One of the problems of many gold projects is to know how to include a gold gravity concentration circuit in a new plant or an existing plant. Basically, there are some basic procedures to evaluate a successful gold recovery circuit. In order of decreasing cost they are large capacity pilot plant; in-plant testing for retrofitting the existing circuit and laboratory testing program. Most the time, this has been also the order of the reliability of the results. The possible errors or mistakes in the testing program are usually natural gold traps in the circuit and the presence of coarse gold in sampling and assaying. It is important to mention that the presence of coarse gold particles is sometimes a result of the low concentration of the valuable minerals coupled with the small number of particles that make up this concentration thereby making it difficult to obtain a representative sample. Also, other typical problems detected in several gold projects are lack of reproducibility, over grinding and the temporary respond of the system.
Many metallurgists consider that the significance of these problems is variable and is influence by the characteristics of the material. Essentially, during in-plant testing, often a gravity device processes a portion of a stream that present circulating load. The benefits of this type of tests are that all the natural gold traps are full so very little new gold is trapped and a large enough mass is processed so the problems related to the presence of coarse gold is minimized. This procedure has the problem that a very small mass is treated when a los capacity machine is used and mechanical problems may make steady state virtually impossible. For example, it is common to select the hydrocyclone underflow for the testing program, and the circulating load of the material may be 250%, but the circulating load of gold may be 1500%.
The gravity unit initially has good performance, but these results may taper off as the gold circulating load changes. The gold that may be recovered in the circulating load decreases at a faster rate than that which recovered less efficiently. This respond makes difficult to predict the performance of a full scale gravity circuit. In some circumstances, the temporary respond condition is minimized or avoided in a pilot plant and there are problems related to the natural gold traps and reproducibility of results. It is important to mention that for the most part gold cleaned from these traps is considered gold gravity recoverable. Now, considering the fact that it is difficult to get a new lot of feed material, the possibility of repeating the pilot test is extremely low. A warning here is that it takes a large sample to build up a full circulating load. Laboratory testwork is not expensive and therefore several tests can be easily performed to assess reproducibility. It is important to consider the size reduction step because there is difference between plant comminution in terms of size distribution of both gold and gangue minerals. Most the time the material is over ground in the laboratory and the recovery may be under estimated.
The main problem in laboratory testing program is the presence of coarse gold particles (nugget effect). If coarse gold particles were to report to the concentrate and all the concentrate assayed, then the problem would not exist in the test. This consideration must be considered when the testing program must indicate the amenability of gold ore to gravity concentration and the possible performance of the circuit at full scale.