Effects of sample mass on gravity recoverable gold test results in low-grade ores

Abstract

Gravity recoverable gold characterisation methodologies are the single- or three-stage tests, which generally use a standard feed sample mass. A case study of low-grade coarse gold-dominated mineralisation is presented that demonstrates the high variability of test results using different sample masses. Samples were processed using the single-stage test and subsequently entire development rounds were batched through a plant for comparison. Unsurprisingly the results indicate that the small (≤50 kg) test samples grossly understate plant gravity recoverable gold and display poor precision. Larger samples display improved precision, but still understate plant gravity recoverable gold. The small mass samples are unrepresentative as they do not contain the full gold particle size distribution. Poor representivity is enhanced by gold particle clustering. Small samples generally capture finer more abundant and disseminated gold particles, but rarely contain clustered gold. The use of standard GRG test sample masses is challenged. Test work should be based on spatially distributed representative field samples, that if required are split to representative sub-samples for testing. An early stage gold particle size characterisation programme is required to optimise sample mass and improve representivity.