The project, that by 2009 will employ a total of 2,500 people, will produce gold at a cash cost of R59,310 per kg, or US$256/oz, (at a rand dollar ratio of 7.2:1) and the recovered grade will be 7.3 g/t. Phakisa will together with the existing Tshepong operation and its sub 66 decline shaft project form a single operational entity, and supply the Free Gold 1 processing plant. The total reserve of this operation is 43.2 million tonnes at 7.2 g/t for 10 million ounces of contained gold, and the resource is 166.5 million tonnes at 9.11 g/t for 48.7 contained ounces of gold. The overall Tshepong operation thus aims for an annual production of 668,600 ounces of gold, with 245,000 tonnes milled a month, a recovered grade of just over 7 g/t and a production cost of US$301/oz.

Harmony new projects director, Bob Atkinson says that the result of the Phakisa and sub 66 decline projects will be a world class gold mine with a 20 year life of mine ahead of it. “Tshepong was always a consistent performer historically even taking into account the change of ownership that took place (the operation was owned by AngloGold prior to Harmony).”

The projects entail low geological and mining risk as the reef that will be mined is an extension of that being currently mined at Tshepong. Issues such as mine call factor, sedementology and hanging wall conditions are already well understood. Efficiencies will be better than those currently being achieved at Tshepong due to significantly shorter travelling and tramming distances. The development and ledging schedules are such as to ensure that face availability does not become a bottleneck.

Having Phakisa and Tshepong as a combined operation has a number of advantages including the draining of water via 55 level from Phakisa to another shaft, the Nyala shaft which is not an operational shaft, and the ventilation of the Tshepong east block from the Phakisa shaft. Backup power can be supplied from Tshepong to Phakisa and the sharing of infrastructure such as the control room, equipment, management and skilled people through a general economies of scale increase will be positive for the combined operation. In addition flexibility will increase with intershaft rock transfer possible via the 66, 69 and 71 levels, and this can improve reserve optimisation through the exchange of ground between the Phakisa and Tshepong shafts.

Phakisa 2

The Phakisa shaft

There is also the possibility of future expansions by extending the decline from the Tshepong to beyond the 77 level, something that would cost about R75 million (US$10 million) to do, and of sinking a decline shaft from Phakisa to the 85 level.

Mining at Phakisa will mainly be carried out on the Basal reef horizon. The Basal reef zone consists of a siliceous quartzite with a thin basal conglomerate, and this zone is overlain by shale. The Tshepong sub 66 decline project will mine the same reef. In addition to the Phakisa project area there is a block of some 600,000 m2, known as the Gift block, that does not make up the current Phakisa reserve. This is part of the old Elands Shaft’ reserve, that was below infrastructure but can now be mined from Phakisa. This is a high grade block and will add to the life of the combined unit.

Phakisa’s Koepe winder, with a braking current of 9,000 Amps, will be fully commissioned at the beginning of August 2006 and the mine’s 10 MW air refrigeration plant will be completed by the middle of November 2006. Surface infrastructure at the mine will be completed by the end of September 2006.

Phakisa, where the cost per tonne of ore, will be R425 (US$59) in 2005/6 terms, will use technology not seen before in South Africa’s gold mines. This cost per tonne figure is based on the R440/t (US$61/t) achieved at Tshepong but because the newer mine will be mining its orebody closer to the shaft as well as the economies of scale achieved by combining the shafts the cost will be lower. Phakisa will also ensure that wet bulk temperature at the rock face is 26o C in order to ensure maximum productivity. Harmony’s capital expenditure on the new mine will come to R644 million (US$89 million), of which R348 million (US$48 million) was already spent by April 2006.

Phakisa 3

The Tshpong shaft.

The Phakisa project was originally initiated in 1993 by AngloGold, as the Freddies No 4 shaft project, and sinking commenced during February 1994, but the project was suspended in 1995 when Freegold’s profitability could not support it. AngloGold continued with the project in 1996 after restructuring its Freegold assets, but it was again suspended in May 1999. The shaft was then mothballed and equipment was moved to other AngloGold projects. At the time the total project cost in 2003 terms was over R1 billion (US$0.1 billion), of which R445 million (US$62 million) has already been spent.

In 2003 Harmony, which bought the Freegold for R1 (US$0.14), re-engineered the project and approved the R550 million (US$76 million) required to complete it, and this included a larger scope of work to exploit an ore reserve containing 132 tonnes of gold compared with the original 94 tonnes. In 2004 Harmony stopped the sinking of the shaft at 77 level, but continued with a twin decline to below 85 level, which is 2,800 metres below collar, and opened the opportunity to expose further ore reserves below 79 level.

The uppermost working level at Phakisa will be the 66 level, with 77 level to be the main service level. One of the main new technologies being used by Phakisa is a combination between a railway and conveyor system for haulage over the 5 km to the Nyala shaft. This system, the world’s first underground rail-veyor system, is located on the 55 level and was due for completion towards the end of July 2006.

The haulage on 55 level was not suited to conventional locomotive or conveyor systems, because the 3.7 by 3 metre dimensions were insufficient for a high speed train. A conveyor system would not be able to negotiate curves as does a rail system, and in the haulage would migrate from left to right and as a result restrict access to the conveyors without more development work being required.

The rail-veyor was first used in the US at a phosphate mine, because of this system’s low risk of fire. It was put on trial there, but the potash market collapsed shortly after that and the system was never used. It did, however, win an innovation of the year award, and in addition to its implementation by Harmony at Phakisa, Gold Fields and Richards Bay Minerals are also considering this technology. This system is expected to see wider use now that patent aspects have been finalised.

Phakisa 4

Work underground on the Phakisa

The rail-veyor has a number of advantages such as lower tramming risk, and in general combines the best of conveyor and rail systems. The rail-veyor is also an automated system, which can tip, load and move faster than a conventional rail and conveyor system. A conveyor runs at a speed of 3 m/s, while a railway runs at 4 m/s. In comparison the rail-veyor loads at 3 m/s, runs at up to 12 m/s and tips at 6 m/s. When Phakisa considered the three options, the rail-veyor also proved to be the cheapest at R21 million (US$3 million) compared with around R31 million (US$4 million) each for a conveyor and rail system. The installation time for the rail-veyor was projected to be 12 months and has turned out to be half that, in comparison with a typical installation time for the conveyor being 18 months and the rail installation likely to have taken 25 months.

The rail-veyor also comes out on top in terms of availability of 98% compared with the 85% availability of a conveyor system and the 70% of a rail system. Unlike a conveyor, which would be a single unit that is powered the whole time, the rail-veyor will have three segments, each of which will only run when it drives a train – there will be three trains. One of the reasons availability will be high is that if one of the three trains encounters a problem it can go to the workshop, while the others continue, whereas, for example, a conveyor problem means that whole system has to shut down.

The rail-veyor drive units will talk to each to increase the efficiency of the system. Overall the operating cost of the rail-veyor will be R2.51/tonne (US$0.35/t), compared with the projected R3.58/t (US$0.50/t) for the conveyor system and R5.68/t (US$0.79/t) for a rail system.

The rail-veyor trains use foam filled truck tyres, these being standard tyres. Standard motors are used. Each hopper, made of ordinary steel, has two wheels and a rubber layer lines the inside of the hoppers, the join of the seals being one of the secrets to the success of such a system. Like a roller coaster the rail-veyor can run upside down, which it does on its return leg, thus ensuring that the haulage is doubled. One hopper weights 500 kg without load and one tonne with load, and each hopper is 1.3 metres in length. With the rail-veyor one only needs rails weighing 15 kg/metre and a one tonne locomotive will be used to do maintenance.

One of the other advantages of the rail-veyor is that it can be implemented in a modular manner with the first train to be in place next year and the second and third trains, which will be 120o apart, to be adder after. Debar is the company undertaking the rail-veyor project.

Phakisa 5

Phakisa winder motor – with a 9,000
Amps breaking current

Another new mine technology Phakisa is using is electric rock drills in the stopes. This is part of the mine’s aim to eliminate the use of compressed air systems. With compressed air systems only 5% of the power required compress the air is used productively, compared with 45% for electric drills. In addition with no one wanting to be a rock drill operator, to the extent that a rock drill operator is now paid at the same level as a team leader to incentivise people to do that job, the target is to look at improved ways of undertaking rock drilling.

The mine is also looking to maximise the positioning efficiency of its fridge plants by ensuring they are closer to the working faces and that the appropriate parts of the mine receive the cooled air. The mine is also investigating the Fogger system for dust allaying and this technology makes the use of filters and fans redundant. Phakisa is also looking at the new cordless LED cap lamp. This technology replaces the use of the 1.2 kg power pack required for the traditional lamps, and which require maintenance worth R15 (US$2)/lamp/month. Taking into account the use of 5,000 lamps at the mine, this comes to R75,000 (US$10,400) a month, and in comparison the new technology which comes from China, though it costs twice as much, recovers this cost over two to three years in maintenance cost savings. This is in addition to the greater convenience and comfort the new lamps afford. These have been tested for some eight months at Phakisa.

Phakisa also intends to implement a standard of 20 lux minimum lighting throughout the mine’s stoping and development areas. And the mine is looking at implementing a high speed mud settling system.

Apart from Phakisa, the other leg of the greater Tshepong mining complex being established is the sub 66 project, the aim of which is to hoist 48,500 tonnes of ore a month to produce 350 kg of gold a month at a cost of R433/t (US$60/t, in 2005/6 monetary terms) by July 2008. This project will replace the existing production at Tshepong and extend the life of the mine by eight more years. The project will be completed in February 2008. This project, which will have a production cost of R60,000/kg (US$8,300/kg), will have a recovered grade of 7.21 g/t and the first gold production is scheduled for August this year, some 50 kg from ledging. The capital cost for the project, which began in April 2003, is R280 million (US$39 million), far less than the R474 million (US$66 million) originally envisaged. One of the ways in which costs were saved is that Harmony undertook the project itself, instead of using a contractor. This project intersects gold at 1.5 km from the Tshepong shaft, and the twin decline, where sinking will be completed in December 2006, goes down from the 65 to the 72 level.

Then the mine will start with a two year build up of production with four raises used. The total length of the twin decline is 1,164 metres and will open up mining at levels 69 and 71 at Tshepong. The 12o declines comprise a material decline with a belt conveyor and monorail system and a chair-lift decline for the transportation of personnel. The 65 level will have surge capacity and will be the tipping level.

Another of Harmony’s projects, the Doornkop project completed its dual shaft sinking process in April this year, and with the main shaft due to be equipped and commissioned at the end of February 2007, this project will be in full production by November 2008. Full production entails just over 10 tonnes of gold produced a year from 1.562 million tonnes milled at a recovery grade of 6.48 g/t. The project’s capital cost is R880 million (US$122 million) and the cash cost at full capacity will be US$308/oz on average. Doornkop will employ 3,579 people, with many of these being new positions created in the industry.

Meanwhile Harmony’s Elandsrand deepening project will look to intersect the 109 level on reef in March next year. It intersected the 102 level on reef in May 2002, the 105 level on reef in May 2004, and anticipates intersecting the 113 level on reef in December 2007. Elandsrand, with an anticipated life of mine of 28 years will employ 4,474 people and produce 509,461 ounces of gold a year from 1.774 million tonnes milled a year at a recovered grade of 8.93/tonne. The capital cost for this project is R761 million (US$106 million) and at full capacity the cash costs will average out to US$241/oz.

Other projects being progressed by Harmony in South Africa include the Orkney 7 shaft project, where the reopening of this shaft that saw stoping start in June this year and will see development to the 36 level completed by June 2009. This project, which is a replacement project for current Orkney operations will employ 704 people and will allow Harmony to keep a footprint in the area and continue to employ the current number of persons at those operations. The mine will produce 68,000 ounces of gold a year from 472,000 tonnes milled annually at a grade of 4.48 g/t over a nine-year life of mine. This mine will produce at a cash cost of US$348/oz and the project has a capital cost of R89.9 million (US$12 million).

Harmony also has a project underway to recover gold from slimes dams in the Free State, with this to be treated by the modified Saaiplaas plant. That currently redundant plant will treat the tailings at a rate of 400,000 tonnes a month at an envisaged operating cost of R10.54/tonne (US$1.50/t). The project has secured quality used equipment from the former Ergo operation and the total system will be commissioned in November this year. The life of this project will be five years and will produce 895 kg of gold a year at a grade of 0.186 g/t. This project will have a capital cost of R35.2 million (US$5 million) and an average cash cost at full capacity of US$270/oz.MRA