemometers) are often sheltered by nearby trees or buildings and they seldom receive the proper preventative maintenance necessary to maintain their calibration.  A very common problem is worn bearings which cause excessive drag on the spinning cups, which, in turn, provides under-reporting of wind speeds.

When the wind data for most meteorological stations is analyzed over a long period, a distinctive trend called “disappearing wind” can be seen.  Figure 9 contains an example of data taken over 20 years at a site in Eastern Indonesia.  In this case, the most recent wind data underestimates available wind resources by 85%! In 1992, the World Bank rejected a $10 million Global Environmental Facility wind project in Eastern Indonesia because it believed that wind resources were inadequate.   Techniques for detecting and correcting the “disappearing wind” misinformation have been developed by the US-DOE National Wind Technology Center.  New, more accurate, wind resource maps are now being produced using these techniques, GIS terrain maps, and recently declassified DOD meteorological databases.

In general, it can be said that most areas of the world have adequate wind resources for small wind turbines.


Implementing a community water supply project using a wind-electric pumping system typically involves the following generalized steps:

  1. Determine pumping head and water volume requirements.
  2. Assess / Determine / Estimate wind resources
  3. Use manufacturers performance curves to bracket design and select pump
  4. Refine design and predictions with spreadsheet models
  5. Choose hardware, prepare site plan, and prepare budget
  6. Find money
  7. Issue bid and select contractor
  8. Install and commission system
  9. Train “operators” and users

Technical assistance is available from manufacturers, trade associations, and government agencies.  Listed below are potential sources for technical assistance and, in some cases, support for pilot projects and training:

  • Bergey Windpower Co., Norman, OK, T: 405-364-4212
  • Energy Unlimited, Philadelphia, PA, T: 610-940-1994
  • American Wind Energy Association, Washington, DC, T: 202-383-2500
  • National Wind Technology Center (US-DOE), Boulder, CO, T: 303-384-6900
  • U.S. Dept. of Agriculture, ARS, Bushland, TX, T: 806-356-5752
  • U.S. Agency for International Development, through local Missions or the Office of Energy & Infrastructure, Washington, DC, T: 202-712-1685