Screening and Sieve Analysis

School of Engineering Department of Materials Science and Metallurgical Engineering NMP 310: Minerals Processing Practical 2 - Screening By Group 4 (Metallurgical & Mining) Nam J. (14268770) Labuschagne J. (14048460) Buthelezi N.N. (14199123) Durgean U. (12274489) Ndimande S.S. (14202604) Khoza I. (13246713) Date of Practical: 04 May 2016 Date of Submission: 10 May 2016 Abstract Screening is the simplest method for sizing particles in minerals processing. It is used to separate particles of a material according to size and shape. Screening usually follows after comminution. The aim of this experiment was to screen and sieve haematite ore and to study its particle size distribution. The screening was done using an industrial screen and the sieving was done using laboratory Tyler sieves. The initial mass of the ore manually fed to the screen was 11.965 kg. However, the total mass of the respective launders (particles <8mm, <10mm, <22mm, and >22mm) after screening was 11.64 kg. The losses are mainly due to blinding, type and condition of panels, and the effects of the 3.03 t/h feed rate. . The industrial screen had a wide particle size distribution (PSD). In addition to the feed rate, the phenomena of stratification and probability may be cause of the relatively low screen efficiency of 91%. Typically, the industry standard is a screen efficiency of at least 95%. As it turns out, the feed rate was too high and formed a thick material bed. It should be lowered in future screening so that the fine material is able to ’see’ the apertures. The under-size from the industrial screen was taken for Tyler sieving. It weighed 2.250 kg, which was more than the carrying capacity of the sieves. Cone and quartering was used to obtained a representative sample of the under-size - weighing 0.525 kg. The material was sieved for 10 minutes on a sieve shaker. The PSD indicated that 80% of the particles were smaller than 3.6 (log size). This implies that the major fraction of the sample was fines. This was supported by the Rosin-Rammler distribution plot. Keywords:[screening, Tyler sieving, particle size distribution, probability, stratification] Contents 1 Introduction 1.1 Aim of the Experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 2 Literature 2.1 Industrial Vibrating Screen . . . 2.2 Tyler Sieving . . . . . . . . . . 2.3 Particle Size Distribution (PSD) 2.4 Stratification & Probability . . 2.5 Dry Screening vs Wet Screening 2 2 3 3 4 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Experimental Procedure 5 4 Results & Discussion 7 5 Conclusions & Recommendations 12 5.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 5.2 Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 6 References 13 7 Declaration on Plagiarism 14 i List of Figures 2.1 2.2 2.3 Classifying vibrating screen (JVI, 2016). . . . . . . . . . . . . . . . . . . . . . . Movement pattern of a particle down a screen (China Suppliers, 2013). . . . . . Typical Tyler sieve stack (MarcTech, 2010). . . . . . . . . . . . . . . . . . . . . 2 3 3 3.1 3.2 The industrial screen used in this investigation. . . . . . . . . . . . . . . . . . . Laboratory (Tyler) sieves. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 6 4.1 4.2 4.3 4.4 4.5 4.6 Worn panels. . . . . . . . . . . . . . . . Particle size distribution after screening. Table of results from laboratory sieving. Particle size distribution after sieving. . . Gaudin-Schuhmann distribution. . . . . Rosin-Rammler distribution. . . . . . . . ii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 . 9 . 9 . 10 . 11 . 11 List of Tables 4.1 Industrial screening results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii 7 Chapter 1 Introduction In minerals processing, screening is considered the simplest and most convenient method of sizing particles. Simply, in this context, screening involves separating particles of an ore according to size and, often, geometry. This is the primary reason why screening is performed. The screening process typically follows after comminution processes such as primary crushing and secondary crushing. There are different types of screens available. Some are grizzlies, vibratory, and wire mesh screens. Simply, the process of screening ore material is such that the material is passed through a screen - or sometimes multiple screens - to classify the particles according to their respective sizes and shapes. Another use for screening is to determine the percentage of fines present in the ore and, as common practice, in the oversize. From these, the efficiency of the screens may be determined. Through screening, there is an observable particle size distribution (PSD) that is used in industry to select appropriate subsequent processes that the material may undergo. For instance, the finer particles may be sent for ’fines dense medium separation (DMS)’ and the larger particle may be sent for ’coarse DMS’ . Likewise, the finer particles may continue to a leaching process and the larger particles be sent back for re-crushing. The efficiency of the screening process is determined by a set of factors: the type of screen used (static or vibrating), type of panels used (steel or rubber), panel size and geometry (square panel or rectangular), feed rate (particle speed and throw), and the phenomena of probability and stratification. The factors are not limited to the above-mentioned. Another, is whether the screening process is dry or wet. Both have their merits, but the latter is preferred because it does not generate dust. In this investigation, 12 kg of haematite ore is sized using an industrial screen. Thereafter, a representative sample of the ore is sieved in the lab using Tyler mesh sieves. Information from both processes can be used by engineers to determine losses and process efficiencies with the intention of not only monitoring, but also optimising the screening process. 1.1 Aim of the Experiment The aim of the experiment is to size haematite ore using an industrial screen and Tyler sieves. In addition, it should be possible to determine the efficiency of the screening process and account for any losses experienced by both systems. Part of the objective is to understand the principles and operations of the two sizing methods and how they can be optimised for future processing. 1 Chapter 2 Literature 2.1 Industrial Vibrating Screen Vibrating screens are primarily used for sizing small rocks. They are very important in the minerals processing industry. Vibrating screens are typically used to classify particles of solidcontaining and crushed ore.They are suitable for both wet and dry feeds. The frequency of the screens vibrations determines separation factors such as particle speed (and throw) , retention time, and the extent of stratification. Vibrating screens generally offer better fines separation, higher purity, and product sizing control than static screens. Other common applications of vibrating screens include scalping, dewatering, desliming, and media recovery (China Suppliers, 2013). Figure 2.1: Classifying vibrating screen (JVI, 2016). In minerals processing, vibrating screen serve an important purpose. Post comminution, the ore is fed onto the screen and the screen classifies the particles of the material according to their respective size ranges. By so doing, material is selected to either enter the next stage of processing or be re-circulated to a crusher. Vibrating screens are able achieve to extensive classification of fine material. This reduces the required capacity of the comminution stage and overall energy consumption. Therefore, both recovery and screening efficiency are generally better than of static screens (China Suppliers, 2013). 2 NMP 310 - Practical Practical 2 - Screening Figure 2.2 illustrates the movement of a particle down a vibrating screen. When the feed is fed onto the screen, it forms a material bed when it moves across the screen panels. The advantage of vibrating screens is that they create thinner beds of particles, improving efficiency and capacity of the screen. Figure 2.2: Movement pattern of a particle down a screen (China Suppliers, 2013). 2.2 Tyler Sieving Sieving is one of the most popular separating techniques for size of particle. It is also called sifter. Before this process, a sample must be dried completely, and it is weighed for sieving. Tyler sieving is conducted as stacking screens in descending aperture size order from top to bottom. The bottom screen will have the smallest opening size. The bottom piece is a pan. These screens will be vibrated and the material will be filtered down through the sieves and oversize particles will be retained on screens. In contrast the size of apertures will increase from the bottom to top. After shaking, material retained on each sieve is weighed and a PSD plotted (911Metallurgist). Figure 2.3: Typical Tyler sieve stack (MarcTech, 2010). 2.3 Particle Size Distribution (PSD) Particle Size Distribution is used to analyse specific size of mineral. When minerals are broken by one of mineral processing methods, a continuous size distribution will be produced by particles. This distribution will show from the finest size of particles to the coarsest particles. There Page 3 of 14 NMP 310 - Practical Practical 2 - Screening are many methods were developed to quantify the size distribution of particle, and one of the most popular equation is Gates-Gaudin-Schuhmann equation. In this equation y-axis shows the fraction of the mineral which is smaller than size (cumulative %), and x-axis normally shows particle size of mineral (TechnologyInforme, 2016). 2.4 Stratification & Probability Stratification This describes a phenomenon when smaller and lighter particles pass in-between larger particles. When a screen is feed, the bed of material is usually layered - light and smaller particles on top and larger particles at the bottom. As the screen vibrates, particles are sorted according to their weight and size. Larger particles generally move more across the screen than downward. On the other hand, the fine particles tend to move more downward than across through the voids between the larger and coarser particles. Probability Probability, in the minerals processing context, refers to the degree of likelihood that a particle will ’see’ a panel opening and fall through it. When the bed of material vibrates across the screen, some particles fall right through the apertures, some bounce off the panels, and some do not even get to see the panel opening due to poor stratification. An excellent screening efficiency, recovery, and minimal losses have an associated corresponding high probability. Poor efficiency and major losses may be indicative of low probability during the screening process. 2.5 Dry Screening vs Wet Screening Dry screening refers to when a dry feed is fed onto a classifying screen. If this step is preceded by a wet process, the feed must first be dried by either baking or by sun. Drying by baking is preferred because it takes less time and is ultimately more economical. Wet screening is when the feed is fed onto the screen along with water or washed across the screen by jets of water. The risk with wet screening is that fine material may be lost as mud/paste on the screen surface or the fines may be suspended in water and carried over to the oversize launder. Wet screening is generally preferred because it does not generate dust (SSSDynamic, 2016). Page 4 of 14 Chapter 3 Experimental Procedure The investigation was divided into two parts, namely: screening and sieving. Both of these methods operate based on the same principle; to separate the feed according to size and shape. The following is a description of the experimental procedure that was carried out during the investigation : The haematite ore was first weighed and then sieved using an 8mm square panel. This was done to approximate the fraction of the ore that was fines i.e. particles <8mm and to be able to later calculated the efficiency of the screening process. The obviously over-size particles were removed by hand from the ore. Stratification could be observed during sieving. Hereafter, the industrial screen was prepared for use. Launders were cleared of any material and particles stuck on screens removed so that the results could be as accurate as possible. The ore was manually fed to the screen and the time it took to do so was record. This was used to determine the feed rate and could later be used to decide, based on the results, if the screen was being overfed. The material in the respective launders (<8mm, <10mm, <22mm, and >22mm) was weighed to construct a PSD and calculate the efficiency of the industrial screen. Figure 3.1: The industrial screen used in this investigation. 5 NMP 310 - Practical Practical 2 - Screening When feeding the screen, caution was taken to make sure that the feed’s initial contact with the panel bed was at the blank panel. This panel serves as a shock-absorber. If the feed were to fall directly onto the classifying panels, they would get damaged. The under-size from the industrial screen was sieved using an 8mm square panel sieve. There were some oversize particles, including flat particles. This was because the screen had rectangular apertures that allowed these flat-shaped particles to pass. However, these particles could not pass through the square apertures of the 8mm panel. The under-size particles(fines) from sieving were weighed and found to be above the carrying capacity of the Tyler sieves. Cone and quartering was used to obtain a relatively representative sample of the fines that was within the carrying capacity of the Tyler sieves so that they are not only overfed, but not damaged as well. Hereafter, nine sieves were stacked in ascending aperture size order from the bottom, as shown in figure 3.2. The bottom-most sieve (above pan) was 425 µm. The ratio √ between the sieves was maintained at 2. Figure 3.2: Laboratory (Tyler) sieves. The stack of sieves was placed on a sieve shaker and the shaker was run for 10 minutes. Hereafter, the masses retained by each of the sieves were weighed and the data was used to construct the second PSD. Page 6 of 14 Chapter 4 Results & Discussion Part A. A sample of haematite ore having a mass of 11.965 Kg contains approximately 2.555 kg worth of fines. This approximation is determined using an 8mm square sieve panel. Flat particles are not expected to be able to pass through this square panel. This approximation of fines is used to give an idea of the results to be expected after screening. That is, it will be used to determine the efficiency of the screening process. The entire mass of the ore is then fed onto the industrial screen. The ore is fed manually onto the screen at a feed rate of 3.03 t/h i.e. 11.965 kg in 14.2 seconds. The masses of the different sized particles are summarised in table 4.1 below: Table 4.1: Industrial screening results. Particle size (mm) Mass passing (kg) <8 <10 <22 >22 2.250 0.990 5.515 2.885 The total mass of the products from the industrial screen sum up to 11.64 kg. By mass balance principles, the mass that is fed onto the screen should be equal to the total mass that is collected after screening. However, this is not true for this screening process. There are losses experienced by the system. The percentage loss is calculated using (1) below: % losses = Where: Ms − Ms′ *100 Ms (1) Ms - mass of the sample ore Ms ’ - total mass after screening sample. Using equation 1, a loss of 2.72% occurs. This loss of material may be attributed to pegging of near-size particles due to their shape and the shapes and sizes of the panel apertures. Also, blinding, probability and throw of the particles. It is possible that the over-size particles were lost when they flew over their launder. The frequency of vibration of the screen may have had an effect on the throw of the particles and whether they had enough time to fall through the apertures. Also, the other launders may not have been well-positioned. One other possible 7 NMP 310 - Practical Practical 2 - Screening reason for losses that was noticed is that, some panels and the panel bed itself, were worn and damaged. Perhaps fine material was able to escape the panel bed from the bottom. This is shown in figure 4.1 below. Figure 4.1: Worn panels. The efficiency of the screen is given by: Efficiency = Where: 100(a − b) *100 a(100 − b) (2) a - % under-size in the feed. b - % under-size in the over-size. 2.555kg *100. Upon sieving the oversize using the 11.965kg 8mm square panel, it was found that 0.285Kg of fines had reported to the oversize. By using 0.285kg *100 . Using the values of a and b and the this mass, ’b’ is determined to be 2.45% i.e. 11.64kg efficiency of this screening process is determined using equation 2 to be 90.75%. The efficiency is acceptable for this investigation, considering there was no minimum required efficiency specified. However, in industry an efficiency of 95% is the general standard. To improve efficiency, the panel material of the screen may be changed, the feed rate adjusted accordingly, and if possible, all flat particles present in the sample should be removed. The feed rate of 3.03 t/h may be too high. A thick bed of the feed material means that the fine particles had a greater chance of ending up in the over-size launder as they would not ’see’ the panel apertures. The feed rate may be lowered to decrease the value of ’b’, thereby increasing the efficiency of the screen. ’a’ was determined to be 21.35% i.e. Below in figure 4.2 is a particle size distribution for this particular screening process. Page 8 of 14 NMP 310 - Practical Practical 2 - Screening Figure 4.2: Particle size distribution after screening. From the profile in figure 4.2, it can be concluded that the particles in the haematite ore received from the lab have a wide particle size distribution. In other words, ore material consists of large particles and fines. The particle size is not uniform. This profile may be indicative of the type of process used to crush the feed material. From the PSD, an educated guess may be made with regard to the breakage mechanism. It appears that the mechanism of breakage would be somewhere between impact and attrition/abrasion, based on the profile of the PSD. If necessary, the responsible engineer may adjust the crusher or select an alternative crusher to have a narrower PSD. Part B. This part of the practical was aimed at investigating the concepts of size separation, mass balancing, and construction of particle size distribution (PSD) curves from laboratory sieving. The lab sieves have a carrying capacity of 1000g (including the mass of sieves). The fine material recovered from the industrial screen weighing 2.250 kg was further sieved using an 8mm square panel. The resulting under-size material was still heavier than 1 kg and could not be fed onto the sieves. Cone and quartering was used to obtain a representative sample of the fine material. The mass of this sample was 525g. Figure 4.3: Table of results from laboratory sieving. Page 9 of 14 NMP 310 - Practical Practical 2 - Screening Table 2 above shows the data used for plotting figure 2, the PSD, Gaudin-Schuhmann distribution, Rosin-Rammler distribution curves. Something to be noted is that the total mass recorded on the sieves is not equal to the mass that was fed into Tyler sieves. The initial mass was 525g, whereas the sum of the total mass recovered from each screen summed up to 514.73g. As with all minerals processing processed, the system experienced losses. In this regard, 10.27g of the material was lost. Theory states that there exists certain characteristics of particles that may influence their behaviour during sieving. These include shape, moisture content, and size distribution of the feed. It may be said that the material was lost by human error i.e. inappropriate handling of the sieves. Some material was lost when the sieves were separated and some when the sieves were lifted and material fell through to the lab floor. Further material was lost when the material was being transferred from the sieves to the weighing container. Figure 4.4 below shows that 80% of the product (P80 ) is distributed below 3.6 (log screen size). Figure 4.4: Particle size distribution after sieving. Most of the product is fines less than 3.6 log size. If the operation requires material larger than 3.7, then 80% of this material would be discarded or not very useful. This would not be economical. On the other hand, if the operation is a sintering plant, then the PSD profile in figure 4.4 may be preferred. Below in figure 4.5, is a Gaudin-Schuhmann distribution that can be used to select a limited number of sieves to determine the size of particles between any two sieves. Perhaps it can be used to simplify the laboratory sieving process for future sieving. Instead of having to use tall stacks of sieves, fewer sieves may be strategically selected. For example, between 8000µm and 4750µm, the cumulative amount passing is about the same. This could imply that the two sieves may be replaced by a single sieve that is an average of the two - linear or geometric average - therefore, simplifying the sieving method/process. Page 10 of 14 NMP 310 - Practical Practical 2 - Screening Figure 4.5: Gaudin-Schuhmann distribution. Apart from that Gaudin-Schuhmann distribution, the Rosin-Rammler distribution profile, as in figure 4.6 below, serves another important purpose; it allows study of the fine particles. The plot is especially used for fines because the ”log of a log” blows up the ends of the profile, making it more convenient to study the distributions of the fines and the coarse particles. From figure 4.6, it is evident that a larger fraction of the sample was fines. Figure 4.6: Rosin-Rammler distribution. All this information from the Gaudin-Schuhmann and Rosin-Rammler could be used to optimise the sieving process or at most, the mining and comminution techniques. Page 11 of 14 Chapter 5 Conclusions & Recommendations 5.1 Conclusions Upon analysis of the experimental results it was found that there were material losses of 2.72% between the feed and product. This can be accounted for by blinding due to near-size particles being pegged in the apertures. The efficiency of the screen was found to be 90.75 %, which is lower than the typical industry standard of 95 %. The screen efficiency can be improved by replacing worn-out panels and adjusting the feed rate - such that throw is lowered. • The feed rate of 3.03 t/h may have been too high. The screen was over-fed and that explains the fines reporting to the over-size launder. Hence, a low screen efficiency. • The feed had a significant fraction of near-size and flat particles, which are reasons for material losses and the lower screen efficiency. • A major fraction of the sieving sample consisted of fines. 5.2 Recommendations • Worn-out panels should be replaced to minimise material losses and to improve screen efficiency. • A longer sieving time may have been appropriate. This is because when the sieves were handled, some particles still fell through the mesh. • Use sieve meshes with rectangular apertures. • The feed rate should be lowered to improve screening. • The ore should not have been fed manually onto the industrial screen. • Perhaps wet screening should have been done. The industrial screen generated dust. 12 Chapter 6 References 1. China Suppliers, High efficiency,(2013),Mineral Processing High Frequency Vibrating Screen for Fine Iron Ore. URL:http : //hcmining.en.made−in−china.com/product/aM GQV dY obCR b/China − High − Ef f iciency − M ineral − P rocessing − High − F requency − V ibrating − Screen − f or − F ine − Iron − Ore.html. (Accessed 6 May 2016) 2. JVI, 2016, Classifying Vibrating Screen. URL:http : //www.jvivibratoryequipment.com/pro ducts/vibrating − screens/classif ying − screens. (Accessed 6 May 2016) 3. 911Metallurgist, 2016. Sieve Analysis Explained. URL: https : //www.911metallurgist.com/ blog/sieve − analysis − explained. (Accessed 6 May 2016) 4. MarcTech, 2010. Tyler Sieve Shakers. URL: http : //www.marctech.com.au/laboratory − products − solutions/particle − sizing − systems/haver − boecker − sieve − shakers − and − sieves − 2/. (Accessed 6 May 2016) 5. TechInforme, 2016. Liberation Size. URL: T echnology.inf omine.com/enviromine/ard/mine ralogy/size%20&20liberation.htm. (Accessed 7 May 2016) 6. SSSDynamic, 2016, Screening Theory. URL: http : //www.sssdynamics.com/docs/def ault− source/def ault − library/screeningtheory.pdf . (Accessed 7 May 2016) 13 Chapter 7 Declaration on Plagiarism Declaration on plagiarism UNIVERSITY OF PRETORIA Faculty of Engineering, the Built Environment and Information Technology Department of Materials Science and Metallurgical Engineering The University places great emphasis upon integrity and ethical conduct in the preparation of all written work submitted for academic evaluation. While academic staff teach you about systems of referring and how to avoid plagiarism, you too have a responsibility in this regard. If you are at any stage uncertain as to what is required, you should speak to your lecturer before any written work is submitted. You are guilty of plagiarism if you copy something from a book, article or website without acknowledging the source and pass it off as your own. In effect you are stealing something that belongs to someone else. This is not only the case when you copy work word-by-word (verbatim), but also when you submit someone else’s work in a slightly altered form (paraphrase) or use a line of argument without acknowledging it. You are not allowed to use another student’s past written work. You are also not allowed to let anybody copy your work with the intention of passing it off as his/her work. Students who commit plagiarism will lose all credits obtained in the plagiarised work. The matter may also be referred to the Disciplinary Committee (Students) for a ruling. Plagiarism is regarded as a serious contravention of the University’s rules and can lead to expulsion from the University. The declaration which follows must be appended to all written work submitted for Literature Survey NSC412 and NSC421. No written work will be accepted unless the declaration has been completed and attached. I (full names) : Student number : Topic of work : Practical 2 - Screening Declaration: 1. I understand what plagiarism is and am aware of the University’s policy in this regard. 2. I declare that this report is my own original work. Where other people’s work has been used, this has been properly acknowledged and referenced in accordance with departmental requirements. 3. I have not used another student’s past written work to hand in as my own. 4. I have not allowed, and will not allow, anyone to copy my work with the intention of passing it off as his or her own work. Signature : 14