I’d suggest that you get in touch with one of the manufacturers such as ThyssenKrupp Polysius, they have pilot plant facilities that can fully evaluate the HPGR if you send a sample to them. There is also a laboratory in Germany that conducts this work; ThyssenKrupp Polysius will supply the contact details as an independent facility. The HPGR can be very ore type dependant.

The cost question is difficult to answer and depends on circuit configuration and processing method. Generally speaking HPGR is a more efficient comminution method than SAG milling particularly if you factor in rising power costs in light of world governments looking to introduce carbon trading schemes as a method of raising taxes.

As mentioned, the cost depends on the circuit design. For hard ores, HPGRs are more cost effective but for softer ore SAG milling is more cost effective. The cross-over point depends on many factors so a pre-feasibility study should be done to determine the most economical path.

Crushing (jaw or cone) does not produce the same product size distribution as an HPGR so the pilot plant must include an HPGR. This will also provide scale-up parameters such as the specific throughput and specific force. Currently, these parameters cannot be obtained by any other means. The HPGR suppliers will have a pilot scale unit that they can use to measure these parameters. You do not need to set up your own pilot plant.

I absolutely agree with the points mentioned by both of you. HPGR is popular in many field, especially in comminution of metal ore with good production indexes. By the way, HPGR also can be used in pre-treatment of raw materials of pelletization and sintering. Therefore it is hard to estimate the cost, unless you do enough experiments to be clear about the characteristics of the porphyry copper ore. As mentioned that crushing (jaw or cone) does not produce the same product size distribution as an HPGR, so you can send some sample to HPGR suppliers, such as Koppern located in Germany and SC located in France, both of which I have visited last two months, to get the experiment parameters or data you need. You do not need to set up your own pilot plant unless you want to research in a long term.

The first issue is whether a HPGR is appropriate for crushing your ore. Then the flowsheet needs to be determined - whether the edge product is recycled and whether a sintered cake is produced (which needs crushing).

With regard to simulation and sizing, some test data is required e.g. the results from a SMC test work programme. An equipment vendor can size, specify, and provide a performance guarantee as well as a quotation based conducting a pilot plant test. This does cost money and a large sample is required.

A number of trade-off studies should be conducted based on a capex/opex comparison between the existing or proposed comminution option and the comminution circuit based around a HPGR. Unless the ore is hard, dry and the production rate is above 5 million tonnes a year, HPGRs may be not viable and the energy saving not realised. Finally there is the issue of maintenance. This needs to be considered.

In Peru there is a HPGR operation (Cu-Mo), as a unitary equipment is really very good, the problem are the belts and screens that are associate with HPGR. Also the HPGR has problems with throughput if you have a high clay ore associate with humidity. As reference you can get SAG conference 2011 proceedings there are very good papers about HPGR vs. SAG

In a term of particles liberation and approximate product granulometry you can simulate the HPGR crushing by a batch crushing in normal hydraulic press (with 4-5 cycles). Capacity and specific power consumption figures as well as lining life time data are possible only on a basis of laboratory scale HPGR.

The scale up between lab tests and full size HPGR unit is very poor. Static tests are simply not good enough for any useful energy information, and the lab size HPGR units give results that (to be honest) don't work. Only with the pilot plant do you start to get close to the real answer. Typically the manufacturers will add in safety factors "just to be sure". If the sample that goes through the pilot plant is not representative then this makes things a great deal worse of course. Best course of action is to approach the manufacturers. The one with the best back-up characterisation laboratory facilities are FL Smidth.

Have you ever performed the test you mentioned? I'm curious of your results.

Yes, it was done with 2 ore types as a preliminary test showing a potential grinding forces as well as better grains liberation.

Metso also offers a line of HPGR with pilot plant & testing capabilities.

This PDF could interest you:

HPGR Process/Overview

• HPGR vs. SAG
• The Manufacturers
• Mine Operations using HPGR
• Projects by Wardrop/Tetra Tech
• Economic Benefits
• Questions