To serve you better, our new website displays information specific to your location.
Please visit the site and bookmark it for future use.

Poor Gold Reconciliation: Acknowledge the Nature of the Ore

SRK News | Issue 53: Metallurgy and Mineral Processing

A4   |   Letter

Gold reconciliation audits are often motivated by the following question:

Did the measured gold ever exist?

• If yes, locate the gold if it is still within the circuit, or identify the likely route by which it left.

• If not, establish reasons for the erroneous estimation and propose remedial measures to rectify.

In essence, gold discrepancies can be reduced to one of sampling/ measurement deficiencies or physical losses/accumulations, or a combination of both. While gold theft is a real risk, this is generally not our focus.

Clearly, one should assess the metal accounting systems and procedures but, in certain instances, one should also acknowledge the nature of the ore.

For example, in a large South African gold mine, two shafts enjoyed the luxury of dedicated plants. Declining tonnages resulted in the decision to process ore from both shafts through the larger, more modern plant. Recovered gold was allocated between the shafts in proportion to the respective head content, with gold grade being based on coarse run-of-mine samples taken by automated crossbelt samplers. Sample size, frequency and size reduction/splitting procedures were considered to be good. Almost immediately, however, the indicated gold recovery of the smaller, lower-grade shaft dropped, bringing its viability in doubt.

Interrogating the systems and procedures could not answer the question. Importantly, it was known that the reef geology and gold deportment of the two ore sources differed (see below).

Nature of ore at a South African gold mine

Video footage showed that the cross-belt sampler over-sampled finer material, which would tend to bias the grade of Shaft 1 Reef upwards.

The aperture of the sample cutter did not respect the basic rule of three times the maximum fragment size. This resulted in rejecting low-grade coarse particles, which further biased the value of Shaft 1 Reef upwards.

Ultimately, it was concluded that the gold had existed but as a consequence of the nature of the reef as well as the sampling methodology, recovered gold had been incorrectly allocated to the detriment of Shaft 2.

Vic Hills:

SRK Latin America