An effective method that we have used in the past is to have a digital camera on the feed conveyor belt and use an image processing package like SPLIT or WhipFrag to tell you the particle size distribution (PSD) on the belt.

Using this data as your baseline measurement you can change the feeders (or ratio of feeders) that you are using to feed the SAG Mill and optimise the feed PSD to find the sweet spot for the PSD or ore type and thus maximise the SAG throughput.

Once you understand what the sweet spot PSD looks like then you can better request this size distribution from the mine.

The Mine can also use this method of digital photos and image processing package of both of the truck load and stockpiles to feed forward to you the PSD being made by the mine, and you can use the feed forward and feedback loops to allow you to maximise the throughput of your plant.

How hard is your hard ore? Has the ore been tested and shown to be suitable for SAG milling? If the hard ore only comes to the plant from time to time the options are many options, including: accept a reduced throughput for a short time (minimum capital case); increase pebble crushing capacity and/or reduce crushed product size; return crushed pebbles to ball mill, bypassing the SAG mill. The extent of the problem will give direction to the best solution.

If you know that your ore will be consistently hard, perhaps consideration should be given to HPGR (High Pressure Grinding Rolls) followed by Ball Milling for final size reduction.

A SAG mill is, after all, just a means of crushing ore. Not all ores are 'SAG-worthy'. There are other ways of tackling the problem if the ore varies in hardness as suggested.

I agree with all of the above, subject to the detail of your circumstances.

If it is now consistently hard then crushing and/or HPGR may be the best long term solution.

If it is only a part of the pit that has irregular higher hardness then reset the crusher for a smaller particle and not overwork the SAG. Alternatively, as some of our customers do, keep a quantity of larger media to load for that period to overcome the short period of mill indigestion.

All good thoughts, from PSD knowledge based blending to using an HPGR, obviously all depending on the ore - and if not yet fully committed to a SAG.

Selective mining and blending of ores before the SAG circuit (with different harnesses / RQD's) may also be useful. Obviously, getting the ore from the specific mine face, through a primary crusher and to the appropriate mill feed reclaim port can be a large part of the issue.

Given the possibility of relatively large circulating pebble loads (˜20% to ˜25%), the use of ore-sorters ahead of a pebble crusher may allow substantial rejection of below-grade material, ultimately leading to substantially increased circuit capacity. Duel energy x-ray transmission working on sized and washed pebbles has been tried by the majors and can often precisely discriminate between ore and waste.

Pre-crushing the SAG feed (crusher before the SAG) will help with this issue.

If the hard ore is consistently more and established by initial research, the circuit need to be suitably designed for higher circulating loads in process circuits. If its intermittent, may be you may control by suitable blasting pattern and reducing the feed rate to the system.. The design circuits also may be balanced with second and third stg ball mill and HPGR.

What exactly do you mean as "hard-ore": high compressive strength, high abrasively, bad grindability, high spec. kW? Depending on the main difficulty corresponding different action have to be done.

Increasing mill speed will help for harder ores. Of course this means a higher energy consumption. Possibly batch the hard ore in to help keep down energy cost.

Adjusting blasting patterns will affect the crusher, Crushing affects milling, speeding up the mill is part of an option but can result in liner breakage if you overthrow the balls.

Mets can look at amount of grinding media and size. Liner and lifter pattern.

Operating team should look at crushing and screening options.

e.g.: No good crushing 1000 tph delivering o-size non quality material to a mill that grinds 150 tph.

Crush a quality 300 tph to give the mills every opportunity to maintain max tonnage with the required grind. Just an example, I've seen a few places crushing tonnage seems to outweigh crushing quality.

Mill densities as well. The more water added the quicker the ore passes through the mill.

You should be able to run high densities with hard ore types. The higher density also allows the worn and broken balls to discharge as they use valuable up energy.

To Improve SAG Mill Tonnage but at the risk of being laughed at, another option not used much is to treat the SAG as an AG mill. Use the hard ore as the main grinding media. A fully autogenous (AG) circuits can operate at much lower operating costs, but with limited throughput. The control of AG mills, fed by hard ores and soft components of oxide ore, can also be challenging, due to the sensitivity of the mill to the ratio of hard to soft components. As mentioned above by others, blending will be very important here as the mill could get overloaded with hard ore.

Spend some money at the front end with reducing the F80 by crusher, allow some of that hard competent rock to by-pass for SAG effect. Then reset the ball size and grade to balance the stream to match the ball mills. The very important lesson that was learnt at Copper Mountain in BC, Canada was not to expect too much to be done in the SAG. That company totally "nailed" the circuit with one brilliant crusher.