8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
1/55
I
Mlardalen 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
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
2/55
II
Copyright Mohammad Tabatabaee Ghomi, 2009ISSN 1651-9256ISBN 978-91-86135-35-5
Printed by Mlardalen University, Vsters, Sweden
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
3/55
3
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 toproduce 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 explosivewelding 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, fluiddynamics, 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.
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
4/55
4
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
5/55
5
Sammanfattning
Impact vgor anvnds i mnga olika branscher och klassificeras beroende p om de orsakar plasteller elastiska deformationer. I plastisk deformation lge, kan dessa vgor anvndas fr att tillverka
speciella elektriska skarvar. I elastisk deformation lge, kan de anvndas fr att upptcka lckageeller att mta tjockleken p rren. Bda lgena har tillmpningar inom offshore-teknik.
I denna avhandling tillmpningen av effekten vgor i plastisk deformation mode och explosivasvetsning diskuteras. I den explosiva svetsning (EXW) behandla en hg hastighet sned anslaget aven noggrant kontrollerad explosion intrffar mellan tv eller fler metaller. Den hga hastigheteneffekten gr att metaller att bete sig som vtskor tillflligt och svetsa ihop. Detta sker under en korttid med hg energi.
EXW r en vlknd metod fr att g olika metaller tillsammans. Det r ett tvrvetenskapligtforskningsomrde och tcker ett brett spektrum av vetenskapliga och tekniska omrden inklusivevgrrelselra, strmningsmekanik, materialvetenskap, tillverkning och modellering. Mnga av deviktigaste resultaten i EXW forskningen erhlls frn experiment.
Denna avhandling r huvudsakligen baserad p experimentellt arbete. Dremot brjar det med engenomgng av grundlggande teori och mekanismer av explosiv svetsning och de olika stegen i enlyckad svetsning operation. Mnga olika EXW tester grs p horisontella och vertikala ytor medojmn yta, och p krkta ytor och rrledningar. terstoden av avhandlingen utvrderar resultaten avdessa frsk, tgrder viktiga parametrar och visar resultaten av simuleringar fr att verifiera de
experimentella resultaten.
Avhandlingen avslutas med ett antal frslag fr att frbttra och optimera EXW processen. En avdessa frbttringar r en modell fr att g metallplattor med ojmna ytor. En Al-Cu gemensamt
baserade p denna modell anvnds i ALMAHDI aluminium fabriken, ett stort fretag i sdra Iransom producerar mer n 200.000 ton aluminium per r. Frbttrade metoder fresls ocks fr att gmed krkta ytor. Dessa metoder kan ha mnga olika tillmpningar i rrledningar i olje-ochgasindustrin, srskilt i vattnet rr.
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
6/55
6
Acknowledgements
The work described in this thesis was carried out at the School of Sustainable Development of Society
and Technology, Mlardalen 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 fromGilan University in Iran for their scientific help.
I would like to express my gratitude to Dr. Adel Karim at Mlardalen 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.
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
7/55
7
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 forwelding Metals of Unequal Surface Area Using Explosive Energy, Submitted to InternationalJournal of Impact Engineering, ISSN: 0734-743XPaper2:
Mohammad Tabatabaee, Jafar Mahmoudi and Gholamhossein Liaghat,Removing Leakage from oiland gas low pressure Pipes and vessels by high energy explosive welding method, ScientificConference on Energy systems with IT in connection with the Energiting 2009, March 11-12 atlvsj fair, Stockholm, ISBN number 978-91-977493-4-3.Paper3:
Mohammad Tabatabaee, Jafar Mahmoudi and Gholamhossein Liaghat, Effect of Explosive Layerthickness on detonation velocity in a high energy process, Submitted to High Energy PhysicsJournal, ISSN: 1126-6708.
Report
Mohammad Tabatabaee,Impact wave process control in explosive welding application, Technicalreport No. 3, 20 July 2008, Mlardalen University
Papers not appended:
Paper 4:
Mohammad Tabatabaee and Jafar Mahmoudi, Finite element simulation of explosive welding, The49th Scandinavian Conference on Simulation and Modeling (SIMS2008), ISBN-13: 978-82-579-46326Paper 5:
Mohammad Tabatabaee and Jafar Mahmoudi, FEM method simulation for Aluminum - Iron -Copper bonding using explosive welding method, IASTED International Conference on AppliedSimulation and Modeling, June 25, 2008, at Corfu, Greece. (ASM 2008), ISBN- 978-0-88986-748-2Paper 6:
Mohammad Tabatabaee and Jafar Mahmoudi,An advanced method for Aluminum - Iron-
Copperbonding using explosive welding method, SSSEC2008 conference, Stockholm, March 12-13, 2008ISBN-978-91-977493-2-9Paper 7:
Mohammad Tabatabaee and Jafar Mahmoudi, An advanced method of explosive weldingsimulation, The 16thAnnual (International) Conference on Mechanical Engineering, ISME2008,May 14-16, 2008, Shahid Bahonar University of Kerman, Iran
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
8/55
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 Youngs modulus
C speed of sound
H hardness
A amplitude
T temperatureT thickness
Greek letters
initial angle
dynamic angle
wave length
tensile stress
density
Abbreviations
EXW explosive welding
BSEW bond strength explosive welded
WW welding window
SEM scanning electron microscope
UTS ultimate tensile stress
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
9/55
9
Table of ContentsAbstract ........................................................................................................................................... 3
Sammanfattning .............................................................................................................................. 5
Acknowledgements ......................................................................................................................... 6
List of Publications ......................................................................................................................... 7Nomenclature and abbreviations ..................................................................................................... 8
List of Figures ............................................................................................................................... 10
List of Tables ................................................................................................................................ 11
1. Introduction......................................................................................................................... 131.1. Background ................................................................................................................................. 13
1.2. Literature review ........................................................................................................................ 13
1.3. Motivation and objective ............................................................................................................ 14
1.4. Research approach ...................................................................................................................... 14
1.5. Limitations .................................................................................................................................. 151.6. Methodology ............................................................................................................................... 15
1.7. Thesis Outline ............................................................................................................................. 16
2. Theory ................................................................................................................................... 172.1. Mechanism and set up ................................................................................................................ 17
2.2. The impact wave in explosive welding ...................................................................................... 18
2.3. Predicting the wavelength .......................................................................................................... 19
2.4. Bonding criteria .......................................................................................................................... 20
2.5. Welding window ........................................................................................................................ 21
2.6. Governing equations ................................................................................................................... 24
2.7. Testing methods .......................................................................................................................... 24
2.8. Simulation of the explosive welding process ............................................................................. 27
3. Experimental Data, Results and Calculations ....................................................................... 293.1. Experiment setup ........................................................................................................................ 29
3.2. Experimental Results .................................................................................................................. 33
3.3. Calculation, numerical and simulation results ............................................................................ 35
3.4. Test Results ................................................................................................................................ 40
4. Discussion and future work .................................................................................................. 454.1. Discussion ................................................................................................................................... 45
4.2. Future work ................................................................................................................................ 47
5. Conclusions ........................................................................................................................... 495.1. Concluding remarks .................................................................................................................... 49
5.2. Practical Output .......................................................................................................................... 50
6. References ............................................................................................................................. 53
7. Papers summary .................................................................................................................... 55
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
10/55
10
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-plane .............................................................. 22Figure 6. Determination of detonation velocity by Dutrich method [ 4] ........................................... 25
Figure 7. The chisel test [ 1]. ............................................................................................................. 26Figure 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 ........................................................ 32Figure 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 ........................ 41Figure 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. AlCu joints transmit the electric power of anode rods in the aluminum factory............ 51
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
11/55
11
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 ................... 30Table 7. Specification of Stage 3 experiments ................................................................................... 31
Table 8. Specification of experiments in Stage 4 ............................................................................... 31Table 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 14. Calculation for line (a-a) .................................................................................................... 36
Table 15. Calculation for line (f-f) ..................................................................................................... 37
Table 16. Calculation for line (g-g) ................................................................................................... 37Table 17. Calculation for line (a-a) .................................................................................................... 38
Table 18. Calculation for line (f-f) ..................................................................................................... 38
Table 19. Calculation for line (g-g) ................................................................................................... 38
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
Table 23. Results of mechanical test in length direction ................................................................... 42
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
12/55
12
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
13/55
13
1. Introduction
Impact mechanics is a branch of applied mechanics that deals with high rates of energy and load ina 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 weldmany 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 hightemperature 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
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
14/55
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 unsteadinessmechanism, 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 varioustypes 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 toweld metals, a process called explosive welding.
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
15/55
15
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 ofunequal 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 papersPaper 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 43 Explosive materials 5
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
16/55
16
1.7. Thesis OutlineThis 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 3Experimental 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 6ReferencesPart 7 Summary of Papers
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
17/55
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
18/55
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 =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 ,,, , etc.) are defined in a specialdiagram 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 toprovide 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 readilyderived from the theory of elasticity [2].
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
19/55
19
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 dFigure 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 platec) 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]
= C . A (7)
Where A is amplitude and C is a constant.
2Axx
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
20/55
20
The plastic work is therefore:
(8)
Where lis 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)
whereis the kinetic energy andis the thickness of the flyer plate.The wavelength is obtained by equating the plastic work and the kinetic energy. The wavelengthcan 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 is critical and it is a
function of (the collision velocity). It has been shown experimentally that and (the velocityof 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 , it can be adjusted bysetting up an initial angle of obliquity (). The speed of sound provides an upper limit for and.These velocities may be increased by increasing . is assumed constant in the welding area in anexplosive welding process [10].
Collision angle()
The collision angle must be between 5 and 25 [1].
5 25 (13)
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
21/55
21
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 and 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 , , , , ,, and the properties of the material determine the WW.
This diagram can be drawn in both the - and - plane and displays an area within which the
weld is available. In this thesis the WW is drawn only in the - plane.
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
22/55
22
Figure 5. An example of a WW diagram in the Vc-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, inpractice 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 cwhich is necessary for jetting.
Abrahamson suggests the following relationship betweenand [24]:= 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 thespeed of sound, and which also limits the other WW parameters.
Lower and upper limit of[lines c-c and d-d]The lower and upper limits of the dynamic angle were experimentally obtained by Bahrani and
Crosslan [21]. They suggested a lower limit of between 2 and 3 and upper limit of 31 for in a
parallel geometry. Suggested minimum and maximum values of the initial angle 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:
22
(17)
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
23/55
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
24/55
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
25/55
25
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 differentpoints to an explosive material of unknown detonation velocity . The distance between thecontact 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= VdL1/ 2 L 2 (33)
L1, L2and 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.
Slanting wire method (for measuring and )
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
26/55
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 )
This is similar to the photographic method except in that it uses Xrays. Radiography can be used to
show the location and the moment of contact. The angle 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 thevolume 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]
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
27/55
27
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]
COMSOL software is used to check results in some sections of this thesis [40].
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
28/55
28
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
29/55
29
3. ExperimentalData, Results and Calculations
3.1. Experiment setup
The set up has been done in 5 stages:Stage 1: EXW experiments on horizontal and flat surfaces with different materialsThe 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 setupNo. Flyer plate Parent plate Stand-off . Explosive . VD Buffer Detonator
1 Fe-
Fe-
0 Ammonium-nitrate:50 gr.
dynamite : 10 gr.
.8
1.2
2000
3000
Al
foil
1
2 Fe-
Fe-
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 . Explosive VD Buffer Detonator
Al-
Cu-
0 Ammonium-nitrate:
110 gr.
Dynamite: 40 gr.
.8
1.2
2000
3000
Al foil 1
Fe-CuTable 4. Specification of Fe-Cu horizontal setup
Flyer plate Parent plate Stand-off . Explosive VD Buffer Detonator
Fe-
cu-
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 thedifferent arrangements (horizontal, vertical and dual). In order to prevent problematic edge effects
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
30/55
when wel
in Figure 1
Horizont
No. Flyer
1 Al 30
100
2 Al 50
120
Vertical u
Flyer plate
Al 50
120
Al 50120
ing unequ
0 and will
l unequal
plate Par
Cu-
20
Cu-
20
nequal sur
Ta
Parent plat
Cu-100
200
10
Cu-100 200
10
l surfaces
e explaine
urface ar
Table 5. Sp
ent plate
100
10
100
10
a
face areas
ble 6. Specificat
e Stand-o
ifferent se
d in subseq
as (Al-Cu)
ecification of Al
Stand-off
Figure 10.a)
b)
(Al-Cu)
ion of Al-Cu Ve
ff .
0
0
tups were
uent sectio
-Cu, unequal su
Explo
0 Dynapowd
75
0 Dynapowd
75
Unequal surfacAl-Cu, unequa
Al-Cu, unequal
rtical unequal s
Explosive
Dynamite po
75
Dynamite po75
sed for so
s.
rface area, hori
ive .
iter
1.2
gr.
iter
1.2
gr.
area set upsurface area, h
surface area, h
rface area and
.
wder
wder
1.
g
1.g
e experim
ontal set up
VD
660m/s
660m/s
b
rizontal set up
rizontal set up
dual method se
VD
.2
./
.2./
660m/s
660m/s
ents. Thes
Buffer
0
1 mm
Paper
0 1 mmPaper
(first test)
(second test)
up
Buffer
0
0
1 mm
Paper
1 mmpaper
30
are show
Detonator
1
1
Detonator
2
2
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
31/55
31
Figure 11. Al-Cu vertical unequal surface area and dual method set up
Stage 3: EXW of curved metal plates
EXW experiments were performed on curved metal plates by applying the results of Stage 2. The
experiments are summarized in Table 7. Figure 12 shows the experimental setup for filling a small
hole in a plate and Figure 13 shows the experimental setupfor welding on a curved surface.
Table 7. Specification of Stage 3 experiments
No. Flyer plate Parent plate Stand-off .
Explosive VD Buffer Fig.
1
0 Detonator(0.7 gr.)
7000 m/s Mastic 12
2 Cu 0.5
thickness
25 mmdiameter
Fe-80 mm
diameter 0 Detonator
(0.7 gr.)
7000 m/s Mastic 13
Figure 12. Set up for filling a small hole
Figure 13. Set up for welding on a curve
Stage 4: EXW on cylindrical surfaces with different materials
EXW experiments were performed on pipes and tubes. The materials and physical specification of
the experiments are shown in Table 8. Figure 14 shows the setup used to fill a small hole to repair a
leakin a pipe.
Table 8. Specification of experiments in Stage 4Flyer plate Parent plate Stand-off . Explosive VD Buffer Fig.
Cu 0.5 thickness25 mm diameter
Fe 80-3 mm Pipe 0 Detonator(0.7 gr.)
7000m/s
Mastic 14
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
32/55
32
Figure 14. Set up for repairing a leak and filling a small hole in a pipe
Key: 1-flyer plate Cu, 2-pipe Fe, 3-stand-off, 4-filler Fe, 5-detonator holder, 6- detonator
Stage 5: EXW Control parameters(explosive materials process parameters, test results)The specification of experiments is shown in Table 9. Figures 15 and 16 show experimental setups
for measuring Vdby the Dutrich method.
Table 9. Specification of experiments in Stage 5 (Dutrich method)
No. Wire
specification
Plate
specification
Explosive
thickness
Explosive
material
. L1 L
1 Cortex 5gr/m
Vd=6500m/s
8 mm AZAR* 1.4
gr./
2 Cortex 5gr/m
Vd=6500m/s
8mm AZAR* 1.4
gr./
3 Cortex 5gr/mVd=6500m/s
10mm AZAR* 1.4
gr./
4 Cortex 5gr/m
Vd=6500m/s
12mm AZAR* 1.4
gr./
5 Cortex 5gr/m
Vd=6500m/s
15mm AZAR* 1.4
gr./
6 Cortex 5gr/m
Vd=6500m/s
20mm AZAR* 1.4
gr./
*AZAR is an explosive mixture that includes TNT and ammonium nitrate.
Figure 15. Dutrich method setup using a thin Aluminum plate
Figure 16. Dutrich method setup using a Brass plate
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
33/55
3.2.
Stage 1
The result
Stage 2
Figure 18(horizonta
Stage 3Figure 19
horizontal
Experi
of explosi
a
d
shows th, vertical a
hows the r
and curved
ental Re
e welding
Fi
e resultsd dual me
a
cFigur
esults of ex
shapes.
ults
on horizon
ure 17. Results
f explosihod).
e 18. EXW of u
a, b) Res
c, d) Res
plosive we
al and flat
b
e
of EXW on hor
a, b, c )Experi
d, e) Experim
f) Experiment
e welding
equal surface a
lts of Al-Cu joi
lts of Al-Cu joi
ding for fil
surfaces ar
izontal flat surf
ents for Fe-Fe
nts for Fe-Fe w
s for Fe-Fe wel
on mater
reas
nt in horizontal
t in vertical an
ling a smal
shown in
ces
welding
elding
ing
als with
b
setup
d dual method s
hole using
igure 17.
c
f
nequal su
d
etup
a detonato
33
face areas
r charge on
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
34/55
Stage 4
Fig 20 sho
Stage 5
The result
the explos
at the mid
aluminum
ws the resu
of the exp
on. The po
le point of
and brass p
No.
1
2
3
4
5
6
lt of an EX
Figure 2
eriments ar
int is mark
the wire ac
lates.
a
T
Thicknes
material
8
8
10
12
15
20
aFigure 19. EX
a) R
b) R
test for
. EXW with a
e summari
d by the e
cording to
Figure 21.
ble 10. Result o
s of explosiv
mm)
tests for fillin
esult of welding
esult of welding
illing a sm
etonator on a p
ed in Tabl
fect of the
utrich me
Dutrich metho
a) Test on a thi
b) Test on a Br
f experiments d
L2
me
----
139
125
116
110
110
ba hole
for filling a sm
with a detonato
ll hole in a
ipe for filling a
10. Figure
xplosion, s
hod. The e
test results
Aluminum pla
ss plate
escribed in Tab
ark accordi
hod (mm)
----------------
.25
ll hole
r on a curved s
pipe.
mall hole
21 shows t
howing the
xperiment i
b
te
e 9
g to Dutrich
--------------
ape
e contact
length L2,
s performe
34
oint after
and M is
d on both
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
35/55
35
3.3. Calculation, numerical and simulation results
Welding windowWWs are drawn for the experiments. Results of some sample calculations are shown below.
Cu-Fe(Tables 7 and 8)
Flyer plate: Cu
0
9
4900
Parent plate: Fe 0
000 Line a-a
Table 11 shows the results of calculations using Equation 15.
Table 11. Calculation for line (a-a)
0 5 10 15 20 25 30 35 40 45
5.5 6 6.5 7 7.5 8 8.5 9 9.5 10
Line b-b
From Section 2.5 we get: 4900 0 Line c-c
From Section 2.5 we get: Line d-d
From Section 2.5 we get: Line e-e
Equation 17 gives
as follows:
0 0 Line f-f
Equation 21 gives as follows: 4 0
94900
00 Equation 22 gives as follows:
min=0 40 0 9
0
For an increased safety margin the higher value (300 m/s) is used and is calculated fromEquation 4. The results are shown in Table 12.
Table 12. Calculation for line (f-f)
3 5 10 20 30 40
5.7 3.52 1.72 0.86 0.57 0.44
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
36/55
36
Line g-g
Equation 25 gives as follows:=
The results are shown in Table 13.
Table 13. Calculation for line (g-g)
0 5 10 15 20 25 30 35
------------- 12.25 7 .15 4.06 3.4 2.95 2.61
The WW for Cu- Fe is drawn using these results and is shown in Figure 22. The weld-able area is
indicated by the crosshatched area.
Figure 22. WW for Cu-Fe joint
Fe-Fe(Table 2)
Flyer plate: Fe
Parent plate: Fe
Line a-a
Table 14 gives the results of calculations using Equation 15.
Table 14. Calculation for line (a-a)
0 5 10 15 20 25 30 35 40 45
5.5 6 6.5 7 7.5 8 8.5 9 9.5 10
Line b-b
From Section 2.5 we get: Line c-cFrom Section 2.5 we get:
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
37/55
37
Line d-d
From Section 2.5 we get: Line e-e
Equation 17 gives as follows:
=1460
Line f-f
Equation 21 gives as follows:
=(
Vc is obtained from Equation 4 and the results are shown in Table15.
Table 15. Calculation for line (f-f)
3 5 10 20 30 40
5.1 3.18 1.55 0.77 0.51 0.4
Line g-g
Equation 25 gives as follows:=
These results are shown in Table 16.
Table 16. Calculation for line (g-g)
0 5 10 15 20 25 30 35
-------- 12.25 7 .15 4.06 3.4 2.95 2.61
The WW for Fe-Fe is drawn from these results (Figure 23) and the weld able area is indicated.
Figure 23. Welding window for Fe-Fe joint
Al-Cu (Tables 5 and 6)
Flyer plate: Al
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
38/55
38
Parent plate: Cu
Line a-a
Table 17 shows the results of the calculation using Equation 15.
Table 17. Calculation for line (a-a)
0 5 10 15 20 25 30 35 40 45 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10
Line b-b
From Section 2.5 we get: Line c-c
From Section 2.5 we get:
Line d-dFrom Section part 2.5 we get: Line e-e
Equation 17 gives as follows:
=1466 Line f-f
Equation 22 gives as follows:= (
= (
Equation 22 gives as follows:min=
For greater safety we use =200 and from Equation 4, can be obtained. The results are shown intable 18.
Table 18. Calculation for line (f-f)
3 5 10 20 30 40
3.8 2.34 1.14 0.57 0.38 0.29
Line g-g
Equation 22 gives as follows:
The result is shown in Table 19.
Table 19. Calculation for line (g-g)
0 5 10 15 20 25 30 35 ---------- 12.25 7 .15 4.06 3.4 2.95 2.61
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
39/55
39
Figure 24 shows the WW for Al-Cu drawn from these results. The weld able area is indicated.
Figure 24. WW for Al-Cu joint
Explosive material
EXW experiments (horizontal and vertical) are arranged in a parallel setup with =0 and =. Byconsidering WW curves, a suitable detonation velocity and collision angle can be obtained.
= = 2500-6000 m/s = 25-10= 6000, = 10The ratio e/m can be calculated from Equations 4 and 32:
0.612
2
2
2 0.1 0.
where e is the explosive mass and m is the flyer plate mass.
For example, for an iron flyer plate weighing 100 grams, the explosive mass is estimated to be 80grams.
An alternative calculation for uses Equation 31 as follows:
= 0.557 C0= = 0.5576000 = 3462 flyer plate Fe= = 0.557 4900 = 2822 flyer plate Cu= = 0.5576400 = 3692 flyer plate Al
can be calculated from using Equation 18 as follows:
For Al with =3692 1.1
.
=.09 radian =5.5
Stand-off distance
As described in Section 2.4, the stand-off distance is selected at half or equal to the thickness of the
flyer plate. The optimum energy for jet formation may be obtained by varying the stand-off
distance.
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
40/55
40
Calculation for Stage 5
is given by Equation 33 according to the results from Tables 9 and 10 as shown below.L1=150 mm, L=950 mm, L2=139 mm, =6500 m/s
= L1/ 2 L2=2
Table 20 shows the results of calculation of for different thicknesses of explosive material.
Table 20. Calculation of VD for different thicknesses of explosive material
No. Thickness(mm)
L2 mark(mm)
VD(m/s)
1 -------- 8 ------
2 139.25 8 3500
3 125 10 3900
4 116 12 4200
5 110 15 4500
6 110 20 4500
Table 21 shows the measured detonation velocities in tests with different explosive materials of
different thicknesses at high, medium and low explosion velocities.
Table 21. Results of testing on measuring detonation velocity for different kinds of explosive materials
Thickness of
explosion (mm)
High velocity
explosionVd(m/s)
Medium velocity
explosionVd(m/s)
Low velocity
explosionVd(m/s)
0 0 0 0
5 5000 2600 1500
10 7300 3800 2200
15 7500 4500 2700
20 7500 5000 2900
25 7500 5000 3100
30 7500 5000 3300
35 7500 5000 3300
Wavelength
Equations 11 and 12 give the wavelength and amplitude for a 3 mm Fe plate and a dynamic angle=10.
2 2
2
2
Similarly, for a 5 mm Al plate and =6 we get
3.4. Test Results
Measuring explosive parameters before welding
Measuring Figures 25 and 26 show the effect of the thickness of explosive material on detonation velocity as
recorded in Tables 20 and 21, found by the Dutrich method.
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
41/55
41
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0 10 20 30
T mm
VD
m/s
Figure 25. Effect of explosive material thickness on detonation velocity
Figure 26. Effect of explosive material thickness on detonation velocity in general
(High, medium and low explosive velocity)
Measuring explosive parameters after welding
Chisel test
Figure 27 shows the chisel test for peeling of the metal pieces.
Figure 27. Chisel test for 2 types of weld
Mechanical tests
Tables 22 and 23 show the results of mechanical tests - bond strength of explosive weld (BSEW) -
in two directions (width and length) for the two setups described in Table 6 and Figure 15. These
results are illustrated in Figures 28 and 29.
0
2000
4000
6000
8000
0 10 20 30 40
T mm
Vd
m
/s
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
42/55
42
Table 22. Results of mechanical test in width direction
L mm 0 5 10 20 30 40 45 50
T(test1) kg/ 0 7 18 18 18 18 7 0
T(test 2) kg/ 15 18 18 19 18 17 18 14
Table 23. Results of mechanical test in length direction
L, mm 0 15 35 60 80 115 135 150
T(test1)
kg/
0 8 18 17 18 18 7 0
T(test 2)
kg/
16 18 18 19 18 18 17 15
Figure 28. BSEW in width directionSeries1. BSEW in simple setup in width direction of flyer plate
Series2. BSEW with groove in width direction of flyer plate
Figure 29. BSEW in length direction
Series1. BSEW in simple setup in length direction of flyer plate
Series2. BSEW with groove in length direction of flyer plate
0
10
20
0 10 20 30 40 50 60BSEW
Kg/mm2 Series1 Series2
L mm
0
10
20
0 20 40 60 80 100 120 140 160
BSEW
Kg/mm2
Series1 Series2
L mm
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
43/55
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
44/55
44
The simulations are performed for flat equal and unequal surfaces. Figure 31 shows a typical
simulation result for pressure distribution and temperature effects in unequal surfaces according to
the setup shown in Figure 18.
a
b
Figure 31. Results of simulation
a) Pressure contour in unequal surface welding
b) Temperature distribution during explosion
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
45/55
45
4. Discussion and future work4.1. Discussion
Requirements for successful bonding
Suitable setups for successful bonding can be identified through tests to measure various variables.Like a series of carriages that make up a train, a series of diagrams can be constructed, and the WW
in this composite diagram describes the test space where a successful weld can occur. In the
absence of this composite diagram, it is assumed that the most pertinent relationship is the one
between the pressure P, the impact velocity , parent plate velocity , and the effective strain. The
flyer plate attains its peak velocity at the collision point. On impact the velocity of the parent plate
at the collision point increases while the flyer plate velocity decreases. Under certain conditions, the
velocities of the flyer and the parent plate are the same. When the situation stabilizes at the collision
point and the pressure is high, inter-atomic bonding occurs.
Metallographic Tests
A fine structure was observed in most of the welding boundaries in SEM tests performed following
the EXW process. Nearly all the experiments with different shapes showed successful bonding.
Microstructure results of bonding are shown in Figures 30, which also shows the wavy pattern. In
the tests performed here, was measured at 0.7 mm and A was 0.14. These values confirmed the
results of the theoretical calculations in Section 3.3.
Chisel or Peel off Test
The bond strengths were measured by peel off and chisel tests. In some experiments, perfect results
were achieved (Figure 27), and there were no physical differences between the metals. The rupture
in these cases was in the weaker metal and away from the contact point. This was a confirmation of
the high quality of welding. Other experiments showed less impressive results, with joints that
separated easily (Figure 17a and c). This is likely to be for one of two main reasons - insufficient
explosive material or dirty surfaces.
Tensile Test
BSEW were measured in two directions in Stages 2 and 3. These tests showed that the bond
strengths were constant in the improved test with grooves throughout the contact surface according
to the setup shown in Figure 10b. In experiments using surfaces that lack grooves BSEW is very
low at the edges and increases towards the middle of the welding area. Figures28 and 29 show that
the strength reaches the maximum value within 1 to 1.5 times the thickness of the flyer plate fromthe edge. In the second test the edges were uniform and produced suitable joints after jumping. This
method is unique and is suitable for jointing two metals with different surfaces and does not have
the edge strength problems seen in Figures 28 and 29.
Discussion of Stage 4 (removing leakage)
The setup is one of the main problems when plugging a hole to stop leakage. The problem is
particularly difficult when the weld has to be applied in a wet area. A simple package is therefore
very desirable. This package includes detonator, a piece of thin copper plate, a piece of mastic as
holder, a piece of iron as filler and small rods to provide the stand-off. Figure 19a shows a
horizontal set up with the parent plate on the ground, which acts as an anvil. Figure 19b shows a
curved set up with a rod as the parent plate. Figure 20 shows a curved set up with a pipe as theparent plate. If the explosive is too high, the pipe may be damaged. A detonator has a small amount
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
46/55
46
of explosive material, is cheap, and is very useful for this application. The use of a detonator alone
is suitable for pipe applications and this is a good method for repairing leaks in metal pipes.
This method is not suitable for large holes and is used only for leakage and for low pressure liquid
pipes. Other methods of explosive welding are better suited to welding high pressure thin pipes.
Hardness test
In most experiments the hardness of the flyer plate is increased in a predictable manner because ofimpact waves.
Discussion of Stage 5 (controlling the parameters), Dutrich method
Aluminum plates were used initially during these tests, but these damaged too easily because of
their small thickness. Although they could be replaced by thicker aluminum plates, 3 mm Brass
plates were used in subsequent experiments.
The Dutrich method is very simple, but an error is inherent in the test because the effect of the
detonator is not taken into account. In these experiments a detonator is used to initiate the
explosion. This influences the result of the test. The undesired effects can be minimized by placing
the detonator outside of the explosive material. The experiments show that most of the explosive
velocity of detonation is constant if there is more than 30 mm of explosive between the detonator
and the plate (Figure 25). This is an important result that should be taken into account in future
setups. 3 types of explosives are tested, and the results of Dutrich method testing on high, medium
and low velocity explosions are shown in Figure 26. This figure can be used to derive a formula for
the region where the detonation velocity is constant. It is divided to three parts (over 10 mm for
high velocity, over 20 mm for medium velocity and over 30 mm for low velocity).
Discussion of simulation
The simulations show 3D maps and profiles of a number of physical parameters, such as contact
pressure, shear stress, normal stress, plastic strain, effective strain, strain rate, internal energy,
kinetic energy, temperature, velocity of the flyer plate at the point of contact and the angle ofcontact. Figure 31 shows the two important parameters of pressure and temperature. Plots of mesh
and material boundaries, and quantities as a function of time and distance for given coordinates are
also available. The maximum element size used is 1/35th (in 2D) of the maximum distance in the
geometry. In this case the element size used is 1/50. However, the maximum element size can be
explicitly specified in the Maximum element size edit field in the Custom mesh size area. A finer
mesh can also be created by selecting Custom mesh size and typing 0.01 in the Maximum element
size edit field. The mesh used here consists of 30758 triangular elements. The highest pressure
predicted is at the collision point and is on the order of 109Pa (Figure 31). In the case of the AZAR
explosive with high detonation velocity (6600 m/s), the pressure wave makes a 30 angle with the
horizontal surface in the base plate. This is because the velocity of detonation is higher than the
speed of sound in the material. Shock waves reflected at the end of the plates produce scabbing andspilling on the edges of metals. This phenomenon reduces the bond strength (BSEW) in the edges,
and is considered an important problem. Scabbing and spilling may be reduced by varying the
velocity of explosive charge and thereby regulating the welding, but the experiments show a
reduction in the strength of bonding in the edges when this is done, and formation of bonding in the
middle of welded metals (Figure 32). The simulation results confirm this and show that the highest
bond strength is in the middle of the plates (see Figure 31).
Figure 32. Bonded areas measured by ultrasonic test for Al-Cu joint
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
47/55
47
4.2. Future workThis thesis considers the impact mechanics area in the explosive welding field in the plastic zone.
More complete studies are needed on explosive welding or impact waves in plastic and elastic
zones. The most important of these future studies are described below.
Future work on explosive welding
-Simulation and numerical study to predict, compare and control parameters and experimental
results.
-Underwater EXW experiments.
- EXW on pressurized pipes.
- EXW on large surfaces with unequal surface areas.
- EXW on special multilayer surfaces.
- Other applications of EXW such as special cladding and stress relief.
Future work on plastic zone of impact waves
- Forming and shaping of metals
- Deep drawing
- Powder metallurgy
Future work on elastic part of impact waves
- Ultrasonic waves
- Measurement by impact waves
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
48/55
48
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
49/55
49
5. Conclusions
5.1. Concluding remarks
As described in this thesis, it is possible to join two or more metal parts using impact waves. Whenexplosive materials are used to generate the necessary impact, the process is called explosivewelding (EXW). This method does not require a heat source and no melt is created during the
process. Theinterface temperature throughout the process is lower than the melting points of either
material. EXW does not have the common problems of welding at the contact point and with
suitable control of the process parameters during the test, the welding is of the highest quality.
These process parameters are predictable and measurable. There are some limitations in EXW at
present and its use in mass production is still at a very early stage of research. This method covers a
multidisciplinary research area, many results have been reported, and many theoretical and
numerical methods have been invented in the last few decades. A number of numerical software
packages have been created to analyze the process but complete simulation of EXW process is not
yet possible.The most important conclusions of this thesis are as follows:
Stage 1:- The EXW joining of CuAl and Al-Al was successful and no fault was formed at the interface.
- A mixed explosion material with lower velocity for joining Al-Cu and Al-Al plates has been used
in the tests.
- The EXW joining of Fe-Fe was not successful at the first attempt, and the joint separated in the
peel off test. A satisfactory joint was obtained after cleaning the surfaces and replacing the
explosive material with a high velocity explosive material.
- For Fe-Fe joints, a medium or high velocity explosive material must be used.
- For Fe-Fe joints grinding the surfaces is recommended.- For all types of joints stress relief is recommended.
Stage 2:- The major result of these experiments was the discovery of a suitable solution for a uniform
contact in explosive welding of two metals with different surfaces. The solution to the problem was
the use of grooves on the flyer plates and jumping the edges during the test.
- An experimental distance to achieve uniform contact strength 1 to 1.5 t was confirmed.
- No joining fault was seen at the interface, and no melting voids or inter-metallic compounds were
observed in the SEM images. The joining of CuAl was successful after the explosive welding.
-The design of a successful vertical setup was another important result of these experiments. In this
set up there is no requirement for an anvil and the impact waves damped each other in the contactzone. Another advantage of this method was savings in time and cost in comparison to the
horizontal set up.
- Simulation results confirmed the edge problem. The groove model is suggested according to
simulation results.
Stage 3:- In curved welding it is best to place the explosion start point at the center of the flyer plate.
- It is possible to use the horizontal stand-off formula for the curved setup.
- In the curved setup the collision angle is variable, but if the flyer plate shape is changed to
resemble the parent plate, this angle remains constant and welding occurs as it does in the
horizontal set up.
- It is possible to use a detonator alone instead of explosive material for thin plates.
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
50/55
50
Stage 4:
-For repairing very thin pipes, a welding fixture must be used, but for pipes and vessels that are
more than 10 mm thick, a plate up to 5 mm thick can be used instead of a fixture.
-A new technique and a simple solution has been used to stop leaking in pipes. This method is
useful only for low pressure and small leaks.- Primary mechanical tests such as peel off have been carried out.
- Both surfaces of the plates must be clean.
- A pressure test is recommended to check the reliability of the repaired leak.
- Suitable fillers must be used to fill the hole before welding. The setup must be arranged so that the
filler does not jump during the explosion. Figure 33 shows a useful set up for this operation. This
system is suggested for use when:
(34)
Where is the diameter of the hole and D is the diameter of the pipe.
Figure 33. A set up for filling a hole before welding
Stage 5:
- There is a relationship between the thickness of explosive material and the explosive velocity.
The experiments described here support this statement and confirm previously published results.
The velocity of material in the small layer is variable and there is a different curve for each material
and a minimum thickness to achieve constant velocity derived from this curve. In this research
work, a curve for AZAR explosive material was obtained by experiment using the Dutrich method,and the results were applied in later experiments.
- Experiments show that high velocity explosive materials reach constant detonation velocity in a
thin layer. Figure 26 shows the relationship is as follows:
High velocity materials thickness>10 = constant
Medium velocity materials thickness>20 = constant (35)
Low velocity materials thickness>30 = constant
- In this thesis brass has been introduced as an alternative plate material for the Dutrich method.
- The location of the detonator can be changed to decrease the error when using the Dutrichmethod. The smallest error was achieved when the detonator was placed outside the explosive
material.
5.2. PracticalOutput
Aluminum, copper, and steel are the most common metals used in high-current conductor systems.
Use of these metals in dissimilar metal systems often maximizes the effects of special properties of
each material. However, joints between incompatible metals must be electrically effective to
minimize power losses. Mechanical connections that include aluminum create high resistance
because of the presence of the self-healing oxide skin on the aluminum component. Because thisoxide layer is removed by the jet in EXW, the interface of an explosion-clad aluminum assembly
offers no resistance to the current.
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
51/55
The result
aluminum
electric po
s of this th
(200,000 t
wer of ano
Figu
sis are bei
ns/year) i
e rods in t
e 34. AlCu oi
ng used in
the south
is factory,
ts transmit the
the ALM
rn Iran. U
and are sho
electric power
HDI alum
equal Al
wn in Figu
f anode rods in
num facto
u joints a
e 34.
the aluminum f
y, a large
e used to t
actory
51
roducer o
ransmit the
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
52/55
52
8/13/2019 Impact Wave Process Modelling and Optimization in High Energy Rate Explosive Welding
53/55
53
6. References
1.Experimental investigation of the mechanics of explosive welding by means of a liquid analogue.
El-Sobky, H., Blazynski, T. Z.Colorado : Fifth International Conference on High Energy Rate
Fabrication, 1975. Vol. 4.
2. B.Crossland.explosive welding of metals and its application. new york : oxford university press,
1982.3. the Academy of Science. Ukraine : E.O. Paton Electric Welding Institute .
4. Darvizeh, A.Research notebooks. Iran : Gilan University.5. Welding Handbook.
6. SME
; Handbook, Tool and Manufacturing Engineers. 1984. pp. 19-1. Vol. L.
7.Metals Handbook, Explosive welding. Vol. 6.
8. welding, Mechanism of wave formation at the interface in explosive. Chemin, C. and , No.2, ,.
May 1989, ACTA Mechanica sinca, Vol. 5. 2.
9.Permissible rang of parameters for interface wave formation in explosive welding; India. Gupta,
R.C and welding.India : s.n.10. Binding criterion for metals with explosive welding combustion, Explosion and shock wave.
Simonov, V.A.1991.
11.Explosive welding status. Phillipchuk, V.1961. ASTME Creative Manufacturing Seminar. pp.
P65100.
12. A. Williams, B. Crossland and J.D.Explosive welding. 1970. pp. 79100.
13. Explosive cladding of large steel plates with lead. Carpente, H. Otto and S.H. 1973. Met.
Mater. pp. 7579.
14.Microstructure of explosively bonded interface between titanium and very low carbon steel as
observed by TEM. Onzawa, T., Iiyama, T., Kobayashi, S., Takasaki, A., Ujimoto, Y.1985.
15.Analytic solutions of liquid-drop impact problems. M.B. Lesser, , Proc. ., ), pp. .London : R.
Soc, (1981. pp. 289308. A 377 .16. The geometric wave theory of liquid impact. Lesser, M.B., Field, J.E.1983. Sixth International
Conference on Erosion by Liquid and Solid Impact.
17. The impact of compressible liquids. Field, M.B. Lesser and J.E.1983. Ann. Rev.Fluid Mech.
Vol. 15, pp. 9712