Residual Moisture Reduction through the Development of an Air-Purged Chute: Vol 1 - Commercial Scale

Volume 1: Air-Purged Centrifuge: Air purging is a novel process that involves the injection of a turbulent stream of compressed air through a bed of coarse coal particles physically displacing adherent water and delivering lower-moisture products. The concept has previously been demonstrated in both the bench and pilot scale centrifuges. This report describes the results of a project that demonstrates the success of air purging in commercial scale centrifuges.

A series of trials was undertaken on one of the centrifuges at the Port Kembla CHPP which process 12 x 0.5 mm coal. Using 2 x 1600 cfm compressors to drive air through a manifold 10 x 9 cm in open area led to a reduction in moisture of 0.7 wt% on Appin coal at a confidence limit above 90 %. Increasing air speed by halving the manifold area produced a moisture reduction of 0.8-1.0 wt% at the 85 % confidence limit on Cordeaux coal (which was slightly coarser than the Appin).

Results

The best results did not quite reach the 1.0 wt% moisture reduction target which had been adopted as the project objective, and which was thought feasible based on the results of the previous pilot trials. The main difference between working at the two scales was the thickness of the bed of coal traveling down the centrifuge basket. Using a new simple device, purpose designed to determine bed thickness inside an operating full-scale centrifuge, the average bed depth was determined to be ca 80 mm, compared with only 30 mm in the pilot scale machine. This extra depth made it much more difficult for the air stream to penetrate the bed and hence remove moisture.

During the project, much progress was made in achieving a reliable and effective way of installing and retaining the manifold within the centrifuge. The preferred option is to construct manifolds from steel plate, coated with a ceramic wear-resistant material, and attached by a screw thread to the air delivery pipe. It is thought that manifolds should be able to last several weeks, and probably only require changeout when the basket is changed.

Conclusions

The lowest-cost route for applying air purging is probably by using blowers. In addition, the costs of applying air purging will be site specific depending on how the system is engineered, lengths of pipe runs etc. One specific example, based on the Appin results, suggests that if one blower is applied to the centrifuge (not the most cost-effective arrangement but a reasonable, conservative starting point for this type of calculation), then the capital cost of the blower will be ca $75,000 with an operating cost of ca 0.5 kWh/tonne of coal.  

Further, more prolonged trials, especially on -50+0.5 mm size coal, would be required to more firmly establish performance and costs..  

Volume 2: Air-Purged Chute: The air-purged chute involves injecting a turbulent stream of air into a chute carrying coarse coal such that the air penetrates the bed of particles, removes water from the coal by physical displacement, and transports the moisture through both the bed and out via a screen located on the reverse side of the chute. The application of air purging in chutes rather than in vibrating basket centrifuges has the potential advantages of:

• The bed height can be set and controlled.
• As the chute runs full of coal, the opportunity for air to pass anywhere else other than through the bed is minimised.
• The chute has no moving parts and thus should not be subject to high wear rates.

The project was conducted at the CSIRO pilot scale facility at Catherine Hill Bay. The feedstock for the trials was coarse coal, nominal size range -50+2 mm, taken from the drain and rinse screen at the Catherine Hill Bay CHPP.

Results

The initial phase of the work involved constructing a rig and devising a procedure for operating a chute at a nominal feed rate of 10 t/h. Three designs of chute were tested, designated Mark 3, 4 and 5, with the objective of reducing moisture by 1 wt% compared with that of a conventional vibrating basket centrifuge. The main conclusions were:

• In three runs using starting moisture between 6.6 to 7.4 wt%, which simulated the plant centrifuge product, the final dewatered materials had mean moisture values of 5.6 to 6 5.wt%. The moisture reduction in these runs was between 0.9 and 1.2 wt%, significant at greater than the 95 % confidence level.
• The best result was achieved by using a bed thickness of 100 mm, air flow rate of 600-700 cfm (1020-1190 m3/hr), coal flow rate between 3 and 8 t/h. These conditions provide the starting point for further development and scale up of the chute.
• There was no evidence of blockage in the chute during the entire testwork program of over 230 runs, perhaps because the air stream was partially fluidising the coal. There was no significant difference in moisture reduction between the chute being choke fed (the original concept) or with the coal in free flow.
• To achieve effective dewatering an airflow to coal feed ratio of approximately 100cfm (170 m3/h) per tonne per hour is required.
• James C Donnelly and Associates were commissioned to provide estimates of capital and on-going operating cost for an air purged chute treating a nominal 100t/h of coal flow. An estimate of payback period based on a cost saving of $0.80/t per 1% moisture reduction was also requested. The analysis showed that a capital cost of $176,192.00 and an operating cost of $0.29/t could be expected for a 100t/h unit. These estimates, of course, made certain assumptions with regard to plant configuration and ease of retrofit. However, a payback period of the order of 9 months seemed a likely outcome. Further, it would seem that the overall economic viability would tolerate potential underestimates in any on the major equipment items.

Conclusions

Although successful at dewatering centrifuge product, the unit has also dewatered centrifuge feed albeit at reduced feed rates. It appears possible to achieve product moisture values 1 wt% below those currently achieved in centrifuges.  

To date, it has only been possible to perform relatively short term (ie ca 1 hour) trials, the limitations being mainly imposed by limits on the quantities of materials that can be handled in a pilot plant situation. Clearly, longer term runs would be beneficial.