This book is based on the industry leading short course of the same name hosted by the GIW Industries Hydraulic Laboratory and founded by Dr. Roland Clift and Graeme Addie, who together with Dr. Ken Wilson continued its development and authored the first edition of this textbook. This fourth edition has been extensively updated by the international team of engineers and authors who inherited this legacy and continue its development to the present day. Focusing on the hydraulic design of slurry pipelines, the pumps that power them, and the interactions between pumps and systems, it retains the classroom tested balance of theoretical development and practical engineering which have made it a slurry transport classic. The topics covered are important to slurry system engineers for the optimization of new designs, as well as the operators of existing systems, who may need to calculate and plan for changing conditions from day to day.

Updates to the fourth edition include:

• Careful formulation of the theoretical concepts, providing greater clarity of slurry flow dynamics, including a new chapter on the principles and characterization of slurry flows.

• Expansion of the 4-Component Models for settling slurry pipeline flow and pump solids effect, based on an extensive series of full-sized tests.

• An expanded treatment of complex slurries, including a broader discussion of non-Newtonian fluids and their interaction with coarse particles.

• A new chapter on test methods, presenting an overview of slurry system instrumentation, moderntechniques for characterizing slurry rheology, and practical advice for planning and executing a slurry test.

• An overview of advances in the computational modeling of slurries, including an in-depth parametric study of slurry pump wear and operating cost.

The authors highlight methods for achieving energy efficiency, which are crucial to the effective use of scarce resources, given the foundational role of slurry transport systems in the energy intensive industries of mining and dredging. Key concepts are supported with case studies and worked examples. Slurry Transport Using Centrifugal Pumps, fourth edition, is both methodical and in-depth. It is ideal as a teaching tool for classroom or self-directed learning domains, and valuable as a design guide for engineer practitioners at all experience levels.

A 4-component model for settling slurry pipe flow has been previously described by Wilson et al (2006) and Sellgren et al (2014) to predict pipeline friction loss (hydraulic pressure gradient) over a range of slurry compositions: from fine to coarse particle size, narrow to broad particle size distribution, and low to high solids concentration. The method applies a weighted average of established pipeline models for various settling slurry flow regimes, according to the volume fraction of solids falling within the applicable size range for each model. Further development of the model was undertaken by Visintainer et al (2017a, 2017b), based on a comprehensive set of laboratory tests in 203 mm (8 inch) and 103 mm (4 inch) pipelines, and it was also adapted to the modelling of slurry pump performance derates. However, this work was all performed with solids having a specific gravity near 2.65, as is typical for many mineral processing and dredging applications.The goal of the present work is to test the applicability of these models for settling slurries having a higher solids specific gravity, as may often be seen in the mineral processing of iron ore deposits. To that end, a test program was carried out in a 103 mm (4 inch) pipe loop using various compositions of an iron ore product having a solids specific gravity of 4.75 and containing both coarse and fine solids. By screening and flushing operations, a range of particle size distributions were created having different proportions of the coarse and fine elements. In all, 19 tests were performed with d50 particle sizes ranging from 50 μm to 3.2 mm and delivered solids concentrations from 10 per cent to 43 per cent by volume. Particle size distributions varied from very narrow to very broad, with d85/d50 ratios ranging from 1.75 to 26. Pipeline pressure gradient and pump performance data were collected and used to test the applicability of the previously developed 4- component models and to propose improvements to the models for the handling of high-density solids.

Luossavaara-Kiirunavaara AB (LKAB), an iron ore company with mines in northern Sweden is continuously considering new technologies for handling, transportation and disposal of waste rock and tailings. The mines and concentration facilities are located north of the Arctic Circle which in Scandinavia means an average temperature of about 0° C. Snow from mid-October to mid-May. In winter the temperature may reach -35 to -40° C during weeklong cold spells.

At the Svappavaara mine early technical-economical feasibility considerations together with expected space limitations in the concentrator area favored location of two thickened tailings thickeners on a hill close to the disposal area about 1600 m away from the concentrating plant. In this way only short distance pumping of thickened slurry is required and warm process water is recovered directly by gravity from the thickener to the concentrating plant.

A thickener of a high-density type with 18 m diameter was first installed. Four years later an additional thickener of paste type with diameter 24 m was put into operation. The design (maximum) capacities were 115 and 275 tph (tons per hour) for the 18 m and 24 m thickeners, respectively, with solids flux rates of 0.45 and 0.6 ton/m2h. Both are planned for common use for 390 tph within a few years.

The tailings product is characterized by an average particle size of about 30 µm with a maximum of about 500 µm and about 40 % passing 20 µm. Solids density about 3000 kg/m3. A solids concentration by mass of 70 % was considered sufficient for deposition at a slope of up to 3 %.

The objective is to present and discuss the performance of the thickening, transportation and deposition systems during the commission stages and first years of operation. The aim is also to describe how initial conditions related to changes in the tailings production rate together with climatic conditions called for robust by-pass arrangements. Furthermore, complicating factors related to the choice of auxiliary equipment and instrumentation for central functions are discussed.

Transported Newtonian settling slurries (mixtures of solid particles and carrying liquid) tend to stratify in a slurry pipeline and the degree of their stratification significantly affects the frictional head loss in a pipeline system. Solid particles can span a broad range of sizes from those typical for fine sand to those typical for coarse gravel. Different fractions of solids have different properties and form different flow patterns. The different patterns are associated with different dominating particle support mechanisms and friction mechanisms in slurry flow. Hence, there are different models describing and predicting the frictional head loss in pipe flows of different settling slurries. In the presented work, we focus on friction-loss models for heterogeneous (partially-stratified) flow (V₅₀-model), and for fully-stratified flow (V_sm-model). The models can serve as tools to scale up information on frictional head loss in flow of specific slurry obtained experimentally in a small laboratory pipe to larger pipes of industrial sizes. So far, the reliability of the scale up has been difficult to verify as an availability of coarse particle experimental data was restricted to small laboratory pipes (an internal diameter of a pipe typically not larger than 100 mm) and data from larger pipes were extremely scarce.

In 2016 and 2019, extensive experimental campaigns were conducted in the GIW Hydraulic Laboratory (Grovetown, GA, U.S.A.) testing flows of Newtonian settling slurries in pipes of 3 very different sizes (103 mm, 203 mm, and 489 mm). We exploit the experimental data to evaluate the pipe-size scale-up ability of the heterogeneous V₅₀-model and the stratified V_sm-model. The evaluation includes an analysis of the pipe-size effect on the characteristic velocities of the models: the suspension velocity V₅₀ and the deposition-limit velocity V_sm.

Pipe flows of bimodal settling slurries exhibit frictional head losses quite different from those determined simply as a sum of loss contributions by the individual fractions. Mechanisms governing flow friction and resulting from an interaction of grains of different fractions in transported slurry are not well understood. This makes a prediction of the frictional head loss in flows of bimodal slurries with Newtonian carrier uncertain. An extensive experimental campaign was conducted in GIW Hydraulic Laboratory in 2016 with slurries of four narrow graded fractions of the virtually same grain densities and very different grain sizes (carrier-liquid fraction, pseudo-homogeneous-, heterogeneous-, and stratified fractions). Besides testing of the individual fractions, different combinations of the fraction mixtures (bimodal, three- and four-component) were tested as well. In our previous work published in 2018, we employed experimental results for bimodal slurry composed of coarse granite rock (the stratifiedfraction) and fine sand (the pseudo homogeneous fraction) to analyze the observed considerable reduction of the frictional head loss caused by an addition of the fine sand to the granite rock slurry. In this work, we extend our analysis to the other bimodal slurries composed of permutations of the four fractions (in total 3 additional bimodal slurries) with a major objective to identify possible mechanisms leading to a modification of the frictional head loss due to an addition of a finer fraction to a coarser mono-disperse slurry, and to quantify this effect for the purposes of a predictive four-component model (4CM). The investigation shows that the frictional loss of bimodal slurry is always smaller than the theoretical loss obtained as the sum of losses of the fractions, although the massive reduction observed in the slurry composed of the stratified rock and fine sand is not observed in any other bimodal slurry. The investigation also suggests that the friction effect obtained by the finer fraction addition is due to different mechanisms for different bimodal slurries although all mechanisms are associated with altering mechanical friction due to granular contacts. It is shown that the observed effects can be well reproduced by the friction loss model 4CM, calibrated by the experimental data set from the 203-mm pipe and validated by the data set from the 103-mm pipe.

Tests were performed in the GIW Hydraulic Laboratory on a milled copper slag tailings slurry in pipe loops with 75 and 100 mm diameter loss sections. Volumetric concentrations of 1 to 14% were tested in the 100 mm loop, and the initial results, derived using magnetic flow meter measurements, showed pressure losses approximately equal to that of water. Chemical Testing of the solid particles was performed, and they were found to be made up of 56% magnetite by weight. Magnetite has been found in the past (Sonar Trac paper) to effect magnetic flow meter measurements, which was the primary method being used in this test program. Secondary measurements using a Sonar Trac instrument indicated, in contrast, the more common and expected equivalent fluid behavior in the 4” loop at the lower concentrations, where the losses, when expressed in height of slurry, now landed on top of the carrier liquid curve, which was found to be equal to that of water. A comparison between the measurements from the magnetic flow meter, and those from the sonar trac, showed a linear dependency between the flow rate ratio (between the magnetic flow meter measurement and the sonar trac measurement) and the solids concentration. These tests illustrate the importance of having knowledge of the makeup of the solid particles in a slurry, and the utility of acoustic based flow rate measurement methods in cases where magnetite (or other solids particles with magnetic properties) particles are present within the slurry.

Industrial slurries transported in pressurized pipelines often consist of particles of broad size distribution. The broad particle size distribution affects slurry flow behavior in a pipe. A four-component model (4CM) predicts the frictional pressure drop in pipe flow of broadly graded slurry. The model considers Newtonian carrying liquid and splits the broadly graded solids into fractions (components) each of which contributes to the pressure drop through its own dominating friction mechanism expressed by a particular sub-model in the 4CM. The sorting of the solids into the components (carrier, pseudo-homogeneous, heterogeneous, fully-stratified) must be based on appropriate criteria. For the sake of simplicity, the 4CM currently uses threshold sizes of particles to split the solids into 4 components.

The goal of the present work is to analyze the existing criteria for the threshold between the pseudo-homogeneous component and heterogeneous component and for the threshold between the heterogeneous component and fully-stratified component. The analysis is based on a description of mechanisms governing particle support (suspension, deposition) of each particular solids component in slurry flow. It shows that the existing grain-size thresholds actually express certain proportions among threshold velocities of flow delimiting different slurry flow regimes. Such threshold velocities are the deposition-limit velocity, the initial-suspension velocity, and the full-suspension velocity. We discuss the proportions and demonstrate how properties (of liquid, solids, flow) and associated parameters additional to the grain size may influence the thresholds.

The analytical results are supported by experimental results for flow of individual components in a laboratory loop.

A basis for a coherent derating approach, which is founded on principal parameters from commonly referred to derating procedures, is discussed. A weighting technique may couple the diminishing drag influences to rheological effects for the intermediate two-component and viscous area here represented by highly concentrated mine tailings slurries. Loop test results with authentic tailings demonstrated the decreasing head derating with increased rotary speed and thus the viscous influence, a dependence which increased with particle fineness. A marked effect on the derating with yield stresses of 200-260 Pa was also found. Maximum reductions in head and efficiency were 15 and 25%, respectively, at solids concentrations by volume of up to about 50%

A 4-component model for settling slurry pipeline friction loss has been previously described by Wilson and Sellgren. The goal of the present work was to collect a comprehensive and accurate data set spanning multiple flow regimes for the purpose of validating and calibrating the model. This was accomplished by obtaining four different graded silica and crushed granite products representing the four model components. These were then combined and tested in permutations from the individual components to the complete mixture at various concentrations. Primary experiments were carried out in a 203 mm (8 inch) pipeline with selected corresponding experiments repeated in a 103 mm (4 inch) pipeline for validation of pipe size scaling. In all, 40 tests were performed with particle sizes ranging from minus 40 m to 12.5 mm, d₅₀ particle sizes from minus 40 μm to 7.5 mm, and delivered solids concentrations from 4% to 38% by volume. Particle size distributions varied from very narrow to very broad, with d₈₅/d₅₀ ratios ranging from 1.3 to 30. Analysis of the data suggested the introduction of two new correlating factors which have resulted in good agreement between the comprehensive data set and the revised model.

Recent loop testing performed at the GIW Hydraulic Lab (Furlan et al. 2013, 2014) has provided pump performance data for two highly non-Newtonian slurries of significantly different character: a high clay content slurry with minimal coarse solids; and a typical, low clay content, two component tailings slurry. The importance of air removal in the sump and pipe loop was demonstrated using a simple, yet novel de-aeration system. In addition to the measurement of performance losses, determination of the upper limit of "pumpability" for these slurries relative to their concentration and associated yield stress was investigated. However, once the slurry was de-aerated, no limits could be found, other than those dictated by suction side losses (NPSHA) or excessive pipeline friction gradients, indicating that the only true limit in practice is one of system economics, i.e. pump operating and capital cost. Experimentally measured pump head and efficiency were compared against corresponding predictions from two different models: the Walker and Goulas technique (Walker and Goulas, 1983), and the Graham et al. technique (Graham et al., 2009), with special focus given to the dependence of the losses on pump rotary speed.