WIND PUMPING

THE STATE OF THE ART WORLDWIDE

J.A. de Jongh

ARRAKIS

(RED, Renewable Energy Development)

Augustus 1996

CONTENTS

1            General overview. 3

2            Applications. 3

3            Wind pumping in relation to other pumping options. 3

4            Manufacturers in developing countries. 5

5            Technology Classification. 5

6            Performance. 12

7            Economy. 20

8            Future developments, R&D and plans. 42

8.1 General 44

8.2    Private sector developments, projects & programmes. 56

8.2.2           Project & Programmes. 74

8.2.3           Wind Pumping in Latin America. 81

8.2.4           Wind Pumping in Asia. 113

8.2.5           Wind Pumping in Africa. 116

8.3    R&D Projects and Programmes: 163

9            Further Possible Improvements. 192

References. 211

1        General overview.

Wind pumps are windmills which drive pumps mechanically to pump water.

There are about 100 wind pump manufacturers in the world. The longest established manufacturers in the traditional countries USA, Australia, South Africa, Spain and France, still produce first generation wind pumps, which were developed from around 1850 onwards. The home market is however gradually declining. At present about 1000.000 are still in operation.

Over the last 5 years about 5 - 8.000 wind pumps are produced per year world­wide. About 20% of this production has taken place in developing countries. The predomi­nant wind pump type produced in the developing countries is the second generati­on wind pump, which were developed after 1973, but low cost wind pumps are made in many countries as well.

2        Applications

There are many potential applications for wind pumps.

The niche for wind pumps is water supply from 20 m⁴ to 2000 m⁴  per day. [1]

The corresponding rotor diameters range from 1 to 7.5 meters. (see fig 1) [2].

         The merits of a wind pump  can  be viewed as serving a multi­tude of users against low energy inputs. E.g. a 3 m diameter windpump can supply 30 m3 water per day (average per year) at a  pumping head of 10 m with a very moderate average wind speed of 3.5 m/sec. This will serve a village with 750 people (with WHO norm of 40 l/d per capita) the average power delivered being only 34 Watts.

         As an example (with some simplifications), it can be said that the 300 wind pumps installed in Kenya deliver water for the equiva­lent supply of about 200,000 people, in remote areas, there­with contributing to a better quality of life.

Table 1 shows the market segments found, with different appli­ca­tions and corres­ponding user categories [1].

3        Wind pumping in relation to other pumping options

Looking at the niches for the various pumping options (fig 1) one sees that at the lower end of the scale hand pumps are the obvious solution and are used up to 100 m⁴/day. But examples are known where wind pumps are used for requirements down to 20 m⁴/day (domestic water and some gardening) e.g. Curaçao and Philippines. The range of mechanical wind pumps is limited by rotor size, from about 1 to 7.5 m diameter. At larger power demands it is more convenient and economical to generate electricity which can be used to drive a motor/pump combination. These are indicated as WEPS, i.e. Wind Electric Pumping Systems. Especially at sites with high wind speeds (  ³ 5 m­/s), they are attractive from diameters of 3 m and up.

         Engine driven pumps are uneconomical at very low requirements, also due to the fact that diesel pumps are not made for power ratings below 2 kW. With the modular character of solar panels, solar pumping can be used from very small scale to very large require­ments. There is no limit as with mechanical wind pumps: the electricity pro­duced can drive small or large pumps.

                               Table 1   Wind pump market segments, worldwide

1

   Fig. 1   The wind pumping niche versus other pumping technologies.

4        Manufacturers in developing countries

In developing countries the manufacturing companies are small and lack finance as well as the necessary technical and marke­ting skills. Many wind pump models in producti­on are unsuit­able for the market segment presently served, while the manu­fac­turer is not aware of more suitable models available elsewhere in the world. Most manufacturers therefore have only small established markets with sales of less than 100 per year. The models in production often need to be developed further. This is mostly done by trial and error.

The basic needs of the manufacturers are; first and foremost capital, second the establishment of networks and bases to support maintenance activities and sales; third assistance for the selection and develop­ment of wind pumps; and fourth technical assistance to improve producti­on engineering and quality control.

5        Technology Classification

A first distinction can be made as regards the type of transmission

  1.    Mechanical wind pumps having a mechanical transmission

                    a)     driving piston pumps, the most common type of wind pump, about one million still in use today.

                    b)     driving rotary pumps e.g. centrifugal or screw pump. The "Bosman" wind pump is an example of the first and is used in the Netherlands for drainage (about 1000 installations). The FDG-5 is an example of the second type and has been devel­oped in China for pumping sea water for prawn culture. The number of installations is very limited sofar.

 2.     Wind Electric Pumping Systems (WEPS) with an electric transmission mostly used in combination with centrifugal pumps. Few installations are spread over the world, but potential is probably high.

 3.     Wind pumps with a pneumatic or hydraulic transmission. They can be useful for remote pumping similar to WEPS. Air lift pumps are used, despite their low effi­ciencies, in far away places as they virtually do not need any maintenance down hole (e.g. northern Brazil)[3].

A second distinction can be made with respect to the type of pump that is driven by the rotor. The two main classes are

  1.    Displacement pumps. The piston pump is the most common. However the Archimedes screwpump which in the past was used for drainage in the Nether­lands in combination with wind power has now also been introduced in China for very low lift (< 3 m), large volume flow applications (see section 4.1).

                   If well designed these pumps have effi­ciencies of about 70% and do not exhibit starting problems as piston pumps do. Another pump which has witnessed a fantastic revival as a hand pump e.g. in Nicaragua where more than 5000 have been sold with no subsidies involved, is the rope and washer pump. This pump is now being used there in combination with a windmill. About 30 are now in the field and experi­ence is very favourable, especially price-wise. The pump is efficient, does not have starting problems, does not constitute a dynamic load as piston pumps do, and is very easy and cheap to maintain although this may be necessary every half year.

  2.    Rotodynamic pumps of which the centrifugal is the most important. Their use with a mechanical transmission is restricted to low heads

A third distinction which is also made reflects the level of technology of the wind pump. [1]

The wind pump product family can be split up in three basic types:

• First generation, classical, multi-blade, American type, wind pumps with the smaller sizes incorporating a back-gearing transmission. These machines have a low tipspeed ratio often less than 1.

Its strong points are: they are highly reliable, long life time (over 20 years), little and easy maintenance.

Its weak points are: heavy weight, high investment cost, complicated installation.

• Second generation, modern lightweight wind pumps which have been developed in the last 20 years. Gearboxes are omitted, many have fewer rotor blades and run faster with tip speed ratios up to 2.

Its strong points are: light weight, easier to produce, more applications and cost effective. Its weak points are: often not so reliable yet, technology often not fully mature, fast running models limited to low head pumping.

The main components of the two generations, i.e. the rotor, the transmission, the safety mechanism etc. show quite some differences, see fig 5. [4]

• Low cost, artisanal wind pumps, a simple design, locally produced with simple tools, from local cheap materials. Both tip speed ratio and efficiencies are low.

Strong points: low initial costs, resources locally available, user is highly involved.

Its weak points are: much maintenance is required, short life times, unit water costs are high. Fig. 4 gives an example.

For an overview of wind pumps available worldwide, see Table 3. [1].  

1First generation wind pumps; Australian southern cross wind pump.

2Low cost wind pumps; the Thai bamboo-mat wind pump.

2     Second generation wind pumps; the CWD 2000, kijto and FDG-5

ig. 2   Various wind pump types

TECHNOLOGY ASSESSMENT

ROTOR:

-        LESS BLADES, STEEL SHEETS + PIPE PARTS

HEAD:

-        REPLACING GEARBOX BY DIRECT DRIVE MECHANISM

-        Rubber bearings

CONTROL SYSTEMS:

-        HINGED SIDE VANE

START HELP MECHANISMS:

-        COUNTERWEIGHTS

-        DOUBLE ACTING PUMPS

-        MATCHING VALVE

-        SPRINGS

PUMPS:

-        SOFT ELEMENTS

-        HYDRAULIC SEALING

-        AIR VESSELS

LEADS TO A TOTAL COST DECREASE OF 4 X  FIRST GEN. TYPE

3

Fig. 3   Components of mechanical wind mills

6        Performance

Water Demand and hydraulic power:

The required water output has to be given by the end-user [in m³/sec]. With the total pumping head given in [m] the required power can be calculated.

The power that is required to pump a certain amount of water per second over a certain height is called the hydraulic power P _hydr . It is given by (1)

1

in which

ρ is the density of water,

g is the acceleration of gravity,

H is pumping height and

   q is average flow rate.

The product H×  is a direct measure of power requirement and can be expressed in m⁴ per unit of time. A nett hydraulic power output of 1 Watt (continuous) is equival­ent to 8.8 m⁴/day. And 1 kWh hydraulic is equivalent to 367 m⁴/day. The require­ments shown in Table 2 lie in the range of 10 - 1000 m⁴/day, equivalent to average hydraulic power outputs of a few Watts to just over a hundred Watts. These require­ments are small explaining that they are not well met using diesel pumps of a few kilowatts power rating!

Wind pump output

The average hydraulic power output P _hydr of a wind pump (at sea level) is given by (2)

in which:

2

  ν is the average wind speed at the site, which is a crucial factor, since it is the most import­ant parameter determining the  power. It should be measured over the pumping season only. This value should be estimated as accurately as possible, preferably based on measured data of nearby wind measurement stations.

A_rotor is swept rotor area with the diameter as the determinating value,

ß is a quality factor expressing the effectiveness with which the particular wind pump converts wind power to nett hydraulic power over a longer period. Normally values of ß range from 0.05 to 0.12, the first being accept­able, the second value being excellent. The value ß = 0.1 can be regarded as an average value for a well-designed wind pump.

7        Economy

Cost prices of wind pumps vary considerably. They depend largely on labour costs and productivity rates of the country (or region) involved. In Asia the

labour costs are very low thus resulting in very low prices of wind pumps. In Africa, labour costs are still lower than Europe, but productivity is also lower, resulting in rather high prices. The following table 3 indicates some of the differences found.

Nevertheless, unit water costs for some of the second genera­tion wind pumps are one of the cheapest, compared with those by other pumping means, such as diesel pumps and PV pumps.

Therefore, wind pumps can be considered as a really economically viable option for water supply.

Table 3 indicates a number of cases, where second generation wind pumps were successfully introduced. One of the reasons for the successful use of wind pumps in the cases mentioned in India and Kenya was that the unit water costs were the least cost option. [6].

Table 3   Macro-economic annualized life cycle costs (ALCC) of some successful

                                               cases of wind pump applications

8        Future developments, R&D and plans

8.1 General

Prospects for market development from a policy point of view:

Growing sales of modern designs in some countries make it clear that wind pump markets can be developed. The following points underline this point:

  -      the need for clean water is growing

  -      wind pumping can compete economically with motor pumps and PV pumps

  -      potential new or larger markets have been identified.

  -      the need for new indigenous industries with job creation is growing

  -      the growing positive attitude towards environmentally friendly technologies

  -      the wind resource is often adequate despite poor or misleading data

A number of governments of developing countries have taken up the policy of implementation of wind pumps, such as India, Philippines and Niger.

Implementation of "multi user service" technologies is however a complicated and often highly underestimated task, of which the limited successes in e.g. woodstove-, hand pumps, PV sys­tems etc. testify.

8.2            Private sector developments, projects & programmes

Especially the manufacturers of second generation wind pumps are still working on improvement of their concepts. A handful is doing R&D sometimes backed up by an international agency, or by Universities. Improvements are still possible, both conceptual as well as in reducing manufacturing costs and improving reliability. A target of investment costs of less than US$ 100/m² swept rotorarea for reliable second generation wind pumps is possible for a number of developing countries. This figure should be compared to investment costs of approximately $ 400/m² (installed) for traditional wind pumps.

Some International Agencies, such as WORLDBANK (GEF), CIDA, DGIS and ODA are supporting implementation programmes albeit on a small scale.

8.2.1 Some new designs with Transfer of Technology

Dutch Industries from Canada

Dutch Industries Ltd from Regina, Canada which has developed the Dutch Delta 16 (16' rotordiameter, weight ca 800 kgf, costs US$ 4,045.-) wind pump has set up a supply line to Africa apart from its regular sales to the USA (Texas). First country involved was Niger, where within the PEEN project a company maintaining wind­pumps of different designs, gradually evolved to a company assem­bling and installing Dutch Delta 16 machines which were shipped in parts from Canada. The company called PTE Sahel Tech at Niamy became a joint venture with Dutch Industries.

The same approach was followed with manufacturers who had started recently in Mali and Gambia. The total amount of Dutch Delta's installed in these Sahel countries might now approach 100. This private sector approach seems to be sustaina­ble. Dutch Industries has recently developed a smaller wind pump of 2.40 m rotor dia­meter, the Delta Junior (weight 160 kgf, costs US$ 1,400.)

Poldaw from UK

Two experienced engineers of Neale Consulting Engineers of the UK have developed a low cost successor of the IT Windpump, called the Poldaw windpump of 3,5 m diameter. It is their intention to provide licences to manufacturers worldwide. Since the first prototype was built in 1993, it has been tested in the Uk for 18 months after which the technology was transferred to the first manufactu­rers; GB Windpumps (UK) who built the first prototype and Hubert Davies & Co. Ltd. Harare (Zimbab­we). Later the design licence was also given to Merin Agrotool Ltd in Pakistan, Nebular Power Equipment Ltd, Pune, India and Darkata Engineering Services, Accra , Ghana. Negotiations are ongoing with about half a dozen other potential manufac­turers. In the design of the 3.5 m. diameter machine all past experiences with the IT Windpump have been incorporated while targeting at reliability as well as low cost. The technical details are therefore quite balanced and interesting. Until now about 50 machines have been produced. Presently they are developing a larger 5 m diameter type.

Aerobomba de Mecate de Nicaragua 

In Nicaragua a new type wind pump was developed by a Dutch engineer who worked in a wind pump project until 1990 and was dissatisfied with the high costs of the designs made in that project. The type is a "rope and washer windpump" (aerobom­ba de Mecate) and is a combination of latest CWD innovations (hinged side vane) with a locally produced modern rope and washer handpump of which more than six thou­sand have been produced in Nicaragua and sold without subsidy. The pump is a rotary displacement pump and thus the reciprocating movement of piston pumps is omitted, giving a fast starting and smooth operating wind pump with the advantage of easy installation and easy level of maintenance. The Aerobomba de Mecate is locally produced with Taller Lopez/A­MEC. There are two models, one with 2.40 m diam and one with 3 m. diameter rotor. Costs are well below US$ 1000,-. The rotorhead is limited in rotation to 270 degrees, which is no problem in Nicaragua with unidirecti­onal winds. Developments on a fully rotating head are ongoing. In lull periods, the pump can be operated by hand. Until now more than 70 wind pumps have been sold and are operating.

Gradually manufacturers from neighbouring countries have become interested and the transfer of this technology has started to one manufacturer in Colombia and one in Argentina.

8.2.2   Project & Programmes

ALIZES programme in Mauretania

In Mauretania, a French NGO, GRET started a programme in 1990, funded by GEF,  introducing both wind pumps and wind chargers in the Trarza region. The programme is called Alizes. The wind pump introduced was the French made OASIS by Poncelet at l"Aube in France. It has a rotor diameter of 2,50 m average pumping depth is around 30 m. A small company in Mauretania, Deyloul at Nouakchott, started assembling and installing the OASIS machines and gradually extended its part to 90 % manufac­tu­ring. Only the forged rotorshaft with crank is still being imported from France. The link with OASIS is a cooperative one.

The villagers are actively involved in obtaining and managing the water tap point, including the collecting of the initial (partly) investment costs and fund for mainte­nance. Presently over hundred wind pumps are supplying water in the villages.

Yearly maintenance contracts by Deyloul have been arranged.

The approach within the project of involving the local village (authorities) right from the start has been very succesful and seems to enhance sustainability.

8.2.3   Wind Pumping in Latin America

The following account of the state of the art is a summary of presentations made by various people during a workshop in Bogota (23-26 November 1994), which concen­trated on a joint research project by three universities in Netherlands, UK, and Colombia (see next chapter)[..] It has been recorded by Dr. John Burton in RERIC news.

1. Chile, Bolivia

Javier Gho and Carlos Garrido (Chile) reported about the experience based upon variants of the Colombian design Gaviotas MV2E. Thirty commerci­al Gaviotas wind pumps were imported through a donation from the UNDP and merged with a local development of the Cretan type wind pump. Eight systems with 4 to 6 m diameter are now working in the field. A similar experience of academic research being translated into commercial wind pump systems, was reported by Emilio Montaño from Bolivia.

2. Panama

At the Universidad Tecnológica de Panamá, a program has started to gather meteo­rolo­gical information. Financial resources are sought for to buy equip­ment for further evaluation at selected sites with known wind energy potential.

3. Peru

Wind pumping is not unknown to farmers in Peru, and a number of indigenous wind pumps have been designed and installed. Most of them had a limited capacity of, at best, comparable to that of a motor pump of about 1 kW. It practically means that the known wind pump models can only serve a very small area of some few hectares. Given a pumping head of 20 metre, one can think of irrigati­on of maximally 2 to 3 ha using a wind pump (depending on the product and the irrigation methods).

At present, the following wind pump types are found in Peru: Miramar (Miramar, Piura), improved Miramar type (Chiclayo); Alborada; ITINTIC; Zimic, Infantes, Segovia, Fiasa ("American" type); Matto; MCTB500 (PUCP, second generation wind pump). Large wind pumps for water flows of 4 l/s and suitable for deeper wells are not available in Perú, and an appropriate wind pump of this size would be very useful. Moreover, farmer interest in wind pumping has recently revived due to the increased fuel prices (from US$ 0.50 per US gallon in 1985 to US $ 2.00 now [2] ).

This makes wind pumping more competi­tive, but short and medium term credits for invest­ments in renewa­ble energy are still rather scarce.

Institutes working on wind pumps in Peru are: Grupo of the Pontificia Universidad Católica del Perú (PUCP - Lima) and a group at the Univer­sidad Nacional de San Agustín in Arequipa, which has recently started to work in wind energy (Rúsbell Zevallos).

4. Nicaragua

Here the "bomba de mecate" is being developed, which is a rope-washer pump combined with a CWD wind turbine design. Important characteristics of the construc­ti­on are its simplicity and the low loads imposed by the pump on the wind rotor resulting in an efficient low-cost system.

5. Colombia

Mauricio Arango from SENA (Colombian apprenticeship training body) presen­ted a general overview of wind pumps installed there about 40 years ago. Main problem is the lack of maintenance and repair. The systems are used by the Wayu indigenous tribes (about 150,000 people). SENA has started a training pro­gramme in conjunction with the Universidad de la Guajira. The region has very strong winds, which has forced local manufac­tu­rers to develop robust wind pump systems, with some success.

Jorge Castro (Jober) was the only manufacturer represented. He gave a brief summary of the interesting development of their wind pumps.

Systems were mainly sold in Colombia but some have been exported to Venezu­ela and Ecuador. Most recent development is a wind pump to suit the conditi­ons of the Guajira region.

Mauricio Gnecco (Fundación para el Desarrollo de Tecnologías Apropia­das - FDTA) showed the interesting experience of the use of wind pumps in the Llanos Orientales. FDTA is a non-govern­mental organisation with the purpose of increasing the use of appropriate solutions for rural needs. Strong emphasis is on energy supply from various sources, one of them being wind. At the time of the seminar, FDTA is working together with Jober to promote the use of wind pumps for water supply for humans and animals in the Llanos region.

6. Venezuela

Vicente Durán and Eliodoro Gómez described the work of Fundación Zumaque, which works in extension and health care in remote regions of Venezuela, and of the Universidad Francisco de Miranda. The latter has a wind energy test site on Paragua­na peninsula, a region not unlike the Colombian Guajira, where a variety of vertical and horizontal axis wind turbines have been installed.

Prospects

Common points are the recurrent necessity to have more international technical cooperation projects as well as financial resources. Education is needed in the field of renewa­ble energies, environ­mental problems and sustainable development in Latin America. The idea was launched for a Latin America based Association for Wind Energy for Rural Areas (WERA). To carry on with this idea, Uniandes undertook the responsibi­lity to manage the begin­ning of such an association for the next two years, perhaps in another country, to share the experience of implementing the 3-S pump concept in wind water pumping and to invite some more institutions from the region.

8.2.4   Wind Pumping in Asia

The state of the art of wind pumping in Asia is not treated here because the work­shop attendants will present an overview of the most active countries; China, India, Vietnam and the Philippines.

8.2.5   Wind Pumping in Africa

Table 1     PRESENT STATUS OF INSTALLED CAPACITY (af­ter 1975) 

                          AND COUN­TRIES WITH P­O­TE­N­TI­AL IN AF­RI­CA

STATUS	AND	POTENTIAL
COUNTRY	Grid con­nec­ted Large WT's	Stand alo­ne      Units	Wind pumps	Remarks
	UNITS kW in­stalled/ Source	UNITS kW inst./  Source	UNITS Number Inst./Source
Egypt	Egypt      ca 200­0kW / 1		>200/1
Lybia	Lybia ?		> 100/2
Tunisia	Tunisia	>90 kW/6	>  40/4
Algeria	Algeria	?	?
Morocco	Morocco	?	> 50/4
Maureta­nia	Maureta­nia		> 30/4
Senegal	Senegal	5 kW/3	> 80/4
Cape Verde	Cape Ver­de	465 kW/3	>90/4
Sudan	Sudan		>10/4
Ethiopia	Ethiopia		>10/5
Djibouti	Djibouti		>10/5
Somalia	Somalia		?
Kenya	Keny­a 600 /2	200 /2	265/4
Tanzania	Tanzania		>10/4
Mozambi­que	Mozambi­que		>100/4
Madagas­car	Madagas­car	?/3	20/3	2 units
Angola	Angola		?
Namibia	Namibia		?
S. Afri­ca	S. Africa	? 1)	>300.000 /3/4/5	1) 300? u­nits/3
Lesotho­	Lesotho
Mali		Mali	78/3
Niger		Niger	16/3
Gambia		Gambia	8/3
Zambia		Zambia	2/4
Zimbabwe		Zimbabwe	>200/4
Botswana		Botswana	>10/5
Seychel­les		Seychel­les	?/3
Burkina Fasso			Burkina Fasso  ?
Tchad			Tchad
Uganda			Uganda ?
Malawi			Malawi
Swazi­land			Swaziland
TOTAL 32 Coun­tries	2600 kW	>760 kW	>301.300

REMARKS:

-        Per application, the countries with theore­tical po­tential, based on windspeed, have been menti­oned in the columns.

-        A question mark indicates that there are probably installa­ti­ons.

-        The number of installations indicated have been done after 1975. Installations no longer in operation have not been mentioned. 

8.3                             R&D Projects and Programmes:

R&D project Auroville, Increasing the Water Output of Wind­pumps

Centre for Scientific Re­search (CSR), Auroville, has done research on loadmat­ching devices for water pumping windmills driving piston pumps. Five load matching devices were tested by CSR, all on the same windpump, the 5.6 m diameter 24-bladed AV55 developed by CSR. This makes a comparison of the results very reliable. The loadmatching devices tested were:

1.      Counterweight balancing the weight of the pumprods and part of the hydraulic load.

2.      Cam mechanism by which the upstroke is lengthened and downstroke shorte­ned in time, thus smoothing the torque.

3.      Spring in tension between the tower and pumprod more or less balancing the loads like the counterweight.

4.      Matching valve. This valve, developed by the wind energy group of Eindhoven University of Technology in The Netherlands, is lighter than water and replaces the normal piston valve of a piston pump. At low wind speeds and so low pumping speeds, the valve by buoyancy remains open; the wind rotor can turn unloaded (except for the weight of the pumprods).

5.      Automa­tic variable stroke me­chanism. This hy­drau­lic system, designed by CSR, adapts the load at different wind speeds by auto­matically adapting the stroke length of the pump.

The conclusions from the test are evident. The output of a windpump can be increased by 30-50 % by installing a spring or a matching valve as load matching device at low or no extra cost. The windpump starts pum­ping at a lower windspeed of  5 km/h instead of 7.5 km/h, meaning that water is pumped during more days than conventionally. [7]

Colombia/Netherlands/UK 3-S Project

International Seminar on New Developments in Wind Pumps (I)

The Universidad de los Andes (Colombia), the Technical University of Eindhoven (The Netherlands) and the University of Reading (United Kingdom) had been working on this concept in a joint EC contract between September 1992 and August 1994 under auspices of the Commission of the European Community DG XII. Title of the project was "Internati­onal Scientific Coopera­tion Project between Colombia, United Kingdom and The Nether­lands: the develop­ment of an innovative 3-S pump".

The focus of the 3-S project has been on the riser, pump rods and piston pump compo­nents:

-        s­moot­hing aims at reducing the dynamic loads in the wind pump either by introdu­cing flexibility into the drive train or by incorporating soft elements below ground level in the pump chamber;

-        sealing aims at eliminating the need for a leather cupseal by replacing it with a hydrodynamic (non-contact) seal, which in principle should reduce mainte­nance;

-        starting aims at solving a classical problem in wind pumping, i.e. the need for a much higher wind speed to start the wind mill than that to keep it running.

The main starting device that has been investigated within the project is the "floating" or "matching" valve. The proper understanding of the basis of the 3-S pump concept is of paramount importance if in the long term these improvements are to be successful­ly implemen­ted in commercially available wind pump systems. Some foreseeable advantages of this develop­ment are an increased reliability and overall efficiency (which includes increased water output) of the wind pump system as a whole, and considerably lower maintenan­ce and repair costs. [8]

UK, Dynamic Behaviour Research

The reduction of lift rod forces in piston pumps

One of the greatest barriers to the uptake of windpump technology is that of reliabili­ty. For example, the French organization GRET reports that 42% of failures on their 2.5 m diameter machines are due to overloading of the lift rod.

Under a project funded by the U.K. Engineering and Physical Sciences Research Council the dynamic behaviour of a reciprocating piston pump has been investigated. The inclusion of varying degrees of softness in several of the windpump components has been shown to greatly reduce the peak lift loads.

The inclusion of a small amount of flexibility into the lift rod, such that it stretches by around 10% of its length when supporting the static head, can be sufficient to almost half the peak lift rod load at high speeds. One major benefit of this technique for improving the longevity of a windpump is that it can be easily retrofitted to any machine.

A more complex alternative is to use an air chamber to smooth the flow fluctuations up the riser pipe and therefore prevent the accelerative loading of the system.

Another solution, which is often prohibitively expensive or too complex for the smaller machines, is to reduce the speed at which the pump operates. A novel realisation of a drive mechanism originally used by the Canadian Brantford windpump at the turn of the century has been built and tested at Reading. This device translates the rotational speed of the rotor shaft into a long slow reciprocating motion using two duplex sprockets and a length of duplex chain with carrier plates.

These simple technologies can greatly reduce the peak load suffered by the lift rod and the above ground windmill structure and offer the possibility of reliable fast running wind powered water pumps.

9        Further Possible Improvements

A number of possible improvements can be made in future with respect to various criteria (which are sometimes contradictory) such as:

-        reducing costs

-        improving reliability

-        easy manufacturing

-        easy maintenance

The following measures can be taken to reduce costs:

-        production engineering or value engineering

-        mass production

-        better matching

-        quality improvement

-        further R&D

-        design adaptation to specific applications and conditions;

         examples: 12PU500 for Bihar (India), NIVA3000 for Sri-Lanka, AUTOSPARES with lateral well and handpump combination, India,

         drip irrigation systems, Gujarat India.

References

 [1]    Windpumps in developing countries: a view of the markets, Halcrow Gilbert Associates, Oct. 1993.

[2]     Wind water pumping: the forgotten option, P.T. Smulders, June 1995.

[3]     Windpumps, a guide for development workers, E. Barlow et al, IT Power, 1993.

[4]     Wind pumping; a handbook, J. van Meel and P.T. Smulders, World Bank Technical Paper No.101.

[5]     Solar Pumping; an introduction for development workers, IT Power, 1993.

[6]     Markets for windpumps in development countries, R. Hacker (HGA) and J.A. de Jongh (RED), paper presented at the ISES World Solar Congress "in search for the sun", Harare, Sept. 1995.

[7]     Field Testing and Monitoring of Load Matching Devices for Auroville - 605 101 - South India, Robert Trunz, Dec. 1995.

[8]     Technical details of the 3-S project are outlined in: Smulders, P.T., et al., "The 3S-Pump Project: Piston Pump Innovation for Wind Pumps". 5th European Wind Energy Conference, pp. 1145-48, Thessaloniki, Greece (1994)