Erik Ögren Oscar Selberg
Mechanical Engineering, master's level 2020
Luleå University of Technology
Department of Engineering Sciences and Mathematics
E7020T
Master of Science, Mechanical Engineering, Machine Design
Transportation of precious metal slurry
Conducted at Outotec Sweden AB
Authors
Erik Ögren, erigre-5@student.ltu.se Oscar Selberg, oscsel-5@student.ltu.se
Supervisor and examiner
Håkan Lideskog, hakan.lideskog@ltu.se, Luleå University of Technology Supervisor from Outotec Sweden AB
David Gustavsson, david.gustavsson@outotec.com, Outotec Sweden AB
Department of engineering, science and mathematics.
May 18, 2020
We would like to express our greatest gratitude towards Outotec Sweden AB for giving us the time and opportunity conducting our master’s thesis towards them. We also like to thank each of the supervisors which all formed the team established by Outotec assessing us with resources in terms of contacts and knowledge concerning the project. These company representatives includes main supervisor David Gustavsson and also Mikhail Maliarik, Joel Lindström and Jimmie Segerstedt. This group provided guidance, helping us push the project towards reality and also helped lay grounds for the great results achieved in this thesis work.
The companies that has provided cost estimations are also recognised in this master’s thesis. These companies includes Lalmek Hydraulics AB, Colly Flowtech AB, Johnson Conveyor Systems AB, Kellve AB, Nordic Plastic Group, GPA Flowsystems and many more.
We would further like to direct our full gratitude towards Luleå University of Technology with special regards to supervisor and lecturer Håkan Lideskog. This project would not have been realisable without his supervision. We would finally like to express our full appreciation towards associate professor Niklas Lehto for providing us with special access to a project room at Luleå University of Technology throughout the thesis work.
Erik Ögren Oscar Selberg
Luleå May 18, 2020
Outotec Sweden AB works in the field of precious metal refining. Silver electrorefining is one of Outotec’s technologies widely applied in numerous silver refining plants worldwide. During this project a specific section in a silver refinery plant has been investigated. Today’s system provided by Outotec utilises gravity as a means of transport of the slurry consisting of refined silver crystals and silver electrolyte. The slurry is directed from the electrolysis cells through pipes mounted in an angle towards a separation tank. This solution requires three floors of the building of the refinery plant. The goal of this project was to develop concepts which would transport the slurry of silver crystals from the electrolysis cells to a separation tank within a single floor of the building. The implementation of such a system would result in lowering the overall investment cost of the refinery by at least 7 %.
Ulrich & Eppinger’s product development process has been utilised in this thesis work which is a six step sequential method for development of products. Through this process, four concepts for transportation of silver crystal slurry were developed, analysed and cost estimated - Syringe, Drop to circulation tank, Suction Pump and Conveyor. The syringe concept eliminated the need for a single floor of the refinery, which translated to total projected investment cost of 337 000 SEK and an overall investment savings of 8.3 %. Drop to circulation tank eliminated the need for two floors which lead to a total estimated cost of 862 000 SEK. This corresponded to an overall investment savings of 16.5 %.
The two final concepts - suction pump and conveyor was estimated to cost 346 000 and 337 000 SEK respectively. Both of the concepts resulted in a total projected savings of the overall investment by 8.3
%, eliminating the need for one of the three floors in the refinery. The conclusion is that each of the concepts developed surpassed the goal of lowering the overall investment of the refinery by at least 7
%. Three of the four concepts eliminated the need for one floor while the final one, drop to circulation tank, eliminated the need for two of the three floors.
The concepts must be tested before implemented. This could either be conducted by approximating the electrolyte as water and silver crystals as metal shavings or by sludge. It would however be beneficial if a test rig is constructed for each concept and they are tested with the same mixture of silver slurry that is transported in Outotec’s existing refineries.
The circulation tank should also be installed on the same floor as the electrolysis cells and the separation tank for the syringe, conveyor and suction pump concepts. This was never investigated by the authors since it was a limitation stated as the project was initiated. This component was however included in the drop to circulation tank concept since it was considered being part of the transportation system. If the circulation tank is installed on the same floor as the other components it will result in eliminating the need for two floors, which would ultimately lead to a substantial decrease in overall investment cost.
Outotec Sweden AB arbetar inom ädelmetall rening. En av de teknologier Outotec erbjuder är elektrolysrening av silver som är applicerad i flera ädelmetallverk runt om i världen. Detta projekt avser att titta på en specifik del av det silverreningsverk som Outotec erbjuder. Dagens system använder gravitation för att transportera en slurry bestående av raffinerade silverkristaller och silver elektrolyt från elektrolysceller genom rör monterade i en vinkel mot en separationstank. Denna lösning förbrukar tre våningar av raffinaderibyggnaden. Målet med detta projekt var att utveckla koncept som skulle transportera silverslurryn från elektrolyscellerna till separationstanken på en våning vilket skulle sänka den totala investeringskostnaden av raffinaderiet med minst 7 %.
I detta projekt har Ulrich & Eppingers metod för produktframtagning har nyttjats som är en sex-stegs metod för utveckling av produkter. Metoden ledde till framtagningen av fyra koncept för transport av silverslurry som har evaluerats, kostnadsestimerats samt analyserats. Dessa kallades - Syringe, Drop to circulation tank, Suction Pump och Conveyor. Syringe konceptet eliminerade behovet för en våning i raffinaderiet. Resultatet var en totalkostnad på 337 000 SEK, vilket medförde att totala besparingen på hela raffinaderiet blev 8.3 %. Drop to circulation tank resulterade i en totalkostnad på 862 000 SEK, vilket ledde till en besparing av hela raffinaderiet på 16.5 % då behovet för två av de tre våningarna avlägsnades. De resterande koncepten, suction pump och conveyor estimerades att kosta 346 000 respektive 337 000 SEK. Båda koncepten eliminerade behovet av en våning på raffinaderiet vilket ledde till besparingar på 8.3 %. Slutsatsen är att samtliga koncept uppnådde målet att minska den totala investeringskostnaden på raffinaderiet med 7 %. Tre av fyra koncept avlägsnade behovet för en våning medan den övriga, drop to circulation tank, eliminerade behovet för två våningar.
Transportlösningarna måste låtas genomgå testning i verkligheten. Sådana tester skulle till exempel kunna genomföras genom att approximera elektrolyt med vatten och silverkristaller med metallspån.
Det skulle dock vara fördelaktigt om en testanläggning kunde konstrueras för att testa samtliga koncept med den silverlösning som idag transporteras i Outotecs raffinaderier.
Cirkulationstanken borde inkluderas i samtliga lösningar, genom att installera denna på samma våningsplan som eletrolyscellerna och separationstanken. Denna konfiguration studerades aldrig av författarna på grund av att de begränsades av avgränsningarna som hade etablerats i början av projektet.
Cirkulationstanken var trots allt inkluderad i drop to circulation tank konceptet. Detta berodde främst på att den bedömdes ingå i själva transportlösningen i konceptet. Om den här komponenten skulle ingå skulle samtliga lösningar kunna eliminera behovet av två våningar, istället för en. Det skulle i slutändan leda till lägre investeringskostnader för hela raffinaderiet.
1 Introduction 1
1.1 Background . . . 1
1.2 Problem Description . . . 2
1.3 Goals and Purposes . . . 3
1.4 Limitations . . . 4
2 Theory 5 2.1 Product Development process . . . 5
2.1.1 Weight Matrix . . . 6
2.1.2 A Five-step Method . . . 7
2.1.3 Brain Writing . . . 7
2.1.4 Screening . . . 8
2.1.5 Scoring . . . 9
2.2 Determining the Kinematic Viscosity of Slurry . . . 9
2.3 Sizing of Hydraulic Cylinders . . . 10
2.3.1 Minimum Force Required . . . 10
2.4 Pump Configuration . . . 12
2.4.1 Limiting Settling Velocity . . . 12
2.4.2 Pump Effect . . . 13
2.5 Finite Element Method . . . 14
2.6 Fluid Flow Analysis - Computational Fluid Dynamics . . . 15
3 Methodology 16 3.1 Planning . . . 16
3.1.1 Literature Review . . . 16
3.1.2 Mission Statement . . . 16
3.2 Concept Development . . . 17
3.2.1 Functional Decomposition . . . 17
3.2.2 Customer Needs . . . 17
3.2.3 Requirement Specification . . . 17
3.2.4 Concept Generation . . . 17
3.2.5 Concept Evaluation . . . 18
3.3 System-Level Design . . . 19
3.4 Detail Design . . . 19
3.4.1 Calculations . . . 20
3.4.2 Design . . . 20
3.5 Testing and Refinement . . . 20
3.5.1 Fluid Flow Analysis . . . 20
3.5.2 Finite Element Analysis . . . 21
3.5.3 Cost Estimations . . . 21
3.6 Production Ramp-up . . . 22
4 Results 23 4.1 Project Planning . . . 23
4.2 Mission Statement . . . 23
4.3 Grounds for Concept Generation . . . 23
4.4 Concept Evaluation . . . 24
4.4.1 Final Concepts . . . 25
4.5 System-Level Design . . . 26
4.6 Syringe . . . 27
4.7 Drop to Circulation Tank . . . 35
4.7.1 Design . . . 35
4.7.2 Testing and Refinements . . . 38
4.8 Suction pump . . . 41
4.8.1 Testing and Refinements . . . 42
4.9 Conveyor . . . 45
4.9.1 Testing and Refinements . . . 46
4.10 Summary . . . 47
5 Discussion 49 5.1 Limitations and Project Scope . . . 49
5.2 Project Execution . . . 49
5.3 Concepts . . . 50
5.3.1 Syringe . . . 50
5.3.2 Drop to Circulation Tank . . . 51
5.3.3 Suction Pump . . . 51
5.3.4 Conveyor . . . 52
5.4 Comparison . . . 52
5.5 Cost Estimations . . . 53
6 Conclusion 54 6.1 Goal Fulfillment . . . 55
7 Future Work 56 7.1 Testing . . . 56
7.2 Refinements . . . 56
APPENDIX A Gantt Chart i B Mission Statement i C Customer Needs i C.1 Customer Needs Matrix . . . i
C.2 Weight Matrix . . . ii
C.3 Questionnaire Outotec Sweden AB . . . iii
D Screening i E Scoring i E.1 Score Intervals . . . ii
F Product Architecture i F.1 Syringe . . . i
F.2 Drop to circulation tank . . . ii
F.3 Suction pump . . . iii
F.4 Conveyor . . . iv
G MATLAB Script i
H Hydraulic Diagram i
Description Denotation Unit Kinematic viscosity of slurry νp [m2/s]
Pulp dynamic viscosity of slurry µp [P aS]
Relative dynamic viscosity µr [P aS]
Dynamic viscosity of fluid µl [P aS]
Volume fraction of slurry φ [−]
Pulp density of slurry ρp [kg/m3]
Density of solid ρs [kg/m3]
Density of fluid ρl [kg/m3]
Density of slurry ρ [kg/m3]
Ratio of densities SS [−]
Ratio of volumes Cw [−]
Pressure P [P a]
Height Z1 [m]
Height Z2 [m]
Height difference h [m]
Force in y-direction Fy [N ]
Force in x-direction Fx [N ]
Sum of forces FT otal [N ]
Pressure in cylinder P1 [P a]
Pressure in pipe P2 [P a]
Velocity of piston U1 [m/s]
Velocity in pipe U2 [m/s]
Limiting settling velocity UL [m/s]
System curve HSystem [m]
Loss term hf [−]
Friction factor f (Re) [−]
Reynolds number Re [−]
Disposable loss coefficient ξ [−]
Durand’s parameter FL [−]
Kinetic energy correction factor α [−]
Flow Q [m3/s]
Diameter of pipe D [m]
Diameter of cylinder d [m]
Cross sectional area of cylinder A [m2]
LTU Luleå University of Technology
CAD Computer Aided Design
FEM Finite Element Method
FEA Finite Element Analysis CFD Computational Fluid Dynamics NDA Non-disclosure Agreement
VAT Value-added Tax
SEK Swedish Krona
TSEK Thousands of Swedish Krona MSEK Millions of Swedish Krona
1 Introduction
This master’s thesis has been conducted as part of the Master’s program in Mechanical Engineering at Luleå University of Technology, corresponding to 30 ECTS credits. The report will present and discuss the results of the project called Transportation of Precious Metal Slurry by students Erik Ögren and Oscar Selberg. The project was conducted as a product development process at Outotec Sweden AB in Skellefteå, Sweden.
1.1 Background
To separate precious metals from noble-metalliferous ore a refinery process is utilised. This process can utilise many different techniques depending on which metal is of interest and the base ore composition.
Outotec is a company specialising in complete metal refining solutions. Today, the company employs approximately 4000 industry experts from all around the world. Their main objective is to develop leading technology of sustaining the planet’s natural resources. The company’s business is divided into three categories; mineral processing, metals refining and services [1].
One of the solutions offered by Outotec is a silver refining plant. This is a complete solution for refining silver from silver rich slurries to 99, 99 % pure silver in the form of bars or granulates [2]. A block diagram for such a plant can be seen in Figure 1.
Circulation tank
Silver Doré anode casting machine
Electrolyte preparation reactor
Cementation tank
Sieve tank
Induction furnace Cleaning and washing equipment
Dryer
Silver bar casting wheel Granulation
equipment
Nutsche filter
Electrorefining cell
Silver bars Silver granules
Silver anode slime to gold refinery Anode scrap to smelting
Cement silver to smelting
Bleed
Solution to effluent treatment
99.99%
silver crystals
Outotec proprietary equipment Other equipment
Figure 1: Schematic of a silver refining plant offered by Outotec [2].
Silver rich metal alloys are delivered to the plant from other refining processes and is melted down to anode plates. The anodes are then placed in the electrorefining cells where the silver is refined through an electrolysis process. After the electrorefining process is complete, refined silver crystals are collected in the bottom of the cells. The slurry cosisting of silver crystals and electrolyte is discharged to a sieve tank where the crystals are separated and washed. The silver crystals are then dried and cast into bars or granules with a silver purity of 99, 99 %, while the electrolyte is recirculated into the system. During the electrolysis process a slime is formed containing valuable impurities. The slime consists of gold and
platinum-group metals. The slime is collected in anode bags for further refining. To ensure sufficient electrolyte flow between anode compartments and cathode it is continuously between the cells and the circulation tank. In order to reduce the temperature of the electrolyte it is passed through a heat exchanger. In the circulation tank, temperature and pH of the electrolyte is monitored and controlled [2].
Outotec uses electrolysis in their silver refining plant to purify the silver from 92 − 95 % purity to 99, 99 % [2]. The layout of such an electrolysis cell can be seen in Figure 2.
Silver Doré
anode Scraper
Cathode Anode bag
Silver anode slime Cell Silver crystals
Cathode
Figure 2: A electrolysis cell in the silver refining plant offered by Outotec [2].
The principal of electrolysis is that direct electric current is passed between anodes and cathodes, submerged in electrolyte. When this happens a electrochemical reactions are forced. In the case of silver electrorefining, the anodes, made of silver-rich alloys, are digested generating silver as well as base metal ions and insoluble residue - silver anode slime. These ions then travel to the cathode where crystals of the anode material is formed. The crystals that are formed have a higher purity than the anode material. As the crystals form a scraper runs over the surface of the cathode and dislodges them from the surface. They then fall to the bottom of the electrolysis cell where they wait to be collected.
The process is completed when the anodes are nearly completely digested [2].
To separate and collect the valuable impurities a fine meshed bag is put around the anode. This is where the slime containing gold and platinum-group metals is collected. These metals have higher standard reduction potential than silver and are not oxidised at these conditions, i.e. remain in metallic form. This is due to them requiring a higher current to dissolve than the silver. This means that as the anode dissolves, the impurities of noble metals (gold and PGMs) fall to the bottom of the bag [2].
1.2 Problem Description
In today’s solution Outotec utilises gravity to transport silver crystals and electrolyte from the electrolysis cell to the sieve tank. This is an easy and reliable solution but the system requires certain fall height to work properly. To accommodate this fall height a multi story building is required with the electrolysis cells on the top floor, sieve tank on the middle floor and a circulation tank on the ground floor [3]. This way, the slurry from the cells is transported to the sieve tank via a common discharge pipe installed at a certain angle to the tank, as shown in Figure 3.
Figure 3: An illustration of the problem description, where the electrolysis cells are located on the top floor, the sieve tank on the middle floor and finally the circulation tank on the bottom floor. The red line represents the scope of this master’s thesis.
As Figure 3 illustrates, the results of this layout is a three story building which are expensive to construct. Each of the floors are approximately 4.5-5 meters in height.
1.3 Goals and Purposes
Outotec wants to evaluate the possibility of transporting the silver slurry horizontally since the cost of building the current refinery is high. The implementation of such a system could eliminate the need for a multi story building which has the potential to decrease the construction cost greatly. Another benefit of constraining the system to one story buildings is that it enables the system to fit within predetermined spaces, such as already constructed buildings.
The purpose of this master’s thesis is to lower the overall cost of building of a silver refining plant.
This means the cost for the transportation system itself may increase as long as the overall investment is lowered. The operating cost also needs to be taken into account to ensure that the profit is not neglected during the refineries life cycle.
The goal of this master’s thesis is to generate and evaluate concepts for a system for transportation of the silver slurry that enables the electrolysis cells and sieve tank to be installed on the same floor.
A concept therefore needs to be able to transport the silver slurry in a safe and effective manner while lowering the overall cost of the refinery solution. At least three solutions are to be evaluated during this project. Cost estimations are also included for the equipment required and operation expenses.
The total investment cost should not exceed 2.3 MSEK resulting in a required minimum overall saving for a new complete silver refinery plant of 7 %.
1.4 Limitations
Some limitations has been established when conducting the project. These limitations are listed below.
This thesis work is limited to a time span of 20 weeks and is concluded on the 20th of May 2020.
• The top of the sieve tank and electrolysis cells needs to be accessible with overhead crane.
• Only the transportation system between the bottom valve of the electrolysis cell to a separation tank is to be considered.
• Flow and energy optimisation is not to be considered, but is beneficial to include to a certain degree.
• The new system needs to be compatible with the existing refinery offered by the company. This means that no changes are to be made to the electrolysis cells. The current separation solution and circulation tank can be changed if a concept requires it.
• No products will be manufactured during this project.
2 Theory
In this section, the theory behind the method and results is presented. In each subsection a different theory regarding the project is explained.
2.1 Product Development process
In product development processes different methods is utilised. One of these methods is the Ulrich &
Eppinger model [4] which is a sequential six step method. Each of the steps are called gates in the development process and consists of deliverables and tasks which defines each gate. These steps are illustrated in Figure 4 which follows.
• ••
• ••
Planning Concept
Development
System-Level Design
Detail Design
Testing and Refinement
Production Ramp-Up
uLr04772_ch02_011-032.indd Page 14 03/03/11 8:36 AM F-501 208/MHBR223/uLr04772_disk1of1/0073404772/uLr04772_pagefiless
Figure 4: An overview of the gates which Ulrich & Eppinger’s product development process is based on [4, p.14].
These six gates presented in Figure 4 are explained and listed below.
1. Planning: The first gate consists of planning the project. This includes identifying opportunities, evaluating and prioritising different aspects and phases of the project. An allocation of the resources provided and a timetable is established. A deliverable from this gate is a complete pre-project plan which includes planning each of the five remaining gates [4].
2. Concept development: In this phase the development of the concept is initiated. Alternative products are generated and the target marked needs are identified. During this gate, an analysis of competitive products is conducted [4].
3. System-level design: The third gate in the product development process consists of decomposing the product into multiple subsystems and defining each of the key components in the product. A deliverable of this gate is usually a geometric layout of the product with each subsystem featured - a product architecture [4].
4. Detail design: This phase of the process consists of a complete specification of the geometry, tolerance and materials of the product. In this phase each of the standard components to be purchased is specified and drawings or computer models are established of the product [4].
5. Testing and refinement: In this gate evaluations of the product are carried out. This includes testing both psychically and virtually and in some cases prototypes of the concept are fabricated [4].
6. Product ramp-up: The final phase in Ulrich & Eppinger’s product development process is the sixth gate. In this gate, the product is constructed by utilising the intended production system.
These products are in some cases supplied to customers and are fully evaluated before launching the final product. In this phase this product is finally launched to the public and a review of the project is implemented which includes an assessment of the project from both technical and commercial perspectives [4].
2.1.1 Weight Matrix
One deliverable of the second gate - concept development is a need analysis. To determine the importance and rank of each need a weight matrix can be utilised. An example of this can be seen in Figure 5 below which is based on Karlberg, [5].
Need 1 1 0 1 1
Need 2 0 0.5 1 1
Need 3 1 0.5 0.5 0.5
Need 4 0 0 0.5 1
Need 5 0 0 0.5 0
Need4
Need2 Need3
Need1 Need5
Figure 5: A weight matrix where the black lines represents the sum of the cells, which always must be equal to one [5].
If a need is more important than another they are given a value of 1, if they are equally important they are given a value of 0.5 and if a need is less important they are given a value of 0. These values are then added up vertically to give a total score of each need [5].
2.1.2 A Five-step Method
When developing concepts in the model presented, different methods can be used. One such method is the five-step method which is presented in this subsection. An illustration of the method can be seen in Figure 6.
1. Clarify the problem.
• Understanding
• Problem decomposition
• Focus on critical subproblems
2. Search externally.
• Lead users
• Experts
• Patents
• Literature
• Benchmarking
3. Search internally.
• Individual
• Group
4. Explore systematically.
• Classification tree
• Combination table
5. Reflect on the solutions and the process.
• Constructive feedback
New Concepts
Integrated Solutions Existing Concepts
Subproblems
Figure 6: The five-step method [4, p.120].
The first step of this process is clarifying the problem. This involves understanding and identifying all the subparts to the problem at hand, decomposing it and focusing on critical subproblems [4].
The process then proceeds into the second part of the method, which is an external search. This search is mainly focused on finding existing solutions to the problem in its entirety and each of the subproblems. The phase includes consulting lead users such as customers and experts. A literature study is conducted on the main subjects and existing patents are studied. After this is done, an internal search is conducted which comprises of personal and team knowledge to finally generating solutions and developing concepts. This type of search is often referred to as brainstorming [4].
A classification tree is established and the concepts subproblems are combined in a table. This way, the concepts are considered combinations of solution fragments. These fragments are then combined to create new concepts which are final solutions. After a couple of concepts have been developed a reflection of the solutions that has resulted in the five-step method is carried out [4].
2.1.3 Brain Writing
A method of generating concepts is brain writing or as it is also called 6-3-5 method. This is a group exercise typically performed in groups of six where one member is the team leader. The team leader
presents the problem to the group and all members are given five minutes to generate three solutions.
Once the time is up the solutions are passed to the next person who builds one the ideas of the previous member. This is done until the group members cant build on the ideas anymore. This is usually done in silence as not to affect the other group members. Once the process are complete the solutions are discussed and classified [6].
2.1.4 Screening
Screening is a method of evaluating multiple concepts at once and selecting the most suited concepts to proceed with. This method provides a quick and approximate evaluation of a large number of concepts and gives a few viable options for further evaluation. A matrix is established with the requirements from the requirement specification used as basis. The next step is rating and ranking the concepts to select them for further evaluation. During this process solutions may be combined to create new and improved ones. Screening is useful since rough estimations can be made and no detailed quantities have to be found which is often hard and may be misleading in the preliminary stage [4].
The screening process is structured in the following manner:
1. Prepare the screening matrix: The first step is to prepare the screening matrix. This is done by taking the requirements established and putting them into a column. Some requirements may be combined to condense the matrix and give a better representation of the concepts performance.
When this is done the concepts are presented and given a column each. For this process a simple sketch and some text is sufficient so the participants understand the concepts to be evaluated [4].
2. Rate the concepts: When rating the concepts one is chosen as a reference. The other concepts are then rated as better than (+), as good as (0) or worse than (-) the reference concept. These are placed in each cell of the matrix to represent how well a concept performs in a specific criterion in comparison to the reference [4].
3. Rank the concepts: When ranking the concepts the pluses, zeroes and minuses are added up.
The reference concept will get a score of zero and the other concepts will be better (positive numbers) or worse (negative numbers) than this concept [4].
4. Combine and improve the concepts: After the concepts have been evaluated team members are allowed to improve or combine concepts. This is generally done if a good concept is degraded by a bad feature or if two concepts can be combined to preserve the good qualities and eliminate the bad. When this is done new concepts are made which can be added to the matrix to be evaluated and ranked along with the previous concepts [4].
5. Select one or more concepts: Once all concepts have been given a score and rank the elimination process begins. The selected concepts will undergo further refinement and analysis.
Based on the resources available a single or multiple concepts may be selected. In this stage the team members should have a good understanding of which concepts are the most promising [4].
6. Reflect on the result and the process: After the process all members should be comfortable with the outcome. If a member is not in agreement with the decision an important criteria may be missing from the matrix, a rating may be incorrect or at least not clear for all participants.
If all the members are in agreement the likelihood of mistakes are reduced and the team will be more committed to the following development activities [4].
2.1.5 Scoring
The second phase in evaluating concepts developed is called scoring. This method is similar to the screening method. The methods are based on the same six steps - preparing a matrix, rating, ranking, combining and selecting a final concept. In this case however the concepts are rated and ranked differently. They are usually rated from 1-5, this score can however vary depending on the project.
The concepts are compared against the reference concept, and number 1 represents the lowest score -
”much worse than reference” and 5 represents the highest rate - ”much better than reference” [4].
Once the concepts are rated, a ranking phase is initialised and ranks of each concepts are calculated.
This is done according to equation (1)
Sj=
n
X
i=1
rijwi, (1)
where Sj represents the final rank of the concept, rij is the score given earlier and wi is the weight criterion [4].
The last step in scoring is combining the concepts and improving them. The result of this method is one final concept to be developed further [4].
2.2 Determining the Kinematic Viscosity of Slurry
The kinematic viscosity of a slurry νp is obtained by
νp =µp ρp
, (2)
where µpis the pulp dynamic viscosity of the fluid and ρpis the pulp density. The dynamic viscosity of the slurry is defined as the product of the fluid and relative viscosity [7]. This is calculated according to
µp= µrµl, (3)
where µr is the relative viscosity of the mixture and µl is the dynamic viscosity of the fluid [7]. The relative viscosity is obtained as follows,
µr= 1 + 2.5φ + 10.05φ2+ AeBφ (4)
where A and B are constants which are obtained by empirical data. These are A = 0.00273 and B = 16.6 [7]. φ is the volume fraction of the slurry. This is calculated as
φ = 1
(1 + (100−CC w
w )ρρs
l), (5)
where Cw is the ratio of the volumes between the fluid and the solid of the slurry. ρs and ρl is representing the densities of the solid and fluid in the slurry [7].
The pulp density is obtained by
ρp= ρLSS
(C + (1 − C )S ), (6)
where SS is the ratio of the densities. νp is determined by combining these equations and computing equation (2) [7].
2.3 Sizing of Hydraulic Cylinders
In this subsection the calculations and theory behind the configuration of hydraulic cylinders is presented.
2.3.1 Minimum Force Required
Here the theory behind the calculations of the minimum force for which the piston in hydraulic cylinders needs to overcome. Figure 7 is illustrating the forces applied in this problem.
Fy FX
D L
h d
Figure 7: A hydraulic cylinder with a piston mounted inside of it and the minimum forces the piston needs to overcome.
As Figure 7 shows the forces the piston needs to overcome are in the x- and y-direction. The pressure in the y-direction is given by Pascal’s law presented in the equation below
P = ρgh, (7)
where h represents the difference in height, as shown in Figure 7, and ρ is the mean density of the slurry [8]. With pressure P given by definition, the force in the y-direction can be obtained by
Fy= P A, (8)
where A is the cross-section area of the piston inside the cylinder. This is given by definition of a circular cross-section area,
A = πd2
4 (9)
where d represents the inner diameter of the cylinder.
To calculate the force required to pull the piston in the x-direction Bernoulli’s equation is utilised.
Bernoulli’s equation for incompressible media is given as follows
P1 ρg +U12
2g + Z1=P2 ρg +U22
2g + Z2+ hf, (10)
where P1and P2 are representing the pressure difference, Z1 and Z2 represents the change in height, U1and U2is the velocity of the liquid and hf represents the losses due to friction [8]. This is illustrated in Figure 8.
Figure 8: The connection between the hydraulic cylinder and a pipe mounted to it.
As Figure 8 shows P, A and U located to the left represents the characteristics of the hydraulic cylinder while the rest represents the characteristics of the pipe connected to it. Solving for the pressure difference P1-P2in equation (10) gives
P1− P2= ∆P = hfU22− U12 2hf + g
ρ. (11)
The loss term hf is obtained as
hf = V2 2g
X(L
Df (Re)) +X ξ
, (12)
where L is the characteristic length, D is the diameter of the pipe, f (Re) is the friction factor due to turbulent flow and ξ is the loss term [8]. The friction factor is given by
f (Re) = 0.3164 × Re−1/4, (13)
where the constant is a friction factor for smooth pipes [8]. Re, Reynolds number is given by
Re = U D
µ . (14)
µ represents the kinematic viscosity of the media [8], which is explained above in its own subsection.
By combining equation (12) with (11) an expression for the required force, Fx, can be obtained by
Fx=V2 2g(X
(L
Df (Re)) +X
ξ) U22− U12
2(V2g2(P(DLf (Re)) +P ξ))+ g
ρA. (15)
Finally the sum of forces, FT otal, is given by
FT otal= Fx+ Fy. (16)
2.4 Pump Configuration
Here the theory behind configurations of pumps is presented.
2.4.1 Limiting Settling Velocity
A slurry being pumped needs a certain velocity to ensure the solids does not settle on the bottom of the pipe. This is given by Durand’s equation for limiting settling velocity, as follows [9]
UL= FLp
2gD(S − 1), (17)
where UL is the limiting settling velocity. FL is Durand’s parameter for limiting settling velocities, D is the pipe diameter and S is the specific gravity of the solid. Durand’s parameter FL is derived from Figure 9 below [9].
Figure 9: A diagram from which Durand’s parameter, solid specific gravity and limiting settling velocity (expressed here as VL) can be derived [9, A1-5].
As Figure 9 shows, Durand’s parameter for limiting settling velocity is derived from the diagram by having the volume concentration and pipe diameter. An example of how the parameter is taken is shown in Figure 9 by the dotted arrows.
The final limiting settling velocity is usually multiplied with a factor 2 for vertical flow which works as a safety factor. A rule of thumb however for the limiting settling velocity is 1-3 m/s [10].
2.4.2 Pump Effect
The minimum required pump effect is given by the following equation
P = ρQgHSystem, (18)
where HSystemis the system curve and Q is the flow of the fluid. HSystem= HP ump is assumed in the pump operating point [8]. This system curve given by (19)
HSystem= hf +hP
ρg+ αU2 2g + zi2
1
, (19)
where turbulent flow is assumed which gives α = 1. The loss term hf is given by equation (12) and z2− z1 = h since this difference is representing the change in the vertical direction [8]. The system curve, HSystem, can be obtained by (12) in (19)
HSystem= h + ( 1 2g)( 4Q
πD2)2(1 + f (Re)(L
D)), (20)
where Q is the flow of the media [8]. Q is obtained by the continuity equation
Q = U A, (21)
where A represents the cross sectional area and U is the velocity given by equation (17) [8]. The required pump effect can be obtained by combining equation (20) and (21) in equation (18). The final expression for the pump effect is presented below
P = ρ(h + (1 2g)( 4Q
πD2)2(1 + f (Re)(L
D))Qg. (22)
There is a decrease in efficiency ratio depending on the pump and particle size the media transported consists of. This efficiency is determined by testing and analysing the pumps. Studies shows however that some suction pumps has a decreased efficiency ratio of approximately 40-50 % depending on the size of the particles. This is something that needs to be taken into consideration when designing a pump [11].
2.5 Finite Element Method
This method of analysing components is used to calculate differential equations through numerical approximation. This method is called the finite element method or FEM. The basics of the method is to establish a discretisation of an entire domain and dividing it into sub-domains called finite elements.
These elements are then connected by nodes and is called a mesh. An example of this is illustrated below [12].
Figure 10: An example of a mesh created, where the finite elements are connected by nodes.
This way, the domain can be solved for a system of equations instead of solving the entire body or component. An expression describing the basics of a structural analysis using the FEM is [12]
[K]{u} = {f }, (23)
where [K] represents the structural stiffness matrix which contains the behaviours of the material.
{u} are the structural displacements and {f } are the structural loads applied. The simulation system solves for different parameters and one of them is the displacement on the elements where the forces are applied. This way, different quantities can be calculated, for example the elongation, pressure and much more which the body is exposed to when applying this force [12].
2.6 Fluid Flow Analysis - Computational Fluid Dynamics
One way of conducting fluid flow analysis is through computational fluid dynamics - CFD. This is a numerical analysis which utilises data structures constructed in different types of computer programs to perform analysis. Through these analysis different types of fluid flow problems can be solved using simulations. These simulations provides an understanding of how fluids interacts with different types of bodies or surfaces which are defined by region boundary conditions. An example of such body is shown below [13].
Figure 11: An example of a meshed body which has assigned region boundary conditions, shown as the arrows in the front of the body.
Figure 11 is illustrating a solid body which is defined by region boundary conditions. In this example an inlet flow and an opening is assigned as the region boundary conditions. The inlet flow is assigned to the front of the body and the opening is defined to the back of the body.
Just as in FEM, discretisation of entire bodies are done by dividing them into finite elements connected by nodes. Region boundary conditions are then established and different parameters are applied, for an example velocities, mass flow, densities and other physical properties for the media. A CFD simulation can finally result in knowledge of how the fluid will interact with the region boundary conditions stated.
These simulation can act as a proof of concept to assess whether or not the input values are reasonable or not, for example initial velocities or other parameters [13].
3 Methodology
In this chapter the methodology used during the thesis work is presented in chronological order, where each part of the project has its own subsection.
3.1 Planning
The project began with gate 1 - planning, where the available time and project deliverables were taken into consideration. A gantt chart was created where the deliverables were separated into six gates based on the Ulrich & Eppingers product development process. To suite this project better the sixth and final gate was altered from the theoretical model. One limitation stated that no products would be manufactured in this project and the deliverables were therefore changed for the final gate, production ramp-up. Instead of having manufacturing of products conducted, a presentation of the project and a final report was implemented in the gantt chart [4].
After the gates had been determined the available time was allocated between them. The time was not divided equally since some parts of the project would be more time consuming than others. When the time frame for each gate was determined the gate time was divided between the elements of each gate.
3.1.1 Literature Review
A literature review was conducted in the purpose of gathering knowledge and insights to the subject at hand, but most importantly provide a theoretical background for which the thesis work could rely on. This information was used throughout the product development process but mainly in concept generation, concept evaluation and detail design phases. The project was by doing this ensured of being based on reliable theory [14]. This study was included in the five-step method as an external search [4].
The study was structured by assuming some components in the system as well as identifying possible alternatives to the current solution provided by the company. A list of search words were thereafter formulated, and these were utilised when the literature study was carried out. These words included slurry, transportation, pump, slurry pump, screw conveyor, tube-chain conveyor, silver refinery and many more. When conducting the study, Luleå University of Technology’s library database was used.
This library database is based on most journals and databases available through the university. The title of the projects and abstracts were used to select suitable articles and reports. These were also used to determine whether or not they were of relevance to this project. Apart from using the databases at LTU, books that previously had been used in courses at the university was collected and studied by the authors. The sources were finally organised and discussed by the authors.
3.1.2 Mission Statement
To conclude the planning phase of the thesis work a mission statement was established. This deliverable was based on the limitations stated for the project by the authors and representatives of Outotec [4].
To formulate key business goals of this project some calculations had to be made, since the initial goal of the project was to lower the overall cost of the refinery. The company themselves did not have an exact figure how much this cost should be lowered. A price for a complete refinery solution was therefore provided to the authors and a estimation of the cost per floor was calculated. The floor price was based on a quote given to Outotec by the company WSP Group. By having a rough estimate of the cost of a standard silver refinery a key business goal of decreasing the overall cost by a certain
percentage was set.
Since the thesis work was conducted at Outotec Sweden AB they were the primary market for the project along with other precious metal refineries. Assumptions and constraints were formulated in the mission statement. This translated into limitations where the entire scope of the project was set.
3.2 Concept Development
The next gate in the product development process was concept development. The methodology behind this gate is presented here.
3.2.1 Functional Decomposition
A functional decomposition was established to separate the problem into subproblems. This was done by starting with the main function and working backwards to separate the problem as far as deemed necessary for the project. The main focus was on the subsystems that the project aimed to change, namely transportation solutions and components for the surrounding control system, such as switches and sensors. To ensure that the project did not get locked into a specific direction, broad terms were used and no solution specific details were implemented [4].
3.2.2 Customer Needs
In the process of concept development a customer needs matrix was established. In order to identify the needs of the customers, the likes, dislikes and finally suggested improvements of the current product. For this project a questionnaire was created, with the standard questions from Ulrich &
Eppinger product development process, and sent to the company employees involved in the project.
This empirical data was then utilised when establishing the matrix of the customer needs. All needs were then interpreted by the authors and assembled alongside the statements from the employees.
These interpreted needs would work as a template for the upcoming requirement specification. To ensure the statements had been interpreted in the correct way the company was consulted for all of the interpretations [4].
When the authors and company representatives were satisfied with the needs of the system these were compared against each other. This was conducted using a weight matrix where a need could be more important, as important or less important, as presented in the theory section. Based on the score each need achieved they where given different weight which would symbolise the importance of each need.
These scores were later implemented in the customer needs matrix.
3.2.3 Requirement Specification
After the customer needs had been interpreted they were converted into measurable requirements for the finished product. These can be in the form of values such as flow in liter per minute or if a specific feature is implemented or not. Each requirement is given an allowed and an ideal value. These differ in that the ideal value is the goal but since some requirements are more important than others the allowed value indicates what the product must achieve to be satisfactory. How well these are achieved lay the basis for the subsequent screening and scoring [4].
3.2.4 Concept Generation
The concept generation was done in multiple iterations to achieve as many concepts as possible. In the first iteration the problem was divided into sub problems. When a satisfactory amount of solutions
for each sub problem were reached they were combined into complete solutions.
For the second iteration a brain writing was conducted. The group members were allowed to draw either complete solution or sub solutions. These were then handed to the other group member who could refine and change the solution. When a satisfactory number of solutions was reached all concepts were taken to the company to be discussed and see if any apparent flaws could be found [6].
With the new information and thoughts from the company, a final brainstorm was conducted. This yielded the final concepts to be evaluated during the final phase of the second gate.
3.2.5 Concept Evaluation
The process of evaluating concepts were done by screening and scoring. This process is illustrated in Figure 12 below [4].
Figure 12: Concept evaluation procedure conducted on the solutions by the authors.
As Figure 12 shows, concept evaluation was done in multiple steps, beginning with a screening of the first draft of the concepts. During screening, all concepts were compared with the solution used by the company today [4]. This was done to quickly eliminate the concepts that would not satisfy the needs and requirements of the company. During screening, the concepts were only considered to be better than, on par with or worse than the system used today. By doing this, no real values had to be considered for the different concepts and the authors could estimate how well a concept would fulfil a specific requirement. To simplify screening, some requirements concerning the concepts were condensed into broader terms. This was also done to prevent the method of being segregated. The screening process was done in multiple iterations during the concept generation and both the company and the supervisor was consulted during this step.
When the screening was complete the scoring process began. All requirements were considered during this phase and each was given a weight from 1 to 5. These were based on the customer need, where each need had been ranked by establishing a weight matrix. The determination of each weight was also conducted in collaboration with Outotec. The requirement satisfaction was then divided into four different intervals between the allowed and ideal value giving different scores based on what the concept achieved [4].
Determining what score a concept was awarded, was decided by finding realistic values for some of the requirements as well as estimations. These estimations included all binary requirements, since no realistic values was able to be found.
The score and weight of the requirement were then multiplied. These were then summed up for all requirements which gave the final score of each concept. At the end of this phase the concepts with the highest scores were the ones ready for detail design and virtual testing. This differentiated from [4] where only one concept proceeds from the concept evaluation phase. Since the project goal states that at least three concepts were to be evaluated, this method had to be altered to satisfy the goals of the project [4].
3.3 System-Level Design
As the third gate was initiated in the product development process, the system-level design began. As explained in the theory section, this step includes decomposing each concept into multiple subsystems and defining each key component. This was conducted for each of the separate solutions that resulted from the scoring process [4].
One deliverable of this gate was the product architecture for the concepts generated. This described the subsystems in each solution and every key component. To do this, the functional decomposition created in the previous phase was utilised. Schematic images were created which was presented to representatives at the company ensuring the authors that the concepts would be possible to implement.
Key components were also discussed and the concepts which started out as superficial ideas translated into total solutions for the entire silver slurry transportation [4].
3.4 Detail Design
To initialise the fourth gate, the detail design phase began. Since this gate included producing scaled models of the concepts some calculations had to be conducted. A list of deliverables was established for each concept of what calculations had to be made in order to construct them. After each concept had a list created the concepts were distributed by the authors and each person were responsible for two of the four concepts. This comprised not only calculations that were to be made, but also development of each concepts using CAD. For this master thesis Siemens NX version 1899 was used.
This gate was conducted iteratively by consulting the supervisors from Outotec and LTU. Each week concepts in development were discussed and any changes were brought up and scrutinised. Topics that were discussed included manufacturing and logistical aspects that could be of concern to the company. The gate eventually concluded when both supervising parties were satisfied with the design and calculations of each concepts. An illustration of the method used in this gate is shown in Figure 13.
Calculating
Configurating Consulting
Figure 13: The methodology of detail design utilised in this project.
3.4.1 Calculations
Characteristics of the silver slurry had to be determined to be able to calculate the requirements for each concepts. This included kinematic viscosity, density and volume concentration of the slurry.
When these characteristics had been computed, calculations concerning the concepts would commence.
To be able to base the concepts on theoretical framework, dimensions had to be determined for each concept. Since no dimensions were known, except for the volume of each cell, a assumption of a 5-cells setup was made. This gave a set volume a solution should handle with regards to total volume of the slurry.
To calculate the velocities and minimum forces acting on a solution, vectors were constructed in MATLAB. These vectors were used in loops calculating the forces and velocities for a given diameter or length of a concept. These physical quantities were then plotted against each other to find the optimum for forces, velocities and dimensions.
3.4.2 Design
The concepts that required detail design were developed using CAD. Preceding theoretical calculations of pipe dimensions, required flow speeds, forces analysed and so on. These dimensions would work as a basis for each of the concepts that were subject to this development process. The concepts that did not undergo this process were researched and discussed with suitable suppliers of the given solution.
The suppliers were given relevant data to understand the problem and asked to make a suggestion for a suitable component.
3.5 Testing and Refinement
As the detail design was concluded the second to last gate of this project could commence - testing and refinement. This gate included conducting analysis on the different types of concepts developed in the product development process. This involved flow and finite element analysis and was conducted as a proof of concept in order to evaluate them. Another deliverable of this gate was a cost estimation of each concept produced. In this section the methodology of these analysis and estimations are presented in chronological order.
3.5.1 Fluid Flow Analysis
To perform flow analysis the concept designed in Siemens NX was utilised. The first step in conducting this analysis was to make simplifications to the geometry of the concepts. This involved suppressing unnecessary features which had no impact on the simulations, for example holes used for assembly purposes. Since simulating entire assemblies was unnecessary only the components that were of relevance was used for the simulation model. The input values such as initial flow velocity were taken from the calculations conducted earlier in the previous gate. The second step in the simulation phase was to establish a control volume of the components. This was done by deleting the outer faces of the bodies in CAD. Doing this resulted in a volume matching the inner dimensions of the object that was to be simulated. This way a control volume of the component had been established. Before simulating a mesh was created of the part. A 3D tetrahedral mesh, TET4 was used with different types of element sizes depending on which component was analysed. An example of the meshing performed in CAD can be seen in Figure 14.
Figure 14: A close up of a mesh of a control volume in CAD with an element size of 20 mm.
Figure 14 shows a mesh of a part with an element size of 20 mm. While meshing, an approximation was established by assigning water as the media to the control volume. The reason for this was that no media which possessed the same properties as silver slurry was available in the simulation program.
The electrolyte however possessed almost the same characteristics as water, which was deemed a good estimation for these types of simulations.
To perform the flow analysis Simcenter 3D Thermal/Flow solver was used, with the analysis type set to flow. The solution type was also assigned flow but with steady state conditions. Finally the simulation object types that were used was region boundary conditions with an inlet flow and an opening to simulate the entire process. Streamlines were used to illustrate how the flow was distributed throughout the control volume.
3.5.2 Finite Element Analysis
Another way of evaluating the concepts designed was using finite element analysis. The method was utilised in this master’s thesis work on the concepts designed using CAD which was of relevance to the matter. This analysis was performed on the components which were deemed to be subjected to the highest stresses and for this method Siemens NX was again used.
The same meshing method was used as in the flow simulations but instead of simulating a controll volume the part itself was assigned a mesh. In Siemens NX the structural solver Simcenter Nastran was used for the FEM simulations. An idealised part was created and features such as holes, chamfers and edgeblends deemed unnecessary for the simulation were removed. Parts were given loads in the form of forces or pressures which would affect the part at specific locations to represent the real world application. The magnitude of the loads came from the theoretical calculations conducted by the authors. Some parts of the component were also fixed in position to see how the forces affected the selected geometry during specific conditions. The results where then analysed and discussed to ensure their validity. The results were illustrated with contour plots showing how different parts of the component was effected with a colour scale.
3.5.3 Cost Estimations
The final step in evaluating the concepts was to conduct a cost estimation of each solution. This was one of the requirements established by Outotec and one of the key steps in assessing whether or not the goal of 7 % in investment savings were met.
To conduct this estimation a list of potential manufacturers was established in terms of contacts. This list consisted among others, companies that had non-disclosure agreement (NDA) with Outotec. This
meant that the solutions could be discussed further without breaking the NDA signed by the authors to Outotec. The companies which had signed these agreements were provided to the authors by Outotec and the rest of the companies were found by the authors themselves.
This research of manufacturers was conducted in similar way of the literature study, using keywords of components, manufacturing methods and potential materials. While researching potential manufacturers, different types of materials such as plastics and metals were discussed and investigated. Finally when enough companies had been found, emails were sent out with some drawings of components that required a manufacturing cost estimation. Some of the companies were also contacted by phone to provide a better understanding of the components such as tolerances and materials that were to be assigned. When the cost estimation of each component was finalised, tables were established where they were compiled. A total sum of the cost of each concept could therefore be estimated. This estimation only included components of the transportation system between the sieve tank and electrolysis cells.
Meaning that components which already existed in today’s solution such as sieve tank, electrolysis cells and valves were excluded in this process.
Since a goal of the project was not only to calculate the total projected investment cost but also the total yearly maintenance cost, estimations for this had to be done. This estimation was also conducted by contacting the companies which provided cost estimations for the components that could be subjected to maintenance intervals. Yearly energy consumption was also investigated by studying the kilowatt hours of the power sources each concept had.
3.6 Production Ramp-up
The concluding gate in this project was the production ramp-up. This gate usually consists of producing the products developed through the product development process. This was not the case however since it was a limitation not to manufacture any concepts or components developed. The gate therefore included finalising the concepts developed, finishing the report and presenting the results of the project at Outotec and Luleå University of Technology.
4 Results
In this part of the report the results from this master’s thesis will be presented in chronological order.
Each of the phases in the project is included in its own subsection.
4.1 Project Planning
The first results to be presented in the project was the planning phase. Since it was conducted using a gantt chart, this can can be seen in Appendix A. A complete list of deliverables was created and these would work as elements to be included in the gantt chart. These deliverables are listed below.
– Gate 1 - Planning: Gantt chart, mission statement, literature study.
– Gate 2 - Concept Development: Customer needs, functional decomposition, requirement specification, concept generation, screening and scoring.
– Gate 3 - System-Level Design: Product architecture.
– Gate 4 - Detail Design: Digital calculations and design.
– Gate 5 - Testing and Refinement: Virtual product testing and cost estimations.
– Gate 6 - Production Ramp-up: Presentation of project.
Since the only date that was set for this master’s thesis was the date of the final presentation, 20th of May this would work as a time reference for each of the six gates.
4.2 Mission Statement
Because one of the key business goals of this project was to lower the overall cost of a refinery, a rough estimation of the savings that could be achieved was calculated. Provided with the investment cost of a refinery of 150 MSEK by Outotec, an estimation of the savings when eliminating the need for a single floor could be calculated. WSP administered the price per square meter of industrial buildings of 16 675 SEK/m2 and Outotec provided a schematic for a previously implemented refinery with a floor size of 22×35 m. This meant that the savings when eliminating a single floor from the refinery could be calculated according to the expression, as follows [15]
16675 × (22 × 35)
150 × 106 =12.8 × 106
150 × 106 = 8.56 %.
Since the new solutions were likely to be more expensive than the current solution, in regards to component cost, a goal of lowering the overall cost of a new silver refinery by 7 % was set. To conclude the mission statement the stakeholders were determined; Outotec AB, Luleå University of technology and the authors. The mission statement established for this project can be seen in Appendix B.
4.3 Grounds for Concept Generation
To enable concept generation for sub solutions, a functional decomposition was created. The functional decomposition for this project is illustrated in Figure 15.
Transport Silver slurry Transport Silver
slurry
Transportation from electrolysis cell to
sieve tank Transportation from
electrolysis cell to sieve tank
Power source
Power source Transportation unitTransportation unit Transportation vessel Transportation
vessel LinkageLinkage
Figure 15: The functional decomposition of the project, where each subsystem is defined.
Through the interpreted needs from the company, which can be seen in the customer needs matrix in Appendix C, a requirement specification was established. All requirements were given allowed and ideal values in collaboration with Outotec. This specification can be seen in Table 1.
Table 1: The requirement specification.
Metric number Customer need number
Metric Unit Allowed values Ideal values
1 1,3 Flow to sieve tank l/s 7 30
2 10 Obstruction frequency F actor 1:100 1:1000
3 2 Modular Binary Yes Yes
4 3 Emptying time per cell Seconds 60 15
5 4 Automated Binary Yes Yes
6 7 Life expectancy Y ears 10 30
7 8,9,11 Height m 3 <3
8 9,11 Cost of implementing
solution
SEK <2 500 000 <1 000 000
9 9 Yearly maintenance
cost
SEK 100 000 0
10 7 Maintenance interval Y ear 1 2
11 9 Yearly energy
consumption
kW h 15 000 0
12 9 Yearly running cost SEK 150 000 0
13 12 Time to access sieve
tank
M inutes Current 0
14 5 Electrically shielded Binary Yes Yes
15 6 Evaporation
temperature of material
◦C 962 100
16 5 Operator safety Binary Yes Yes
17 2 Flexibility Binary Yes Yes
The allowed and ideal values were based on the capabilities of today’s system, simple calculations done by the project group as well as values given by the company. When all values were decided a final review was done with both the company and supervisor to ensure their validity.
4.4 Concept Evaluation
To quickly eliminate some of the concepts generated a screening was conducted. This was done to enable the scoring process to focus on the most suited solutions for the specific task. The screening matrix for this project can be seen in Appendix D.
