Search Documents

By Donald Barton

THIS paper outlines a time-saving graphical method of calculation for application to the interpretation of observations with the Eötvös torsion balance. Preliminary to the treatment of this short cut method which has been found practicable, the principal formu-las by which the student of interpretation can calculate the gradient and differential curvature produced by different types of bodies in different types of situa-tions are presented and discussed. It is necessary to split most geologic bodies into a series of simpler ones for which not too complicated formulas are avail-able. The gradient and differential curvature effects are calculated for each of the simple bodies and then summed to the total effect for the whole body. The calculations are considerable in number and, though more tedious than difficult, should be' entrusted to an experienced calculator. Considerable facility in qualitative interpretation can be acquired by the geophysicist from experience and from a thorough study of the types of gradient and differential curvature anomalies produced by the dif-ferent types of possible structural situations. How-ever, semi-quantitative calculations' of the probable structure bring out many relations that are not evident in a simple inspection of torsion balance results ; it throws out as impossible or improbable certain of the alternative structural possibilities that may seem equally probable under simple inspection of the results; it will determine more accurately the position of the crest of the structure; and reduce considerably the un-certainty in the interpretation and it is necessary to intelligent interpretations. The calculation of such cross-sections consists of a trial-and-error calculation of a body that will give gradient and differential curvature effects as closely as possible the same as those observed. From his knowledge of the gradient and differential curvature produced by different types of bodies, the geophysicist sketches a tentative cross-section of the structure as indicated-to him-by the observed results of the torsion balance survey, and the calculator then splits the struc-ture into blocks, and calculates the gradient and differ-ential curvature block by block. If, as rather usual, the calculated .gradient and differential curvature pro-files diverge considerably from, the respective observed ones, it is necessary to add and subtract blocks from the cross-section until a close agreement is obtained .between the calculated and observed profiles. The com-plete calculations have to be gone through for each block that is added or subtracted.

Jan 9, 1928

By E. J. Magri

Consideration is given to the problem of the estimation, from limited data, of the likely grade and tonnage for a new mining property that is to be exploited by the selective mining of two metals that have substantial contributions to make to the total revenue. In particular, the case of a new gold and uranium mine in South Africa is analysed as follows. (I) The applicability of the underlying lognormal bivariate model for different support (i.e. ore unit) sizes is examined. (2) The necessary parameters for the bivariate lognormal models for different block sizes are estimated from abundant chipsampling data from a section of the Hartebeestfontein Gold Mine using alternative approaches, and the results are compared. (3) A method is given for obtaining the necessary parameters for a tonnage-grade relationship relative to a joint pay limit from the very limited information likely to be available at the end of the exploration stage of a gold mine, and the results are compared with those obtained from the large volume of data available from a minedout area.

Jan 1, 1982

By Henry Howland

INTRODUCTION DURING the last decade no topic has created more interest or received more thought among blast-furnace men than colic. One reason for this is, undoubtedly, the remarkable increase in the use of byproduct coke. Formerly our coke was made at a distance from the furnaces and by a distinct organization. There was little cooperation or development possible under these conditions. Today the situation is greatly changed. Much of our coke is made by the organization operating the furnaces and the coke oven is operated with the primary object of obtaining the best results in the blast furnace. The effect of changing the method of coke manufacture, even in some minor detail, is often felt almost immediately at the blast furnace. There naturally results one of the most valuable assets of the byproduct coke oven, -i.e., the possibility of increased blast-furnace efficiency clue to close observance and correct operation of the ovens. The results are evident, but it often is difficult correctly to account for them. In attacking the solution of a problem of as much importance as why one coke is better than another, there are facts which should be kept clearly in mind The author hopes, in this paper, to bring out some points relative to the use of carbon in the blast furnace which will be of value in solving this question.

Jan 3, 1916

By A. H. Lee

Discussion of the paper of H. P. HOWLAND, printed in Bulletin No. 111, March, 1916, pp. 627 to 650. A. H. LEE, Buffalo, N. Y. (communication to the Secretary*). The statements and results recorded in Mr. Howland's paper can be accounted for by the use of coke with a low sulphur content (0.60 and under). This allows the furnaces to be run on a small volume of acid slag, which it is presumed was done in Mr. Howland's practice. H. A. Brassert has admitted that this was the largest factor in low fuel consumption in his practice. Where the coke contains from 1.00 to 1.25 per cent. of sulphur the situation is quite different and I doubt whether the results mentioned by Mr. Howland could be obtained under such circumstances. ` N. M. LANGDON, Mancelona, Mich. (communication to the Secretary + ).-Gruner's ideal working (which is but a statement of the law first enunciated by Bell) as given in Gordon's Translation of Studies of Blast Furnace Phenomena, "supposes that the reduction of the oxide of iron is effected by the CO only, without intervention of solid carbon." This ideal, if reached, would mean theoretically perfect or 100 per cent. efficiency of reduction. The workings of the large number of furnaces as given in Bell's and Gruner's great works do not show that this standard of perfection ever was reached. This problem has been worked out by the writer for a number of furnaces, and his results, as well as those given by others, all show a failure to reach perfection. Some furnaces, however, approach nearer to perfection than others. The "ideal" as given by Gruner was not stated as a solved problem in blast-furnace practice, but as an ideal to the attainment of which furnace managers should strive. A study of the factors involved from every angle is interesting and instructive and the paper on this subject contributed by H. P. Howland is no exception to the rule. This article is fairly open to criticism, not because of the unique method of presenting the subject, but for the reason that, as the writer conceives, erroneous assumptions are taken which lead naturally to incorrect conclusions.

Jan 7, 1916

By Richards J. W.

J.W. RICHARDS, South Bethlehem, Pa. (communication to the Secretary?).-Mr. Howland's paper, data and conclusions have interested me intensely, and have led me to study carefully the record-of his 26 furnaces therein given. Mr. Howland is undoubtedly right in his conclusion that when expressing the amount of carbon burned at the tuyères in percentage of the total

Jan 1, 1917

Prof. Mathesius analyzes the running of Howland's furnace No. 19, according to the method described by him in Stahl and Eisen, 1916, Nos. 30 and, 31, and in his work "Die physikalischen und chemischen Grundlagen den Eisenhüttenwesens" (Spamer, Leipzig). By assuming 500 Cal. lost from the hearth by radiation and cooling water per kilogram of pig iron made, a throat-gas temperature of 600° C., and taking the ratio of C02/CO as 0.64, he finds that his diagram would show 0.68 kg. of carbon oxidized per kilogram of pig iron, while Howland gives 0.66 kg. He expresses himself as well satisfied with this test of the accuracy of his theory. Making the further assumption that the heat losses from the shaft are 1,000 Cal. per kilogram of pig iron (twice as

Jan 5, 1917

By W. H. Blauvelt

W. H. BLAUVELT, Syracuse, N. Y. (communication to the Secretary*).-I have read Mr. Howland's paper on Calculations with Reference to the Use of Carbon in Modern American Blast Furnaces with much interest, and believe that his method of analysis of the subject will be of material assistance in our study of the action of coke in the blast furnace. Mr. Howland quotes Professor Richards' statement that from the standpoint of the generation of the maximum quantity of heat in the furnace, Grüner was right in saying that all the carbon burned in the furnace should be first oxidized at the tuyères to CO and all reduction of oxides above the tuyères should be caused by CO, which thus becomes CO2. It is clear that the maximum quantity of heat is generated in the furnace when the greatest amount of carbon is burned to CO2 irrespective of whether it is burned at the tuyères or further up the furnace. So it would appear that Grüner's theory is sound regarding the maximum production of heat only because if all the carbon is burned to CO at the tuyères the reduction of the ores will have to be effected by the further oxidation to CO2. Whereas, if the reduction of the ores is effected in part by solid carbon, part of this carbon might be oxidized only to CO, resulting in a higher percentage of CO in the escaping gases than from Grüner's "ideal conditions.? But in studying the combustion of coke in the furnace, it is clear that the production of the maximum quantity of heat is not of the first importance in blast-furnace operation, or in the utilization of the fuel charged into the furnace. To my mind the production of a high thermal head at the tuyères is of the first importance, and the best coke is that which reaches the tuyères in proper condition to produce the highest temperature at the tuyères, and in just sufficient quantity to do the amount of work required there under the conditions produced by this maximum temperature. The combustion of a much larger amount of fuel at the tuyères, under conditions that will fall short of producing the highest possible temperature, cannot produce as good results, either in fuel economy or output. If this statement is correct, does it not correspond with Mr. Howland's statement that too much wind is inimical to low coke consumption? Nothing is more fatal to obtaining the highest temperatures than an excess of air for combustion. In the blast furnace an excess of air dilutes and cools the products of combus-

Jan 10, 1916

By William W. Edgerton

The California Department of Transportation (Caltrans), the Contra Costa Transportation Authority (CCTA), and the Alameda County Congestion Management Agency (Alameda CMA) are involved in a unique partnership on the planning and design of the fourth bore of the Caldecott Tunnel, located on State Route 24 in Alameda and Contra Costa counties. The project will add a two-lane bore north of the existing three bores. The project will alleviate traffic congestion in the non-peak direction. This paper provides background on the existing facility, summarizes the need for the new fourth bore, describes the proposed improvements, sets forth the approach for project delivery, and lists some of the challenges that are expected during construction.

Jan 1, 2008

By Rolando Campos

Until 1970, the Caletones Smelter of CODELCO-CHILE, División El Teniente, operated under the conventional pyrometallurgical process used by most of copper smelters. The annual copper production varied around 165.000 tons. The failure in the commissioning of the "oxygen smelting" technology in Peirce- Smith converters (1970-1972), forced further introduction of transcendental changes in the Caletones smelter operating scheme. Implemented technological innovations were based on wide studies and research works performed by División El Teniente. This paper presents a summary of technological modifications introduced, their impact on production and energetic performances, and short and medium terms improvement plans.

Jan 1, 1993

By Luis Suescun, Robin Dufour, Jorge Zafra

La minería a tajo abierto es una actividad económica relevante en varios países del mundo, incluyendo Perú. Sin embargo, esta actividad puede tener un impacto significativo en el medio ambiente y en la calidad del agua subterránea, especialmente si no se toman medidas adecuadas de manejo y monitoreo de la hidrogeología local. Por esta razón, los estudios hidrogeológicos son esenciales para comprender la fase operación de la minería a tajo abierto y los efectos que con la ejecución se pueden presentar en los recursos hídricos superficiales y subterráneos, y para diseñar estrategias de mitigación, optimización y operación de los proyectos mineros.

Sep 29, 2023

By C. Banff, E. C. F. Hamman, J. Venter

Open pit slope stability analyses for large scale instabilities are often carried out using limit equilibrium, finite element or finite difference computer programs. All of these require empirical estimates of rock mass shear strength when calibrated values are not available, which is a common occurrence. A common method of estimating rock mass shear strength is the Hoek-Brown failure criterion, which often over- or underestimates rock mass shear strength when used in limit equilibrium or elastic linearly plastic slope stability analyses, resulting in published alternative estimates for the coefficients of the Hoek-Brown failure criterion. This paper investigates the differences between limit equilibrium and elastic linearly plastic slope stability analysis based on the Hoek-Brown failure criterion. The investigation consisted of a comparison of 3840 RS2 finite element analysis models completed with strain softening, and 216 Slide2 limit equilibrium models. The models were used to evaluate a methodology to determine equivalent limit equilibrium and elastic linearly plastic shear strength parameters that provide the same stable slope angles and instability back break distances as the more representative strain softening stress analysis methods. This paper presents the methodology and compiles the results into design charts and equations for equivalent limit equilibrium Mohr-Coulomb shear strength parameters, that simplify calibration against actual case studies. The results compare well with older empirical shear strength estimates, however, a detailed calibration of the results against case studies will need to be carried out. The analysis also did not consider pore pressure effects; this is left for future publication.

Sep 1, 2025

Judicious placement of a range of geotechnical instruments was employed at the Tasmania gold mine, Beaconsfield, Tasmania, as a key strategy in the monitoring of rock mass conditions over the past four years. The range of instruments employed included vibrating wire stress-meters and hollow inclusion cells to monitor changing stress conditions, resistance wire and multipoint borehole extensometers to monitor the impact of stress changes on the integrity of the rock mass, and instrumented cable bolts and threadbar bolts to monitor the static and dynamic impact of rock mass changes on the installed ground control system. Initially the monitoring was of a qualitative nature, monitoring change from the time of installation. The value of the output was considerable when considered in relation to the ongoing seismic record which was also being monitored. It was clear from the seismic record and the monitoring results that stope blasting was the primary trigger for changing rock mass conditions. However, it was unclear how well observed behaviour was reflecting conditions anticipated by elastic numerical modelling. The latter had predicted significant stress increase in the near field of man-access footwall development. Qualitative monitoring and visual inspection did not support the modelling prediction, and this led to a program of quantitative measurement being undertaken. Measurement of the absolute stress condition in the backs of footwall development immediately adjacent to monitoring hollow inclusion cells provided the benchmark for tracking quantitative stress change. The program provided insight into the evolution of principal stress field magnitude and orientation in relation to rock mass change observed with extensometers and instrumented ground control elements, as well as providing a degree of calibration between the elastic numerical modelling and the virgin stress field. Calibrated monitoring data has also provided a useful benchmark for finite element modelling which is to be used for future mine-wide design.

Nov 25, 2010

By Thierry Lavoie, Mark K. Larson

"Engineers at the National Institute for Occupational Safety and Health (NIOSH) and Itasca Consulting Group, Inc., conducted a study to determine the procedure for calibrating a caving model implemented in FLAC3D. The model is calibrated to measured surface subsidence. This sedimentary caving model was developed by Itasca for NIOSH for the purpose of more closely simulating mechanical processes during caving. When a model is considered calibrated to surface subsidence, the load between the extracted panel's roof and floor is a reasonable estimate of gob loading. Such information can also provide an estimate of the amount of overpanel weight that is transferred to the abutment. This load distribution is a parameter needed, for example, to estimate loading of pillars in longwall gate roads. For the current case, a calibrated model resulted in a calculated abutment angle of 30.6°considerably higher than the assumed value used in empirical methods, but reasonable considering the presence of massive stratigraphic members in the overburden. Because of additional information available after this study that the extracted seam height was actually greater in the first panel, a corrected range of abutment angle was estimated to be from 27.7° to 30.6°, still significantly higher than the empirical assumption. The expectation is that better estimates of abutment loading profiles will result in improved evaluations of mine layout design and, therefore, reduce the number of fatalities and injuries resulting from falls of ground.INTRODUCTIONThe safety of miners in underground coal mines depends on careful consideration of the mechanisms involved in failure and transfer of load. The successful design of pillars and panels in longwall and room-and-pillar coal mining requires thorough attention to any information and measurements available that help understand these mechanisms. Although empirical methods, such as Analysis of Pillar Stability (ALPS) Mark (1987); (Mark, 1992) or Analysis of Retreat Mining Pillar Stability (ARMPS) (Mark and Chase, 1997), can provide a sense of success or failure, complex geology and failure in the roof or floor material or complex material behavior of the coal can make evaluation by an appropriate numerical modeling tool necessary. The numerical tool must simulate important mechanisms and appropriate detail that underlie the mechanism. For example, Starfield and Cundall (1988) advised using ""the simplest model that will allow the important mechanisms to occur, and could serve as a laboratory for the experiments you have in mind."" To that end, data gathering about the geology, geometry, and material behavior is very valuable. Such data facilitate selection of an appropriate model and model inputs. Moreover, this data facilitate improved assessment of risk to miners and how to minimize that risk."

Jan 1, 2016

By A L. Zawadski, B G. Putland

"This paper discusses a method to build a calibrated comminution circuit model, using historic operating data and blend information that is useful for forecasting and optimising plant performance, as well as evaluating expansion opportunities. A preferred method for building a calibrated power-based comminution model is discussed and justified based on two case studies, offering low-cost and effective strategies to utilise existing information for the maximum possible benefit.The first example is the Lefa Gold Mine located in Guinea, West Africa operating a semi-autogenous grinding (SAG) and ball milling circuit with recycle crushing (SABC). Daily blend information and operating data was utilised to build and calibrate a power-based model of the comminution circuit. The least squares regression method was used to solve the SAG mill specific energy requirement and the operating work index (Wio) for each individual ore type. The calibrated model was used to evaluate expansion strategies to sustain throughput at higher proportions of fresh ore based on the proposed life-of-mine schedule.The second example is the Ok Tedi Copper-Gold Mine in Papua New Guinea operating a SAG and ball milling circuit (SAB). The circuit limitation would often shift between the SAG mill and the ball mill due to variable ore characteristics and feed blend. Typical SAG mill relief strategies, such as recycle crushing or secondary crushing, required significant capital expenditure and the potential throughput benefit would vary depending on ore type, making economic justification difficult. Daily blend information and operating data was used to build and calibrate a power-based model of the comminution circuit to identify ore types that might benefit from higher blast intensities to increase fines content, thus relieving the SAG mill limitation and increasing overall throughput without a major capital investment.Citation:Zawadski, A L and Putland, B G, 2016. Calibrating Comminution Models Using Operating Data to Forecast and Optimise Plant Performance, in Proceedings 13th AusIMM Mill Operators’ Conference 2016, pp 17–26 (The Australasian Institute of Mining and Metallurgy: Melbourne)."

Oct 10, 2016

By Keith A. Heasley, Jian Yang

"LaModel uses a laminated overburden boundary-element model and can calculate not only seam-level stresses and displacements but also surface subsidence for thin tabular deposit such as coal seams (Heasley, 1998). Up to this point in time, the material property wizards in LaModel have primarily been designed for calculating accurate stress distributions in single- and multipleseam situations, and for investigating and optimizing pillar sizes and layouts in relation to overburden, abutment, and multipleseam stresses. However, the critical input parameters that will give the most accurate seam-level stress distribution do not necessarily produce the best surface subsidence prediction. The objective of this paper is to develop a methodology for calibrating the critical input parameters in LaModel to optimize surface subsidence prediction.For optimum surface subsidence prediction, it was found that the gob convergence (as defined by the final gob modulus) and the overburden stiffness (as defined by the lamination thickness) were the two most critical parameters that needed to be calibrated (Heasley, 2016). By calibrating numerous panels with various widths and depths, a formula that provides the optimum final gob modulus to use for subsidence prediction was developed. Also, three different empirical formulas for optimizing the lamination thickness as a function of overburden depth and/or the panel width to- depth ratio were determined for the three cases: supercritical panels, and subcritical panels with and without offsets. Further, if the user has measured data for subsidence factor and angle-of draw, the optimum final gob modulus and lamination thickness can be determined from the measured data. These new subsidence prediction formulas are being implemented into new material wizards in LaModel.INTRODUCTIONIn the past few decades, with the increased need for highly productive mining techniques, full extraction mining methods have been increasingly employed in the U.S. coal mining industry. These total extraction mining methods, which include both longwall mining and room-and-pillar retreat mining, normally cause immediate roof caving and the associated surface subsidence. These methods can be highly productive and cost-effective, but the associated surface subsidence can cause damage to surface structures and negatively impact the surface environment (Peng, 2008). Therefore, a reliable program for predicting the surface subsidence associated with full extraction mining is required for appropriately engineering and managing the surface subsidence impacts of full extraction mining."

Jan 1, 2016

By Andre C. Zingano

It is a very difficult task to estimate the immediate roof and support behavior in a room-and-pillar mining section in the initial phase of the mining project. There is little information available at this phase, and it is restricted to geomechanic characterization of the rock layers from cores samples, and the in-situ instrumentation of experimental entries and intersections. Even being a pilot excavation for roof support test a good indicator for the behavior of the immediate roof, its behavior for a full excavation mining section is different from a single intersection of two entries, mainly because the stress redistribution caused by the multiple entries excavation and face advance for a full excavation section. The objective of this paper was to calibrate a 3-D numerical model of an experimental intersection, based on the in-situ instrumentation of an experimental intersection excavation, and the geomechanic description of the rock layers. The numerical model was built with the same specifications as the excavation test, including the entry geometry and roof support. The numerical model calibration was based on wire extensometers that were installed at different positions in the roof of the pilot excavation, and on the geomechanic characterization of the rock layers that form the roof, coal seam, and floor. Therefore, it is possible to estimate whether the roof support specification adopted in the test phase will be good enough for the development of the entire mining section.

Jan 1, 2009

By Morgan Sears

The LaModel program (Heasley, 1998) is a displacement discontinuity, boundary element model (BEM) for analyzing stresses and displacements in single and multiple-seam coal mines. After the Crandall Canyon mine collapse, the need for better design guidelines for LaModel was observed and a new calibration method for deep cover pillar retreat was developed (Heasley et al., 2010). However, the scope of this initial calibration was limited to single-seam mines deeper than 750 ft. This limited application of the initial calibration method lead directly to the research presented in this paper, which attempts to define appropriate LaModel calibration guidelines for pillar design where shallow cover and/ or multiple-seam mining scenarios are encountered. As part of this research, a database of shallow cover case histories has been developed, and this database is used to help validate the shallow cover calibration method. By adding the shallow cover database to the previous deep cover database, the LaModel calibration method has now been extended to the complete range of mining depths, as well as to multiple-seam cases. With this shallow cover calibration, the utility of the LaModel program continues to increase.

Jan 1, 2013

By Keith A. Heasley

The LaModel program has been used successfully in the United States for many years to design pillars to ensure global stability of pillar recovery operations. However, the recent Crandall Canyon Mine collapse showed that further research was required to improve the design of pillar recovery under deep cover. Toward this end, a new calibration procedure was implemented in LaModel. This calibration procedure prescribes techniques for calibrating the critical LaModel input parameters based on empirical abutment loading and pillar strength, as utilized in the time-tested ALPS and ARMPS programs. The calibration procedure is intended to help guide a novice modeler in choosing calibrated input parameters that produce a standard model that fits within the verification database. However, the rigid parameter specification incorporated in the calibration procedure also removes most of the flexibility in LaModel for specific calibration to the unique conditions at a given mine site. In this paper, a number of the LaModel input parameters are varied outside the bounds of the suggested calibration and the effects on the model results are analyzed. Ultimately, it is shown how the LaModel user might vary certain input parameters to adjust the model to site-specific conditions, while still maintaining the general accuracy and philosophy of the recommended calibration.

Jan 1, 2012

By S. C. Oliphant, C. R. Houssiere, George H. Fancher

DRILLING mud can be an item of great expense on the Gulf Coast. In order to understand the nature of a drilling mud, it is desirable to know the composition and size of the mineral particles that are present, The supercentrifuge has been adapted to the fractionation of mud, and this paper discusses the calibration and methods of operation and control. An alignment chart has been prepared, which was found useful in this work. Typical data are included, showing the comparative reliability of the method.

Jan 1, 1942

By N L. Turner, D J. Robinson, A J. McFarlane, N M. Chapman, L L. Kuhar, R L. Meakin, R Martin

A number of geometallurgical tests have been developed within the AMIRA P843A Geometallurgical Mapping and Mine Modelling (GeMIII) project, an interdisciplinary collaborative research project involving four research groups and over 20 global mineral companies. Researchers at the Commonwealth Scientific and Industrial Research Organisation (CSIRO) have been focusing on the development of predictive hydrometallurgical indices for the GeMIII project via rapid, small-scale, cost-effective tests and protocols for comparative ranking of attributes relevant to leach performance. One of the focus areas has been the development of a leaching scheme for determination of the mineralogical deportment of copper in a large number of samples for geometallurgical applications.Various leaching schemes have been used over the last 20 years to identify the mineralogical deportment of copper. Typically, weak sulfuric acid and cyanide are used to target copper oxides and æsecondaryÆ copper sulfide minerals, covellite, chalcocite and bornite, respectively. Residual copper is assigned to chalcopyrite and/or bornite. Adaptations of this scheme have included the use of a weak organic acid to target oxides or ferric sulfate to target secondary copper sulfides. Testing of the weak acid/cyanide scheme in the GeMIII project using head and residue QEMSCAN mineralogical data analysed with a statistically reliable comparison method indicated that the inferred mineralogical deportment from various leaches could differ markedly from the theoretical or assumed value. Based on this initial work, test development and leach tests on samples from multiple sites has resulted in the development of a scheme to target copper oxides, chalcocite, covellite, bornite and chalcopyrite. This paper presents results from leach tests conducted on a large sample suite from two pits in a Cu-Au ore deposit with results calibrated using mineral liberation analyser (MLA) mineralogical analysis. This study has assisted in the development of a protocol for determining the mineralogical deportment of copper in a large number of samples for geometallurgical applications.CITATION:Kuhar, L L, McFarlane, A J, Chapman, N M, Meakin, R L, Martin, R, Turner, N L and Robinson, D J, 2013. Calibration and testing of a geometallurgical leaching protocol for determining copper mineralogical deportment, in Proceedings The Second AusIMM International Geometallurgy Conference (GeoMet) 2013 , pp 177-186 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Sep 29, 2013