Coal Preparation » Dewatering
The objective was to assess the effectiveness of the techniques being used in plants for conditioning of filter feed with flocculant, and establish whether improvements in conditioning could lead to improved filter performance. The criterion of success adopted in this project for good conditioning was the weight percent of the flocculated feed passing 25 mm since it is the ultrafine particles which essentially control filter performance. This approach, which analysed the state of the flocs themselves generated in plants, was thought to be novel.
The four plants assessed in this project were selected to give as broad a range as possible of fines circuit configuration and equipment type, in terms of flotation conditioning and filtration. Two were in NSW and two in Queensland.
The proportion of ultrafines below 25 mm was determined using a laser diffraction analyser and an experimental procedure aimed at avoiding any break up of the flocs prior to analysis. It was found that the flocculated filter feed in all four plants still contained significant amounts (from 10 to 23.5 wt%) of ultrafines passing 25 mm. This was much greater than the level of zero thought to give optimum filtration.
In a series of bench scale tests a link between product moisture and the weight percent of material in the filter feed below 25 mm was established on the feedstocks from all four plants. This was achieved by altering flocculant dose or Camp number, a measure of the agitation received during conditioning. The basic postulate underlying the project was thus substantiated. For Plants 1 and 2, the relationships were almost identical.
An attempt to realise the bench scale predictions at full scale in Plant 1 was unfortunately not successful. The proportion of ultrafines in the filter feed was reduced from 17.6 to 10.4 wt% passing 25 mm by increasing flocculant dose from 8 to 20 g/tonne. However instead of reducing moisture, the reduction in ultrafines led instead to a substantial reduction in vacuum. Apparently the filter vacuum pumps could not cope with the greater air flow caused by the increase in cake permeability.
The results of the work have been encouraging but there are several aspects that are still not understood. The next stage of the work is recommended to be an assessment of a new laser back-scattering device which is available as an on line instrument and therefore could overcome many of the vagaries of sampling and off line measurements. If this instrument can produce meaningful floc size data then this would help in optimising flocculant selection at a given plant. It would also ensure that the flocculant dose/conditioning can be maintained close to optimum to maximise efficiency of flocculant usage and filter performance, commensurate with the plant being able to maintain vacuum with higher permeability filter cakes.