The Afridev is inherently designed to be a village level operation and maintenance pump. It is also designed to be easy and cheap to manufacture, considering its intended use. It is basically a very good design, and when the components are of good quality it can provide 2 or 3 years of service without the need for replacement parts (usually low cost items such as U seals, bobbins, plastic bush bearings, etc). Even then, after many years, major components do not need replacing (24).
Fast moving parts include “U” seals (K15 each, 186 sold in KGP over a 30 month period) and “O” rings (K15 each, 42 sold); bush bearings (K186 a set, 48 sold); rod centralizers (K40 each, 40 sold); as well as pump rods (K1600 each, 25 sold); fulcrum pins (K559 each, 12 sold); hangar pins (K468 each, 6 sold); and riser mains (K900 a length, 21 sold).
It is said that the design life of the Afridev pump is 7 years (25), but experience shows that with good training, routine maintenance and the availability of parts, this is easily extended by many more years.
Another problem with riser mains can occur if the joints are not properly made during installation. In this case, the mains can break when an attempt is made to remove it for repairs.
Making repairs on the riser mains involves keeping the timing of the pump to the original setting. Repairs involving the removal of the riser mains are often referred to the District maintenance teams. However, as with many Government departments, these are not functioning effectively. In order to enable the users to be independent of Government services as far as possible, Area Mechanics are trained by the project to deal with all repairs affecting the riser mains.
Availability of pump spare parts
Changing the Take out all tubes for repairs
Unavailability of spares
The system was to operate as a revolving fund. Nevertheless, after only two years the system collapsed and there were no longer any fast moving spares in the Chipiku stores. Chipiku management26 confirmed that they have the interest and the capacity to undertake spares distribution on a national level, so what went wrong?
Several assumptions and miscalculations were made:
a) In the initial purchase, it appears that there was no relationship made between the purchase order and those component parts that are in demand. SKAT gives someguidelines on the life span of fast wearing parts as well as the required quantity of slow moving parts for a 5-year operation (SKAT manual, pages 68-69). In addition, there is now a database that can also be used so that some sort of relationship can be established (c.f. Hankin).
b) It was assumed that a revolving fund involving the government actually works. In fact, it is well known that once a fund is paid into a government department it requires nothing short of a miracle to get it back. The system failed not because of a failure by the distributor, but because there was no subsequent procurement.
c) It is assumed that users are reluctant to pay for spares. In fact, we have found out that users are willing to pay reasonable prices in order to have access to safe water. From October 1997 to March 2001, over 420 items of spare parts have been procured by the borehole users in KGP, with several purchases of over K2000 worth of spares.
d) It is assumed that spares have to be available within the shortest distance possible. Indeed, some management within the public sector have expressed concern that the Chipiku distribution system is inadequate. In fact, given that parts for a particular hand pump are not required very frequently somebody is always prepared to travel a reasonable distance to purchase the spares. Many user groups are concerned that if spares are made available to small scale entrepreneurs (e.g., grocer stall, vendors, etc) then an illicit market in stolen parts would develop making their own hand pumps vulnerable to theft.
Attempts have been made to rectify the situation.
b) Local companies have started to manufacture some fast moving parts, such as the nylon bush bearings. The question has to be asked whether they can compete on cost, and whether other fast moving parts can be manufactured locally.
c) Hardware and General Dealers (part of the Press Conglomerate) have decided to venture on a trial basis, ordering directly from the manufacturers. Success will depend on market demand. They have a smaller distribution network than Chipiku.
d) The National Water Development Project has engaged two local companies for the supply of hand pump spares; this is on three-year renewable contracts. These suppliers would then have to find outlets.
Given the assumption that prices need to be kept low, Government controls the price of spares. If the profit margins are too low, then there is little incentive for private enterprise to become fully engaged. It is encouraging to see public-private partnerships. Since a spares supply chain may have to be subsidized in some form or other for some years to come, it will be interesting to see how Hardware and General Dealers’ fully private venture will fair.
An interim solution
If the initiatives discussed above do not result in a sustainable spares supply chain at a national level, contingencies will have to be set up. This may involve church based groups active in the area. Advantages may include a long term presence; non-profit oriented; ability to order directly from abroad; access to foreign exchange; and a desire to increase their profile in development activities. Two such groups have been identified and they have good accounting procedures and proven capacity to manage a rotating fund.
Incidentally, the spares are available at only two outlets, both near to existing Chipiku stores. The weakness of this approach is that it depends on the presence of the author (or similar project manager) in the area.
Obviously, a long-term solution needs to be found which ensures that the required spare parts are readily available at prices that are affordable to the users.
(25) Verbal communications from SKAT participants at HTN workshop, Hyderabad, India, March 2000.