5. Technical Account of a 3 Tonne/Day Limeburning Operation in Moshaneng, Botswana[edit | edit source]
5.1 Background[edit | edit source]
The purpose of this section is to provide technical and operating information of a specific project experience using a kiln design which has been well tried and tested. It should be noted that the information given will not necessarily apply in other circumstances. The description and recommendations are intended to serve as a stimulus for the fieldworker and as a general guide through the trials procedure. .
The original plan was to supply the local building industry with lime for use in mortars and plasters. However, in the course of investigations, it was discovered that a demand for agricultural lime, lime for soil stabilization and a small amount of lime for the local tannery and gold mining operations existed.
The project had two major aims: firstly, to provide a quality product which would substitute a portion of the South African imported product and secondly, to concentrate on labour intensive methods to maximize employment creation. Further, the limitations were to use locally available materials for the construction of the kiln, and to design it so as to orient the operating, repair and maintenance manpower requirements to the local skills levels.
The location of the site was determined by the fact that already crushed material was available on a tailings dump of an abandoned mining operation. Thus the quarrying and dressing aspects of the projects would be limited to hand selection and transporting to the kiln. Local demand was estimated at 1000-1500 tonnes anually.
Raw materials -dolomite, coal
Dolomite: The tailings dump contained 100-150 thousand cubic metres of usable stone. The dump was variable as to stone size and quality but selection was relatively easy. 20 villagers were employed on a "piece rate" basis to select and transport stone to the kiln.
Chemical analysis of the usable stone was as follows:
Average CaO content 31.00 %, Average MgO content 21.50 % Theoretically this could produce 89.96 % quicklime. This would be an acceptable quality lime for the purposes required.
Coal: Low volatility 22 MJ/kg.
5.2 Production process[edit | edit source]
Dolomite lumps selected and prepared at the tailings dump were brought to the kiln ready for firing. At regular intervals a 20 litre sample was taken, visually checked for quality and weighed. A 5 % variance in weight from a predetermined mean was set as acceptable. The purpose of this was to control the grading of the dolomite lumps.
The stone feed and coal were then brought to the top of the kiln by means of wheel barrows and loaded in alternate layers.
A layer of firewood was placed at the base of the column of material to assist the ignition of the fire. The fire was lit (with difficulty) and after a period of approximately 24 hours it reached the firing zone. The burnt lumps could then be extracted from the outlet ports. The temperature in the kiln was monitored by means of thermocouples. The quicklime produced was transported by means of wheel barrows to the slaking floor where it was slaked, and then screened by a barrel screen and bagged in 25 kg bags. A labour force of 14 men was employed on a full-time basis and of 20 on a temporary basis.
Plant and equipment
- Khadi Village Industries (KVI) type, 3 tonne per day vertical shaft kiln.
- 3 thermocouples.
- Wheel barrows, shovels, sledgehammers and extraction forks.
- Slaking floor.
- Hand operated barrel screen with a 5 mm and 30 mesh screen.
5.3 Trials[edit | edit source]
Following the chemical analysis of the raw materials by the Geological Survey Department and their suggestions, the stone was fired in the on-site testing kiln (see section 4). Positive results were obtained. The construction of the full size kiln and the execution of the trials followed.
The objective of the trials was to determine the best operating conditions under the prevailing circumstances. It must be noted, particularly for this size of project and level of technology that the most appropriate and effective operating conditions can only be determined after continuous trial and error by the operator during the course of production. The first trial runs, conducted as scientifically and systematically as is practically possible will serve as a means of obtaining the basic operating information on which to continue. Extensive trials, prior to going into production, conducted with strict scientific discipline, will not necessarily result in a proportional improvement in quality or cost effectiveness. It will almost certainly burden the project with a disproportionately high implementation cost relative to the total investment cost.
The method adopted for the trials was as follows:
1. Representative samples of stone feed and coal were weighed and the volumes charged recorded.
2. Firing commenced and the temperature at the three kiln zones was recorded at half hourly intervals (need only be at hourly or two hourly intervals).
3. Quicklime was extracted and representative samples weighed.
4. Output was inspected visually and physically.
5. Output was slaked and a representative sample sent for chemical analysis.
6. Results were considered and appropriate adjustments made.
The trial results must provide the following information:
1. The correct proportion of coal to stone, volume of layers, stone size.
2. The temperature levels required to produce a wellburnt, uniform quality product.
3. Rate of extraction.
4. Water requirement for slaking.
5. On going quality control requirements.
5.3.1 Observations during firing trials
- In practice, the range of feed size (hand sorted) could not be limited to better than 10 % each side of the mean.
- The stone shape was found not to vary to any noticeable extent.
- The coal size varied from between 5 to 50 mm diameter lumps.
- Strong gusts of wind, as indicated by the thermocouple, caused the temperature to drop back by 25 °C.
- Wind blew in consistently from the west through the western discharge opening.
- Temperature, where it exceeded 950 °C caused an increase of overburnt material with no obvious changes in the amount of core material.
- The fire rose up the shaft at a rate of around 100 mm per hour through a column of material having a stone feed size of approximately 25 mm diameter.
- The material classified as well-burnt had a 10-15 % portion of unburnt core.
- The quicklime lumps became progressively darker from underburnt to overburnt.
- Well-burnt stone weighed less and was softer than overburnt stone.
- The loss on ignition (LOI) was around 28 % for wellburnt material, having a 10 % core.
- The rate of extraction was between 600 700 litres every 4 hours.
- A large variance existed in the degree of calcination of the quicklime lumps.
- The bulk density of the stone feed was 1430 kg/m3 and the bulk density of coal 980 kg/m3.
5.3.2 Batch versus continuous operation
Trials were conducted both on a batch and a continuous basis. It was found that the major advantage of the continuous operation was that it had a greater output per unit time. Consequently the fixed cost per unit output was lower. Other advantages were that a slightly more even quality product resulted, the difficulty and headache of starting a fire was avoided and although it was not possible to gauge exactly to what extent the amount of fuel required could be reduced, i.e. a more thermally efficient operation was possible. It was observed that, in addition to the considerable amount of heat lost in heating up the kiln each time a new batch was started, a substantial amount of heat was lost to the atmosphere when the fire reached the upper level of the kiln. The batch operation is definitely wasteful of heat.
However, if the demand is irregular and unreliable or labour cannot be employed on a continuous basis, e.g. in a subsistence farming environment, a batch kiln may be appropriate. The economic implications of the choice must be carefully defined before a decision is made.
Trials Results
- The percentage under-and overburnt lime was determined by handsorting a representative sample of output and visually estimating the proportions. Handsorting was based on differences in colour and weight.
- The percentage core was determined by slaking a representative sample of the under burnt stone. A portion of it remained unslaked which constituted the core. The proportion were visually estimated.
5.4 Hydration[edit | edit source]
It was decided to hydrate the quicklime to a powder on a concrete slaking floor and then screen it using a barrel screen and bag in 25 kg multilayer paper bags. The quicklime was slaked by being spread in a layer 200 mm thick on the slaking floor and water was sprayed on it. After the initial spray the material was mixed, piled in a mound in the centre, respread and sprayed again. This procedure was continued until the pile no longer absorbed the water i.e. the slaked lime produced (pebbles, sand and powder mixture) was slightly damp. It was found that if one waited a few minutes between each spray and then piled the material up high, the quicklime was broken down to powder more effectively. Further, if it was left to lie two days before screening, a portion of the slow slaking, overburnt quicklime, slaked.
The material was screened to material over 5 mm diameter, between 5 mm and 30 mesh. The amount of material in the minus 30 mesh range was around 50 % that between 5 mm and 30 mesh around 35 % and the remainder over 5 mm diameter. It was observed that the material in the range 5 mm-30 mesh slaked further in the stock pile and a fair amount of minus 30 mesh lime resulted after screening. The minus 30 mesh slaked lime was bagged in 25 kg bags and stored in a dry area.
5.5 Conclusions and recommendations[edit | edit source]
5.5.1 Firing
The wide variance in quality seemed the most urgent matter to be dealt with. The irregular supply of air sometimes cooled the fire and sometimes provided too little air for combustion.
To alleviate this problem it is recommended that gates are built at all four discharge openings which can be opened and closed by the operator as required. This will provide a slightly better control of the draught.
Other adjustments which would make for a better distribution of heat are:
- Reduction in the volumes of the stone and coal charged in each layer from 160 litres stone and 40 litres coal to 80 litres stone and 20 litres coal;
- Making the coal size more uniform and distributing it evenly over each layer of stone. The weight ratio must then be checked to ensure that the necessary adjustment in volume is made so that the proportion of coal used is not increased;
- Keeping a careful control of the quality of stone fed into the kiln. The fine crystal variety is preferable, but more important, the quality fed into the kiln should be consistent.
Another problem was the large amount of underburnt material comprising a considerable proportion of unburnt core. It is recommended that the following measures be taken:
- Reduce the stone size to 100 mm to reduce the draught through the kiln.
- Fire with a stone to coal volume ratio of 4:1. The temperature should be kept at between 900 -950 °C.
5.5.2 Hydration
Under the circumstances slaking should be continued for the time being on the slaking floor, screened and bagged as described. The change recommended is to try the production of lime putty from the screened oversize material. Small, trial slaking pits should be built in which the oversize material, with an excess amount of water can be allowed to slake for a period of 2-3 weeks. The over 0.5 mm* material should then be screened out. The screened putty must then be tested chemically, including a soundness test, as described in section 4, to ascertain the extent to which the overburnt core remaining in the lime putty causes popping.
(See the fig. above for a schematic representation of a hand operated hydration plant that could be used.)
If these tests prove successful, before going into full production of the lime putty' the saleability of lime in such a form should be established and the economic implications calculated and evaluated.
