The variations created by control and procedural problems in the testing program have an impact on the projected and actual metallurgical results in a gold operation. For this reason the laboratory work must be done at detail. In the case of flotation tests, the base or standard for metallurgical predictions is the locked cycle test in which recirculation of intermediate streams in the flowsheet is practiced. Basically, locked cycle test produce better results than a full scale flotation plant can obtain on a realistic ore. This difference is produced due to loss of process control occasioned by laboratory operator error, mechanical problems or by variability in gold ore type.
It has been noted a quite separate group of problems that promote an optimistic difference in the laboratory procedures. These differences are accidentally included into the laboratory procedures by the technicians performed the metallurgical tests and could be overlooked by the project Metallurgist. Probably the most common and typical of these problems is that of employing a technique in the laboratory which cannot be repeated in full scale operation. For example, perform bottle roll tests of high grade gold ores (e.g. 15 g/t). These kind of gold ores are processed by agitated tanks and since the rheological properties of the slurry has an impact of the results; cyanidation test must be done using a mechanical agitator. Many times, metallurgists overlook the effect of agitation on particle suspension and dissolved oxygen. Other important is the related to the possibility of scaling laboratory conditions to the design of leaching tanks.
If we consider flotation tests, it is common to consider extremely short flotation times in conjunction with starvation quantities of collector that produces unusually high grade froth products. Considering the physical size of flotation cells, these laboratory results cannot be obtained in the full scale plant. Other common mistake is to bypass material around selected process stages to limit circulating load build up and to employ multiple reagents with very short half lives. This action improves process selectivity and obtains some good flotation results over few minutes, but reproducibility is not obtained if the reagents remain in contact with the slurry for longer periods.
Basically, these problems can be avoided by considering cyanidation, batch flotation and locked cycle tests are performed according to strictly regulated protocols. The metallurgical laboratory must be preceded by modal studies conducted on master composites to determine the required primary and regrind sizing parameters and provide useful information about the reagent scheme to obtain high gold recovery. In atypical metallurgical testing program about half of the costs are generated by batch tests which are performed through the labyrinth of process variables. This consideration wastes time, money and induces errors in the selection of treatment variables. If there is not order or a logical reasoning, results will be not representative and the geometallurgical projection will have errors at full scale.