Search Documents

By Koppe J. C

This paper describes the methodology developed to establish a sequentia mining plan applied to a coal mine, using computers. Firstly, isocontou. maps were generated from the drill holes and outcrop description information, representing the thickness and depth of the rock formation: present in the mine site. The procedure relies on the repeated application of the cross validation techniques, until the standard deviation of thi estimation is reduced to a minimum.From this map the coal reserves of the two coal seams, the overburdet (separated in hard and soft material) and the interburden were calculated After that, the maps were transferred in digital files to AutoCAI software, where different mining plan simulations were performed. 7b ore evaluation in each case was done automatically, including tonnages volumes and stripping relations. The final mine configuration consider the economic aspects of the operation.

Jan 1, 1993

By A McEwan, J Masters, C A. Jackson, A Catanzano, V Sharp, M Millen, S Rainey, S J. Spencer, G Roberts

A passive acoustic emission (AE) monitoring system has been developed and trialled on two M10000 IsaMillsTM. The analyser detects AE stress waves generated by grinding media impacts with the mill liner during mill operation. Locally propagating stress waves can be detected on the exterior liner surface using strategically mounted broadband accelerometers.The analyser has been developed into a robust commercial prototype system, and has been designed, installed and commissioned with Xstrata Technology at two mineral processing plants.A six-month field trial showed large-scale variation in AE power signature with process conditions. Analysis of the enveloped AE power output demonstrated a strong correlation with mill process conditions – more specifically grinding media relative loading, feed pulp density, feed volumetric flow rate and feed type.Placement of accelerometers along the length of the mill body provided key information on the grinding media relative deportment. The enveloped AE power detected by the mill body, discharge pipe and bearing housing sensors all captured information strongly correlated with mill operating parameters and process state.The prototype analyser demonstrates suitability as a long-term online tool for stirred mill performance monitoring, and potentially for optimisation and control, in beneficiation plants.CITATION:Jackson, C A, Spencer, S J, Masters, J, Sharp, V, McEwan, A, Rainey, S, Catanzano, A, Roberts, G and Millen, M, 2014. Monitoring of M10000 IsaMill™ process performance by passive acoustic emissions, in Proceedings 12th AusIMM Mill Operators’ Conference 2014 , pp 241–254 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 1, 2014

By S McKnight, L Koroznikova, S Hall

Mineral sands represent an important new resource being developed in the Murray Basin, including parts of western Victoria. This paper will outline a simple methodology for mineral sands characterisation, developed as part of the AMIRA-managed project P777 æThe Development of Heavy Suspension Techniques for High Density Separations (Replacement of ClericiÆs Solution)Æ. This project is currently sponsored by three multinational mining companies (De Beers Consolidated Mines, Iluka Resources Limited and Rio Tinto Limited) and is developing an innovative laboratory mineral characterisation procedure that will allow the replacement of the currently employed highly-toxic chemicals.Mineral sand resources almost always contain more than one valuable (and relatively heavy) mineral. Titanium minerals are found with a large range of titanium contents, giving rise to density variation and often subjective mineralogical descriptions. Companies tend to rely on laboratory heavy liquid separation in the evaluation of samples arising from exploration, mining or metallurgical processes. Unfortunately, there are only a limited number of high density (æheavyÆ) liquids and these tend to be more toxic as their density increases. Low-toxicity inorganic solutions, based on tungsten compounds, have been developed that can be utilised at relative densities (RD) up to 3.0. However, beyond this value currently only organic liquids can be used. Diiodomethane (methylene iodide) having a relative density of 3.31 is commonly used; however, this presents significant health and safety hazards. Mixtures of thallium formate and thallium malonate were found in the early 1900s by Clerici to provide liquids having specific gravities between 4.0 and 5.0, hence æClericiÆs solutionÆ. For the characterisation of the heavy components of mineral sand deposits (eg anatase sg 3.9, rutile sg 4.2, ilmenite sg 4.4 û 4.7 and zircon sg 4.6 û 4.8) there is currently no heavy liquid alternative to ClericiÆs solution. ClericiÆs solution is highly toxic and testing is now conducted by very few laboratories worldwide with costs reflecting the chemical costs (though extensive efforts are made to recover and reuse the liquid, plus the requirement of its removal from the mineral samples), the infrastructure costs and health and safety regimes (eg blood testing of exposed staff, inventory management). A simple laboratory technique of density fractionation is being developed, employing suspensions of fine tungsten carbide particles in lithium heteropolytungstate (LST) solutions, that can replace ClericiÆs solution in the evaluation of fine mineral sands samples (eg -250+150 microns). The developing methodology that can achieve low-cost, low-toxic separations at relative densities above 5.0 will be outlined and the comparison of results with ClericiÆs solution presented.

Jan 1, 2006

By Stevak D, Plesingerova B, Ellias K, Tkacova K

This paper presents information on the processing of minerals by decrepitation that can be an integral part of their mineral dressing. In the case of carbonates, processing by decrepitation is a special operation that has its objective as concentration of the oxide. Selective fragmentation of magnesite brought about by decrepitation uses pseudomorphosis which was observed at thermal decomposition of carbonates. By decrepitation of magnesite high-grade MgO concentrates were obtained with MgO contents above 80 per cent. With the sideritic ores a very important phenomenon was observed consisting in the removal of quartz in the fragmented fine-grained product which favourably affect pelletisation of the fines. The selective concentration of the MgO product brought about by thermal decomposition of dolomite accompanied with decrepitation was negligible.

Jan 1, 1993

By Sewell L. G, White H. S

OWING to two large falls of ground which resulted in serious accidents occurring towards the end of 1908, it became evident that the "open chamber" system then in use was unsuitable for the conditions existing in the Mount Morgan mine. It will be admitted that in all mining operations the first consideration is the economy of the system adopted, combined with its safety and ability to give the desired output. A sectional system with square sets has been introduced into the mine, in which due attention has been paid to these factors. In this paper a brief description of this system will be given. An outline of those previously used will be included, as it is necessary to understand the conditions which existed in the mine at the initiation of the sectional system. Outside the actual mining methods employed, various points that might prove of interest will be touched on.The square set system of mining has been in use throughout, the stopes in the gold ore in the upper levels of the mine since underground operations were started in Mount Morgan. In the oxidized zone the ore was of a friable nature, and in some cases rImning ground was encountered. In the lower levels the ground became heavier, and although the stopes were for the most part long, wide, high, and irregular, and not sectionalized, no difficulty was experienced in winning the ore, and no accident of a serious...

Jan 1, 1911

THIS excellent volumetric method commends itself as a means of estimating the fineness of silver firstly, owing to the large number of possible impurities which do not affect its accuracy (these impurities include Cu-up to 70 % Au, Ph, Zn, Bi, Cd, Fe, Mn, Sb, and As); secondly, the method combines accuracy with speed. The above impurities are the only ones likely to be present in the B. H. A. S. Pty. Ltd. refined silver, and, with the exception of Cu and Au, the others, if present at all, would be well under 0.1 parts per 1000.Briefly stated, the method consists in the solution of the refined silver in HN03 , addition of ferric alum (indicator), and titration with an alkaline (either ammonium or potassium) thiocyanate solution until the brown ferric thiocyanate is formed, and gives a definite tint to the solution. The method is used at the Port Pirie smelters, and the outstanding feature in the method as there applied is the type of burette employed. This burette enables the operator to read 0.01 of a cc. with ease, whereas, with the ordinary burette fitted with a float, 0.05 of a cc., when a number of readings have to be taken, becomes very straining to the eyes. Moreover, waiting for the float to rise, and dislodging air bubbles from it, hinders rapid working.A description of the burette and fittings is appended with fig. 2. DETAILS OF THE METHODSolutions required are:a) Stock Solutions:(1) Ammonium thiocyanate, 40 grm. per litre of water; 30 litres are prepared and placed in a carboy with glass syphon attached.(2) Ferric alum, 250 grnl. per litre; two litres are prepared.

Jan 1, 1922

By K Hortin

Kidd Creek mine in north western Ontario is developing phase 1 of the deep mine project for mining between 2100 m and 3100 m below surface. Heat problems were experienced during the 2003 summer in working places about 2300 m below surface. At the time the mine had a æcold stopeÆ for cooling the air supplied to the mine below 1500 m from surface. The æcold stopeÆ is where during winter, ice is created in both the open pit and its connecting airways to the mine and used to cool air in summer. The operation and limitations of the existing cold stope were evaluated and led to the requirement for additional mechanical cooling to be available for the summer of 2004. The first stage of a surface refrigeration plant providing 7.5 MWR of cooling was approved, the compressor set was ordered in early January 2004 and the plant commissioned six months later in early July 2004. Based on the current modifications to the mining plan, further plant expansions are required in 2006 and 2009 leading to a total of 19 MWR of mechanical refrigeration and 8.5 MWR equivalent from the extended cold stope. The refrigeration system is described with emphasis on the specific adaptations necessary to suit Canadian conditions. Additional investigations were; using the refrigeration compressors in winter for heating and, extending the performance of either cold stopes or alternative ice storage systems. The winter use of refrigeration involve removing heat from the mine exhaust air to provide a cooling load and to use the hot water in the condensing circuit to heat the intake air and reduce reliance on mine air heating with natural gas. This paper follows on from an earlier one presented at the Seventh International Mine Ventilation Congress where the background and planning of the mine ventilation and cooling requirements were described for extending the mine at depth to 3100 m.

Jan 1, 2005

By I Sucholutsky, D M. First, D Mogilny, F Yusufali

Machine learning is creating value in all facets of the mining industry, from exploration to production. The authors provide an accessible, high-level introduction to artificial intelligence (AI), machine learning (ML), and deep learning (DL); the latter being widely recognised as one of the most powerful forms of ML. In particular, the authors will introduce deep learning models known as convolutional neural networks (CNNs), how they are applied, and the economic considerations necessary for determining when DL may be the right solution to de-risking complex block modelling problems. The authors present preliminary results from the mineral resource modelling study of the Jundee orogenic gold deposit, Yandal belt/gold province, Western Australia. The primary goal was to identify the direction and location of the narrow gold veins more accurately and demonstrate how non-linearly correlated elements are used as direct inputs to the resource model to assist with the target element’s grade prediction. This demonstrates that: (1) existing techniques for finding correlations between assayed elements do not adequately reflect the complex geology of the asset, (2) nonlinear correlations that are difficult to model as simple mathematical functions are representative of geological patterns in a deposit, and (3) non-linearly correlated assayed data, fed as inputs, increase the performance of the resource model as reconciled through blind tests. To conclude, the authors hypothesise that the patterns represented by the DL block models may be revealing the results of overprinting geological processes that generated mineral deposits. For example, the primary hydrothermal processes that deposited metals created depositional patterns that become particularly complex as a result of being overprinted, in part or in whole, by secondary physicochemical processes. This may explain why non-linear geochemical relationships have the capacity to generate more accurate block models of mineral deposits.

May 24, 2023

By S W. Kingman, D Whit, I S. Lowndes

The use of compressive crushing equipment such as gyratory crushers within minerals processing plants can potentially generate large quantities of dust. Remedies to this problem include the retrofitting of shrouds, enclosures, local exhaust ventilation (LEV) systems and water suppression systems. The single or combined application of these systems must be optimised to ensure they operate efficiently. It is desirable that the future design planning of such facilities include integrated dust suppression and/or removal systems to ensure material delivery rates are maintained and the welfare of the workforce is protected. The characteristic release of dust and the disturbance to the surrounding ventilation airflow is investigated by the construction and solution of computational fluid dynamic (CFD) models validated by data obtained from scale experiments. A 3D computational model was constructed using the FluentTM CFD software to represent the background ventilation regime within an underground crushing installation. The falling ore is represented as a continuum granular fluid falling under gravity from the raised bed of the dump truck. Transient path lines are computed to simulate the updraft airflow generated by the tipping process and indicate the dispersion path of any entrained dust. The relationships between the falling ore material, the displaced air and the shear induced flow are determined by an analysis of the computed volume/mass flow rates predicted across the crusher opening. An analysis of the airflow patterns predicted by the CFD simulation models indicates strong shear flows induced between the tipped material and the surrounding airflow in the vicinity of the crusher during tipping, and a strong influence of the background ventilation regime on the path taken by this shear flow. Scale experimental studies were conducted to characterise the induced shear flow and visualise the potential entrainment and transport of dust. An analysis of these experiments is presented.

Jan 1, 2005

By Bell I. F

In 1985, ZC Mines (now Pasminco South) had been unprofitable for some years due to a combination of high operating costs, low metal prices, low head grade and an aging and inadequate mining infrastructure. However, the resource of metal in the ground was still of world class, even after some 80 years of mining. A major study, which was to be of two years duration, was therefore commenced to establish a practical and profitable long term mining strategy and to cost the infrastructure options needed for its support and thus form the justification for a capital expenditure program to take the mine profitably into the 21st century and its second century of operation. This paper discusses the rationale behind the methodology adopted and the nature of the results achieved in one aspect of the study - the determination of the mineable resource and the relationship between the head grade/quantity of the economically mineable resource and the variables of cut-off grade, cost structure, annual production rate and metal prices/exchange rates. The basis of the study was the designing of some 5000 scopes to mine out the entire resource for a range of cut- off grades and the assembly of the detailed physical data thus obtained in a stope-by-stope data base for financial and other analysis. Those scopes which were shown to be economically mineable were then scheduled for two optimum annual production/cut-off grade/metal price combinations to provide life-of-mine cost and value of production cash flows.

Jan 1, 1992

By J S. Laskowski

Froth flotation process requires the use of frothers. These important flotation agents are commonly characterised as either æselectiveÆ or æpowerfulÆ and are chosen following general guidelines and verification by laboratory and/or pilot plant tests. Fundamental properties of the flotation frothers have been extensively studied over the last few years. These studies have led to the development of standardised procedures to characterise frothers in terms of their ability to reduce bubble size and to increase foam stability. In this research project, the performance of five frothers in flotation of coal is evaluated and related to the fundamental properties of these agents. Since the recovery of water in the concentrate is closely related to the non-selective transport of solid particles by entrainment, the tested frothers are also assessed in terms of their ability to promote the transport of water towards the froth collection zone, both in absence and in presence of solids.

Jan 1, 2005

Conventional treatment of oxide and oxidised sulphide minerals by flotation involves the use of sodium sulphide for sulphidisation. Sodium sulphide is generally applied in a "slug" dosing fashion with manual regulation. However, it is recognised that "over addition" or "under addition" of sulphidising reagent can result in poor flotation recoveries. Thus, control of sodium sulphide additions is an important aspect of a sulphidisation circuit, both commercially and metallurgically. In any plant where the flotation feed is "constant" it is possible to operate at a constant rate of addition of sodium sulphide. In actual plant practice, however, it is difficult to maintain a truly constant feed for any reasonable duration because this implies not only a constant feed-rate, but also constant mineral composition and particle size range. These factors can vary significantly and ideally require corresponding variations in the additions of sodium sulphide to maintain optimum recoveries. To cope with variations in the feed, it is essential to have automatic rather than manual regulation of the sulphide reagent addition. The CPS process achieves this control by utilising a sulphide ion-selective electrode as a control instrument in sulphidisation. The CPS process which uses a selective electrode, enables the sulphide ion concentration in the flotation pulp to be monitored and controlled continuously throughout the stage of sulphidisation.

Jan 1, 1989

By S B. Sheldon

Wall control programs tend to be expensive and consequently require financial justification in order to get approval for the expenditure required. The difficulty faced in getting this approval is that the outcomes of a wall control program are all subjective – a prediction of wall stability and what the wall would do with and without the program. Determining a net benefit is difficult and often justification of a wall control program is based on gut feel, history and a good selling campaign to build a perceived need.This paper outlines a simple solution to this problem by applying risk management principles to wall control. The process is to identify the risks to the operation associated with current wall conditions and then build a risk database for different potential scenarios for each risk. A financial impact is assigned to each risk scenario to complete the database. A consequence and likelihood analysis of each of the risks is undertaken and a formula applied to estimate the projected cost to the operation for the current risk profile. The process is repeated for the predicted wall control conditions after the application of the wall control program. A successful wall control program will reduce an operation’s risk profile so this projected cost should be less than that calculated for the current scenario. The difference in the two projected costs is the value that the wall control program has produced and this value can be used to determine the net financial benefit potential for the wall control program.This methodology can also be applied to other improvement programs that focus on risk reduction such as vibration control.

Nov 8, 2011

By A E. Annels, M D. Raine, I M. Platten

Lode-gold systems are often characterised by a high to extreme nugget effect (>50 per cent), and the presence of coarse gold particles (>100 ¦m in size). They rank amongst the most difficult of ore deposits types, in terms of producing an accurate and precise resource estimate. Their effective sampling is generally difficult because of the relatively low concentrations involved and erratic nature of the gold particles. Grade continuity is usually less consistent than gross geological continuity, and is typified by the localisation of grades within discrete oreshoots. In most cases drilling, particularly diamond drilling is an effective measure of global geological continuity, whereas grade can only be reliably resolved from underground development. Depending upon drill density and geology, it is unlikely that anything above an Inferred Mineral Resource can be estimated from surface drilling alone. Underground development, in-fill drilling and, potentially, bulk sampling or trial mining will be required to delineate Indicated and Measured Mineral Resources and subsequent Ore Reserves. Grade estimation will have to account for highly skewed and often multi-modal populations, extreme values, a high nugget effect and low geostatistical ranges. A clear geological model that honours both geological and grade continuity must drive the resource estimation process.

Jan 1, 2003

With the iron ore industry in Australia producing hundreds of millions of tonnes of ore per annum, the associated costs of these operations extend into the billions. It is no surprise in the current economic climate that more and more companies are looking into different ways of optimizing their costs to keep their sites operational. One way of improving operations is the use of electronic initiating systems like i-kon and Uni tronic detonators. These systems have been used around the world to improve current practices and optimize drill and blast costs. However, these systems have yet to make an impact upon the iron ore sector in Australia. Some of the potential improvements include:pattern expansion and a subsequent reduction in drilling and explosive costs;improved fragmentation and hence increased dig rates, thereby driving $/tonne down;embracing the 'mine-to-mill' concept by reducing the downstream costs of crushing and milling;improving the overall condition of pit walls and floors, leading to a safer working environment in which optimal haulage performance can be achieved; and improvements in vibration management around heritage sites and critical infrastructure, thereby saving the cost of relocation works. This paper introduces some of these concepts and the process of making the changes leading to potential savings from the improvements, using various examples as proof of the concepts.

Jan 1, 2009

The continuous casting of zinc alloys has been of fundamental importance for many years in the production of a number of products, including sheet for architectural applications, battery cans for dry cells and wire for metallising. The intrinsic benefits of continuous casting, including efficiency and high tonnage capability, uniformity of metallurgical structure and properties, and absence of voids and other defects, have contributed to the success of these products. Less well known, but of growing importance, is the more recent application of continuous casting technology to the production of zinc alloy bar stock products, including bearing stock, machining bar stock, forging stock and cathodic protection anodes. A wide range of zinc alloys are now continuously cast in a variety of cross sectional shapes and sizes to suit the diverse characteristics of these products. Of particular significance are the introduction of hollow zinc-aluminum bearing stock, and the availability of zinc alloy machining bar stock for hydraulic applications and for producing prototypes of zinc alloy pressure die castings. In addition to the traditional benefits, continuous casting provides special properties which are important in the application of these bar stock products. The list of successful applications of continuously cast zinc alloy bar stock is growing rapidly, and the market potential for resultant new zinc applications appears to be significant.

Jan 1, 1993

By J W. Lee, E Baafi, N Aziz

Ground subsidence due to mining is a common dilemma confronting the underground coal mining industry. The effects of underground longwall mining on river valleys have come under scrutiny, especially when mining is located underneath catchment areas. As public awareness of mining impacts on the environment increases, there is the need to develop damage mitigation strategies. Results of numerical modelling show that river valley response to underground longwall mining can be simulated and evaluated.

Jan 1, 2005

By A M. Mueller

St Helena tunnel is the centrepiece of the Pacific Highway upgrade from Tintenbar to Ewingsdale, abbreviated and commonly known as T2E. At the deepest point the twin tunnels run 45 m below the ridge line of St Helena Hill, located about 5 km west of Byron Bay. The tunnel breakthrough was achieved early in 2014 and the works for permanent lining will be completed in early 2015.Once the project will be handed over to the public in 2015, T2E will deliver a safer highway and uninterrupted traffic flow between New South Wales and Queensland. The design speed in the project area has been set to 110 km/h by the client Roads and Maritime Services (RMS), which will reduce the travel times between Brisbane and Ballina to just over two hours.The two tubes of the tunnel have been designed to accommodate three lanes in each direction with shoulders and walkways on either side of the carriageway. These requirements result in a huge cross-section of about 20 m width and an excavated area of about 210 m2.With the tunnel alignment being situated in basaltic rock, St Helena tunnel is conventionally driven. The selected excavation method was predominantly drill-and-blast with temporary support being installed immediately after the excavation. Despite the wide span of the tunnel, the geology allowed the tunnel perimeter to be supported by only rock bolts and a layer of shotcrete over the majority of its length.Due to environmental reasons one of the approval criteria of the project was that the tunnel must not have a detrimental long-term impact on the groundwater system within St Helena Hill. Consequently the tunnel has been designed as a fully tanked structure with a waterproofing sheet membrane installed on the temporary shotcrete.The permanent structure of the tunnel has to meet a 100-year design life and it is designed as a cast in situ concrete lining. Considering the water pressure on the lining the concrete is reinforced with a significant amount of bar reinforcement.The presented paper provides a more detailed explanation of the geology and the consequential temporary and permanent support measures together with the waterproofing system.CITATION:Mueller, A M, 2014. Pacific Highway upgrade from Tintenbar to Ewingsdale (T2E), St Helena Tunnel, in Proceedings 15th Australasian Tunnelling Conference 2014 , pp 419–424 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 17, 2014

By J Sargeant, T Levi, J Theiler, B Beamish

Two coal samples (810 Panel and Area 900) collected from Clarence Colliery have been tested under a controlled heating procedure to evaluate the gas evolution trends corresponding to increasing temperature. Gas evolution data are normally used to identify key gas indicators and validate the trigger action response plan settings for the mine as part of the Principal Hazard Management Plan for Spontaneous Combustion. The samples were located some 6 km apart and there is a significant coal-type difference between them. At low temperatures (<100°C), the Graham’s ratio (GR) values for the two samples are almost identical. However, at elevated temperatures, there is a significant difference, with the 810 Panel coal having higher GR values than the Area 900 coal. This is most likely due to a difference in the coal reactivity. The 810 Panel coal does not evolve ethylene until temperatures in excess of 140°C are reached, whereas Area 900 coal evolves ethylene once temperatures in excess of 120°C are reached. The 810 Panel coal also produces significant ethane evolution (200 ppm) at low temperatures (50°C) due to the presence of retained ethane seam gas, which decreases as the temperature increases due to desorption from the coal. Once the coal temperature exceeds 130°C, the ethane evolution increases again due to coal oxidation reactions, reaching levels of 50 ppm. In contrast, the Area 900 coal has no significant low-temperature ethane evolution, and ethane evolution only increases once the coal temperature exceeds 120°C. There are also noticeable differences in the trends of other gases evolved from the coal. These findings have implications for the interpretation of gas monitoring results from the two different areas of the mine. They also illustrate the need for obtaining results from each new area of a mine as it develops so that changing gas evolution trends can be identified in advance of mining.CITATION:Beamish, B, Sargeant, J, Levi, T and Theiler, J, 2015. Gas evolution trends for coal from two different areas at Clarence Colliery, in Proceedings The Australian Mine Ventilation Conference, pp 273–278 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Aug 31, 2015

By J Bolen, I Cameron, M Huerta, M Okrutny, K O’Leary

As greater amounts of fine iron ore concentrates enter the marketplace, new investments in iron ore pelletising capacity are inevitable to deliver these concentrates to steel industry consumers. When investors assess the two well-proven pelletising technologies, the straight grate and grate-kiln, the most appropriate choice may not be immediately evident. Certain technical, market and site-specific factors will favour one technology over the other; investors need to be aware and understand these factors in order to make the best investment decision. Hatch has compared the merits of both technologies regarding plant configuration; type of iron ore processed; energy consumption and flexibility to use a variety of fuels; ability to produce the required pellet quality; required plant capacity and scale; operating and capital costs and environmental performance. A technology assessment checklist that investors should consider when making a new pellet plant investment decision is presented.CITATION:Huerta, M, Bolen, J, Okrutny, M, Cameron, I and O’Leary, K, 2015. Guidelines for selecting pellet plant technology, in Proceedings Iron Ore 2015, pp 45–52 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Jul 13, 2015