Cover Image
close this bookBoiling Point No. 29 - December 1992 (ITDG - ITDG, 1992, 40 p.)
View the document(introduction...)
View the documentHousehold Energy Developments in Southern and Africa
View the documentCookstoves in East & Central Africa
View the documentTanzanian Stoves
View the documentCharcoal & Woodfuel Health Hazards
View the documentFrom Clay & Wood to Cast Iron & Coal in South Africa
View the documentHousehold Energy Activities in Uganda
View the documentGTZ Section
View the documentBurundi Institutional Peat Stove Programmes
View the documentWood, Charcoal or Coal for Cooking in Southern Africa
View the documentEnergy & Environment in Zimbabwe
View the documentA New Environmentally Sound Energy Strategy for the Development of Sub-Saharan Africa
View the documentKang-Lianzao Bed Stove
View the documentField Trials of Electrical Heat Storage Cookers in Nepal
View the documentNEWS
View the documentR & D NEWS

From Clay & Wood to Cast Iron & Coal in South Africa

by Grant Ballard-Tremeer, University of the Witwatersrand, School of Mechanical Engineering, Johannesburg, P O Wits, 2050 South Africa

The University of the Witwatersrand has been involved in the development of appropriate stoves for a number of years. The present project on low-cost woodburning cookstoves was started in 1990. The focus of the work has been on designing a stove for the predominantly Tsonga-speaking people who live in this area. In South Africa it has been found that 79% of household energy requirements in rural areas are met with fuelwood.

Initial Survey

In 1990, H Kennedy carried out a survey into energy usage in the village of Cottondale. A sample of 50 out of an estimated 600 households was interviewed. Kennedy found that 74% of the total household energy consumption came from the burning of wood. Meals usually consisted of a large serving of maize meal porridge with aside serving of meat or vegetables. Monthly income was predominantly (62%) below R400 per household per month (1 US$ = R2.83). The average number of people per household was 7.2. Only 2% of people interviewed possessed a wood or coal stove, but most desired one (86%). Kennedy designed a two pot stove made predominantly from clay and laboratory tested two scale prototypes.

Refining the Design

A number of requirements of the wood-burning stove were identified by two students, G Ballard-Tremeer and S Trickett in 1991. Of particular importance were the need for low cost (around R50), high durability and ease of manufacture by rural artisans.

To meet these requirements the design focused on a stove similar to the one designed by Kennedy - a two-pot stove, built mainly from clay, without a chimney, and similar to the Teri Improved Chimneyless Woodfuel Cookstove from India. The general shape can be seen in figure I

Fig 1 - Original Clay Stove

A local rural potter was approached. Samples of prepared wet clay were analysed for grain size distribution and plastic behaviour and fired specimens were tested for compressive strength. Three variations were considered: normal clay; clay with an organic additive (wheat husks); and samples with the largest particles removed (using a 2mm sieve). This removal of large stones resulted in an increase of 10% in the fired strength. The addition of wheat husks (50% by volume) was found to decrease the strength by 34%, but to increase the resistance to thermal shock. In the final design the firebox was made from clay with sawdust added. The outside wall of the stove was made from the strongest and toughest material - sieved clay. To ensure correct dimensions a set of 20 templates was used by the potter to manufacture the stove.

Fig 2 - Improved Clay Stove

There is no fuel door - the pot is lifted for refuelling. The fire is observed through the secondary air hole. From the field trials it was found that the smaller families (5 or 6 people) could cook their whole meal without refuelling. Larger families said that having to remove the pot was a problem.

In the laboratory, Specific Fuel Consumption (SFC) tests predicted a wood saving of around 50% compared with an open fire. The Percentage Heat Utilized (PHU) in the clay stove for 5 tests, calculated according to the formulae given by Stewart et al in "Wood, Waste and Charcoal Burning Stoves.." are as follows:

· Boiling Phase: 19.3%
· Simmering Phase: 24.4%
· Overall: 21%

The consumption rate is 53%.

The stove was tested in the field to determine the response of the community to the design. Detailed measurements of wood, paraffin and dung usage in 8 families in a nearby village were made over 24 days. The stove was found to decrease wood consumption by 10% as well as virtually eliminating the use of paraffin. However, since cooking frequency had increased, the energy saving per cooking task was higher than this figure (though not quantitatively determined).

In subsequent visits to the families (two months later and again after six months) problems with long-term acceptance were apparent. In one household the stove was only used when cooking food to sell. One stove had been replaced by a cast-iron coal stove and one had broken when a child dropped it - the owner intended to have it repaired.

The Latest Developments

This year (1992), E Harvey and S Sithole examined long term acceptance and manufacturing techniques Preliminary work into the design of a rural 'factory' for the ceramic stove was carried out. In order to address the problems of long term acceptance, interviews with the families involved in the earlier field study, stove demonstrations and a workshop with community women were held. This led to a re-evaluation of some of the design requirements - the need for a stove to heat 2 or 3 large pots and to be at waist height. The short-term enthusiasm for clay stoves had diminished - clay being seen as a primitive material. This is possibly a result of migrant labourers' exposure to electric stoves in urban areas. A low-cost metal stove, as opposed to the clay ones previously designed, was proposed. Again suitability for ease of manufacture was considered a priority.

The new design has not yet been manufactured. The estimated cost is about R200 - the people interviewed were prepared to pay considerably more for a metal stove than for a clay one. Cast iron stoves currently on the market cost many times this amount. They have poor efficiencies and usually only operate effectively with coal.

Emission Measurement

The author is carrying out a research project in the field of emission testing. Current measurement techniques the carbon balance method; the artificial flue method; and the chamber method (the flue and the chamber methods were reported in BP No. 28) - all require specialized and costly equipment. Simpler and less expensive methods are being developed for both laboratory and field situations and suggestions for standard test procedures similar to those proposed for efficiency tests will be considered.


Many people have been involved in these projects. Mr D Banks and Dr H Jawurek are the driving force behind the research. The communities involved in the studies deserve great thanks. The potter Sarah and the interpreter Edith Ngobeni helped enthusiastically and the staff of the Wits Rural Facility have always been ready to help with advice and direction.

Ed Note: If the proposal for a rural factory to produce clay or metal stoves at an approximate cost of R200 is to be implemented, then ITDG's stoves work in Sri Lanka etc. suggests that the possible socio-economic effects should be carefully investigated. Such a study should consider the effects of the scheme on local potters and metal workers and on the poorest village people who may not be able to pay R200 for a stove.