The essential requirement of a good control system of gold recovery plants is efficient and representative sampling of the process streams. Metallurgical accounting is a key part of all gold operations for plant process control and to determine the distribution of gold in the various mill products. Basically, most concentrators produce a metallurgical balance showing the plant performance for each operating shift, the shift results being averaged over a daily, monthly and yearly period to show the overall performance.
Modern gold beneficiation plants employ computer control of mineral processing and this need continuous measurement of process parameters and the use of on-line sensors such as magnetic flowmeters, nuclear density gauges, on stream analyzers, and particle size analyzers. In this way, the process sampling for all main stream flows in a gold plant is performed with automatic sampling devices linked to an on stream analyzer system. Basically, this type of system ensures the rapid and accurate measurement of the overall plant efficiency. Nevertheless, sampling of secondary process streams within the plant circuit will become important and necessary from time to time. This operation needs the use of manually operated sampling devices such as belt cutters for primary mill feeds, and bottle samplers and slurry cutters for slurry streams.
Sampling is the means whereby a small amount of material is taken from the main bulk in such manner that it is representative of the larger amount. In gold assay sampling, for instance, a flotation feed stream of 2,000 t/h of dry solids will be represented by an amount of 50 grams or less. For this reason, there is a great responsibility rests on a very small sample. Essentially, the fundamental sampling error is the minimum sampling error produced by the differences between individual particles, and is dependent on several aspects such as sample weight, particle shape, mineral and gangue specific gravity, particle size, gold content, and gold liberation.
The grouping and segregation error is a complementary sampling error generated by the differences in gold content between sampling increments, and is dependent on the number of increments per sample and the stage of homogenization of the mineral. Errors in stream sampling will be a function of non-uniformity in the flow stream due to segregation by specific gravity and stratification by size. Classification will be particularly evident in solids with the changes in particle densities and sizes. Fines and heavy minerals tend to settle through the burden during conveyor transport and accumulate next to the belt. Slurry will classify during flow in a launder as a result of velocity variations, with larger and higher density particles settling to the bottom of the stream.
Many factors come into play in the procurement of a good sample. In practice, the most satisfactory method of minimizing variables during sampling is to sample the material while it is motion at a point of free discharge. Two sampling procedures can be considered, one is to divert the entire stream to sample collection for a specific time interval; and the other more realistic procedure for large stream flows is to take a portion of the stream to sample using a cutter. It is important to define the objective of the work, specify the point to be sampled in terms of the quantity of flow, the quantity of sample required to achieve the desired level of precision, selection of equipment to obtain a representative sample and the possibility of reducing the primary sample volume by size reduction and secondary sampling methods.
Coarse Material Sampling | Slurry Sampling | How to Take Slurry Samples | Slurry Density Adjustment | How to Obtain a Representative Sample | How to Obtain a Sample in Steady State | Sampling Error | Sample Processing of Gold Circuits | Precautions for Size Analysis |