Novel small-scale manufacturing of
welding consumables
Ny småskalig tillverkning av svetstillsatsmaterial
B.Sc. Thesis Project Report
Summary
Linnaeus university (LNU) Welding Mechanic Laboratory (WML) is currently undertaking the naval architectural research project “R/S Snöfrid”, and this ship shall be constructed in the high strength steel SSAB Strenx 1300. The different section shall be joined by arc welding processes. No suitable welding consumable is commercially available, so a novel welding consumable must be researched and developed.
Current research and development processes of novel welding consumables constitutes a great cost and requires a lot of time and resources. LNU WML has requested a novel small batch GTAW welding consumables manufacturing process that is cost-, time-, and resource-efficient.
The study was conducted in accordance to Design For Six Sigma (DFSS), a product developing process under the overarching scientific engineering method “Systems Engineering”. The DFSS process consists of four stages, where stage one and two where completed.
This has resulted in the development of three novel concepts, which fulfils the requirements of LNU WML, and key parameters have been established. One novel concept has been selected by the client and an experiment has been conducted that proves the principle functionality of the novel concept.
Recommendations on how to proceed to a finished, usable product is expressed as a
recommendation of a continuation of the DFSS-process, to conduct a secondary
experiment and suggestions for key parameters.
Sammanfattning
Linneuniversitetet (LNU) Welding Mechanics Laboratory (WML) genomför för närvarande det marina arkitektoniska forskningsprojektet ”R/S Snöfrid”, varav detta fartyg ska byggas i det höghållfasta stålet SSAB Strenx 1300. De olika sektionerna ska förenas med bågsvetsningsprocesser. Inget lämpligt svetstillsatsmaterial är kommersiellt tillgängligt, så ett nytt svetstillsatsmaterial måste utvecklas.
Nuvarande forsknings- och utvecklingsprocesser för nya svetstillsatsmaterial utgör en stor kostnad och kräver mycket tid och resurser. LNU WML har begärt en ny tillverkningsprocess för småskalig tillverkning av TIG-svetstillsatsmaterial som är kostnads-, tids- och resurseffektiv.
Studien genomfördes i enlighet med Design For Six Sigma (DFSS), en
produktutvecklingsprocess under den övergripande vetenskapliga ingenjörsmetoden
”Systems Engineering”. DFSS-processen består av fyra steg, varav ett och två slutfördes.
Detta har resulterat i att tre nya koncept har utvecklats, som uppfyller kraven specificerade av LNU WML, och nyckelparametrar har fastställts. Ett koncept har valts ut av klienten och ett experiment har genomförts som bevisar det nya konceptets principiella funktionalitet.
Rekommendationer om hur man går vidare till en färdig, användbar produkt
uttrycks som en rekommendation om en fortsättning av DFSS-processen,
genomförande av ett sekundärt experiment och förslag på nyckelparametrar.
Abstract
Current research and development processes of novel welding consumables constitutes a great cost and requires a lot of time and resources. LNU WML has requested a novel small batch GTAW welding consumables manufacturing process that is cost-, time-, and resource-efficient.
The study was conducted in accordance to DFSS, a product developing process under the overarching scientific engineering method “Systems Engineering”. The DFSS process consists of four stages, where stage one and two were completed.
This has resulted in the development of three novel concepts, and key parameters has been established. One novel concept has been selected by the client and an experiment has been conducted that proves the principle functionality of the novel concept.
Recommendations on how to proceed to a finished, usable product is expressed as a recommendation of a continuation of the DFSS-process, a conduction of a
secondary experiment and suggestions for key parameters.
Keywords: welding consumables, GTAW, manufacturing, novel process,
Acknowledgments
This study is conducted as a degree project and final assignment for the university education in Mechanical Engineering at LNU. The scope of the study was first formulated by LNU WML which requested a novel welding consumables
manufacturing process, that is time-, cost-, and resource-efficient, and that can aid in the research and development (R&D) which occurs there. As the author has close to a decade of work experience as an international welder (IW), and had expressed a desire for the Bachelor of Science to involve; product development, welding and solving “real world problems”, this where an excellent match.
Ordinary, Bachelor of Science are co-authored, this study being the exception. The decision was validated as correct as the Covid-19 restrictions would prevent and/or limit the benefits of co-authoring and would instead render it a liability. The technical report “Structural analysis of critical parts of a novel small batch manufacturing process of welding consumables and rolling mill” is referenced and used in this study. The technical report was made as an assignment for the “Solid Mechanics – Advanced course”. The report is co-authored with Maja Håkansson, and for this she has the author’s thanks, praise and gratitude.
The author would like to thank LNU WML for the interesting and educational assignment. The author wishes to extend a special thanks to supervisor Per Lindström, who went above and beyond the line of duty, to support in this endeavour.
The author would like to thank Meltolit AB and AXSON Teknik AB.
The author wishes to thank Oskar Cederfjord for his insight and advise in his role as an opposing group.
Finally, I would like to thank my family and friends, whose support made this possible.
Jonathan Fager Stark Älmhult, Date 2021-05-17
Table of contents
1 Introduction 1
1.1 Background 1
1.2 Problem definition 2
1.3 Purpose and goal 2
1.4 Hypothesis 2
1.5 Research questions 3
1.6 Limitations 3
1.7 Research Method 3
1.8 Relevance for the society 3
2 Nomenclature 4
2.1 Introduction 4
2.2 Acronyms 4
2.3 Nomenclature 4
3 Method 6
3.1 Systems Engineering 6
3.1.1 Design For Six Sigma 6
3.2 Research design 8
3.3 Validity and reliability method 9
3.4 Research quality and ethics 9
4 Identification of client’s requirements 10
4.1 Acceptance criteria of welding consumables 10
4.2 GTAW 10
4.3 Voice of the customer/Acceptance criteria 12
4.4 Quality function deployment (QFD) 13
4.5 Preliminary concepts 16
4.6 Risk assessment 16
4.6.1 Design FMEA 16
4.7 Stage Gate 1 16
5 Stage 2 DFSS, design 17
5.1 Concept development 17
5.1.1 Concept 1 18
5.1.2 Concept 2 19
5.1.3 Concept 3 21
5.2 Evaluation and selection of concept 23
5.3 Experiment - Proof of concept 24
5.3.1 Experiment 24
5.4 Preliminary designs 26
5.5 Structural integrity analysis of the mould 26
5.5.1 Pressure calculations on mould 26
5.5.2 Mould extraction force 27
5.6 Thermodynamic analysis of the mould 28
5.6.1 Transition thickness 28
5.6.3 Heat transfer of a turbulent boundary layer in a forced fluid 34
6 Results 35
6.1 Acceptance criteria 35
6.2 Selected concept 36
6.3 Results of experiment 37
6.4 Identification of key parameters 38
6.5 Construction recommendations 38
7 Discussion 39
7.1 Method discussion 39
7.2 Acceptance criteria 39
7.3 Concept selection 40
7.4 Experiment 40
7.5 Construction recommendations 41
8 Conclusion 42
8.1 Conclusions of the study 42
9 Recommendations 43
9.1 Recommendation to LNU WML 43
10 Reference list 44
11 Appendices 48
Appendix 1, Design FMEA 1
1 Introduction
This chapter presents an overview of research and development of welding consumables and why it is of interest.
1.1 Background
Current research and development (R&D) process of novel experimental welding consumables constitute a great cost, is time consuming, requires a lot of resources and is labour intense [1] - [2][3].
The conventional R&D procedure is characterized by an experienced trial and error approach that is grounded in the principles of physics, chemistry and metallurgy [3]
- [4]. Manual Metal Arc (MMA) welding electrodes are used for test welding of novel alloys [1][5].
The MMA electrodes consist of two main components, a solid metal core and a flux that covers the core. The flux has several key functions; stabilizing the arc, reduce spatter, create slag, create gas, protect the melt pool – and add alloying elements to the melt pool and consequently aid in controlling the chemical composition and the microstructure of the finished alloy [6].
The desired chemical composition of the welded novel alloys is adjusted by adding alloying elements to the flux that covers the core of the MMA electrodes [1].
Changing the chemical composition of the flux, instead of the core of the MMA electrode, is both more easily implemented and beneficial from a cost perspective.
The flux chemical composition, which is measured in percentage per weight unit, is mixed to the desired admixture, and is pressed onto the metal core, forming a complete MMA electrode ready for test welding [7]. The result of the interaction between the flux, solid metal core and the welding process is hard to predict – hence the requirement of numerous trials [8] - [9].
When the desired chemical composition of the welded alloy, regarding operational and mechanical properties, has been achieved, a bigger production batch is
produced.
The raw material and the cost of heating the raw material to its melting point constitutes a high cost, as the batches of melted material often is in the weight range of 500-1000kg [2], [10]. In addition, there is a need for a wire drawing plant with a capacity to reduce the wire diameter from ů10m to 3,2mm, in the case of gas tungsten arc welding (GTAW) welding rods [2]
In practice, this narrows the number of companies and groups that have the economic capacity to operate R&D of novel welding consumables to but a few.
High research costs can therefore stifle creativity and innovation pace [8]. At the
same time, there is a high demand for new alloys in industries such as oil and gas,
aerospace, refining and petrochemical processing, chemical processing, thermal
trends for requirements for novel welding consumables are higher strength, higher temperature resistance, higher toughness, cleaner, more energy efficient, more environment friendly, higher efficiency and enable higher levels of automatization [12].
Reduced overall cost, resources, and time, at development of novel welding consumables is desirable [8], [10]. Several approaches have been tried in service of the aforementioned desire, for example, but not limited to;
I. Reduction of number of welding tests required through employment of a novel welding design and optimization methodology [8].
II. Statistical modelling tool to predict welding flux design with higher accuracy [2].
III. Batch size reduction [9].
1.2 Problem definition
Linnaeus University (LNU) welding mechanics laboratory (WML) has received a request to develop a novel welding consumable for the new high strength steel of 1300 MPa, namely SSAB Strenx 1300.
LNU WML does currently not possess a method or process for in-house small-scale production of small batches of experimental welding consumables, nor is there a product commercially available for this purpose.
Therefore, a need has locally emerged at LNU WML for a novel process that allows for a time-, resource- and cost-efficient small-scale production of experimental welding consumables to develop a novel welding consumable for the new high strength steel.
1.3 Purpose and goal
The purpose of this project is to identify, pursue a greater understanding and
propose time-, resource- and cost-efficient solution for small-scale manufacturing of experimental welding consumables along with the acceptance criteria for the novel process.
The ultimate goal of this study is to generate and propose at least one novel concept which can be used for small-scale manufacturing of experimental welding
consumables at LNU WML.
1.4 Hypothesis
The present author has postulated the following hypothesis:
It is possible to manufacture small batches (0,5 – 1,0 kg) GTAW welding
consumables by the use of a novel process.
1.5 Research questions
The main research question is:
I. How can one proceed at small batch manufacturing of Ø 3,20 mm GTAW welding consumables?
To be able to answer the main research question, the following questions has to be answered:
I. What are the acceptance criteria?
II. How shall this novel manufacturing process be constructed?
1.6 Limitations
This study is limited by and/or to:
- Stage Gate 2 in the design for six sigma (DFSS) process.
- GTAW single torch as a heat source.
- Inert atmosphere.
- Time-, resource- and cost-efficient.
- Small batches defined as about 0,5 – 1 kg.
- Covid-19.
- No cost or time calculation for developing proposed novel process.
1.7 Research Method
The research method used is Systems Engineering which is an interdisciplinary field of engineering that gives a methodology in how to approach complex systems, and design for a product’s whole life cycle. Systems engineering takes on a holistic approach, as it focuses on the system as a whole instead of the individual parts that make up the system [13].
1.8 Relevance for the society
This study is expected to primarily contribute with a novel process that enable small-scale production of experimental welding consumables for LNU WML and aid in the R&D that occurs there.
Secondary this study is expected to contribute with a process that drastically reduces the time-, resource- and costs needed for R&D for novel welding consumables for society in general.
An expected consequence of the result of this study is a possibility of increase of
innovation pace in welding consumables and in extension, that of the industries that
currently labour under a deficit of suitable welding consumables.
2 Nomenclature
2.1 Introduction
This chapter provides the acronyms and nomenclature used in this Bachelor of Science.
2.2 Acronyms
DFFS Design For Six Sigma
FMEA Failure Modes and Effects Analysis
FR Functional Requirement
GTAW Gas Tungsten Arc Welding
IEEE The Institute of Electrical and Electronics Engineers ISO International Organization for Standardization
IW International Welder
LNU Linnaeus University
MMA Manual Metal Arc
QFD Quality Function Deployment
R&D Research and Development
SG-1 Stage Gate 1
SG-2 Stage Gate 2
VOC Voice Of the Customer
WML Welding Mechanics Laboratory
2.3 Nomenclature
A Area [m
2]
Cp
fSpecific heat capacity, constant pressure [J/kg∙°C]
c Specific heat capacity [kJ/kg∙°C]
d Diameter [m]
F Force [N]
F
sFriction force [N]
g Gravity of earth [m/s
2]
h Height [m]
I Current [A]
k
fThermal conductivity [W/m∙ °C]
m Mass [kg]
N Normal force [N]
Nu Nusselt´s number [ - ]
Nu
dNusselt´s number for tube [ - ]
P Pressure [N/m
2]
P
ePower [kW]
P
hHydrostatic pressure [N/m
2]
P
rPrandtls number [ - ]
P
wWeld heat power [W]
Q
wWeld heat input [J/mm]
Re
dReynold´s number [ - ]
r Radius [m]
T
0Temperature [°C]
T
1Temperature [°C]
T
2Temperature [°C]
t Thickness [m]
t
kTemperature cold fluid [°C]
t
vTemperature warm fluid [°C]
U Current [V]
V
fFlow velocity [m/s]
v Velocity [mm/s]
ϴ
mLog mean temperature [°C]
α
fCoefficient of heat transfer [W/m
2∙°C]
δ
3DTransition thickness [mm]
η Thermal efficiency factor [ - ]
µ Friction coefficient [ - ]
ν
fKinematic viscosity [m
2/s]
ρ Density [g/mm
3]
σ
1Hoop stress [MPa]
ṁ Mass flow [kg/s]
3 Method
This chapter describes the scientific engineering methodology used in the study.
3.1 Systems Engineering
This study is executed by the well-established scientific engineering DFSS [14]
process as defined by:
I. International Council on Systems Engineering Website (INCOSE)[13],[15].
II. National Aeronautics and Space Administration (NASA) [16] - [17].
III. American Society of Mechanical Engineers (ASME) [18].
DFSS constitutes a branch of the overarching engineering method “Systems
Engineering”. This study uses engineering-accepted reference format: “The Institute of Electrical and Electronics Engineers” (IEEE) [19].
3.1.1 Design For Six Sigma
DFSS process is a branch of the well-established scientific engineering method
“Systems Engineering”, a process comprising of interdisciplinary knowledge areas and their mutual relations. The knowledge areas consist of empirical, statistical, mathematical models, categorizations of phenomena, and propositions and
hypotheses and more. Through understanding of different knowledge areas and their different relations, predictions of observations can be made [14].
The prediction of observation is the DFSS process. The DFSS process is used in product development, where its major objective is to “design it right the first time”, this to avoid costly experiences [14]. DFSS process is illustrated in Figure 1.
Figure 1, The DFSS process illustrated, author’s contribution [20].
The DFFS process consists of four stages [20]:
1. Identify 2. Design 3. Optimize 4. Validate Stage 1, Identify.
The main purpose of the first stage of the DFSS process is to identify the VOC and translate it into Functional requirements (FR). This stage establishes the
Requirements for the product development process. For this to occur, the following must be realized:
• To validate the business case.
• Identification of the customers, and identification and prioritization of their needs.
• Review and assess applicable technologies.
• Derive functional requirements to fulfil critical customer requirements.
Stage 2, Design.
The purpose of this stage is to generate and select concepts, in the Conceptual Design, that fulfil the VOC. In the Preliminary Designs the concepts are further explored through identification of key parameters and development of layouts. For this to occur, the following must be realized:
• The technical and functional requirement for systems and subsystems must be established.
• Concepts that fulfil the technical and functional requirements must be developed and selected.
• The preliminary design is prepared.
• Evaluate the capabilities of the supporting processes.
Stage 3, Optimize.
The purpose of this stage is to optimize input factor in relation to cost, quality and time-to-market. For this to occur, the following must be realized:
• Finalization of the transfer functions.
• The design tolerances need to be established.
• The detailed design shall be prepared.
• The system and subsystems shall be tested through simulation, prototyping etc, in relation to previously set requirements.
• The design shall be pre-launch validated.
Stage 4, Validate.
The purpose of this stage is to demonstrate that the design satisfies the requirements derived from the VOC. For this to occur, the following must be realized:
• A transition plan for qualification to productions must be implemented.
• Implement design and process controls.
• Gather data which support that the novel process/product satisfies the customers.
3.2 Research design
The research design starts with identification of the problem definition through conversation with supervisor Per Lindström. A brief overview of the relevant literature is conducted so that the purpose of the study, the goal and the research questions can be specified with precision. A more in-depth examination of the relevant literature is conducted and is continuous through the next steps in the research design. The methodology DFSS is chosen. DFSS is a product development methodology that is so comprehensive, that to fulfil the time requirements of this study, only stage one and two are completed.
In stage one in the DFSS process the requirements are identified. The requirements are identified through an establishing of a VOC [14]. Secondary data, literature, and primary data, dialog with client, will provide the data needed for the establishing of the VOC.
The VOC is translated into functional and requirements using a Quality Function Deployment (QFD).
In stage two, from these requirements, concept alternatives that fulfils the
requirements are generated. Furthermore, the concepts are evaluated, presented to client LNU WML, and one is selected. Key parameters are identified. An
experiment is constructed that acts as proof of concept and stage two of the DFSS
process, the preliminary design, is completed [14]. The research design ends with an
analysis and discussion of the results, see Figure 2.
Figure 2, Research design.
3.3 Validity and reliability method
Validity is a term that describes the relevance of a measurement or observation in relation to what is intended to be examined. It is also a term that describes the context for which a measurement or observation is valid [21].
Reliability is a term that describes consistency of a measurement and its repeatability [21].
Both traits are highly desirable in a study, and steps have been taken to assure both high validity and reliability are reached. To achieve high validity through this study, verifiable sources and peer-reviewed references are used, documented, and
displayed in the reference list in accordance to IEEE. For high reliability, tests, calculations, and experiments are thoroughly documented to ensure its repeatability by an external part.
3.4 Research quality and ethics
This study has been done in accordance to the research ethics stipulated by Swedish Research Council [22].
Identification of problem
Overview of relevant literature
Specify purpose, goal, and research
questions
Choice of methodology
Identify and specify functional
requirements
Generate concepts
Evaluate and select concept
Proof of concept, conduct experiment
Analysis and diskussion of
results
4 Identification of client’s requirements
This chapter is stage 1 in the DFSS-process. It identifies the VOC and provides the information for and presents the requirements for the stage 2, Design phase.
The applicable technologies allowable are restricted to GTAW by the limitation set in this Bachelor of Science by request of LNU WML.
4.1 Acceptance criteria of welding consumables
Welding consumables are the material added to the welding process. There exists a great variety of welding processes so there also exists a great variety in welding consumables. Solid wire, powder, flux covered electrodes to name a few [23]. As the welding consumables play a key part in the quality of the finished weld, the manufacturing and quality of welding consumables are regulated through international standards, such those composed by ISO [24], see Figure 3.
Figure 3, Standards regarding manufacturing of welding consumables, author’s contribution [24].
Nevertheless, the aforementioned requirements are not applicable to this case as the process shall be used for manufacturing of experimental welding consumables for client LNU WML only, with the exception stated in the VOC.
4.2 GTAW
Gas tungsten arc welding (GTAW) is a welding process where an arc is generated, in an inert atmosphere created by a shielding gas, between a non-consumable tungsten electrode and the workpiece. The tungsten electrode is in the torch, and the torch is moved in the desired direction during welding. Consumable electrodes are feed into the melt pool created by the arc either manually or automatically, see Figure 4. The welding process positive characteristics are [25]:
• High quality of weld goods.
• Concentrated arc.
• Easy to automate.
• Easy to control heat input into base material.
• No welding slag or spray.
GTAW, if automated, can be used in additive manufacturing, where it has a high deposition rate and a low cost as a positive, and high heat input and residual stresses as a negative [26][27].
The weld heat power, 𝑃
𝑤, is the power inserted into the welding process, the current, U, and amperage, I, multiplied with the thermal efficiency factor. Weld heat power can be calculated with Equation 1 [28]:
Equation 1 𝑃
𝑤= 𝐼 ∙ 𝑈 ∙ 𝜂
Weld heat input, 𝑄
𝑤, is weld heat power per mm weld, v , and is calculated with Equation 2 [29]:
Equation 2 𝑄
𝑤=
𝐼∙𝑈∙𝜂𝑣Figure 4, Principal parts of GTAW, author’s contribution [30].
4.3 Voice of the customer/Acceptance criteria
The following requirements have been specified by LNU WML for the novel process:
• Produce novel welding consumables.
o Shall be able to produce novel welding consumables.
o Called “Novel consumables” in QFD.
• GTAW.
o The novel process shall use GTAW as heat source as a Tetrix 551 Synergic is available at LNU WML [31].
o Called “GTAW” in QFD.
• Small batches (≈0,5-1kg).
o It shall be able to produce batches of minimum ≈0,5-1kg.
o Called “Small batches” in QFD.
• Ø 3,2 mm rod for GTAW.
o Shall produce a rod of Ø 3,2 mm for GTAW.
o Called “Ø 3,2 mm rod” in QFD.
The following requirements are specified by the International Organization for standardization (ISO):
• The process shall enable uniform production [32].
o The process shall enable uniform production of novel welding consumables, regarding chemical composition, dimensions etc.
o Called “Uniform” in QFD.
In addition, more generalized requirements are:
• Time efficient.
o The process shall be more time efficient than the conventional process.
o Called “Time efficient” in QFD.
• Cost efficient.
o The process shall be more cost efficient than the conventional process.
o Called “Cost efficient” in QFD.
• Resource efficient.
o The process shall be more resource efficient than the conventional process.
o Called “Resource efficient” in QFD.
These requirements are added into the QFD, under the “What’s”, and each requirements are assigned a number for its importance.
4.4 Quality function deployment (QFD)
QFD is a method to, in the early stages of product development, set engineering specifications. QFD do, in an organized way, help develop crucial pieces of information, such as [33];
• The VOC.
• Specify and develop specification and goals.
• Finding out how the specifications measure customer’s desires.
• Benchmark the competition.
• Develop numerical target to work towards.
The QFD method consists of different steps, and each step make up one of the rooms in the house of quality that is constructed as part of the QFD method, see Figure 5. The steps, and their corresponding room as displayed in Figure 5, are [33]:
1. Customer identification – Who.
2. Determine the customer’s requirements – What.
3. Determine relative importance of the requirements – Who versus What.
4. Identify and evaluate the competition – Now.
5. Generate engineering specifications – How.
6. Relate customers’ requirements to engineering specifications – What versus How.
7. Set targets and importance to engineering specification – How much.
8. Identify relationships between engineering specifications – How.
The VOC covers steps 1-2 wholly, and 3-4 partly. The relative importance of the
customers’ requirements a given a numerical value and entered the QFD. The
competition is benchmarked, and it is the conventional process for R&D. In step 5
engineering specifications are generated, functional characteristics of the novel
process.
Figure 5, House of quality, author’s contribution [33].
The following product characteristics have been identified as FR´s – How, see Figure 6:
• Controlled distribution of the alloying elements.
o The alloying element of the novel welding consumable shall easily and reliably be distributed.
o Called “Distribution” in QFD.
• Controlled melting of the alloy.
o The melting of the novel welding consumable shall easily and reliably be controlled.
o Called “Melting” in QFD.
• Produce an ingot/rod of a suitable size.
o The process shall be able to produce an ingot/rod of suitable size.
o Called “Ingot” in QFD.
• Cross contamination control.
o The process shall not cross contaminate the finished novel welding consumable to such an extent that it is rendered unusable.
o Called “Contamination” in QFD.
• Shaping of novel alloy.
o The novel alloy shall be shaped to desired geometry.
o Called “Shaping” in QFD.
• Inert atmosphere.
o The process shall occur in an inert atmosphere.
o Called “Inert” in QFD.
The relationship between the requirements and the FR`s are explored and evaluated
in step 6. Step 7 is not applied. In step 8 the relationship between the FR´s are
explored.
4.5 Preliminary concepts
From the VOC and QFD, three preliminary concepts are created.
Preliminary concept 1 is constituted of a movable GTAW torch which melts added alloying elements into a melt pool in a mould. When the melt pool reaches the size of 0,5-1 kg, the mould’s content is poured into a cast, forming ingots during its solidification.
Preliminary concept 2 is constituted of a fixed GTAW torch which melts added alloying elements into a melt pool in a mould. The melt pool is continuously moved downward, through a hole in the mould. From this hole, an ingot is extracted.
Preliminary concept 3 is constituted of a fixed GTAW torch which melts added alloying elements into a melt pool in a mould. When the melt pool reaches the size of 0,5-1 kg, the melt pool is continuously moved downward, through a hole in the mould. From this hole, a wire is extracted.
4.6 Risk assessment
The risk assessment is done to illustrate the risks associated with the different concept ideas. The risk assessment is done by a design Failure Modes and Effects Analysis (FMEA).
4.6.1 Design FMEA
A FMEA is a method to identify weaknesses, their causes and their implications in a product or process. FMEA consist of 5 steps [34]:
1. Identify the affected function.
2. Identify failure modes.
3. Identify the effect of the failure.
4. Identify the failure causes.
5. Identify the corrective actions.
Appendix 1 showcase the performed design FMEA for the 3 novel concept ideas.
4.7 Stage Gate 1
At the Stage Gate 1 (SG-1) meeting the three preliminary concepts are presented and proposed to the head of LNU WML, Per Lindström February 2021.
The three preliminary concepts are approved for further stages of concept
development.
5 Stage 2 DFSS, design
This chapter is the Conceptual Design step of the DFSS-process, where concepts are developed and evaluated.
5.1 Concept development
The concepts are developed with the aid of:
• Literature.
• The QFD.
• Experts, supervisor Per Lindström.
• The author’s education and work experience as an IW.
The concepts fulfil the requirements set in the VOC and possess the necessary FR´s.
5.1.1 Concept 1
This concept consists of 7 essential parts, see Figure 7:
1. GTAW torch – which provides the heat source.
2. Wolfram electrode – non consumable.
3. Arc – which controlls the melting of the alloy.
4. Rods with alloying elements – which controlls the distrubution of the alloying elements.
5. Molten material – new alloy.
6. Mould – which contains the melt pool.
7. Cast – which forms the ingots during is solidification.
Figure 7, Concept 1.
The concept is a direct welding to casting process, where the alloying elements will go from;
I. Solids to a liquid alloy to,
II. Homegenized solid ingot of desireable diameter.
The process starts with an arc being generated between the wolfram electrode and
the mould. The arc creates a melt pool which rods with alloying elements are fed
into. The rods have different alloying elements and when mixed in the melt pool,
generates a new alloy. The melt pool expands with the continously feed alloying
elements untill the melt pool reach 0,5-1kg and the mould is full. The molten steel is
poured into the cast, where ingots are formed. The ingots need to be processed in a
hot rolling mill to achive the correct diameter.
5.1.2 Concept 2
The concept consist of 8 essential parts and are depicted in Figure 8:
1. GTAW- torch, - which provides the heat source.
2. Wolfram electrode – non consumable.
3. Arc – which controlls the melting of the alloy.
4. Rods with alloying elements – which controlls the distrubution of the alloying elements.
5. Mould – which produces an ingot of suitable size.
6. Molten material – new alloy.
7. Starting cylinder – the process is started on a cylinder of same size and geometry as the hole in the mould.
8. Screw – which rotates downwards and therefore controlls the distribution of the alloying elements as well as extracts the ingot from the mould.
Figure 8, Concept 2.
The concept is a direct continuous casting process, where the alloying elements will go from;
I. Solids to a liquid alloy to,
II. Homegenized solid ingot of a desireable diameter.
The process starts with an arc being generated between the wolfram electrode and the start cylinder. The arc creates a melt pool which rods with alloying elements are feed into. The rods have different alloying elements and when mixed in the melt pool, generates a new alloy. The melt pool diameter is controlled by the mould´s hole diameter. The mould do cool the melt pool, as the mould itself is cooled by internal cooling ducts. The start cylinder is attached to a screw which by its
downward rotating motion, continuously extracts the ingot out of the mould. As the
process progress, the cross contamination of the start cylinder into the new alloy
ingot decrease as the melt pool moves further away from the start cylinder.
5.1.3 Concept 3
The concept consist of 8 essential parts and are depicted in Figure 9.
1. GTAW- torch, – which provides the heat source.
2. Wolfram electrode – non consumable.
3. Arc – which controlls the melting of the alloy.
4. Rods with alloying elements – which controlls the distrubution of the alloying elements.
5. Mould – which has internal cooling ducts, and ability to produces a wire of Ø 3,2 mm.
6. Molten material – new alloy.
7. Mixing chamber – added alloying elements are mixed in this chamber.
8. Bottom plug – the bottom plug extracts Ø 3.2mm wire out of the mixing chamber.
Figure 9, Concept 3.
The process starts with an arc being generated between the wolfram electrode and the mixing chamber. The arc creates a melt pool which rods with alloying elements are feed into. The rods have different alloying elements and when mixed in the melt pool, generates a new alloy. The melt pool diameter is controlled by the mixing chambers diameter. The mixing chamber is insulated to mitigate heat losses, and to allow for a large melt pool to form (≈0,5-1kg). When the melt pool is of such a size as to ensured proper mixing of alloying elements, the bottom plug is released. The plug is pulled downward while simoultaniously being roteted, shearing itself lose.
The plug extracts a Ø 3.2mm liquid alloy out of the mixing chamber and into the
mould, where the alloys solidification begins. The alloy is cooled by the mould, and
the mould itself is cooled by its internal cooling duct. As the process progress, the
cross contamination of the start plug into the new Ø 3.2mm wire alloy decrease as
the melt pool moves further away from the bottom plug.
5.2 Evaluation and selection of concept
The concepts are evaluated in accordence to their ability to satisfy the requirements set in the VOC. The previous QFD is expanded and the proposed concepts are added, evaluated and are benchmarked against the conventional process, see Figure 10.
Figure 10, QFD expanded with the concepts rated against the requirements.
Concept 3 do fulfil the requirements of the VOC most accurately. This is confirmed
when the concepts are presented to client LNU WML, as they do favour Concept 3.
5.3 Experiment - Proof of concept
This chapter provides an overview of the experiment done as a proof of concept for the proposed novel process, concept 3.
5.3.1 Experiment
The experiment consists of 9 essential parts, as illustrated in Figure 11. The parts are:
1. Molten tin.
2. Copper pipe.
3. Still water.
4. Container.
5. Ingot.
6. Glue.
7. Drill.
8. Protective plate.
9. Drilling machine.
The container is prepared by drilling a hole in it where the copper tube is inserted, and the copper tube is held in place by glue. The container is filled with water until it envelops the copper tube but for the top 5mm. From the underside a drill of the same diameter as the copper tube’s inner diameter is inserted into the copper tube.
The drill is retracted 15mm from the other end of the tube to give room for the
solidification of an ingot. The drill has a protective plate attached to it and the drill
is attached to a drilling machine that can rotate the drill. The flutes of the drill are
resulting in a stronger bound between the drill and the ingot, aiding it in shearing
and extracting the ingot from the copper tube.
Figure 11, Proof of concept experiment.
The experiment will evaluate and/or measure:
I. Solidification to ingot.
II. Extraction method.
III. Ingot length.
The experiment is conducted by the following steps:
I. The drill is inserted into the copper tube.
II. The drill starts to rotate.
III. The tin is heated to its melting point.
IV. The tin is poured into the copper tube.
V. The drilling machine is manually lowered at a speed which matches that by
which the tin is poured.
5.4 Preliminary designs
This chapter provides an overview of the expected thermal and mechanical loads the mould is subjected to. The structural integrity analysis chapter presented here is based on the calculation method presented in the technical report “Structural analysis of critical parts of a novel small batch
manufacturing process of welding consumables and rolling mill”[35]. All calculations are done in Excel.
5.5 Structural integrity analysis of the mould
5.5.1 Pressure calculations on mould
Hydrostatic pressure is the pressure at any given point in a static fluid and is expressed as Equation 3.
Equation 3 𝑃
ℎ=
𝑚∙𝑔𝐴
The mould is exposed to a pressure from the liquid melt pool. Maximum pressure is located at the bottom surface area. The model for approximation of the loads the mould is subjected to, utilizes the maximum pressure as consistent throughout the mould, see Figure 12.
Figure 12, Model of the pressure on the mould.
The pressure is regarded as uniform throughout the form, and as the geometry of the mould is of a tube, the mould is regarded as a pressure vessel and Equation 3- Equation 4 is used to calculate the stresses in the mould caused by the pressure of the molten metal.
Equation 4 𝜎
1=
𝑃∙𝑟𝑡
The essential data for making calculations on the mould is presented in Table 1.
Table 1, Mould as pressure vessel - essential data.
Density of molten steel, 7040 kg/m
3Height of mould, h 0,05 m
Gravity of earth, g 0,05 m
Radius of mould, r 0,03m
Thickness of walls, t 0,02 m
Pressure, P 0,1 MPa
Hoop stress, 𝝈
𝟏0,005 MPa
5.5.2 Mould extraction force
The casting process involves a screw in the bottom of the ingot, that rotates and pulls down the ingot, see Figure 13. This process gives rise to a friction force between the ingot and the mould, and can be expressed as Equation 5:
Equation 5 𝐹
𝑠= 𝑁 ∙ 𝜇
The contact area and the required extraction force is calculated with Equation 6- Equation 7.
Equation 6 𝐴 = 𝑑 ∙ ℎ ∙ 𝜋
Equation 7 𝐹
𝑟𝑒𝑞𝑢𝑖𝑟𝑒𝑑= 𝜇 ∙ 𝑃 ∙ 𝐴
Figure 13, Extraction force of ingot
The essential data for calculating the required extraction force is presented in Table 2.
Table 2, Required extraction force - essential data.
Pressure, P 0,1 MPa
Diameter of mould, d 0,1 m
Height of mould, h 0,05 m
Friction coefficient, 𝜇 0,15
Contact area, A 0,00942 m
2Extraction force, F
required4,9 N
5.6 Thermodynamic analysis of the mould
The mould’s surface is in direct contact with the molten steel and the solidified ingot, rendering a heat transfer to occur between the surfaces. To secure that the mould doesn’t overheat and melt, cooling ducts in the mould are needed, see Figure 14. The cooling ducts need to transfer the same amount of heat that the GTAW inputs. This chapter will provide an overview of the dimensioning and placement of the cooling ducts. All calculations are done in Excel.
5.6.1 Transition thickness
It is of crucial importance that the heat can transfer from the melt pool, through the walls of the mould and into the water in the cooling ducts, therefore a 2D heat source approach is desired.
Figure 14, Cooling ducts in the mould.
If a 2D or 3D heat source approach should be used is determined by the transition thickness, so the placement of the cooling ducts is determined by the transition thickness. The transition thickness is approximated by Equation 8:
Equation 8 𝛿
3𝐷≥ √
2∙𝜌∙𝑐𝑄𝑤∙ (
1𝑇2−𝑇0
−
1𝑇1−𝑇0
)
The transition thickness approximation is illustrated in Figure 15.
Figure 15, An approximation of when a 2D or 3D approach can be applied.
5.6.2 Heat transfer in heat exchanger
Heat exchangers are a system that allows for heat transfer to occur between two fluids, for example molten metal and water. This can be used for both heating and cooling. The heat exchanger’s effect can be calculated with Equation 9 [35]:
Equation 9 𝑃
𝑒= 𝐴 · 𝑘 · 𝛳
𝑚In Equation 9, A, is the heat transferring area, k, is the coefficient of heat transfer and ,ϴ
m,¸is the logarithmic mean temperature.
The temperature of the fluids in a heat exchanger differs dependent on its location in the heat exchanger as the fluid is either warmed or cooled. The fluids temperature difference is greatest if measured at the inlet and outlet. ϴ´ and ϴ´´ are the
temperature difference between the different fluids at the inlet and outlet of the heat exchanger [35]. The logarithmic mean temperature can be approximated with Equation 10.
Equation 10 𝛳
𝑚=
𝛳′−𝛳´´𝑙𝑛𝛳
𝛳´´
The water in the cooling duct face a boiling risk if the heat transfer is too great. The temperature of the cooling duct outlet is approximated with Equation 11 [35]:
Equation 11 𝑃
𝑒= ṁ
𝑘· 𝑐
𝑘· (𝑡
𝑘2− 𝑡
𝑘1)
With the flow velocity of the water, V
f,set to 1m/s, and 5 different sized cooling ducts, and Equation 11, Figure 16 is constructed.
Figure 16,Temperature of cooling water at outlet at 1 m/s flow velocity.
19 20 21 22 23 24 25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Temperature of water at outlet (°C)
Effect (kW)
Outlet temperature (P, d)
Ø 0,01m Ø 0,02m Ø 0,03m Ø 0,04m Ø 0,05m
The approximate temperature of the wire at the point of extraction from the mould is calculated with Equation 12 [35]:
Equation 12 𝑃
𝑒= ṁ
𝑣· 𝑐
𝑣· (𝑡
𝑣1− 𝑡
𝑣2)The temperature is calculated for different wire extraction speeds and cooling effects and is illustrated in Figure 17.
Figure 17, Wire extraction temperature.
0 200 400 600 800 1000 1200 1400 1600 1800 2000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Temperature of wire at extraction (°C)
Cooling effect (kW)
Wire extraction temperature (P, V
fv)
100 mm/min 200 mm/min 300 mm/min 400 mm/min 500 mm/min 600 mm/min 700 mm/min 800 mm/min 900 mm/min 1000 mm/min
The logarithmic mean temperature, ϴ
m, is approximated for different wire extraction speeds, with a cooling duct Ø 0,01m and a flow velocity of 1m/s, with Equation 10, and is illustrated in Figure 18.
Figure 18, Logarithmic mean temperature at flow velocity 1 m/s and cooling duct of Ø 0,01m.
500 700 900 1100 1300 1500 1700 1900
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 ϴm
Effect (kW)
Log mean temperature (P, V
fv)
100 mm/min 200 mm/min 300 mm/min 400 mm/min 500 mm/min 600 mm/min 700 mm/min 800 mm/min 900 mm/min 1000 mm/min
Equation 9, the logarithmic mean temperatures for different wire extraction speeds, with a cooling duct Ø 0,01m and a flow velocity of 1m/s, is used to approximate and illustrate the required thermal conductivity in Figure 19.
Figure 19, Required thermal conductivity for Ø 0,01 cooling duct with 1 m/s cooling water flow rate.
0 5000 10000 15000 20000 25000 30000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Coefficent of heat transfer (W/m2∙°C)
Effekt (kW)
Required thermal conductivity (P, V
fv)
100 mm/min 200 mm/min 300 mm/min 400 mm/min 500 mm/min 600 mm/min 700 mm/min 800 mm/min 900 mm/min 1000 mm/min
5.6.3 Heat transfer of a turbulent boundary layer in a forced fluid
The heat transfer coefficient of a turbulent boundary layer of a fluid, 𝛼
𝑓is calculated with the aid of several equations. Prandtl number, Pr ,which describes the thermal properties of a boundary layer, is needed and is calculated with Equation 13 [28]:
Equation 13 𝑃𝑟 =
𝜈𝑓∙𝜌𝑓∙𝐶𝑝𝑓𝑘𝑓
As is Reynold´s number, 𝑅𝑒
𝑑, which describes the kinematic boundary layer, which is calculated with Equation 14 [28]:
Equation 14 𝑅𝑒
𝑑=
𝑉𝑓∙𝑑𝜈𝑓
When these numbers are known, Nusselt´s number, 𝑁𝑢
𝑑, which describes the sub- viscous layer where the heat transfer occur, can be calculated with Equation 15 [28]:
Equation 15 𝑁𝑢
𝑑=
0.038∙𝑅𝑒𝑑3 4∙𝑃𝑟 1+1.5∙𝑃𝑟−16∙𝑅𝑒𝑑−
1 8∙(𝑃𝑟−1)
Which leads to the heat transfer coefficient of a turbulent boundary layer of a fluid can be calculated with Equation 16 [28]:
Equation 16 𝛼
𝑓(𝑉
𝑓, 𝑑) =
𝑘𝑓∙𝑁𝑢𝑑𝑑
The heat transfer coefficient of a turbulent boundary layer in a forced fluid is calculated with Equation 13 -Equation 16 and with additional data from Data och diagram [37], and 5 different sized cooling ducts, Figure 20 is constructed.
Figure 20, Coefficient of heat transfer of a turbulent boundary layer internally of pipes.
0 2000 4000 6000 8000 10000 12000 14000 16000 18000
0 1 2 3 4
Coefficent of heat transfer (W/m2∙°C)
Velocity of water (m/s)
Heat transfer of a turbulent boundary layer internaly of pipes (Vf, d)
Ø 0,01 m Ø 0,02 m Ø 0,03 m Ø 0,04 m Ø 0,05 m
6 Results
In this chapter the results of this study are presented. The selected concept is presented, along with the result of the experiment and key parameters.
6.1 Acceptance criteria
The proposed novel process has several sources for its acceptance criteria. The following are from the VOC but are originally specified by LNU WML:
• Shall be able to produce novel welding consumables.
• The novel process shall use GTAW as heat source as a Tetrix 551 Synergic is available at LNU WML.
• It shall be able to produce batches of the size ≈0,5-1kg.
• Shall produce a rod of Ø 3,2 mm for GTAW.
The following requirements are specified by ISO-standards:
• The process shall enable uniform production of novel welding consumables, regarding chemical composition, dimensions etc.
In addition, more generalized requirements are:
• The process shall be more time-, resource-, and cost-efficient than the conventional process.
There is in addition several functions the process shall be able to perform satisfactory, FR´s. They are:
• The alloying element of the novel welding consumable shall easily and reliably be distributed.
• The melting of the novel welding consumable shall easily and reliably be controlled.
• The process shall be able to produce an ingot/wire of suitable size.
• The process shall not cross contaminate the finished novel welding consumable to such an extent that it is rendered unusable.
• The novel alloy shall be shaped to desired geometry.
• The process shall occur in an inert atmosphere.
6.2 Selected concept
The concept that client LNU WML favours is Concept 3. This concept also meets the product requirements set in the VOC most satisfactory in comparison to the other concepts generated, and the conventional process.
Concept 1 constitutes a 3-step process. The steps are:
1. Heating/melting and mixing.
2. Pouring into the cast to create an ingot.
3. Hot rolling to correct geometry.
The clear separation of the different steps provides the opportunity for great process control.
Concept 2 is a development of Concept 1, where Concept 1´s first and second step is joined to one, rendering Concept 2 to be a 2-step process. The steps are:
1. Heating/melting, mixing, and extracting an ingot.
2. Hot rolling to correct geometry.
Concept 3 is a further development of the previous concepts and is a single step, but 2 phase process.
The step is:
1. Heating/melting, mixing, and extracting a Ø3,2mm wire.
With the first phase being: heating/melting, mixing until mixing chamber/correct
batch size is acquired. Second phase is: heating/melting, mixing, and extracting an
Ø3,2mm wire.
6.3 Results of experiment
The experiment resulted in the extraction of a Ø10mm and length ≈ 18 mm tin ingot through the principal extraction method of Concept 3, see Figure 21.
Figure 21, Tin ingot from experiment.
The experiment proved that the tin would go from:
I. Solid to a liquid to,
II. Solid ingot of desireable diameter.
The experiment proved the principle functionality of the extraction method.
6.4 Identification of key parameters
The structural integrity analysis of Concept 3 results in an approximation of the forces, pressures, and stresses. It fails to discover any key parameters, as all approximated forces, pressures and stresses are of such magnitude that they are neglected.
The thermodynamic analysis discovers several primary key parameters. The primary key parameters are:
• Transition thickness.
• The heat transfer coefficient of a turbulent boundary layer in a forced fluid.
The primary key parameters are affected by other parameters, secondary key parameters. They are:
• Effect input.
• Diameter of cooling ducts.
• Flow velocity of water in cooling ducts.
• Extraction speed of wire.
6.5 Construction recommendations
The construction recommendation is based on the found key parameters, showcased in Figure 19 and Figure 20, and client LNU WML available welding equipment’s specifications.
The construction recommendation is presented in Table 3.
Table 3, Construction recommendations for key parameters.
