By Paul Staples, VP and Global Practice Lead, Minerals and Metals
In 2009, I was part of a group of established comminution practitioners that wrote a paper on the design pitfalls of SAG milling circuits treating competent ores. We were particularly concerned by the number of designs we were being asked to review that appeared, in our view, to be overly optimistic. The main conclusions of the paper were:
1.Comminution circuit design relies on input and interpretation of comminution test work data. Understanding of the pitfalls inherent with these tests is critical to ensure robust designs are achieved in the end.
2.Comminution circuit design for average competence ores is more straight forward than for very competent ores, many of which have been studied recently. The application of a methodology which works well for lower competence ores does not guarantee accurate predictions for all ores.
3.It has been observed that engineers and consultants have, at times, under estimated the SAG mill specific energy for treating competent ores due to lack of relevant data for comparable benchmark operations, lack of understanding of the test work methods used by various laboratories, and poor use of packages such as JKSimMet.
Now here in 2017 we can see that a reasonable number of recent SAG mill based projects have indeed under-performed and not achieved throughput targets under their design conditions.
As described in 2009, the recent problems originate either from unrepresentative sampling, poor testing protocols, inappropriate interpretation of data and definition of design criteria, wrong mill sizing methodologies, project management decisions or a combination of these.
The consequences of this are:
At Ausenco we take great pride in our track record and ability to model and design grinding circuits for most ore types. Ask us how we provide comminution solutions for competent ore at projects such as Cadia East, or click here.
A link to the 2009 paper, "What Can Go Wrong in Comminution Circuit Design", can be found here.