Impact wave process modeling and optimization in high energy rate explosive welding

(1)

Mälardalen University Press Licentiate Theses No. 106

IMPACT WAVE PROCESS MODELING AND OPTIMIZATION IN HIGH

ENERGY RATE EXPLOSIVE WELDING

APPLIED MECHANIC

Mohammad Tabatabaee Ghomi

2009

School of Sustainable Development of Society and Technology

(2)

II

ISBN 978-91-86135-35-5

(3)

ISBN 978-91-86135-35-5

Printed by Mälardalen University, Västerås, Sweden

Abstract

Impact waves are used in many different industries and are classified according to whether they cause plastic or elastic deformations. In the plastic deformation mode, these waves can be used to produce special electrical joints. In the elastic deformation mode, they can be used to detect leakage or to measure the thickness of pipes. Both modes have applications in offshore technology.

In this thesis the application of impact waves in the plastic deformation mode and explosive welding are discussed. In the explosive welding (EXW) process a high velocity oblique impact produced by a carefully controlled explosion occurs between two or more metals. The high velocity impact causes the metals to behave like fluids temporarily and weld together. This process occurs in a short time with a high rate of energy.

EXW is a well known method for joining different metals together. It is a multidisciplinary research area and covers a wide range of science and technology areas including wave theory, fluid dynamics, materials science, manufacturing and modeling. Many of the important results in EXW research are obtained from experimentation.

This thesis is mainly based on experimental work. However, it begins with a review of the fundamental theory and mechanisms of explosive welding and the different steps of a successful welding operation. Many different EXW tests are done on horizontal and vertical surfaces with unequal surface areas, and on curved surfaces and pipes. The remainder of the thesis evaluates the results of these experiments, measures the main parameters, and shows the results of simulations to verify the experimental results.

The thesis ends with a number of suggestions for improving and optimizing the EXW process. One of these improvements is a model for joining metallic plates with unequal surface areas. An Al-Cu joint based on this model is used in the ALMAHDI aluminum factory, a large company in southern Iran that produces more than 200,000 tons of aluminum per year. Improved methods are also suggested for joining curved surfaces. These methods may have extensive applications in pipelines in oil and gas industries, especially in underwater pipes.

(4)

(5)

Sammanfattning

Impact vågor används i många olika branscher och klassificeras beroende på om de orsakar plast eller elastiska deformationer. I plastisk deformation läge, kan dessa vågor användas för att tillverka speciella elektriska skarvar. I elastisk deformation läge, kan de användas för att upptäcka läckage eller att mäta tjockleken på rören. Båda lägena har tillämpningar inom offshore-teknik.

I denna avhandling tillämpningen av effekten vågor i plastisk deformation mode och explosiva svetsning diskuteras. I den explosiva svetsning (EXW) behandla en hög hastighet sned anslaget av en noggrant kontrollerad explosion inträffar mellan två eller fler metaller. Den höga hastigheten effekten gör att metaller att bete sig som vätskor tillfälligt och svetsa ihop. Detta sker under en kort tid med hög energi.

EXW är en välkänd metod för att gå olika metaller tillsammans. Det är ett tvärvetenskapligt forskningsområde och täcker ett brett spektrum av vetenskapliga och tekniska områden inklusive vågrörelselära, strömningsmekanik, materialvetenskap, tillverkning och modellering. Många av de viktigaste resultaten i EXW forskningen erhålls från experiment.

Denna avhandling är huvudsakligen baserad på experimentellt arbete. Däremot börjar det med en genomgång av grundläggande teori och mekanismer av explosiv svetsning och de olika stegen i en lyckad svetsning operation. Många olika EXW tester görs på horisontella och vertikala ytor med ojämn yta, och på krökta ytor och rörledningar. Återstoden av avhandlingen utvärderar resultaten av dessa försök, åtgärder viktiga parametrar och visar resultaten av simuleringar för att verifiera de experimentella resultaten.

Avhandlingen avslutas med ett antal förslag för att förbättra och optimera EXW processen. En av dessa förbättringar är en modell för att gå metallplattor med ojämna ytor. En Al-Cu gemensamt baserade på denna modell används i ALMAHDI aluminium fabriken, ett stort företag i södra Iran som producerar mer än 200.000 ton aluminium per år. Förbättrade metoder föreslås också för att gå med krökta ytor. Dessa metoder kan ha många olika tillämpningar i rörledningar i olje-och gasindustrin, särskilt i vattnet rör.

(6)

6

Acknowledgements

The work described in this thesis was carried out at the School of Sustainable Development of Society and Technology, Mälardalen University, Sweden.

I would like to thank my supervisor, Professor Jafar Mahmoudi for his encouragement, guidance, scientific help and unlimited support.

I would like to express my gratitude to Professor Erik Dahlquist and Professor Jinyue Yan.

I would like to express my gratitude to Professor Gholamhossein Liaghat from Tarbiat Modares University, Professor Mohammad Mahjoob from Tehran University and Professor A. Darvizeh from Gilan University in Iran for their scientific help.

I would like to express my gratitude to Dr. Adel Karim at Mälardalen University.

The author would like to acknowledge Professor Dobroshin from PATON institute in Ukraine, Mr. Chavideh from Chime-Tec Company in Germany, ACECR, and the TDI organization in Iran for help with experiments.

(7)

Acknowledgements

The work described in this thesis was carried out at the School of Sustainable Development of Society and Technology, Mälardalen University, Sweden.

I would like to thank my supervisor, Professor Jafar Mahmoudi for his encouragement, guidance, scientific help and unlimited support.

I would like to express my gratitude to Professor Erik Dahlquist and Professor Jinyue Yan.

I would like to express my gratitude to Professor Gholamhossein Liaghat from Tarbiat Modares University, Professor Mohammad Mahjoob from Tehran University and Professor A. Darvizeh from Gilan University in Iran for their scientific help.

I would like to express my gratitude to Dr. Adel Karim at Mälardalen University.

The author would like to acknowledge Professor Dobroshin from PATON institute in Ukraine, Mr. Chavideh from Chime-Tec Company in Germany, ACECR, and the TDI organization in Iran for help with experiments.

List of Publications

This thesis is based on the following papers and technical reports:

Appended Papers: Paper 1:

Mohammad Tabatabaee, Jafar Mahmoudi and Gholamhossein Liaghat, An Applied Method for welding Metals of Unequal Surface Area Using Explosive Energy, Submitted to International Journal of Impact Engineering, ISSN: 0734-743X

Paper2:

Mohammad Tabatabaee, Jafar Mahmoudi and Gholamhossein Liaghat, Removing Leakage from oil and gas low pressure Pipes and vessels by high energy explosive welding method, Scientific Conference on “Energy systems with IT” in connection with the Energiting 2009, March 11-12 at Älvsjö fair, Stockholm, ISBN number 978-91-977493-4-3.

Paper3:

Mohammad Tabatabaee, Jafar Mahmoudi and Gholamhossein Liaghat, Effect of Explosive Layer thickness on detonation velocity in a high energy process, Submitted to High Energy Physics Journal, ISSN: 1126-6708.

Report

Mohammad Tabatabaee, Impact wave process control in explosive welding application, Technical report No. 3, 20 July 2008, Mälardalen University

Papers not appended: Paper 4:

Mohammad Tabatabaee and Jafar Mahmoudi, Finite element simulation of explosive welding, The 49th_{Scandinavian Conference on Simulation and Modeling (SIMS2008), ISBN-13:} 978-82-579-46326

Paper 5:

Mohammad Tabatabaee and Jafar Mahmoudi, FEM method simulation for Aluminum - Iron - Copper bonding using explosive welding method, IASTED International Conference on Applied Simulation and Modeling, June 25, 2008, at Corfu, Greece. (ASM 2008), ISBN- 978-0-88986-748-2

Paper 6:

Mohammad Tabatabaee and Jafar Mahmoudi, An advanced method for Aluminum - Iron -Copper bonding using explosive welding method, SSSEC2008 conference, Stockholm, March 12-13, 2008 ISBN-978-91-977493-2-9

Paper 7:

Mohammad Tabatabaee and Jafar Mahmoudi, An advanced method of explosive welding

simulation, The 16th_{Annual (International) Conference on Mechanical Engineering, ISME2008,}

(8)

8

Nomenclature and abbreviations

Latin letters

P pressure

VD, Vd detonation velocity Vp, Vf velocity of flyer plate Vc, Vw collision velocity Ei strain energy Re Reynolds number E Young’s modulus C speed of sound H hardness A amplitude T temperature T thickness Greek letters α initial angle β dynamic angle λ wave length σ tensile stress ρ density Abbreviations

EXW explosive welding

BSEW bond strength explosive welded WW welding window

SEM scanning electron microscope UTS ultimate tensile stress

(9)

Nomenclature and abbreviations

Latin letters

P pressure

VD, Vd detonation velocity Vp, Vf velocity of flyer plate Vc, Vw collision velocity Ei strain energy Re Reynolds number E Young’s modulus C speed of sound H hardness A amplitude T temperature T thickness Greek letters α initial angle β dynamic angle λ wave length σ tensile stress ρ density Abbreviations

EXW explosive welding

BSEW bond strength explosive welded WW welding window

SEM scanning electron microscope UTS ultimate tensile stress

List of Figures

Figure 1. a) Basic set up for explosive welding [1] b) The explosive bonding process [4] ... 17

Figure 2: Geometry of the process during the collapse [1] ... 17

Figure 3. Impact wave reflection [ 1] ... 19

Figure 4. Shape of the wave at the interface of the plates [ 1] ... 19

Figure 5. An example of a WW diagram in the Vc-β plane ... 22

Figure 6. Determination of detonation velocity by Dutrich method [ 4] ... 25

Figure 7. The chisel test [ 1]. ... 26

Figure 8. Pressure simulation by AUTODYN software [ 34] ... 27

Figure 9. Setup for Fe-Fe welding ... 29

Figure 10. Unequal surface area set up ... 30

Figure 11. Al-Cu vertical unequal surface area and dual method set up ... 31

Figure 12. Set up for filling a small hole ... 31

Figure 13. Set up for welding on a curve ... 31

Figure 14. Set up for repairing a leak and filling a small hole in a pipe ... 32

Figure 15. Dutrich method setup using a thin Aluminum plate ... 32

Figure 16. Dutrich method setup using a Brass plate ... 32

Figure 17. Results of EXW on horizontal flat surfaces ... 33

Figure 18. EXW of unequal surface areas ... 33

Figure 19. EXW tests for filling a hole ... 34

Figure 20. EXW with a detonator on a pipe for filling a small hole ... 34

Figure 21. Dutrich method test results ... 34

Figure 22. WW for Cu-Fe joint ... 36

Figure 23. Welding window for Fe-Fe joint ... 37

Figure 24. WW for Al-Cu joint ... 39

Figure 25. Effect of explosive material thickness on detonation velocity ... 41

Figure 26. Effect of explosive material thickness on detonation velocity in general ... 41

Figure 27. Chisel test for 2 types of weld ... 41

Figure 28. BSEW in width direction ... 42

Figure 29. BSEW in length direction ... 42

Figure 30. Metallographic results ... 43

Figure 31. Results of simulation ... 44

Figure 32. Bonded areas measured by ultrasonic test for Al-Cu joint ... 46

Figure 33. A set up for filling a hole before welding ... 50

(11)

List of Figures

Figure 1. a) Basic set up for explosive welding [1] b) The explosive bonding process [4] ... 17

Figure 2: Geometry of the process during the collapse [1] ... 17

Figure 3. Impact wave reflection [ 1] ... 19

Figure 4. Shape of the wave at the interface of the plates [ 1] ... 19

Figure 5. An example of a WW diagram in the Vc-β plane ... 22

Figure 6. Determination of detonation velocity by Dutrich method [ 4] ... 25

Figure 7. The chisel test [ 1]. ... 26

Figure 8. Pressure simulation by AUTODYN software [ 34] ... 27

Figure 9. Setup for Fe-Fe welding ... 29

Figure 10. Unequal surface area set up ... 30

Figure 11. Al-Cu vertical unequal surface area and dual method set up ... 31

Figure 12. Set up for filling a small hole ... 31

Figure 13. Set up for welding on a curve ... 31

Figure 14. Set up for repairing a leak and filling a small hole in a pipe ... 32

Figure 15. Dutrich method setup using a thin Aluminum plate ... 32

Figure 16. Dutrich method setup using a Brass plate ... 32

Figure 17. Results of EXW on horizontal flat surfaces ... 33

Figure 18. EXW of unequal surface areas ... 33

Figure 19. EXW tests for filling a hole ... 34

Figure 20. EXW with a detonator on a pipe for filling a small hole ... 34

Figure 21. Dutrich method test results ... 34

Figure 22. WW for Cu-Fe joint ... 36

Figure 23. Welding window for Fe-Fe joint ... 37

Figure 24. WW for Al-Cu joint ... 39

Figure 25. Effect of explosive material thickness on detonation velocity ... 41

Figure 26. Effect of explosive material thickness on detonation velocity in general ... 41

Figure 27. Chisel test for 2 types of weld ... 41

Figure 28. BSEW in width direction ... 42

Figure 29. BSEW in length direction ... 42

Figure 30. Metallographic results ... 43

Figure 31. Results of simulation ... 44

Figure 32. Bonded areas measured by ultrasonic test for Al-Cu joint ... 46

Figure 33. A set up for filling a hole before welding ... 50

Figure 34. Al–Cu joints transmit the electric power of anode rods in the aluminum factory ... 51

List of Tables

Table 1. Summary of papers ... 15

Table 2. Specification of Fe-Fe horizontal setup ... 29

Table 3. Specification of Al-Cu horizontal setup ... 29

Table 4. Specification of Fe-Cu horizontal setup ... 29

Table 5. Specification of Al-Cu, unequal surface area, horizontal set up ... 30

Table 6. Specification of Al-Cu Vertical unequal surface area and dual method set up ... 30

Table 7. Specification of Stage 3 experiments ... 31

Table 8. Specification of experiments in Stage 4 ... 31

Table 9. Specification of experiments in Stage 5 (Dutrich method) ... 32

Table 10. Result of experiments described in Table 9 ... 34

Table 11. Calculation for line (a-a) ... 35

Table 12. Calculation for line (f-f) ... 35

Table 13. Calculation for line (g-g) ... 36

Table 20. Calculation of VD for different thicknesses of explosive material ... 40

Table 21. Results of testing on measuring detonation velocity for different kinds of explosive materials ... 40

Table 22. Results of mechanical test in width direction ... 42

(12)

(13)

1. Introduction

I

mpact mechanics is a branch of applied mechanics that deals with high rates of energy and load in a very short time. This important process has applications in many different industries. One of the most useful ways of producing high energy rate impacts is by use of explosive materials. A small quantity of explosive material can shape a tank, build a large crankshaft, has the power to weld many parts of a heat exchanger and improve the mechanical properties of a rail.

Explosive Welding (EXW) is one application of impact waves. The impact waves are of the tension wave types that produce elastic and plastic deformations in the solid material. Explosive welding and shaping occurs in the plastic deformation region. Measurement by ultrasonic and impact waves is done in the elastic deformation region.

In EXW, an oblique impact occurs between two parts such that they behave like fluids and weld firmly together. Because of the high velocity of impact, a jet is formed that cleans the two surfaces, presses them together and produces a joint. This joint has an acceptable resistance that is equal or greater than the resistance of the weaker plate.

1.1. Background

Explosive materials were first used in manufacturing shortly after the Second World War. However, the first observations of their potential uses in manufacturing date back to the First World War. It had been observed that a bullet did not only pierce metal but also welded to it. This phenomenon was subsequently reproduced in the laboratory and applied commercially in industry. Advances in the aerospace industry and the close tolerances necessary for manufacturing complex parts drove the use of the EXW method on an industrial scale. By the mid 1950s, EXW was being applied in manufacturing.

In the following years, it was quickly accepted that EXW methods could be applied to a number of other industries. EXW processes were adapted and refined to serve the needs of the automotive, shipbuilding, material processing, mining, and construction industries, among others. Over three hundred joint between similar and dissimilar materials have been produced until now. The first experiments with the EXW technique were carried out on horizontal surfaces, but many commercial tests have subsequently been done on curved surfaces such as pipelines and heat exchanger components.

1.2. Literature review

This section reviews research that predates this thesis.

One of the major reference works in this field is the book by Blazynski [1]. In the work, Blazynski described clearly the method of explosive welding, explained wave phenomena and the overall EXW procedure. The basic method is also described in the book by Crossland [2]. The PATON Institute [3] and professor Darvizeh [4] have performed many EXW experiments. The fundamentals of the EXW process have also been explained in a number of handbooks [5], [6] and [7] and the mechanism of the wave interface has also been described in the literature [8], [9] and [10]. A number of researchers consider EXW to be fundamentally a fusion welding process (Phillipchuk, [11]) which relies on the kinetic energy at the interface. Crossland and Williams [12] look at the method as a pressure weld process.

Otto and Carpenter [13] proposed that interfacial shear occurs during welding, and attributed the weld to the result of heat generated by shearing at the boundary. The process reaches a very high temperature at the interface, above the melting point of the welded parts, for a short period on the order of microseconds. Onzawa [14] reached a similar conclusion in his study. He performed

(14)

14 interface observations using scanning electron microscopy (SEM). It is generally accepted, based on experimental data, that jet creation makes an important contribution to welding. The jet cleans the surfaces by removing a thin layer of metals and other contaminants. The investigative solutions of the pressure and jet velocity of the impact of liquid drops were found by Lesser [15] and Lesser and Field [16], [17], and Field [18] provided the first photographic evidence of the effect. Wilson and Bronzing [19] studied the waves that form in the interface. The theories proposed for the mechanism of the wave formation can be classified as indentation mechanism, flow unsteadiness mechanism, vortex shedding mechanism and stress wave mechanism (Reid, [20]. Bahrani and Crossland [21], Bergman [22], Bahrani et al. [23] and Abrahamson [24] have worked on groups of these categories. Another theory of wave formation was proposed by Hunt [25] and Robinson [26], who suggested that the explosive welding wave forms when there is a velocity difference between adjacent streams. The flow instability mechanism was described by Robinson, who proposed that the waves are created behind the collision zone because of a velocity across the interface which involves a jet. This is different to the flow instability mechanism expressed by Hunt. Cowan [27] and Kowalik and Hay[28] pointed out the parallels between the waves in explosive welding and the Von Karma's vortex generated by a barrier. A stress wave mechanism of wave formation was proposed by El-Soky and Blazynslki. This wave formation mechanism was recognized by Plaksin [29].

Lazari and Al-Hassani [30] studied the behavior of metal plates under explosive pressure using a finite element method. They used the theory of virtual displacement of the Lagrangian deformation to derive the equations of motion. Oberg [31] simulated the explosive welding process using Lagrangian finite difference computer code. The process was also modeled by Akihisa [32]. Finally, the results of simulation provided by Alhasani [33] and Akbari Mousavi [34], [35], have been reviewed.

1.3. Motivation and objective

• Application of explosive welding in industry is the main motivation:

The southern Iranian aluminum company ALMAHDI had a requirement for special copper-aluminum joints with unequal surface areas. Joints they had made previously were unsatisfactory. As a result of this thesis, more than 1000 successful EXW joints have been made and confirmed by the factory. Another motivation for conducting this thesis was the problems faced by the oil and gas industry in repairing and preventing the leakage in pipelines.

• Improving some explosive welding method is the main objective of the project:

During the course of this research, many experiments have been performed on materials of various types and shapes, and a number of new techniques have been applied to improving the EXW process, including a new method for horizontal welding, a new method for curve welding, and a useful curve of the velocity of explosive material versus its thickness.

1.4. Research approach

The main hypothesis of this thesis is that “Materials can be bonded together by the high energy transient pressure or impact waves produced by oblique collision at high velocity’’.

To test this hypothesis, explosive material was used to produce the high velocity or impact waves to weld metals, a process called explosive welding.

(15)

interface observations using scanning electron microscopy (SEM). It is generally accepted, based on experimental data, that jet creation makes an important contribution to welding. The jet cleans the surfaces by removing a thin layer of metals and other contaminants. The investigative solutions of the pressure and jet velocity of the impact of liquid drops were found by Lesser [15] and Lesser and Field [16], [17], and Field [18] provided the first photographic evidence of the effect. Wilson and Bronzing [19] studied the waves that form in the interface. The theories proposed for the mechanism of the wave formation can be classified as indentation mechanism, flow unsteadiness mechanism, vortex shedding mechanism and stress wave mechanism (Reid, [20]. Bahrani and Crossland [21], Bergman [22], Bahrani et al. [23] and Abrahamson [24] have worked on groups of these categories. Another theory of wave formation was proposed by Hunt [25] and Robinson [26], who suggested that the explosive welding wave forms when there is a velocity difference between adjacent streams. The flow instability mechanism was described by Robinson, who proposed that the waves are created behind the collision zone because of a velocity across the interface which involves a jet. This is different to the flow instability mechanism expressed by Hunt. Cowan [27] and Kowalik and Hay[28] pointed out the parallels between the waves in explosive welding and the Von Karma's vortex generated by a barrier. A stress wave mechanism of wave formation was proposed by El-Soky and Blazynslki. This wave formation mechanism was recognized by Plaksin [29].

Lazari and Al-Hassani [30] studied the behavior of metal plates under explosive pressure using a finite element method. They used the theory of virtual displacement of the Lagrangian deformation to derive the equations of motion. Oberg [31] simulated the explosive welding process using Lagrangian finite difference computer code. The process was also modeled by Akihisa [32]. Finally, the results of simulation provided by Alhasani [33] and Akbari Mousavi [34], [35], have been reviewed.

1.3. Motivation and objective

• Application of explosive welding in industry is the main motivation:

The southern Iranian aluminum company ALMAHDI had a requirement for special copper-aluminum joints with unequal surface areas. Joints they had made previously were unsatisfactory. As a result of this thesis, more than 1000 successful EXW joints have been made and confirmed by the factory. Another motivation for conducting this thesis was the problems faced by the oil and gas industry in repairing and preventing the leakage in pipelines.

• Improving some explosive welding method is the main objective of the project:

During the course of this research, many experiments have been performed on materials of various types and shapes, and a number of new techniques have been applied to improving the EXW process, including a new method for horizontal welding, a new method for curve welding, and a useful curve of the velocity of explosive material versus its thickness.

1.4. Research approach

The main hypothesis of this thesis is that “Materials can be bonded together by the high energy transient pressure or impact waves produced by oblique collision at high velocity’’.

To test this hypothesis, explosive material was used to produce the high velocity or impact waves to weld metals, a process called explosive welding.

The work described by this thesis includes studies of the process, review of previous work in the field, design of experiments using different materials and shapes, manipulation and control of parameters before and after welding, calculation of the weld parameters, process optimization, simulation and comparison of the results. Following numerous experiments on different materials with different geometries, such as horizontal and vertical alignment, different surfaces and curved shapes, several methods for improving the process are proposed.

1.5. Limitations

EXW has several limitations in theory and in practice. Working with explosives is very dangerous and the high levels of sound produced can be harmful to hearing. The plate surfaces must be clean and the process is best performed in a vacuum. EXW is at present a manual process and has not been automated. There are various analytical methods for calculating the process variables, and many formulas are obtained empirically. Therefore, simulation and calculation of these methods is very difficult. All of the experiments performed for this thesis were performed in a vacuum chamber.

1.6. Methodology

This experimental work described in this thesis is divided into five stages: Stage 1:

Study, calculation and experimental work on explosive welding together of horizontal surfaces of equal surface areas and different materials such as Fe-Fe, Al-Cu and Fe-Cu

Tests are first performed on flat surfaces because EXW of flat surfaces is easier than on rods and curved surfaces.

Stage 2:

Study, calculation and experimental work on explosive welding together of horizontal surfaces of unequal surface areas and different materials such as Fe-Fe, Al-Cu and Fe-Cu

Here experimental tests (horizontal, vertical and dual method) are carried out on flat surfaces with different dimensions.

Stage 3:

Study, calculation and experimental work on explosive welding of curved shapes

Tests are performed on curved surfaces based on the results from the tests on flat surfaces. A steel rod is used at this stage.

Stage 4:

Study on explosive welding of cylindrical surfaces with different materials such as Fe-Fe, Al-Cu, and Fe-Cu

Experiments are performed on cylindrical surfaces such as pipes and tubes. Stage 5:

Control of explosive parameters (explosive materials – process parameters, mechanical testing)

EXW parameters such as explosion velocity and flyer plate velocity are measured. The work is presented in papers 1-7 and the areas covered are summarized in Table 1.

Table 1. Summary of papers

Paper Area discussed Stages

6, 7 Horizontal EXW 1

5, 4, 7 EXW Simulation 1

1 EXW of unequal surfaces 2 2 EXW of curved shapes 3 and 4 3 Explosive materials 5

(16)

16

1.7. Thesis Outline

This thesis describes research during which more than 100 experiments were performed and the results were applied in a large aluminum production company. There are at least 3 patentable technologies described in the results. The thesis is organised as follows:

Part 1 Introduction: including background, objectives, motivation, limitations of the studies, presentation of the methodology, formulation of the problem and outline of the thesis.

Part 2 Theory: including definitions and expressions used in the thesis, and presentation of theory in preparation for the scientific discussion.

Part 3 Experimental Data and Results: including the experimental set up, results and test reports. Part 4 Discussion: including discussion of results, calculations, and future work.

Part 5 Conclusions: including conclusions and practical output. Part 6 References

(17)

1.7. Thesis Outline

This thesis describes research during which more than 100 experiments were performed and the results were applied in a large aluminum production company. There are at least 3 patentable technologies described in the results. The thesis is organised as follows:

Part 1 Introduction: including background, objectives, motivation, limitations of the studies, presentation of the methodology, formulation of the problem and outline of the thesis.

Part 2 Theory: including definitions and expressions used in the thesis, and presentation of theory in preparation for the scientific discussion.

Part 3 Experimental Data and Results: including the experimental set up, results and test reports. Part 4 Discussion: including discussion of results, calculations, and future work.

Part 5 Conclusions: including conclusions and practical output. Part 6 References

Part 7 Summary of Papers

2. Theory

2.1. Mechanism and set up

E

xplosive welding occurs under oblique impact at high velocity. The explosive energy is used to form a cold pressure weld. The basic setup for welding two plates is shown in Figure 1. The flyer plate is mounted at a small angle to the parent plate which is held on an anvil which can be either a metal plate or sand. The top surface of the flyer plate is covered with a layer or buffer, which may be rubber or a thick coat of plastic. A layer of explosive material in the form of a sheet or powder is laid on top of the buffer and it explodes from the lower edge. A specific distance - the stand-off distance - and an initial angle α between the flyer and parent plates are necessary for the flyer plate to obtain the required acceleration. The impact angle (also called the collision or dynamic angle) β can be variable depending on the plate arrangement. In a parallel setup α is zero and β is constant, but in an oblique and a curved set up, β can vary during the process [4].

.

a b

Figure 1. a) Basic set up for explosive welding [1] b) The explosive bonding process [4]

The EXW process can be divided into three basic phases: (I) Detonation of the explosive charge

(II) Acceleration and deformation of the flyer plate (III) Collision of the plates

It is accepted that jet formation at the contact point is an essential condition for the welding. This jet sweeps away the oxide films on the surfaces of the metals and forms a metallurgical bond that causes the atoms of the two materials to meet at atomic scale distances. The pressure has to be sufficiently high and last for a sufficient time for these inter-atomic bonds to be formed. The velocity of the collision point Vc sets the time available for bonding. The pressure also causes local

plastic deformation of the metals at the interface. The strength of the bond is stronger than the strength of the weaker plate material, and the quality of the bond depends on careful control of parameters such as surface grounding, plate separation and the detonation velocity Vd (see Figure 2). The impact velocity Vp and the dynamic angle β determine the pressure and the shear stress at

the collision point.

Figure 2: Geometry of the process during the collapse [1]

(18)

18 �� sin�� cos�� _{2 �}� �� sin�� cos�� _{2 �}� 2 sin�� _{2 � cos�}� � �_{2 �} cos�� _{2 �}

�

��

� sin�

��

�

(1)

From the sine equations in triangle SBD we can write: �� cos �� _{2 �}� �� cos �� _{2 �}� �� sin � �

�

_�

� �

_�

� �

_�� (2)

�

� �

�� (3)

Where �� is the velocity of the flyer plate in relation to point s and �� is the velocity of welding, equal to the collision velocity ��.

In a parallel set up α=0 and �� = �� = ��, and from the previous equations:

�

_{2 sin� 2}��_�

(4)

The selection of parameters is based on the mechanical properties, density, and shear wave velocity of each component, and many of these are determined experimentally. Considerable progress has been made in setting up the optimum parameters required to produce an acceptable bond.

The parameters involved in the process (such as ��, ��, ��, α � �, etc.) are defined in a special diagram called the Welding Window (WW) that has been proposed by various authors [36].

2.2. The impact wave in explosive welding

In EXW the pressure created in the region of the detonation front of the explosive charge is used to provide rapid acceleration of the flyer plate to a high velocity prior to impact on the parent plate. The flyer plate velocity depends on the amount of explosive charge and the stand-off distance. The pressure produced in the detonation front transmits into the flyer plate as a stress compression wave. When the compression wave reaches the back surface of the metal slab, it is reflected as a tension wave, and the velocity of particles is doubled [1] (Figure 3a). The same phenomenon occurs at the edges of the plates where the impact wave propagates horizontally. This edge effect creates problems for the quality of welding at the edges.

The pressure waves in the flyer plate are oblique and when they are reflected from the back surface of the plate, they give rise to both a dilatational wave and a shear wave (Figure 3b). Many authors have written on this subject. Examples of their conclusions include:

a) A low stress wave transmits through the solid at the speed of sound, which can be readily derived from the theory of elasticity [2].

(19)

�� sin�� cos�� _{2 �}� �� sin�� cos�� _{2 �}� 2 sin�� _{2 � cos�}� � �_{2 �} cos�� _{2 �}

�

��

� sin�

��

�

(1)

From the sine equations in triangle SBD we can write: �� cos �� _{2 �}� �� cos �� _{2 �}� �� sin � �

�

_�

� �

_�

� �

_�� (2)

�

� �

�� (3)

Where �� is the velocity of the flyer plate in relation to point s and �� is the velocity of welding, equal to the collision velocity ��.

In a parallel set up α=0 and �� = �� = ��, and from the previous equations:

�

_{2 sin� 2}��_�

(4)

The selection of parameters is based on the mechanical properties, density, and shear wave velocity of each component, and many of these are determined experimentally. Considerable progress has been made in setting up the optimum parameters required to produce an acceptable bond.

The parameters involved in the process (such as ��, ��, ��, α � �, etc.) are defined in a special diagram called the Welding Window (WW) that has been proposed by various authors [36].

2.2. The impact wave in explosive welding

In EXW the pressure created in the region of the detonation front of the explosive charge is used to provide rapid acceleration of the flyer plate to a high velocity prior to impact on the parent plate. The flyer plate velocity depends on the amount of explosive charge and the stand-off distance. The pressure produced in the detonation front transmits into the flyer plate as a stress compression wave. When the compression wave reaches the back surface of the metal slab, it is reflected as a tension wave, and the velocity of particles is doubled [1] (Figure 3a). The same phenomenon occurs at the edges of the plates where the impact wave propagates horizontally. This edge effect creates problems for the quality of welding at the edges.

The pressure waves in the flyer plate are oblique and when they are reflected from the back surface of the plate, they give rise to both a dilatational wave and a shear wave (Figure 3b). Many authors have written on this subject. Examples of their conclusions include:

a) A low stress wave transmits through the solid at the speed of sound, which can be readily derived from the theory of elasticity [2].

b) At high pressure, as in a shock wave, the shear strength becomes negligible compared to the pressure and the material behaves like a fluid [3].

c) El-Sobky and Blazynski [1] suggested a mechanism for the stress wave, represented in Figure 3d. In their model, a single compression wave (symbolized by a full circle) is generated at the collision point, while successive reflections from both the flyer and parent plate (represented by the dashed circles) generate the wavy shape at the contact point that can be seen under the SEM microscope.

a

b

c _d

Figure 3. Impact wave reflection [1]

a) Reflection of a compressive stress wave from back surface of the metal plate b)Detonation of a layer of explosive in contact with a metal plate

c) Stress wave mechanism of surface wave formation d) The wavy shape of the contact point

2.3. Predicting the wavelength

Figure 4 shows the shape of the impact wave at the interface of the plates. The wavelength can be derived from this figure as follows:

Figure 4. Shape of the wave at the interface of the plates [1]

λ = C . A (7) Where A is amplitude and C is a constant.

λ

2A x

(20)

20 The plastic work �� is therefore:

�� _�� (8) Where l is the length of the plate and Y is the ultimate yield point of the metal. The kinetic energy can be expressed as:

� �� 1 2� ��₍₉₎ where �� is the kinetic energy and � is the thickness of the flyer plate.

The wavelength � is obtained by equating the plastic work and the kinetic energy. The wavelength can therefore be determined from Equations 4, 8 and 9we could:

� � �� 2� � (10) � � ��_��

k has been measured experimentally at 28, therefore:

� � 2�� 2� � (11) In addition, we know from experimental data that is obtained as follows:

� � ��2 � (12)

2.4. Bonding criteria

Previous experiments have shown that there are critical values for the geometry and the collision parameters which have to be observed for successful welding. These are summarized below. • Velocity limit

Because a jet must be formed at the collision point, the collision angle β is critical and it is a function of �� (the collision velocity). It has been shown experimentally that �� and �� (the velocity of the flyer plate) must be less than the speed of sound in both metals. It is known that at supersonic velocities, the dynamic pressure is not sustained for long enough to support the changes of inter-atomic flow and stability in the collision area. As �� is dependent on �� and β, it can be adjusted by setting up an initial angle of obliquity (α). The speed of sound provides an upper limit for �� and��. These velocities may be increased by increasing α. �� is assumed constant in the welding area in an explosive welding process [10].

• Collision angle (β)

The collision angle must be between 5° and 25° [1].

(21)

The plastic work �� is therefore:

�� _�� (8) Where l is the length of the plate and Y is the ultimate yield point of the metal. The kinetic energy can be expressed as:

� �� 1 2� ��₍₉₎ where �� is the kinetic energy and � is the thickness of the flyer plate.

The wavelength � is obtained by equating the plastic work and the kinetic energy. The wavelength can therefore be determined from Equations 4, 8 and 9we could:

� � �� 2� � (10) � � ��_��

k has been measured experimentally at 28, therefore:

� � 2�� 2� � (11) In addition, we know from experimental data that is obtained as follows:

� � ��2 � (12)

2.4. Bonding criteria

Previous experiments have shown that there are critical values for the geometry and the collision parameters which have to be observed for successful welding. These are summarized below. • Velocity limit

Because a jet must be formed at the collision point, the collision angle β is critical and it is a function of �� (the collision velocity). It has been shown experimentally that �� and �� (the velocity of the flyer plate) must be less than the speed of sound in both metals. It is known that at supersonic velocities, the dynamic pressure is not sustained for long enough to support the changes of inter-atomic flow and stability in the collision area. As �� is dependent on �� and β, it can be adjusted by setting up an initial angle of obliquity (α). The speed of sound provides an upper limit for �� and��. These velocities may be increased by increasing α. �� is assumed constant in the welding area in an explosive welding process [10].

• Collision angle (β)

The collision angle must be between 5° and 25° [1].

5 ≤ β ≤ 25 (13)

• Pressure limit

A minimum impact pressure (ܸ௣) is required to impart sufficient impact energy to produce a weld. It has been suggested by Wylie et al.[1] that the impact energy required is related to the strain energy and the dynamic yield strength of the flyer plate. The upper limit of the impact energy is determined by the need to avoid excess heat and possible melting by viscous dissipation and the consequent formation of weak layers. This upper limit depends to the lower melting point of the weld.

• Stand-off

The stand-off distance must be sufficient to allow the flyer plate to accelerate to the required impact velocity. The minimum stand-off distance has been empirically determined as half the thickness of the flyer plate. An empirically determined formula for the suggested stand-off distance [1] is:

S=3 K Xe C/M (14)

Where S is the stand-off distance, K is between 4 and 7, depending on the impact velocity, M is the flyer plate mass, C is the explosion mass and Xe is the thickness of explosive material.

• Surface flatness

The flatness is very important in the EXW process. This is because imperfections in the surface cause the jet to be concentrated at a point, which produces high temperatures at the contact points, thereby preventing high quality welding. For a successful test, surface flatness in the range of 2 to 3 microns is sufficient.

• Explosive material

High velocity explosive materials are not used in the EXW process because of the risk of damage to the flyer plate during the test.

A special diagram called the welding window or weld-ability window is used to describe the state of plates at the interface and in the weld area. The critical parameters used to establish a weld-ability window are

• The critical impact angle for jet formation • The collision velocity ܸ௖

• The kinetic energy and impact pressure that is indicated by ܸ௣

2.5. Welding window

The welding window (WW) includes straight and curved regions. In order to draw the WW the relationships between the initial conditions (the angles α and β and the characteristics of the explosive) must be established. The WW lies within the boundaries of 7 parameters as shown in Figure 5. The parameters α, β, ܸௗ, ܸ௣, ܸ௙,, ܸ_௖and the properties of the material determine the WW. This diagram can be drawn in both the ܸ௖ - β and ܸ௣ - β plane and displays an area within which the weld is available. In this thesis the WW is drawn only in the ܸ௖- β plane.

(22)

22

Figure 5. An example of a WW diagram in the Vc-β plane

Critical angle limit for jet formation [line a-a]

The most important condition for welding is jet formation. This must occur at the contact point for successful welding to occur. Theoretically, jetting will occur if �� remains subsonic. However, in practice a minimum angle is necessary to satisfy the pressure requirements. Jetting occurs to the left of the line a-a in Figure 5, which represents the critical angle βc which is necessary for jetting. Abrahamson suggests the following relationship between β and �� [24]:

β = 10(��-5.5) (15) Upper limit of �� [line b-b]

Line b-b in Figure 5 describes the upper limit of ��, which is predictable at 1.2 to 1.5 times the speed of sound, and which also limits the other WW parameters.

Lower and upper limit of β [lines c-c and d-d]

The lower and upper limits of the dynamic angle β were experimentally obtained by Bahrani and Crosslan [21]. They suggested a lower limit of between 2° and 3° and upper limit of 31° for β in a parallel geometry. Suggested minimum and maximum values of the initial angle α in an inclined set up are 3° and 18° respectively.

Lower limit of �� [line e-e]

Equation 16 defines the lower limit of �� for bonding as proposed by Simonov [10]. �� 2�� (16) Cowan [27] defined the lower limit of �� according to the fluid hypothesis as follows:

(23)

Figure 5. An example of a WW diagram in the Vc-β plane

Critical angle limit for jet formation [line a-a]

The most important condition for welding is jet formation. This must occur at the contact point for successful welding to occur. Theoretically, jetting will occur if �� remains subsonic. However, in practice a minimum angle is necessary to satisfy the pressure requirements. Jetting occurs to the left of the line a-a in Figure 5, which represents the critical angle βc which is necessary for jetting. Abrahamson suggests the following relationship between β and �� [24]:

β = 10(��-5.5) (15) Upper limit of �� [line b-b]

Line b-b in Figure 5 describes the upper limit of ��, which is predictable at 1.2 to 1.5 times the speed of sound, and which also limits the other WW parameters.

Lower and upper limit of β [lines c-c and d-d]

The lower and upper limits of the dynamic angle β were experimentally obtained by Bahrani and Crosslan [21]. They suggested a lower limit of between 2° and 3° and upper limit of 31° for β in a parallel geometry. Suggested minimum and maximum values of the initial angle α in an inclined set up are 3° and 18° respectively.

Lower limit of �� [line e-e]

Equation 16 defines the lower limit of �� for bonding as proposed by Simonov [10]. �� 2�� (16) Cowan [27] defined the lower limit of �� according to the fluid hypothesis as follows:

��

� �� _{2��}��2 (17)

Where Re is the Reynolds number, H is the Vickers hardness (� �� and F and B stand for flyer � and base plates respectively. The lower limit of �� can be determined at the transition limit which occurs at Re=10.6.

Lower limit of impact critical pressure (��) [line f-f]

Equation 18 shows the lower limit for welding of β in radians, where H is the Vickers hardness in N/m2_{and ρ is the density in kg/m}3_.

β=1.14 �

_{� �}��

��

�

�_�_�

₍₁₈₎ The equation simplifies to:

β ��= cte (19)

Hardness is derived from tensile stress as follows [1]:

Hv=5 � (20)

Wittman [7] also proposed a lower limit for �� as follows:

��

=�

��_�

�

��

(21) Where σTU is the ultimate tensile stress and C is the bulk sound velocity.

Equation 22 gives another formulation for the lower limit of �� [4]:

�� (min)� �� _{�� }� � (22) Where A is a symbol for the plate with higher strength, � �� yield stress and U is the speed of sound in the metal. Equation 23 gives �� for the plates with the same material:

��

(min)�

��_��

(23) Upper limit of �� [line g-g]

Wittman gives the upper limit for welding as:

sin

�_�

�

��

�� (24)

�

�� 2

�

, �� , � � � �� 2��

Where t is the thickness of the flyer plate and k3 is experimentally determined with �� set to half the compressive wave velocity Cf.

Wittman also suggested the following experimental formula:

(24)

24

2.6. Governing equations

The main governing equation for the EXW process is the equation of motion [37]: � � � �� (26) A number of useful equations follow below.

The collision velocity, dynamic angle and thickness of flyer plate are related to each other as follows:

f =f (β,��, t) (27) From plastic theory [37], the maximum pressure is expressed as:

P= ρ β C �� (28) Using the hydrodynamic method, the maximum pressure is:

� � ��

� (29) Power in the EXW process is given by:

� � �� 1 � �� C� � � �� (30)

Where Ei is strain energy, and m1 and m2 are the masses of flyer and parent plates [38]. From equation 30, �� is calculated as follows:

�� Opt. =C_√��=0.557C0 (31)

Equation 32 is an experimentally determined formula that relates the flyer plate velocity to the explosive velocity.

��

�

0.612 e/m_2�e/m �� (32)

Where e/m is the ratio of mass of the explosive material to the mass of the flyer plate and buffer

2.7. Testing methods

The main requirement to achieve the best quality welds is accurate calculation of EXW parameters. These parameters can be estimated from theory and experimentally. The most important of these are ��, ��, ��, β, explosion power and explosion sensitivity. Some of the important parameters can only be measured after welding. Other operations that are carried out after the welding process include mechanical testing, metallographic testing, NDT testing, machining, stress relief, cutting and sizing.

(25)

2.6. Governing equations

The main governing equation for the EXW process is the equation of motion [37]: � � � �� (26) A number of useful equations follow below.

The collision velocity, dynamic angle and thickness of flyer plate are related to each other as follows:

f =f (β,��, t) (27) From plastic theory [37], the maximum pressure is expressed as:

P= ρ β C �� (28) Using the hydrodynamic method, the maximum pressure is:

� � ��

� (29) Power in the EXW process is given by:

� � �� 1 � �� C� � � �� (30)

Where Ei is strain energy, and m1 and m2 are the masses of flyer and parent plates [38]. From equation 30, �� is calculated as follows:

�� Opt. =C_√��=0.557C0 (31)

Equation 32 is an experimentally determined formula that relates the flyer plate velocity to the explosive velocity.

��

�

0.612 e/m_2�e/m �� (32)

Where e/m is the ratio of mass of the explosive material to the mass of the flyer plate and buffer

2.7. Testing methods

The main requirement to achieve the best quality welds is accurate calculation of EXW parameters. These parameters can be estimated from theory and experimentally. The most important of these are ��, ��, ��, β, explosion power and explosion sensitivity. Some of the important parameters can only be measured after welding. Other operations that are carried out after the welding process include mechanical testing, metallographic testing, NDT testing, machining, stress relief, cutting and sizing.

Tests to be carried out before welding: a) Measuring ܸௗ and ܸ௖

Dutrich method

One of the easiest and oldest methods for measuring the detonation velocity is the Dutrich method. In this method wire of length L with a detonation velocity ܸ௖ is held in contact at two different points to an explosive material of unknown detonation velocity ܸௗ. The distance between the contact points is L1 (shown in Figure 6). The middle of the wire is marked on a thin aluminum, brass or lead plate. The explosive wave passes through both ends of the wire. They collide at a point a distance L2 from the middle of the wire. This collision creates a mark on the abovementioned plate.

Figure 6. Determination of detonation velocity by Dutrich method [4]

The time that it takes for the explosive wave to travel from each contact point to the collision point is the same. t1=t2 L1 / VD+ (L /2 – L2)/ Vd= (L /2) / Vd + L2 / Vd L1 / VD =2 L 2 / Vd ⇒ VD= Vd L1 / 2 L 2 (33)

L1, L2 and ܸௗ are known - ܸ஽ can therefore be calculated from equation 33. Pin contactor method

ܸௗ can be measured using a complex recording system to record electrical pulses. In the EXW process when the flyer and parent plates are parallel, ܸ௖ is equal to ܸௗ. This provides the basis for the pin contactor method.

Velocity probe method

ܸௗ and ܸ௖ can be calculated directly by using measuring probes to draw the location-time curve. Each probe includes a closed-ended aluminum tube and a sensitive insulating wire. The voltage variation after the explosion, caused by the decrease in the length of the insulating wire is measured and indicated by oscilloscope.

(26)

26 This is a simple electrical method which is used to obtain contact parameters using electrical probes and an oscilloscope.

Photographic method (for measuring ܸௗ )

This method can be used to measure ܸௗ continuously. Radiography method (for measuring β)

This is similar to the photographic method except in that it uses X–rays. Radiography can be used to show the location and the moment of contact. The angle β is therefore measured directly.

b) Measuring explosive power Ballistic mortar method

In this method an explosive charge of 10 gram is used to fire a standard shot at the end of a pendulum. The angle of recoil following the explosion is measured and used to calculate the explosive power.

Trauzl method

In this method the explosive is placed in a cylindrical hole in a lead block. The remainder of the hole is filled with sand. The explosive power can be calculated by measuring the increase in the volume of the hole after detonation.

c) Measuring sensitivity

The sensitivity is used to determine the necessary load to initiate detonation. A small booster charge is placed between the detonator and the main charge. The mass of the booster charge required to initiate a stable detonation is used to calculate the sensitivity.

Tests to be carried out after welding a) Primary tests

These including sizing, cutting, machining and stress relief. b) NDT tests

Non destructive tests, including ultrasonic and radiography tests. c) Mechanical tests

Impact resistance test

The impact test is used to check the strength of a weld under dynamic load. Chisel or peel off test

This is a simple but very important test, and is illustrated in Figure 7. A chisel is used to try to separate two the plates. Easy separation of the plates indicates incomplete welding.

(27)

This is a simple electrical method which is used to obtain contact parameters using electrical probes and an oscilloscope.

Photographic method (for measuring ܸௗ )

This method can be used to measure ܸௗ continuously. Radiography method (for measuring β)

This is similar to the photographic method except in that it uses X–rays. Radiography can be used to show the location and the moment of contact. The angle β is therefore measured directly.

b) Measuring explosive power Ballistic mortar method

In this method an explosive charge of 10 gram is used to fire a standard shot at the end of a pendulum. The angle of recoil following the explosion is measured and used to calculate the explosive power.

Trauzl method

In this method the explosive is placed in a cylindrical hole in a lead block. The remainder of the hole is filled with sand. The explosive power can be calculated by measuring the increase in the volume of the hole after detonation.

c) Measuring sensitivity

The sensitivity is used to determine the necessary load to initiate detonation. A small booster charge is placed between the detonator and the main charge. The mass of the booster charge required to initiate a stable detonation is used to calculate the sensitivity.

Tests to be carried out after welding a) Primary tests

These including sizing, cutting, machining and stress relief. b) NDT tests

Non destructive tests, including ultrasonic and radiography tests. c) Mechanical tests

Impact resistance test

The impact test is used to check the strength of a weld under dynamic load. Chisel or peel off test

This is a simple but very important test, and is illustrated in Figure 7. A chisel is used to try to separate two the plates. Easy separation of the plates indicates incomplete welding.

Figure 7. The chisel test [1]

Tensile tests

Including shear and tensile tests.

d) Metallographic tests

Metallographic tests are performed using scanning electron microscopy (SEM). SEM is a useful method for evaluating the quality of welding. Under SEM, a perfect weld has a uniformly wavy geometry. Irregularities or cracks indicate problems with the weld.

2.8. Simulation of the explosive welding process

As most results of EXW have been obtained by explosive experiments, repetition of similar experiments can be avoided by using modeling and simulation. In addition, result prediction, parameter selection and wave distribution determination can be performed by simulation software. The most important simulation software packages in impact mechanics are ABAQUS, ANSYS, LS-DYNA, AUTODYN, RAVEN [39] and COMSOL [40]. Stress, strain, pressure, temperature distribution, behavior of materials and displacement of energy can be simulated in two or three dimensions. Figure 8 shows a 2D simulation of pressure distribution in an EXW process in AUTODYN software.

Figure 8. Pressure simulation by AUTODYN software [34]

(28)

(29)

3. Experimental

Data, Results and Calculations

3.1.

Experiment setup

T

he set up has been done in 5 stages:

Stage 1: EXW experiments on horizontal and flat surfaces with different materials

The materials used and their specifications are listed in Tables 2, 3, and 4. The setup for an experiment with a Fe-Fe joint is shown in Figure 9.

Fe-Fe

Table 2. Specification of Fe-Fe horizontal setup

No. Flyer plate Parent plate Stand-off .α Explosive .ρ VD Buffer Detonator

1 Fe-ͷͲ ൈ ͺͲ ൈ ͵݉݉ Fe-ͷͲ ൈͺͲ ൈ ͵݉݉ ͺ݉݉ 0 Ammonium-nitrate: 50 gr. dynamite : 10 gr. .8 1.2 2000 3000 Al foil 1 2 Fe-ͷͲ ൈ

ͺͲ ൈ ͵݉݉ Fe-ͷͲ ൈͺͲ ൈ ͵݉݉ ʹǤͷ݉݉ 4 dynamite : 40 gr. 1.5 4000 Al foil 1

Figure 9. Setup for Fe-Fe welding

Al-Cu

Table 3. Specification of Al-Cu horizontal setup

Flyer plate Parent plate Stand-off .α Explosive ߩ VD Buffer Detonator

Al-ͷͲ ൈ ͺͲ ൈ ͵݉݉ Cu-ͷͲ ൈ ͺͲ ൈ͵݉݉ ͺ݉݉ 0 Ammonium-nitrate: 110 gr. Dynamite: 40 gr. .8 1.2 2000 3000 Al foil 1 Fe-Cu

Table 4. Specification of Fe-Cu horizontal setup

Flyer plate Parent plate Stand-off .α Explosive ρ VD Buffer Detonator

Fe-ͷͲ ൈ ͺͲ ൈ ͵݉݉ cu-ͷͲ ൈͺͲ ൈ ͵݉݉ ͺ݉݉ 0 Ammonium-nitrate : 50 gr. Dynamite : 10 gr. .8 1.2 2000 3000 Al foil 1

Stage 2: Experimental work on EXW experiments on unequal flat surfaces with different materials

Horizontal, vertical and dual experiments were performed on unequal flat surfaces. These tests were performed prior to the subsequent stage of welding on curved shapes because explosive welding on cylindrical shapes is effectively welding on surfaces of unequal surface areas. The specifications of the experiments are shown in Tables 5 and 6. Figures 10 and 11 show the experimental setups of the different arrangements (horizontal, vertical and dual). In order to prevent problematic edge effects

(30)

when weld in Figure 1 Horizonta No. Flyer p 1 Al 30� 100 � 2 Al 50� 120 � Vertical u Flyer plate Al 50� 120 � �� Al 50� 120 � �� ding unequa 10 and will b al unequal s plate Pare � � �� Cu-200 � � �� Cu-200 nequal sur Ta Parent plat Cu-100 � 200 � 10�� Cu-100 � 200 � 10�� al surfaces d be explaine surface are Table 5. Sp ent plate S -100 � 0 � 10�� -100 � 0 � 10�� a rface areas able 6. Specificat e Stand-of �� different set ed in subseq eas (Al-Cu) ecification of Al Stand-off �� Figure 10. U a) b) (Al-Cu

)

tion of Al-Cu Ve ff .α E � � 0 0 D 7 D 7 tups were u quent section ) -Cu, unequal su Α Explos 0 Dynam powde 75 �� 0 Dynam powde 75 �� Unequal surface Al-Cu, unequal Al-Cu, unequal rtical unequal su Explosive Dynamite po 75 �� Dynamite po 75 ��

used for som ns.

urface area, horiz

sive .ρ mite er �� 1.2 gr./ mite er �� 1.2 gr./ e area set up l surface area, ho surface area, ho

urface area and

.ρ owder owder 1. gr 1. gr me experim zontal set up VD /�� 660_m/s /�� 660_m/s b orizontal set up orizontal set up

dual method set

ρ VD .2 r./�� .2 r./�� 660 m/s 660 m/s ments. These Buffer 0 1 mm Paper 0 1 mm Paper (first test) (second test) t up Buffer 00 00 1 mm Paper 1 mm paper 30 e are shown Detonator 1 1 Detonator 2 2 0 n

Impact wave process modeling and optimization in high energy rate explosive welding