Search Documents

By Ralph J. Holmes

INTRODUCTION Although sampling techniques used by the mineral industry continue to improve, sampling is still an area which is often neglected. Frequently, sampling and sample preparation requirements are left to personnel who do not fully appreciate the significance and importance of sampling. Consequently, plant evaluations are often based on seriously biased samples, leading to mass balance problems and incorrect assessments of plant performance. The basic rule for correct sampling is that each particle of ore or concentrate must have an equal probability of being collected and becoming part of the final sample for analysis. If this is not the case, bias is easily introduced and the sample is not representative. For example, when ore is travelling on a conveyor belt, the lumps may come to the surface. Consequently, a grab sample taken from the top layers only will contain a greater proportion of lumps, i.e., the sample is biased. There seems little point in making major investments in precise analytical equipment, if the sample provided to the analytical laboratory is not representative in the first place. This paper describes the sources of plant sampling errors, reviews the methods of estimating their variance, presents some worked examples, and provides some practical advice for reducing sampling errors and their variance. SAMPLING BASICS It is generally accepted that the discharge point of an ore or concentrate stream is the most suitable sampling location. The stream can be intersected at regular intervals, and representative samples, referred to as increments, can be obtained by taking a complete cross-section of the ore stream. On the other hand, sampling devices that take part of the stream on an intermittent or continuous basis, e.g., a bleed from a pipe, may introduce serious bias, and should be avoided at all cost. Increments may be taken at completely random times or tonnages during the sampling operation, in which case it is called "random sampling". However, as shown below, random sampling results in the largest sampling variance, so it is preferable to divide the ore or concentrate stream up into strata of equal time or mass and take one increment from each stratum. This operation is called stratified sampling and may be either systematic sampling, in which increments are taken at the same point in each stratum, or stratified random sampling in which the increments are taken at random within each stratum. Systematic sampling results in the smallest sampling variance, provided periodic variations in quality or quantity are not present which may coincide with, or approximate to, any multiples of the proposed sampling interval. In such cases, it is strongly recommended that stratified random sampling within fixed time or mass intervals be carried out.

Jan 1, 1991

A wide range of drill holes and process streams are sampled for resource estimation, grade control, and contractual purposes in the minerals industry. However, despite the availability of training courses, conferences and both national and international standards on correct sampling practices, it is still surprising how often little attention is given to ensuring that representative samples are collected for analysis. The reason for this is that the responsibility for sampling is often entrusted to personnel who do not appreciate the significance and importance of sampling, with cost being the main driving force rather than whether the sample is representative of the material from which it was extracted. This seriously undermines the precision and accuracy of the analyses subsequently generated and can render the analysis process a total waste of time and money and expose mining companies to serious, potential, financial losses. Company management needs to reverse this situation and ensure that sampling is given the attention it deserves to generate representative samples for analysis.

Jan 1, 2010

By Alvin H. Lense

In the spring of 1967, after a careful study of existing geology along the Pacific Coast including Alaska, the USBM, in conjunction with the USGS, decided that the offshore area around Nome, Alaska, presented an unparalleled land-to-sea comparison site for the evaluation of marine core drilling techniques and sampling equipment. Considering the geological evidence available and the production history of the Nome Beaches (5-6 million oz. of gold recovered), it appeared likely that gold in commercial quantities could occur in the offshore area. A 180 sq mi area of Norton Sound, extending roughly 14 miles to the east and west of Nome and averaging 6 miles to the south, was tested by approximately 530 line-miles of sub-bottom profiling. A total of 4043 ft (56 holes) were drilled in unconsolidated sediments, yielding mixtures of gravel, sand, silt and clay. Of the total holes drilled, 16 went to bedrock and six were terminated short of their intended depths due to boulders which halted penetration, or rough seas which caused the ship to move as the anchors slipped over hard sea-floor surface. The rest of the holes were terminated when the penetrative capabilities of the drills were reached, or when the required amount of geological data were attained.

Jan 8, 1968

By S. N. Dixon

Sampling and assaying procedures were developed to obtain unbiased and reasonably accurate Au grade estimates in drill samples fm a coarse gold deposit. Gold grade comparisons were made by assaying 3 (1 assay ton) splits from each drill interval, then processing the entire reject thru a pilot plant for a blocked out zone. The result comparing the assay splits to the pilot plant results gave a ratio of 1 to 1.07. Another technique was developed to detect drill samples which has a coarse gold content by monitoring the varience. Testwork could be done on a 1 kg sample to improve the accuracy of the grade estimation if the variance was high for a drill interval.

Jan 1, 1984

By S. J. Aresco, George E. Keller, J. Visman

INTRODUCTION The accurate sampling of coal, as with most minerals, is a difficult task. Coal is a very heterogeneous material made up of different types of coal and varying amounts of mineral matter. The product as mined may contain all of the layers of coal and impurities found in the seam as well as portions of the strata above and below the coal seam. The preparation plant is the device to size, crush and/or remove impurities so that the coal may be shipped as a saleable product. Sampling may be needed at many stages in the preparation process. Although this chapter is mainly concerned with sampling in connection with preparation processes, the principles will apply to all coal sampling. The main reasons for sampling in connection with coal preparation are: 1. To determine the washability characteristics of the raw coal by means of float and sink tests. These may be used for the design of a preparation plant, prediction of results to be obtained or for monitoring raw coal delivered to the plant. 2. To check the performance of various operating units in the plant, either by float and sink tests or chemical analyses. 3. To determine the analytical characteristics of the products produced, such as moisture, ash, Btu, sulfur, etc. HISTORY The earliest accepted methods of sampling coal in the United States were given in U. S. Bureau of Mines publications by J. A. Holmes,1 G. S. Pope2 and A. C. Fieldner.3 The American Society for Testing and Materials more or less adopted the methods proposed by the U.S. Bureau of Mines in their Standard D-21-1916, and it remained a standard for many years. Briefly, the standard called for a 500 pound sample for coals less than. 3/4 -in. in size and a 1,000 pound sample for all coals over ?4 -in. in size. No mention was made of increments but there was an instruction that the sample should spread over the coal to be sampled. Sample preparation was by hand crushing with the alternate shovel method for reduction to 250 pounds followed by coning and quartering below 250 pounds. Obviously, with the amount of work involved, this method of sample preparation was rarely followed. This method of preparation resulted in samples that were practically air dried because of long exposure while crushing, shoveling and quartering. In the late nineteen twenties and early nineteen thirties, studies for more practical methods were begun. Dr. E. S. Grummell4, Imperial Chemical Industries, Ltd, London, summarized the work to date at the

Jan 1, 1968

By Jan Visman, S. J. Aresco

INTRODUCTION The accurate sampling of coal, as with most minerals, is a difficult task. Coal is a heterogeneous material made up of different types of coal and varying amounts of mineral matter. The product as mined may contain all the layers of coal and impurities found in the seam as well as portions of the stratas above and below the coal seam. The preparation plant is the device to size, crush, and/or remove impurities so the coal may be shipped as a salable product. Sampling may be needed at many stages in the preparation process. Although this chapter is primarily concerned with sampling in connection with preparation processes, the principles apply to all coal sampling. The main reasons for sampling in connection with coal preparation are: 1) To determine the washability characteristics of the raw coal by means of float and sink tests. These may be used for the design of a preparation plant, prediction of results to be obtained, or monitoring raw coal delivered to the plant. 2) To check the performance of various operating units in the plant, either by float. and sink tests or chemical analyses. 3) To determine the analytical characteristics of the products produced, such as moisture, ash, Btu, sulfur, etc. HISTORY The earliest accepted methods of sampling coal in the United States were given in US Bureau of Mines (USBM) publications by J. A. Holmes,l G. S. Pope,2 and A. C. Fieldner.3 The American Society for Testing and Materials (ASTM) more or less adopted the methods proposed by USBM in their Standard D-21-1916 and it remained a standard for many years. Briefly, this standard called for a 500 lb (227 kg) sample for coals less than 94 in. (19 mm) in size and a 1000 lb (454 kg) sample for all coals over 94 in. (19 mm) in size. No mention was made of increments, but there was an instruction that the sample should spread over the coal

Jan 1, 1979

By A. L., Bailey

All who are even generally aware of the tremendous rate of increase in coal washing operations must realize the growing importance of the float-and-sink test. I believe it is conservative to estimate that, in the past decade, the dollar volume of float-and-sink testing has increased tenfold. It is a simple matter of economy, then, to examine the factors that determine the cost of adequate float-and-sink testing. When the Coal Preparation Section of the Bureau of Mines entered upon a greatly expanded program of such work in connection with the synthetic liquid fuels investigations, it seemed advisable to examine these factors experimentally. The principal consideration that differentiates float-and-sink test sampling from general purpose sampling, is that the original particle size must be preserved. Therefore, the total cost of the test will be directly affected by any standard that might be proposed to limit sample bulk reduction at any given particle size. For this reason, the relationship of sample size to variability of results was the first factor to be studied experimentally.

Jan 1, 1949

Automatic sampling and weighing of copper sulphide concentrate produced on Bougainville Island (PNG) is described. Contained metal weights gave satisfactory agreement with buyer's results after the introduction of polyethylene lined sealed aluminium foil bags for transport and storage of assay samples. Paper bags used previously for this purpose permitted oxidation of the sulphide concentrate and thus caused lower assay values.

Jan 1, 1976

By Pierre-François Bareel

EU legislation imposes ambitious recycling targets for various consumer goods. End-of-Life Vehicles (ELVs), scrap iron and various proportions of Waste Electrical and Electronic Equipment (WEEE) are processed by shredding plants which recover most of the metals but produce Shredder Residues (SR) which are landfilled. They represent 20 to 25 wt.% of the shredder input. Energy issues for the coarse fraction of SR are now largely studied. However, below 2 mm, the calorific value falls off drastically. These Fine Shredder Residues (FSR) represent up to 55 wt.% of the total SR. To meet European objectives other alternatives to landfilling must be considered for the FSR. The present paper describes a study carried out to characterise the FSR and investigates ways to process them and produce marketable products. Applying Gy's theory, a sampling procedure was conducted over two years to statistically quantify the variability of the physical and chemical properties of FSR produced by a Belgian post-shredding treatment plant. Macroscopic characterisation has led to the identification of four material categories which could be separated according to their physical properties: ferrous metals (more or less oxidised), non-ferrous metals, an organic fraction (plastics, rubber, textiles, foams, wood) and a minerals fraction (glass, ceramic, concrete). The entrainment of small metallic particles within the pores of organic fragments will affect the quality of the separations.

Jan 1, 2006

By U. Meyer, A. Alonso, P. Banios

The sampling of cathodes has changed with the introduction of the ISA-process in copper refineries. A Jernkontoret joint working group of refineries and rod plants from both primary and recycled copper plants has been investigating whether the actual sampling of cathodes is representative for this type of cathodes. As a first stage, samples were sent to the members of the working group with the objective to find an easy basis for a comparison of the results of the next stages. All samples show the same tendency for nearly all elements. In some cases, there may exist differences in calibration and for some elements the detection limits are too high, especially when Optical Spectrometry is used. In the second stage, discs were compared with melted samples to answer the question whether the punching of discs is sufficient for the quality control of the cathode. The analyses of the discs represent fairly well the cathode quality. There are differences between cathodes from different crops. Three punches are sufficient to determine the cathode quality. Finally, samples from various lots of cathodes were compared with samples from the rod produced from these cathodes. The impurity content of the rod corresponds to the impurity content of the cathode.

Jan 1, 2000

Sampling is an integral part of gold bullion analysis. Gold mining industries sell millions of dollars worth of gold on the basis of analyses. Biased sampling can cause a serious loss in terms of monetary value to a mining company; hence accurate sampling is of great importance. The sampling procedures for gold bullion will be described. Emphasis is given to the most reliable method of obtaining a representative sample from a gold bar. The problems associated with sample preparation will also be discussed.The sampling procedure at the Kalgoorlie Metallurgical Laboratory will be evaluated in the light of comparative results from the Australian Gold Refineries (Perth, Kalgoorlie). The most common sampling practices adopted by different gold mining companies in the Goldfields and Australian Gold Refineries, Western Australia will also be presented.

Jan 1, 1976

While sampling techniques used by the mineral industry continue to improve, sampling is still an area which is often neglected. It is often left to personnel who do not fully appreciate the significance and importance of sampling. Yet sampling and sample processing are of vital importance. There is little point in making major capital investments in the latest analytical equipment and spending considerable time obtaining the best analytical precision, if the sample provided for analysis is seriously biased in the first place.

Jan 1, 1976

By C. H. J. Beaven, M. P. Jones

Since the various new techniques of automation and computing have enabled vast numbers of mineralogical observations of relatively small samples to be made cheaply and rapidly, the statistical assessment of these measurements has become of vital importance. Much of the theory and most of the procedures of sampling that are used in the mineral industry are based on Gaussian statistics, but modern measuring techniques show that many mineralogical parameters do not necessarily follow the Gaussian distribution function. Consequently, it has become necessary to examine the application of other, more suitable statistical methods in order to derive the maximum amount of useful information from the examination of mineralogical samples. Some of the errors that can be introduced by the use of incorrect distribution functions are discussed and the value of determining the function that is appropriate to the particular parameter that is being measured is shown. Examples are provided of non-Gaussian distribution functions in mineralogical materials that were encountered during prospecting and during mineral treatment operations: a Canadian uranium ore, a Thai tin ore, and gold ore tailings.

Dec 1, 1971

By K. I. Afewu

A modified approach to the implementation of Gy?s sampling theory is discussed and used to design a procedure for sampling crushed gold ore feed to metallurgical process plants. The approach involves calibration of certain ore-specific parameters. These parameters were calibrated for two different crushed gold ores en route to grinding mills in process plants. The model predictions were compared with the predictions from Gy?s original model. Far smaller masses of samples are required by the modified model than by Gy?s model at the same relative variance. The representativeness of the sampling procedure if normally distributed errors are assumed was also calculated. These procedures and evaluations lead to significant savings on cost of handling and preparation of the samples and to a quantifiable degree of confidence in the analytical results for the process plant.

Jan 1, 1998

Prior to 1975, most geochemical drainage surveys for gold were based on analyses for pathfinder elements, such as As and Cu, rather than direct analysis for gold, for two reasons. Firstly, analytical methods lacked the required sensitivity and precision to determine gold abundance at background levels in stream sediments (usually in the range 1 - 10 ppb). Direct analysis was also inhibited by concern about the potential for gross sampling error, brought about by the dispersion of gold as rare discrete particles of the extremely dense native metal. This sampling error, commonly termed the 'nugget effect' has two components: erratic distribution of the element within the stream environment, related to hydraulic factors, and the difficulty in taking representative samples, both in the field and later in sub-sampling for analysis. In more recent years the collection of very large samples was commonly advocated to overcome the sampling problem (Hams, 1982; Nichol, 1986).Two methods have been widely used to evaluate the gold content of bulk samples (Elliott and Towsey, 1989). The preparation and analysis of heavy mineral concentrates has in recent years been largely superseded by the leaching of gold contained in bulk samples by cyanide, commonly termed the BLEG (Bulk Leach Extractable Gold) or BCL (Bulk Cyanide Leach) methods. The terms BLEG and BCL are essentially interchangeable and the latter will be adopted for the remainder of this paper. The strengths and weaknesses of these two techniques were compared with conventional total gold analysis on fine fraction (

Jan 1, 1976

By Allchurch P. D

At Perseverance a reserve of 45 million tonnes at 2% Ni was indicated by drilling. Holes were drilled in trellis pattern on section lines 61m apart to assess shallow mineralization. To explore at depth holes were spaced 122m apart, collared vertically, then allowed to deflect under controlled drilling conditions. The mineralization may be divided into the three homogeneous classes of soarsely disseminated, heavily dissem- inated and massive sulphides. In the standardised sampling and assaying procedures used to assess the deposit, attention was always paid to these fundamental ore types. All sampling and assaying processes were continuously monitored by checks. Sampling and sample preparation errors were small, to judge from the differences between results of original and check samples. Assaying errors, determined from analyses made by independent laboratories using different methods, slightly exceeded sampling errors. Comparison of underground core, c'iannel, and grab sampling, suggests that core sampling tends to undervalue.

Jan 1, 1976

The discovery of economic uranium miner alisation in unique geological conditions at Yeelirrie, 700 km NNE of Perth, presented Western Mining Corporation Limited with a num- ber of sampling and evaluation problems. Carn- otite mineralisation occurred in a porous cal- crete-clay lithology in close proximity to the water table; background information on such a deposit was non-existent and sampling in these conditions could be difficult. The deposit was sampled initially by aug- er drilling with X-ray fluorescence analysis of drill cuttings, followed by reverse circulation and diamond drilling with analyses by X-ray fluorescence and gamma logging of the holes. Two costeans provided lithological and grade distribution data and mining experience in the deposit. Spoil from the costeans was stockpiled according to location in the first costean and according to location and grade in the second costean. Samples from the various stockpiles were analysed by X-ray fluorescence. No. 1 costean was extensively sampled by channel sampling of the walls and by pit sampling of the floor of the benches of the costean by back hoe. No. 2 costean was sampled by Geiger counters mounted in T-probes and by the gamma logging of auger holes to establish the valid- ity of the T-probe results. Samples were taken

Jan 1, 1976

By Francis F. Pitard

The presence of elements, minerals, or chemical compounds in materials at very low concentrations is becoming a constantly increasing concern for several segments of the Industry. Some trace elements or chemical compounds may be considered as hazardous, some are simply considered as undesirable In complex processes where they could be constantly recycled and end up as the major factor of decreasing quality for some valuable products. In some other cases, the presence of a trace element or mineral may be highly desirable because of its economical value: precious metals are falling into this category, they probably were the first trace elements to draw our attention. Tremendous quantity of money is spent for analysis at the trace level, also the analytical "tool" has been considerably improved during recent years, but still the sampling aspect of the problem remains obscur for many, and because of it, large quantity of analytical efforts become meaningless. Not only the absence of sampling strategies is badly affecting the representativeness of a large majority of samples, but furthermore for the same reasons, many laboratories are failing In their endeavor to prepare reliable standard reference materials. The reason of this failure has its roots in the unability to estimate effectively the sampling characteristics of the element, mineral, or compound of interest, not to mention the unability to perform precise and accurate complete analysis, so essential for the instrumental analyst (1). In this paper a strategy to perform a better sampling operation is recommended. SELECTION OF A PROBABILISTIC APPROACH AS SOON AS PRACTICALLY POSSIBLE It Is essential to differentiate probabilistic sampling from non-probabilistic sampling. As Illustrated in Figure 1, the sampling operation Is a selection process which could be performed In two different ways: I. A probabilistic selection process according to which all elements of a lot of material to be sampled are submitted to the selection with a same probability of being selected. 2. A non-probabilistic selection process according to which all elements of a lot of material to be sampled are submitted to the selection with different probabilities of being selected.

Jan 1, 1986

By Baafi EY, Appiah HK

The Prestea Goldfields Ltd of Ghana, West Africa has produced over 170 t of gold with average grade of 11 g/t since 1912. The deposits occur in a wide shear zone which closely follow the Birimian-Tarkwaian contact. Both the Birimian and Tarkwaian systems are of Precambrian age. Underground exploration is carried out to delineate ore bodies by means of AX size diamond drilling at 50 m sampling intervals. After the orebody has been delineated, development work is carried out by exposing the gold reef. Channel sampling is then carried out at about 3 m intervals in both the stope and stope drives. A channel sample length is commonly 0.6 in, but may vary depending on physical characteristics of the reef. The values of channel samples are linearly combined to obtain the average reef or slope values and/or widths of the exposed sections. The sampling practice and methods of improving the efficacy of sampling at the mine are described.

Jan 1, 1976

The efficient and profit-maximising daily and long-term management of the Vera Nancy resource depends on decisions based upon assay data. This paper examines the range of sample sources, the methods of sample collection, assay techniques and Quality Assurance used by the mine. The orebody has a relatively low nugget effect for a gold mine (around 25 per cent) and this greatly assists in reducing sampling error and producing reliable assay data for production decision-making. Reconciliation between mine and back-calculated mill grades generally sees the mine under-calling mill grades by around four per cent.

Jan 1, 2000