| Main knowledge bank page |
Recent additions |
Recent changes |
What links here |
Categories |
Category cloud How-to guides | Organisation profiles | Project profiles | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Clay Testing - 5 Country Reports
In April 1992, at the first ITDG International "Clays for Stoves" seminar at Sheffield University, Dr Messer and Dr Gaspe proposed that the C/NC ratio be accepted as a good predictor of stove failure resulting from thermal stress. This was based on a surprisingly strong correlation found between the C/NC ratio and the reported tendency of the stoves to fracture when in use. Over the past year testing has been carried out in Ethiopia, Kenya, Nepal, Thailand and Indonesia to evaluate the clay/non clay ratio test procedure. The results show the effectiveness of adjusting the C/ NC ratio to around 1:1 and whether this improves the durability of pottery stoves. The country papers were presented at the second "Clays for Stoves" seminar hosted by Dian Desa, ARECOP in Indonesia in April 1993. Below are summaries of 5 papers. [top] [end]ConclusionsIt seems clear that the C/CN ratio is an important factor to be taken into account when deciding the suitability of a clay for making stoves. However, the rather general rule that the C/ NC ratio must be lower than 1:1 is probably an over- generalisation and is not very helpful for many clay sources.In Kenya a C/NC ratio above one seemed to be successful, while in Thailand C/NC ratio were as low as 0.7:1 and in Nicaragua as low as 0.4:1. It therefore seems sensible that when selecting clays a good starting point is a ratio of 1:1 but other clay ratio samples should also be considered. The choice of types of additive that can be used to mix with clays in order to bring the C/NC ratio down to 1:1 is still somewhat confusing. So far very few tests have been carried out which exclude other compounding factors such as method of moulding. In Sri Lanka fine sand seems better than grog but sand and grog are better than just sand. In Thailand grog seems to be very effective. Since there is very little clear evidence it is likely that the decisions on which additives to use will be made according to cost and availability. If lime or montmorillonite* clay mineral are present in the raw clay body the C/NC ratio is not a good predictor of the clay's suitability for making stoves. In such cases the clay is best abandoned and an alternative used. Lime can be detected by dripping an acid solution onto a freshly cut clay surface. Bubbling indicates lime. Detecting the presence of montmillonite is much more difficult and can only be verified in a well equipped laboratory. However, as a simple guide very high shrinkage may be an indication of montmorillonite. In Kenya and Nepal cutting through the bridge above the stove's entrance was found to be a very effective way of providing stress relief. This was also found to be effective in Sri Lanka and India. Sulpya in Nepal has recommended that the use of a detachable clay ring fitted over the bridge gap helps to reduce distortion in drying. Another factor mentioned in the country reports is 'beating' - a technique used to shape cooking and water pots which is also being used to shape some stoves. Koopman concludes that beating is likely to induce greater particle alignment and therefore more anistropic shrinkage and residual stress. This may be counteracted by its tendency to increase the strength of the pottery through closing up small fissures. * Montmorillonite is a clay allied to koalinite. It has finer grains and a weak cell structure, exceptionally plastic and exhibits high shrinkage Full seminar proceedings can be obtained from Hofman Engineering Systems - Greenfields, P O Box 167, Chiangmai 50000, THAILAND. [top] [end]1. Kenya Clay TestingSummarized from "Clay Test Results" by James Muriithi, Ministry of Energy/GTZ SEP, Nairobi, KenyaClay for pottery work is found in many parts of the country but its quality and suitability for making stoves varies enormously. The cracking of ceramic stoves when in use and during the production process has been a major problem, particular!) during the start up period in many production units, and has persisted for some time. It was decided to measure the C/NC ratio in some clays and then adjust this ratio to near 1:1 as recommended at Sheffield to see what improvement this had on the durability of the stove. In total 14 clays were tested for the C/NC ratio. Out of these clays 5 were selected to be tested further by making 3 types of samples. Sample 1 Sand was added to the selected clays in sufficient quantity to bring the C/NC ratio to 1:1. They were then used to make stove liners and test bars so that the shrinkage rates could be measured and to see how well the clays moulded. Sample 2 Stove liners were made with the selected clays using the C/NC ratio as made by the potters. Sample 3 Stove liners were made from clay no.7 and after 2 days of careful drying cuts were made through the bridge over the door way. Finally all the samples were fired in a kiln at 815-835°C. [top] [end]ResultsShrinkage (additive: pottery/feldspar sand)
Clay No's 4, 7 and 13 produced better specimens in terms of physical appearance and firing quality. [top] [end]Stress Relieving CutsIn total there were 5 liners with the cuts. 4 of the liners developed off-sets as follows:
Table A - Clay/Non Clay TestResults
No 14 was a mixture of 2 clays of grey plastic clay and brown clay in the ratio of 1:3 respectively. The ratio is the one used for making the ceramic liners. The brown clay alone is used by the producer for making bricks. No 4 during the clay fraction separation it sedimented poorly possibly indicating the presence of rnontmorillonite. Key - P clays used to make pots; L clays used to make liners, B clays to make bricks; NS new clays used to make liners.
[top] [end]Liner ProductionThe liners made from the samples with C/NC ratio of 1:1 and existing ratio for clays No. 4, 7 and 13 were all good quality and with no noticeable differences between them. The liners made from clay no. 6 with C/NC ratios adjusted to 1:1 all developed cracks during drying. Clay no. 11 had very little plasticity and attempts to make liners failed.[top] [end]ConclusionsFurther testing of the liners has yet to be completed and they need to be tested in service. It has not been possible to verify that by adjusting the C/NC ratio to 1:1, durability is improved. As far es can be seen only clay no. 12 appeared worse when the adjusted C/NC ratio was 1:1 compared to the as-used condition which had a C/NC ratio of 1:1.14.Certain clays are slow to sediment which can make the whole process of measuring the C/NC ratio rather slow. In general the C/NC ratio test procedure is a good starting point but not all clays necessarily need to be adjusted to 1:1. Stress relieving cuts are an effective means of reducing cracking but the offsets that develop as a result of warping will detract from the liners appearance and will effect how well the pot sits on the stove. [top] [end]2. Indonesian Pottery Production & Strengthby Aryanto Sudjarwo, Dian Desa, P O Box 19, Bulaksumar, Yogyakarta, IndonesiaThree different sites eg. Kasongan (Ka), Balangan (Ba) and Kwoso (Kw) were chosen where pottery is a way of life and where charcoal and/or wood stoves are made. In all three villages clay from the rice fields is used. In two places, Ka and Ba the wet clay preparation process is used to mix clay with sand in a ratio of 3:1. In Kw the dry method is used. The dry method results in a better prepared mixture as all foreign matter as well as larger particles can be removed during sieving. Mixing of clay and sand in the dry state is also easier and more thorough and the end result is a well prepared mixture. The production process in all three places is the same. Coiling is combined with beating the rough coils into the final stove shape with a wooden spatter and a smooth stone. The drying process in all three villages is the same. Drying initially is carried out in the shade while final drying takes place in the full sun. Frequent turning of the stoves during drying ensures that the stoves are dried evenly. Breakage rates are more or less uniform: 5-7%. Table 1 - Cracking rates of stoves from different areas during testing (4 samples each)
The firing processes show some differences. In Kw a kiln is used while in Ka and Ba an open fire is used for firing the stoves. In Ba firing is a two stage process: The stoves are pre-fired in an open fire using leaves (banana leaves mainly). This firing process which lasts about two hours results in the stoves becoming completely black, indicating that carbon is deposited on the outside and/or that the firing is carried out in a reducing atmosphere (not enough oxygen). The stoves are then cooled and stacked in a second open fire kiln using fuelwood as a source of energy. The potters claim that using this two stage firing process, the pots and stoves are better fired ea. give a clear ringing sound when knocked. The breakages in the different places and firing processes show some differences with average breakage rates ranging from 5-15%. The two-stage firing process at Ba has a breakage rate of about 15%, the open fire in Ka about 5% while in Kw the breakage rate is about 10%. Besides breakages there is also cracking of stoves but these are repaired by just filling and covering the cracks with a mixture of grog and rice. Cracking rates are normally of the order of 10%. The parts which cracked most were found to be the base and around the door opening as is shown in table 2. Both can be explained as deficiencies in the production process. Different parts of the stove will dry at different rates which will result in differences in drying shrinkages and hence in built-in tensions. Cutting the door opening after forming will result in small cracks as the clay will be partly sheared. Although not directly visible they form weak points and any stress formed during heating and cooling may result in cracks starting from these tiny micro-cracks. Table 2 - Parts cracked during tests
Note: Unlike the charcoal stoves the woodstoves do not have a grate. Both wood and charcoal stoves have a erased base [top] [end]3. Clay Testing in Ethiopiaby Genene Endale, Ministry of Mines, Ethiopia Energy Authority, Addis Ababa, Ethiopia[top] [end]BackgroundEthiopia is one of the countries in Africa where pottery is a tradition and clays are widely used by potters to make charcoal stoves. A mud-straw mix is used for making wood stoves where there are no organized producers. The Cooking Efficiency Improvement & New Fuels Marketing Project/ CEINFMP tried a number of possibilities to replace the mud-straw with other materials. One of the successful trials used a pumice-cement-clay mix. This stove is light in weight and is suitable for mass dissemination but the pumice and cement are difficult to obtain in most urban areas so an all ceramic stove appeared to be the remaining option. A ceramic, electric, injera stove had already been developed by the project but the large scale production of this had proved difficult.[top] [end]Testing clays used for making stovesA sample of clay suitable for making liners used incharcoal stoves was taken to test its C/NC ratio. The test procedure presented by Dr Gaspe and Dr Messer at the Clay for Cooking Stoves seminar in May 1992 was followed.[top] [end]The test took 4 days:Day 1The clay was crushed and sieved to 1mm size and tested for the presence of carbonates but the procedure used did not specify the concentration of the acid used for testing. Day 2 The sedimentation procedure used was based on Stoke's law* of settling (see page 4 for Stokes Law) Day 3 The suspension containing the 2µm clay fraction was siphoned off the top 10cm of the setting tank. The bulk of the water was then removed from the suspension by boiling in a metal pan using a charcoal stove. Day 4 The organic material was removed from the raw clay and from the sub 2pm fraction by heating to 375°C in an electric furnace. The dehydroxylation weight losses from the raw clay and the clay fraction were then determined by reheating to 700°C for 3 hours. The amount of clay mineral in the raw clay was estimated by calculating the ratio of dehydroxylation weight loss of the raw clay to that of the sub 2pm fraction. By knowing the amount of clay present, the C/NC ratio was determined. The sample proved to have C/NC ratio of 1:0.95. It was mixed with sand to make a C/NC ratio of 1:0.75 by volume. Liners made from this mix took only two days to dry and proved to be thermally and mechanically strong. During the rainy season the ceramic liners took quite a long time to dry. To make drying faster sawdust was tried as an additive to replace sand. The drying period improved remarkably but made the clay mechanically very weak after firing. Thermally-induced cracking was improved by avoiding sharp corners and substituting curved ones. To build stoves in sections was also very successful as a thermal stress relieving method by giving an expansion gap at the joints. This was tried on a pumice-cement-clay biomass stove and appears likely to work for a ceramic stove. This method is similar in principle to the stress relieving cut tried in Sri Lanka. [top] [end]Conclusions and RecommendationsThe stresses in clay stoves in service are a resultant effect of two components, namely i) the thermal stress and ii) the mechanical stress. Their degree of contribution may differ depending on the use of the stove. In the kind of biomass stoves used to cook Injera (a staple food in Ethiopia) the cook has to apply force to the ceramic cooking plate to remove food remaining from the previous cooking.For a fast drying, mechanically and thermally strong clay, a Clay/Non Clay ratio of about 1:1 needs to be mixed with a non-clay mineral like sand to give a ratio of 1:0.75 of wet clay to sand by volume. For stove programmes that aim to assist potters, the following major points are the necessary areas of study.
[top] [end]4. Thai Production Methods and Durability of Stovesby Auke Koopmans, HSE Greenfields, P O Box 167, Chiangmai 50000, ThailandIn 1990 a study was carried out by Hofman Systems Engineering in Thailand to assess to what extent production methods have an influence on the thermal shock resistance of stoves. For this test, small stoves were made resembling the Thai Bucket stove. The stoves are 10cm high, had a top diameter of 12cm, a bottom diameter of 9cm and a wall thickness of 1.5cm. The samples were mace et the workshop of the Northern Industrial Promotion Centre (NIPC) in Chiangmai, using different production techniques (hand thrown, coiling, jiggering, casting and pressing).
The results indicate that production methods do have an influence on thermal resistance, although the results are ambiguous. Stoves made by the jiggering and pressing methods both appear to have a lower durability Table 1: Effect of production methods and firing temperature of lifetime. The figures indicate test survival times in minutes.
(a) these examples were thinner (about 10mm as compared to 15mm for the other production methods) due to production problems (saturation of mould) The low durability of stoves made by jiggering be common, at least from the evidence available up to now. For instance, the CA e in El Obeid in the Sudan distributed stoves made in different ways. Stove liners made by hand lasted for two years while with jiggering no satisfactory liners could be made. The same experience was gained at another place where JIKO stove liners were produced both by hand and on a foot operated jigger-jolly. After firing, with both types distributed at random over the kiln to allow for firing discrepancies, there was a marked difference in the failure rates. Liners made on the jigger-jolly showed a much higher failurer rate than the handmade liners. The same happened with different mixtures (personal comunication Tim Jones, ITDG, see also Boiling Point No 22 of August 1990, Clay Testing Centre -Sudan) There may be several reasons for the low strength with jiggering. The production process can result in differences in compaction if the clay is not centred properly in or on the mould. If too much water is used during the jiggering process, there may be weak planes within the body due to water being locked up between clay layers. Another, and probably the most important factor, could be that during the jiggering process the clay particles which are in the form of tiny platelets, become aligned. As Dr Messer and Dr Gaspe (see Clays for cooking Stoves, Seminar Report, avalable from ITDG) showed during the first seminar this will re-enforce anisotropic shrinkage ie. shrinkage is different in different directions.These differential shrinkage rates result in internal stress with the outside of a circular stove developing a tensile hoop stress while the inside may be under compression. It is known that in order to decrease drying shrinkages, non-clay materials such as sand, ash, groug, etc. should be added. Having less shrinkage, it may be assumed that the amount of tension built up will also be less This has been proven in the work carried out in Sheffield where extruded cylindrical tubes with different amounts of non-clay, in this case silica, were added. After cutting the tubes radially, the samples with larger amounts of silica added to the clay oppened up less than those made out of clay only. [top] [end]5. Clays for Nepali Cooking Stovesby K.M. Sulpya, RECAST, Tribuvan University, Kathmandu. NepalPottery stoves have been promoted in Nepal for many years mainly because clay is readily available and there are potters with good skills. However, the number of stoves disseminated remains low for a number of reasons - but it is considered that their poor durability has been a major contributing factor. Recently a study on suitable clays for producing stoves was carried out to try to improve their strength and durability Determination of C/NC ratio in Naural Clay Bodies The C/NC ratios of 6 samples were determined using the procedure set out in the second article in this issue. Table 1- C/NC Ratio by Clay Type
Of the five clay samples rested, only Janakpur clay contained a majority of non-clay materials. [top] [end]Adjustment of C/NC RatioNatural clays contain more clay materials with a ratio greater than 1:1 as shown in Table 1. The addition of a predetermined amount of a non-clay material such as fine sand or paddy husk ash reduced the ratio. This was found by measuring the weight of raw clay samples before and after drying at about 110°C overnight (Gaspe and Messer 1991). Table 2 shows the adjustment of C/NC ratio.Table 2 - Adjustment of Clay/Non-clay Ratio
[top] [end]Drying and Firing ShrinkageThe drying and firing shrinkage rates are determined by making 4 test bars 130 x 25 x 20 mm for each clay. See Table 3. Table 3 - Drying and Firing Shrinkage by Clay Type.
Table 4 - Influence of Production Method on Stove Fracture by Clay type. Water Boiling Tests
[top] [end]Fracture TestingTwo procedures were used to test for fractures in stoves made by hand throwing and by beating.1. Water Boiling Tests Table 4 Water was boiled on the stove for 1 hour, twice daily, for 6 days and the total number of large and small fractures was noted. 2. Field Testing Table 5 Stoves were distributed to families who used them to prepare their normal meals 3 times per day for 6 days. The total number of large and small factures was noted. [top] [end]Results & Discussion
Table 5 - Field Test Results
[top] [end]Contents: Boiling Point 31: Clays for Stoves
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Page created:
28 August 2007; Last edited:
30 August 2007; Version: 0 | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pagename: ClayTesting-5CountryReports @HEDON: DCHA | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
