An on-stream iron ore analyser has been developed in collaboration with Hamersley Iron Pty. Limited for determining the iron content of both lump (- 30 + 6 mm particle size) and fines (- 6 mm particle size) on conveyor belts. The analyser, which is called IRONSCAN, is based on pair production, and is now commercially available from Mineral Con- trol Instrumentation Limited (MCI) in Adelaide. It can be mounted under existing conveyor belts with minimal modifications to the conveyor struc- ture, and the presence of steel cables in the belt does not interfere once the analyser has been correctly calibrated. The analyser has been exten- sively tested on both lump and fines on the shiploading conveyor at Damp- ier, and typically the root mean square (r.m.s.) deviation between single IRONSCAN measurements and conventional chemical analyses is better than 0.5% Fe. It is currently being evaluated on - 150 mm ore from the primary crusher at Mount Tom Price, and initial results are encourag- ing. The principal advantage of IRONSCAN is that it provides rapid infor- mation on ore grades. But perhaps its greatest potential is at the primary crusher where conventional sampling and analysis is very expensive to Implement. at the University of Melbourne, and those of us who were his students had learnt of the application of physical chemistry to pyrometallurgical processing, pioneered by Chipman in the USA, Wagner in Germany, and Richardson in the UK. It was possible, by applying this theoretical insight to the equilibria between metal, speiss or Legierung, matte and slag phases, to find a practical and effective way to recover metal values in some residues which had been accumulating for many years at the NA, but which had defied all attempts to treat them to that date. There was also an opportunity to study the batch lead refining methods still in use at all lead refineries apart from Port Pirie, and to devise optimis- ing techniques which greatly improved the operations and effected considerable savings. Next the writer spent some six years alternating between the BHAS, Port Pirie, and Imperial Smelting at Avonmouth in the UK, concerned with lead or lead-zinc smelting, and the develop- ment of several further continuous lead refining processes, before returning to Australia as Section Leader, Process Development in the CSIRO Division of Chemical Engineering, Fisherman's Bend, Victoria. Metallurgical opportunities which arose during this last period included the problem of tin smelting which had developed over about a decade: the world's tin smelters had been accustomed to treating fairly high grade tin concentrates, con- sisting of almost pure cassiterite, with very little impurity-mostly quartz, limestone, iron oxides, and sundry other silicates. Most such concentrates had been produced from alluvial deposits, where nature had already done some of difficult part of the mineral processor's work-at who knows what recovery-so that he could now produce high grade concentrates at high recovery._x000D_
The smelters were not really set up for the treatment of lower grade concentrates, such as could be made with reasonably high recoveries from rock mining. Thus they imposed such penalties on impurities and lower than desired tin grades, that whenever a mineral processor improved tin recovery at the expense of lower- ing grade, such an improvement entailed a lower financial return to the miner. There was clearly a need for improved understand- ing of tin smelting, (at that time shrouded in secrecy) and develop- ment of techniques better suited to the treatment of lower grade concentrates. As well as the practical development, a start was made with the computer modelling of the smelting process, with the obiect of optimising the process parameters.