The following report highlights the pertinent conclusions of a feasibility study by Pyromet Technologies (Pty) Ltd. for a new 250,000tpa Manganese Ore Sinter Facility in South Africa. As part of the study, sinter testwork was performed at Kumbas pilot plant in Pretoria, South Africa, on typical South African Manganese ores, as well as on some unique blends of Manganese ore, Ferromanganese baghouse dust and scrubber sludge from a local producer.
Important design considerations for a smaller scale plant are highlighted and the testwork that was performed as part of the study is described. The report also draws comparisons between the test results, literature and current operating practice. Wherever the conclusions were found to have an impact on the process design of a sinter plant, these factors are pointed out.
THE SINTERING PROCESS
It is assumed that not everyone who read this paper would be familiar with sintering, so a short description is given below:
The process of sintering is an agglomeration technique for fine ore that relies on heat to melt the surface of smaller particles together to form larger agglomerates. A typical sinter plant consists of a number of sequential operating units with the sinter strand at the heart of the plant. The simplified process sequence is as follows:
Raw materials such as ore fines, coke/coal, dust/sludge and in some cases slag modifiers and additives, are batched and conveyed to a blending system. The raw materials are blended in a rotating mixing drum with Sinter Fines and water to achieve a Green Feed in a process commonly referred to as Nodulizing. The nodulized green feed is introduced to the sinter strand on top of a sized Hearth Layer to form the Sinter Bed. This bed now passes through the Ignition Hood to initiate the reaction. Burners in the hood ignite the carbon in the green feed and the reaction is propagated by chemical reaction between the carbon and air sucked through the sinter bed by the offgas fans. The sinter burns through vertically while the bed moves horizontally towards the discharge end. The sintered material is discharged through a finger crusher onto a cooling strand, where ambient air is blown through the crushed material from below. After cooling, the sinter is conveyed to a crushing and screening station where it is sized and finally conveyed to product storage.
Despite all other demands which the design of a modern strand-type sinter plant possess, such as materials preparation and handling, as well as reasonably sophisticated controls and instrumentation, the sinter strand proper remains the core of the plant. Below are therefore outlined the design requirements and engineering approaches for the main components of the sinter strand. To a certain extent, very sophisticated computer software is utilised for these developments, such as Flo++ for Computational Fluid Dynamics (CFD) and MSC.Marc® for dynamic thermal stress modelling.
1 Pallet Cars
The pallet cars serve to convey the sinter along the strand and above the windboxes (wherein negative system pressure prevails), while the sintering process takes place.
The sinter strand can kinetically be regarded as an unlinked endless chain. The pallet cars are therefore subject to stresses resulting from:
- exposure to cyclic thermal variations due to the high temperatures experienced on the upper (sintering) strand and cooling taking place on the lower (return) strand sections,
- exposure to cyclic static loads from the mass of the green feed/sinter, • exposure to cyclic dynamic loads from the forces imparted by the drive sprockets, as well as by the pallet cars against one another.
While the development of strand-type sinter plants and their component materials have been going on for well near a century, the fact remains that the above described arduous duty will fatigue the most suitable materials (e.g. HS Meehanite® and similar cast irons) within a limited number of cycles. The major European operators of sinter plants therefore base their pallet cost estimates upon an average service-life of 10 years, at somewhat less than 8000 hours per year.
The choice of materials and shapes of the components of the pallet cars is further governed by the following requirements:
- minimal pressure drop through the grate bars,
- maximum abrasion resistance of the grate bars,
- maximum ductility and abrasion resistance of the cheekplates in respect of sliding motion of green feed and sintered material against same,
- quick exchangeability of worn or otherwise unserviceable components by unskilled labour.
2 Sinter Strand Drive
The (non-linked) pallet cars are pushed along the top strand of the machines frame by the drive sprockets, which are fitted with shrink-discs on a common shaft. The sprockets are equipped with replaceable tooth segments, precision-cast from special steel. The teeth impart a rolling action upon the inner wheels of the stub axle assemblies, of which four are attached to each pallet car. Note that the outer wheels of the stub axle assemblies serve to guide the pallets at their return points, i.e. at the drive and discharge stations, while the inner wheels carry the static and dynamic loads as the pallets are pushed along the strand.
The drive of the sinter strand is normally not located at the discharge end of the strand, for reasons of heat and serviceability. Prime mover options are:
- electro-mechanical, with Variable Speed Drive (VSD).
- electro-hydraulic, with variable displacement pump or motor.
Dual or single drives can be fitted. The main considerations in the choice of drives and drive arrangements are given to:
- reduction of overhung loads, by use of shaft-mounted planetary gear-boxes,
- speed range,
3 Take-Up Mechanism
This serves to compensate the differential thermal expansion between the moving pallet cars and the machine frame with the rails and windboxes, while maintaining adequate pressure to avoid separation of the pallet body faces. Take-up mechanisms are generally automatic by means of counterweight/pulley systems, or hydraulic.
The benefits of a hydraulic system are:
- minimum pressure can be dialled-in to reduce frictional wear between the pallet body faces,
- replacement of single pallet assemblies (opening the strand) is facilitated by the use of a double-acting cylinder (or cylinders).
While some suppliers have in the past for very valid technical reasons provided large strands, where the takeup station is located at the discharge end of the strand, for smaller machines it is more feasible to provide that facility at the cold drive end. In either case, the respective station must be designed as a mobile unit, either mounted on a wheel/rail arrangement, or suspended from it. An accurate guiding mechanism, which allows alignment of the drive station on the centre-line of the strand, is essential.
4 Crash-Deck and Finger Crusher
The crash-deck serves to guide the sintered material, as it is discharged out of the pallet cars, into the finger crusher. As the crash-deck is subject to severe impact and abrasion, it is heavily lined. In one case at Sidmar in Belgium, on a 5,8 m wide machine of about 8 million tpa capacity, the crash-deck is lined with ceramic cubes of 300 mm square cross-section, 500 mm high. The life span of these liners is quoted to be one year i.e. the duration between the yearly scheduled plant maintenance periods. For plants of smaller capacity (in the range of several 100 000 tpa) the lining of the crash-decks with hardwearing plate (Hardox®, Ti-Hard), or even cast (e.g. Titanium Carbide alloys) is normally adequate.
The finger crusher reduces the lumps of sinter cake to a size smaller than 150 mm, in preparation for cooling and secondary crushing. As the sinter cake retains a temperature of about 850 degrees C upon leaving the strand, the finger crusher operates in a very hostile environment.
This fact is addressed with the following design features:
- exchangeable fingers/finger-wheels,
- shielded bearings,
- water-cooled shafts (on larger machines),
- quick exchangeability of complete shaft assembly, incl. fingers, bearings and bearing shields. This necessitates the facility for easy removal of the dust canopy, as well as easy disconnection of the crusher drive. On some existing plants, the operation is performed in a couple of minutes with the help of an overhead crane.
5 Ignition Hood
Ignition hoods can be described as refractory lined steel boxes, in which two or more horizontally opposed burners are arranged. Any kind of fuel, like fossil fuels or furnace off-gas can be used as heat source. Vertically operable doors shut off the faces of the ignition hood down to the top level of the green feed, in order to minimise heat losses. The duty of the ignition hood has been explained above.
To meet these requirements, the ignition hood must be equipped with the following features:
- the burner flames must operate with low velocity, to avoid disturbance of the green feed bed,
- a flat flame shape is beneficial for fast and even ignition of the green feed,
- adequate provision must be made for the controlled supply of cooling air to the burners, so that the requisite flame temperature can be dialled-in,
- the operation of the burner controls must be easily understandable,
- all controls must be fail-safe,
- pilot flames must be reliable, e.g. if the burners are operated run with gas of low and/or fluctuating calorific value, like furnace or coke-oven gas, it is recommended that the pilot flames be operated on Liquid Petroleum Gas (LPG).