A Feasibility Study of Cold Metal Transfer Gas Metal Arc Welding (CMT-GMAW) Nickel Base Superalloy Inconel 718

by Timothy Patrick Hasselberg A Thesis Submitted to the Graduate Faculty of Rensselaer Polytechnic Institute in Partial Fulfillment of the Requirements for the degree of MASTER OF SCIENCE Major Subject: MECHANICAL ENGINEERING

Approved: _________________________________________ Ernesto Gutierrez-Miravete, RPI Thesis Adviser _________________________________________ Samuel Christy, Pratt and Whitney Thesis Adviser Rensselaer Polytechnic Institute Hartford, Connecticut April 2009

 Copyright 2009 by Timothy Patrick Hasselberg All Rights Reserved

ii

CONTENTS
Cold Metal Transfer Gas Metal Arc Welding: CMT-GMAW Material Joining Characterization of Nickel Base Superalloy Inconel 718......i LIST OF TABLES ......v LIST OF FIGURES ...vi LIST OF EQUATIONS..xii NOMENCLATURE..xiii ACKNOWLEDGMENT ...xv ABSTRACT......xvi 1. Introduction 1 1.1 Fusion Welding Overview .3 1.2 Arc Welding Precipitation-Hardenable Nickel-Base Superalloys .5 2. The Cold Metal Transfer-Gas Metal Arc Welding Process 7 2.1 Arc Length ...11 2.2 Wire Feed Speed ..12 2.3 Traverse Speed .....13 2.4 Electrode Orientation ...13 3. Method of Approach ..15 3.1 Pre- and Post-Weld Thermal Treatments .17 3.2 Weldment Specimen Fabrication .18 3.3 Post-Weld Non-Destructive Evaluation ...21 3.4 Material Property Characterization .24 3.4.1 Micro-Analysis .....24 3.4.2 Specimen Mount Preparation26 3.4.3 Mechanical (Hardness) Analysis ..27 3.4.4 Mechanical (Tensile) Analysis .28 4. Results and Discussion ..31 4.1 Non-Destructive Analysis of CMT-GMAW Weldments 31 4.2 Destructive Analysis 34 4.2.1 Representative Weldment Microstructure 34 4.2.2 Macrostructural Comparison to GTAW ...38 iii

4.2.3 Microstructural Comparison to GTAW 39 4.3 Hardness Properties .42 4.4 Tensile Properties 43 4.4.1 Room Temperature Properties ..45 4.4.2 Elevated Temperature Properties ..47 4.4.3 General Comments and Property Comparison .50 5. Conclusions 52 6. References ..54 Appendix A .A-1 Appendix B. B-1 Appendix C.. C-1 Appendix D .....D-1

iv

LIST OF TABLES
Table 2-1: Effect of changes in process variables on weld attributes [7]..10 Table 3-1: Material specification for PMET 818 (AMS 5832).....15 Table 3.2-1: Various parameters examined for CMT-GMAW representative geometry weldments ..19 Table 3.2-2: Parameters used for tensile weldments.....21 Table 3.4.2-1: Grinding and polishing procedure used for micro-sections ..27 Table 3.4.4-1: Machining procedure for tensile specimen extraction...29 Table 4.1-1: Visual observations made during processing of representative geometry weldments ..31 Table 4.3-1: Material hardness results for GTAW and CMT-GMAW weldments ..43 Table 4.4.1-1: Room temperature (65F) tensile properties .45 Table 4.4.2-1: Elevated temperature (1100F) tensile properties ....48 Table A-1: Parameters used for weld trials.....A-1

LIST OF FIGURES
Figure 1-1: High speed photography image of a CMT cycle with real time stamps [1] ...2 Figure 1.1-1: Fusion welding processes tree diagram [2] ..3 Figure 1.1-2: Arc welding processes tree diagram [2]....4 Figure 1.1-3: Basic arc welding circuit diagram [2] ...5 Figure 1.2-1: Diagram showing the effect of aluminum and titanium hardener content on the tendency to PWHT cracking [3].6 Figure 2-1: Fronius CMT-GMAW process control diagram [1] ....7 Figure 2-2: Schematic representation of short circuiting metal transfer [8] ...8 Figure 2-3: CMT-GMAW wire retraction process [1]........9 Figure 2-4: Comparative thermal inputs for various metal transfer processes [1]10 Figure 2.1-1: Constant-potential power source illustration [8].12 Figure 2.4-1: Effect of electrode position and welding technique [8] ..14 Figure 3-1: Illustration of welding equipment used for robotic trials [4] 16 Figure 3.2-1: Illustration of straight single-line beads, straight-line-weave pattern beads, circular weld build-ups for CMT-GMAW representative geometry weld trials ...20 Figure 3.2-2: Illustration of edge weld build-ups for CMT-GMAW representative geometry weld trials...20 Figure 3.2-3: Stock geometry used for tensile weldment fabrication ...21 Figure 3.2-4: Illustration of tensile weldment post-welding ....21 Figure 3.3-1: Types of gas porosity commonly found in weldments [9] .....22 Figure 3.3-2: Lack of fusion in various weld joints [9]....22 Figure 3.3-3: Identification of cracks according to location in weld and base metal [9]..22 Figure 3.4.1-1: Illustration of CMT-GMAW representative geometry weldments with indicated cut planes ....25 Figure 3.4.1-2: Illustration of micro section plane definitions [10]..25 Figure 3.4.1-3: Illustration of tensile weldments with indicated tensile extraction and cut planes .26 Figure 3.4.3-1: Illustration depicting the various locations for hardness testing ..28 Figure 3.4.4-1: Illustration of tensile specimen used for testing...29 Figure 3.4.4-2: Illustration of tensile specimen extraction ...29 vi

Figure 3.4.4-3: Illustration of tensile specimen loading scheme ......30 Figure 4.1-1: White light macro image (post thermal treatment) of representative CMTGMAW weldment specimen #3.....32 Figure 4.1-2: White light macro image (post thermal treatment) of representative CMTGMAW weldment specimen #5 ....32 Figure 4.1-3: X-ray image of representative geometry CMT-GMAW weldment specimen #3....33 Figure 4.1-4: X-ray image of representative geometry CMT-GMAW weldment specimen #5....33 Figure 4.2.1-1: Transverse micro-section location 3 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid....34 Figure 4.2.1-2: Planar micro-section location 8 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid ..35 Figure 4.2.1-3: Longitudinal micro-section location 12 from single straight-line pattern, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid ...35 Figure 4.2.1-4: Transverse micro-section location 4 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid36 Figure 4.2.1-5: Planar micro-section (post thermal treatment) showing primary dendrite growth, etchant: electrolytic 10% oxalic acid 37 Figure 4.2.1-6: Longitudinal micro-section (post thermal treatment) demonstrating the randomness of dendrite growth, etchant: electrolytic 10% oxalic acid .38 Figure 4.2.2-1: Macro image analysis of tensile weldments (post thermal treatment) showing weld profile and distortion of manual GTAW and automated CMT-GMAW 39 Figure 4.2.3-1: Longitudinal micro-sections of both CMT-GMAW and GTAW tensile weldments (post thermal treatment), etchant: electrolytic 10% oxalic acid ..40 Figure 4.2.3-2: Transverse micro-sections of both CMT-GMAW and GTAW tensile weldments (post thermal treatment) exhibiting deposit height, etchant: electrolytic 10% oxalic acid ..41 Figure 4.2.3-3: Transverse micro-sections of both CMT-GMAW and GTAW tensile weldments (post thermal treatment) exhibiting deposit depth, etchant: electrolytic 10% oxalic acid...42 vii

Figure 4.4-1: Macro image of flat specimen used for testing, specimen CMT-1 .44 Figure 4.4.1-1: Representative macro photographs, room temperature specimens GTAW1 and CMT-2 .46 Figure 4.4.1-2: SEM photographs of fractured surfaces, room temperature specimens GTAW-1 and CMT-2 47 Figure 4.4.2-1: Representative macro photographs, elevated temperature specimens GTAW-6 and CMT-4.49 Figure 4.4.2-2: SEM photographs of fractured surfaces, elevated temperature specimens GTAW-6 and CMT-4 49 Figure A-1: Specimen 25-1 (20, Pull Angle). ...A-2 Figure A-2: Specimen 25-3 (0, Neutral Angle) ...A-2 Figure A-3: Specimen 25-6 (20, Push Angle)...A-2 Figure B-1: White light macro image (post thermal treatment) of representative weldment specimen #1 (wire brushed). ...B-2 Figure B-2: White light macro image (post thermal treatment) of representative weldment specimen #2 (wire brushed)....... .B-2 Figure B-3: White light macro image (post thermal treatment) of representative weldment specimen #3 (as welded). B-3 Figure B-4: White light macro image (post thermal treatment) of representative weldment specimen #4 (wire brushed). ...B-3 Figure B-5: White light macro image (post thermal treatment) of representative weldment specimen #5 (as welded). B-4 Figure C-1: Planar micro-section location 1 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid. C-2 Figure C-2: Longitudinal micro-section location 2 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid. C-2 Figure C-3: Transverse micro-section location 3 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid. C-3 Figure C-4: Transverse micro-section location 4 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid. C-3

viii

Figure C-5: Transverse micro-section location 5 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid. C-4 Figure C-6: Longitudinal micro-section location 6 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid. C-4 Figure C-7: Longitudinal micro-section location 7 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid. ...C-5 Figure C-8: Planar micro-section location 8 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid. C-5 Figure C-9: Transverse micro-section location 9 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid ....C-6 Figure C-10: Transverse micro-section location 10 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid ....C-6 Figure C-11: Transverse micro-section location 11 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid ...C-7 Figure C-12: Longitudinal micro-section location 12 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid ...C-7 Figure C-13: Planar micro-section location 13 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-8 Figure C-14: Transverse micro-section location 14 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-8 Figure C-15: Transverse micro-section location 15 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid ...C-9 Figure C-16: Transverse micro-section location 16 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid ...C-9 Figure C-17: Longitudinal micro-section location 1 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid ...C-10 Figure C-18: Transverse micro-section location 2 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid ...C-10 Figure C-19: Transverse micro-section location 3 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid ...C-11

ix

Figure C-20: Transverse micro-section location 4 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acidC-11 Figure C-21: Longitudinal micro-section location 5 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acidC-12 Figure D-1: Post-test macro image of room temperature specimen CMT-1 .D-2 Figure D-2: Macro image of fractured surface, specimen CMT-1 ....D-2 Figure D-3: Post-test macro image of room temperature specimen CMT-2 .D-3 Figure D-4: Macro image of fractured surface, specimen CMT-2 D-3 Figure D-5: Post-test macro image of room temperature specimen CMT-3 .D-4 Figure D-6: Macro image of fractured surface, specimen CMT-3 ....D-4 Figure D-7: Post-test macro image of elevated temperature specimen CMT-4 D-5 Figure D-8: Macro image of fractured surface, specimen CMT-4 ....D-5 Figure D-9: Post-test macro image of elevated temperature specimen CMT-5 D-6 Figure D-10: Macro image of fractured surface, specimen CMT-5 ..D-6 Figure D-11: Post-test macro image of elevated temperature specimen CMT-6 ..D-7 Figure D-12: Macro image of fractured surface, specimen CMT-6 ..D-7 Figure D-13: Post-test macro image of room temperature specimen GTAW-1 ....D-8 Figure D-14: Macro image of fractured surface, specimen GTAW-1 ...D-8 Figure D-15: Post-test macro image of room temperature specimen GTAW-2 ...D-9 Figure D-16: Macro image of fractured surface, specimen GTAW-2... D-9 Figure D-17: Post-test macro image of room temperature specimen GTAW-3 ..D-10 Figure D-18: Macro image of fractured surface, specimen GTAW-3 D-10 Figure D-19: Post-test macro image of elevated temperature specimen GTAW-4 .D-11 Figure D-20: Macro image of fractured surface, specimen GTAW-4 .D-11 Figure D-21: Post-test macro image of elevated temperature specimen GTAW-5 .D-12 Figure D-22: Macro image of fractured surface, specimen GTAW-5 .D-12 Figure D-23: Post-test macro image of elevated temperature specimen GTAW-6 .D-13 Figure D-24: Macro image of fractured surface, specimen GTAW-6 .D-13 Figure E-1: Stress-Strain curve of room temperature CMT specimens .E-2 Figure E-2: Stress-Strain curve of elevated temperature CMT specimens ....E-3 Figure E-3: Stress-Strain curve of room temperature GTAW specimens ..E-4 x

Figure E-4: Stress-Strain curve of elevated temperature GTAW specimens .E-5

xi

LIST OF EQUATIONS
Equation 4.2.1-1 .38 Equation 4.4-1 44 Equation 4.4-2 44 Equation 4.4-3 42

xii

NOMENCLATURE
A Af Ao AC AMS CG CMT CMT-GMAW DC E EGW/ESW e FCAW FPI G gf GMAW GTAW HAZ HRC HV HT ipm L Lf Lo LOF LOP LLC LST Arc Current, amps Final Area Original Area Alternating Current Aerospace Material Specifications Columnar Grain Cold Metal Transfer Cold Metal Transfer-Gas Metal Arc Welding Direct Current Elastic Modulus Electrogas Welding/Electroslag Welding Elongation Flux Core Arc Welding Florescent Penetrant Inspection Temperature Gradient (C/cm) gram-force Gas Metal Arc Welding Gas Tungsten Arc Welding Heat Affected Zone Rockwell C Hardness Vickers Hardness Heat Treatment Inches Per Minute Arc Length, inches Final Length Original Length Lack of Fusion Lack of Penetration Limited Liability Company Local Solidification Time xiii

N NDE ODS Pmax PAW PWHT R RA RPM Su SAW SEM SMAW TL Ts V

Newton Non-Destructive Evaluation Oxide Dispersion Strengthening Maximum Force Plamsa Arc Welding Post-weld Heat Treatment Solidification Rate (cm/min) Reduction in Area (%) Revolutions Per Minute Ultimate Strength (Tensile) Submerged Arc Welding Scanning Electron Microscope Shielded Metal Arc Welding Liquidus Temperature (C) Solidus Temperature (C) Arc Voltage, volts Gamma Prime Precipitate Gamma Double Prime Precipitate

xiv

ACKNOWLEDGMENT
I would like to express my heartfelt appreciation to my Pratt and Whitney advisor, Dr. Samuel Christy, for providing continuous technical guidance and support throughout this research thesis. I would like to extend my appreciation to my Rensselaer Polytechnic Institute advisor, Dr. Ernesto Gutierrez-Miravete, for his additional support and acknowledgement of this research. I would also like to thank the following people who provided support and encouragement throughout this thesis, all of whom are my colleagues at Pratt and Whitney: Mr. Bruce Saxton, Mr. Russell Melnick and Mr. William Rose for providing the ideas and funding for this research; Mr. David Rutz for providing endless amounts of technical guidance; technician, Mrs. Daria Palladino, for providing the various weldment mounts; Mr. David Gaudreau for providing micro-imaging analysis, and finally Mr. John Finn for providing his GTAW artisanship. I would also like to extend my appreciation to Mr. Shaun Relyea of Fronius USA LLC for providing guidance on weld parameter development and optimization. Finally, a special thanks to my friends and family for without their support and direction, I would not have been able to complete this research endeavor. Lastly, I would like to extend a special thanks to Erin Altman for editing this thesis.

xv

ABSTRACT
The subject of this research is the metallurgical and structural characterization of welding wrought nickel base superalloy Inconel 718 (AMS 5596) using a new Gas Metal Arc Welding (GMAW) process dubbed Cold Metal Transfer (CMT). It was hypothesized that CMT-GMAW would provide an equivalent, if not increased, weld joint yielding improved metallurgical and structural capabilities when compared to the Gas Tungsten Arc welding (GTAW). The CMT-GMAW weldment in question was placed though a series of tests, varying weld parameters with the objective of obtaining a standardization of the processs critical inputs, Non-Destructive Evaluation (X-ray and Florescent Penetrant Inspection [FPI]); micro hardness; metallographic review (macroscopic and microscopic, including scanning electron microscopy); tensile (room and elevated temperature). Prior to welding, the wrought AMS 5596 substrate material was solutioned. After welding, the material was placed through a conventional post weld heat treatment cycle including a solution and precipitation heat treatment. Under the various test conditions previous mentioned, the CMT-GMAW process was directly compared to the current industry standards for GTAW. According to the parameter optimization study it was noted that the key attribute of the CMT-GMAW process is its electronically controlled short circuit droplet detachment method, which is dictated by the weld synergic line. The synergic line is a linear mathematical relationship, proprietary to Fronius International LLC, which incorporates voltage and amperage process controls into the wire feed speed. All non-destructive and destructive analyses of the CMT-GMAW weldments showed little evidence of the porosity that is commonly inherent when using a conventional GMAW process. Macro and micro-analysis of the CMT-GMAW weldments exhibited a columnar grain microstructure similar to those obtained with conventional GTAW with the exception of a reduced substrate consumption, Heat Affected Zone (HAZ), and less distortion weld. Mechanical hardness properties of the CMT-GMAW were evaluated at various metallographic locations, including the substrate, HAZ, and weld filler locations. All measurements confirmed equivalency to those obtained from GTAW and were consistent with the current AMS standards. xvi

Tensile properties of the CMT-GMAW weldments were at both room temperature (65F) and a serviceable elevated temperature of 1100F. The results obtained indicated equivalency to GTAW, with the exception that the CMT-GMAW specimens exhibited approximately 2x increase in the percentage of Reduction in Area (RA) at both room and elevated temperatures. On the basis of overall evaluation, it was concluded that CMT-GMAW provides the following benefits: excellent weld quality on wrought Inconel 718; comparable metallographic structure to those commonly seen in fusion welded deposits; increased deposition rates when compared to GTAW; a reduction in overall thermal input by achieving almost current-free metal transfer; virtually spatter-free metal transfer by controlling the short circuiting; comparable material hardness to GTAW weldments; and analogous tensile strength with increased RA when compared to GTAW.

xvii

1. Introduction
The purpose this research is to metallurgically and structurally characterize the welding of nickel base superalloy Inconel 718 using a Cold Metal Transfer-Gas Metal Arc Welding (CMT-GMAW). During the research there were several milestones or objectives that were sought to be met including: welding parameter optimization; solution and precipitation heat treatment; Non-Destructive Evaluation (X-ray and Florescent Penetrant Inspection [FPI]); micro hardness; metallographic review (macroscopic and microscopic); and tensile (room and elevated temperature). Until now, aircraft engine manufacturers have faced a choice between GTAW or GMAW as a low cost option for arc welding gas turbine components during manufacturing and repair. GTAW has successfully captured much of the market, as the limiting factors of traditional GMAW have always been the joining of unique material types (including non-ferrous materials) and consistent quality (i.e. the weld spatter inherent in traditional GMAW). Fronius International LLC, a European market leader in arc-welding technology, introduced the concept of CMT-GMAW welding to the public in 2005. By incorporating the traditional concepts of the GMAW method that is, applying a wire consumable Fronius was able to produce a process of arc welding with high levels of accuracy on various materials which were typically reserved for GTAW welding. The CMT-GMAW is a fully digital, micro-processor-controlled inverter welding process that results in the introduction of a reduced amount of residual heat to the workpiece and produces a virtually spatter free weld. The improved weld quality is obtained via a digital process-control that detects a short circuit, and then retracts the wire being feed so as to help detach and deposit a single molten droplet at a time, as shown in figure 1-1. Figure 1-1 shows one cycle of the CMT-GMAW wire retraction process using high-speed photography. It is important to note the short duration of arcing period, approximately 1/3 of the total cycle time (14.31ms). [1]

Figure 1-1: High-speed photography image a CMT cycle with real time stamps [1]

While it is well known that GTAW is the industry standard, it also requires a highly skilled operator when performed manually. The process typically necessitates an operator to use both hands to execute the weld: one for holding the torch and the other for feeding in filler to ensure maximum precision. GTAW also uses a tungsten electrode from which an arc is generated to provide the heat for the addition of the filler material, creating a large temperature differential between the filler and the workpiece. Additionally, there is a potential for transfer of molten tungsten from the electrode to the weld causing contamination, resulting in a hard and brittle inclusion. However, the CMTGMAW uses a consumable electrode that acts not only as the filler, but also the heat conductive electrode. This coupled to the computer controlled wire retraction mechanism results in a process that greatly reduces the amount of heat applied to the workpiece during joining. Additionally, the thermal input levels of CMT-GMAW are much lower and more controllable than conventional GMAW or GTAW welding, since each of these processes cannot go below a certain heat level to create an arc. Speed and spatter control are the two key benefits of Fronius CMT-GMAW. CMT-GMAW welding offers a speed that is four to five times faster than conventional GTAW, with levels of consistent quality matching those of robotic automation. [1]

1.1 Fusion Welding Overview

Fusion welding is any process that uses thermal energy to produce melting of the metals to be joined to create a weld pool that ultimately creates a solid joint when solidified. The melting process allows the materials to flow and mix together to form a sound metallurgical bond upon solidification. Typically, a filler material, generally of suitable material composition, is added in a molten form to enhance the strength of the bond. Most fusion welding processes generally require a heat source of a sufficient temperature and intensity to produce a localized melting of both the substrate material and the subsequent filler material. Fusion welding of materials can be accomplished by both chemical and electrical means, figure 1.1-1. Figure 1.1-1 is a process tree diagram that represents the various fusion welding processes, including the break down into sub categorical processes (e.g. flame, arc, radiation, etc.). However, the specific focus of this research is based on arc welding processes. [2]

Figure 1.1-1: Fusion welding processes tree diagram [2]

A formal definition of arc welding is a group of joining processes that produces a coalescence of workpieces by heating them with an electrical arc. Figure 1.1-2 is a process tree diagram that represents the various arc welding processes, including the break down into sub categorical processes. However, the specific focus of this research is based on Gas Metal Arc Welding (GMAW) and Gas Tungsten Arc Welding (GTAW). A

fusion arc welding process basically consists of a power source; workpiece lead, electrode lead, and a workpiece (see figure 1.1-3). The power source provides the welding voltage and current. Welding current may be either AC (Alternating Current) or DC (Direct Current). DC is by far the most common power source; although, many power sources contain circuitry for converting AC electricity to DC. In order for an arc to take place, there must be some means for attaching the welding electrode lead to the power source. This may be initiated through fixed connection or sliding contact, depending on the specific arc process. The workpiece lead must be connected to the workpiece, which is usually accomplished by a clamp. Welding electrodes are either consumable (GMAW) or non-consumable (GTAW). Consumable electrodes melt and transfer to the weld pool as the source of filler material as well as the electrode. In contrast, non-consumable electrodes when used properly do not melt and transfer to the substrate material during the arcing process. Tungsten is generally used for this purpose since it has a high melting point (6147F). [2]

Figure 1.1-2: Arc welding processes tree diagram [2]

Figure 1.1-3: Basic arc welding circuit diagram [2]

1.2 Arc Welding Precipitation-Hardenable Nickel-Base Superalloys

The precipitation-hardened alloys distinguish themselves by exhibiting superior mechanical properties after undergoing precipitation treatment (aged). These materials are characterized by their distinctively high strength at room temperature to approximately 1300F. Fusion often leads to dissolution of the hardening phases and their reprecipitation in a less desirable physical form in the matrix. The objective of employing joining processes on precipitation-hardened superalloys, particularly nickel-base superalloys, is to find a way to keep the high strength associated with the hardening phases, or the long-term strength associated with oxide dispersion strengthening, or ODS from being lost due to the welding process. Depending on the actual composition, these alloys may have good weldability, and most are formed, machined, and welded in the solutioned condition. The processed materials are then re-solutioned after welding and aged to obtain the desired properties. The post-weld processing of these alloys after fusion welding may be considerably difficult, caused by a tendency to cracking. The susceptibility to hot cracking is directly related to the aluminum and titanium contents. Figure 1.2-1 shows a plot of weldability as a function of (Ti + Al) content, a number that will reflect the expected level of precipitates. Here it can be seen that with increasing Ti and Al content one can expect the weldability of the material to decrease. [3]

Figure 1.2-1: Diagram showing the effect of aluminum and titanium hardener content on the tendency to PWHT cracking [3]

It is well known that some precipitation-hardened nickel-base and iron-nickelbase superalloys are considerably less weldable than cobalt-base superalloys. Because of the presence of the strengthening phases when fusion welded, nickel-base alloys tend to be susceptible to hot cracking (weld cracking) and Post-Weld Heat Treatment (PWHT) cracking, sometimes referred to as strain-age cracking. Fortunately, alloys such as Inconel 718 and Inconel 706 are much less susceptible to PWHT cracking since the age hardening develops via the Ni, Nb, , precipitate, but the precipitate is formed at a much slower rate then those -hardened superalloys. This allows superalloys such as Inconel 718 to be heated into the solution temperature range without suffering significant aging and the resultant PWHT cracking. [3]

2. The Cold Metal Transfer-Gas Metal Arc Welding Process

In most aerospace applications, it is highly desirable that the fusion arc welding process has low thermal input. CMT stands for Cold Metal Transfer. Of course, the term cold has to be understood in terms of a welding process: when set against conventional GMAW, CMT-GMAW is indeed a cold process with its characteristic feature of alternating thermal arc pool, i.e. hot when an arc is initiated and cold when the arc is extinguished and the wire is retracted. This alternating hot and cold treatment has been made possible by a new technological development from Fronius International LLC that incorporates the wire motions into the process control via a computer monitoring system (see figure 2-1). Some of the other features that make this unit unique when compared to conventional GMAW units include: two separate wire-drives (front and rear) that are separated by the wire buffer. The front drive, located in the torch, (see figure 2-1A), moves the wire back and forth in a dabbing motion at a rate of up to 90 times per second. Simultaneously, the rear drive pushes the wire directly from the filler spool located in the wire drive. It is important to note that both drives are digitally controlled by the process control. To ensure a constant wire feed, a wire buffer (see figure 2-1B) is interposed between the two drives to decouple them from one another. [1, 4, 5 and 6]

Figure 2-1: Fronius CMT-GMAW process control diagram [1]

In the world of GMAW processes, there are essentially three types of metal transfer mechanisms that can be used: (1) short circuiting transfer, (2) globular transfer,

and (3) spray transfer. The type of transfer is determined by a number of factors, the most influential of which include (1) magnitude and type of welding current, (2) electrode diameter, (3) electrode composition, (4) electrode extension, and (5) shielding gas. CMTGMAW in its most elementary form can be considered a short circuit GMAW process. Short circuit GMAW consists of the lowest range of welding currents and electrode diameters associated with GMAW. This type of transfer produces a small, fast-freezing weld pool. During the arcing process, metal is transferred from the electrode to the workpiece only during a short period when the electrode is in contact with the weld pool, hence the term short circuiting transfer. It is important to note that no metal is transferred across the arc gap. [7 and 8] The sequence of events in the transfer of metal and the corresponding current and voltage for a typical short circuit transfer GMAW process can be seen in figure 2-2. As the wire touches the weld metal, the current increases [(A), (B), (C), (D), in figure 2-2]. The molten metal at the wire tip pinches off at D and E, initiating an arc as shown in (E) and (F). It is here that the rate of current increase must be high enough to heat the electrode and promote metal transfer, yet low enough to minimize spatter caused by violent separation of the drop of metal (one of the major disadvantages of conventional short circuit GMAW). Finally, when the arc is established, the wire melts at the tip as the wire is fed forward towards the next short circuit at (G). Overall, the open circuit voltage of the power source must be low enough that the drop of molten metal at the wire tip cannot transfer until it touches the base metal. [7 and 8]

Figure 2-2: Schematic representation of short circuiting metal transfer [8]

Unlike its short circuit GMAW counterpart, CMT-GMAW incorporates a digital process-control that detects the short circuit at the workpiece, and then mechanically retracts the wire to help detach the molten droplet. The wire retraction greatly reduces the spatter that is typically associated with conventional short circuit GMAW. Reduced spatter is the first essential difference from conventional short circuit GMAW processes; the second most notable difference is a reduction in thermal input since there is virtually current-free droplet detachment, hence the term Cold Metal Transfer. Lastly, unlike conventional short circuiting GMAW -where there is a constant push motor driven system, the CMT-GMAW uses a two motor drive system that pushes the wire forward, and as soon as the short circuit occurs, it pulls it back (see figure 2-3). [1]

Figure 2-3: CMT-GMAW wire retraction process [1]

The rearward movement of the wire assists droplet detachment during the short circuit. In this way, the arc itself only inputs heat very briefly during the arcing period. The thermal input is immediately reduced after arc is extinguished, creating an oscillating hot/cold weld pool. During the CMT-GMAW process, the average current is kept very small by controlling the short circuit, resulting in virtually spatter free metal transfer. Figure 2-4 illustrates the reduced thermal input required for metal transfer as compared to other conventional metal transfer processes. Also worth mentioning, precision dropletdetachment of the CMT-GMAW ensures that after every short circuit, a near-identical quantity of filler metal is melted off. [1, 4, 5 and 6]

Figure 2-4: Comparative thermal inputs for various metal transfer processes [1]

The most important variable of any GMAW process, including CMT-GMAW, which affects the weld penetration, bead geometry, and overall weld quality are: (1) welding current (wire feed speed), (2) polarity, (3) arc voltage (arc length), (4) travel (traverse) speed, (5) electrode extension, (6) torch angle, and (7) electrode diameter. Knowledge and control of these variables are essential in order to consistently produce welds of acceptable quality. However, changing one variable generally requires altering additional parameters to retain an acceptable quality weld because the variables are not completely independent of each other. The effects of these variables on deposit attributes are shown in table 2-1. [7]

Table 2-1: Effect of changes in process variables on weld attributes [7]

10

2.1 Arc Length

Arc voltage and arc length are related terms that are often used interchangeably, however they are quite different in practice. Arc voltage is the electrical potential between the electrode and the workpiece. Arc voltage is generally lower than the voltage measured directly at the power source because the voltage drops at the connections and along the length of the welding cable. Consequently, an increase in arc voltage will result in a longer arc length. It is also important to note that excessively high arc voltage can cause weld porosity, spatter, and undercut; therefore, arc length is a variable of interest and should be controlled as it can have a profound impact on the overall weld quality. [7 and 8] In conventional GMAW, the surface of the workpiece (i.e. jagged or flat) and the welding speed can both have a marked effect on the stability of the arc. The arc length is acquired and adjusted mechanically with the CMT-GMAW. This means that the arc remains stable, regardless of the surface condition of ones workpiece. In addition to a mechanical response, the CMT-GMAW utilizes a self-correction mechanism via a constant-potential power source as illustrated in figure 2.1-1. Where L is the arc length, the length between the melting electrode tip and the base metal. As the contact-to-work distance increases, the arc voltage and arc length would tend to increase with a conventional GMAW power source. However, with the CMT-GMAW power source, the welding current decreases with a slight increase in voltage, while the mechanical drives in the torch adjust appropriately to maintain a consistent arc length. Conversely, if the distance is shortened, the lower voltage would be accompanied by an increase in current and a mechanical adjustment in wire feed speed to compensate for the shorter wire stickout. [7 and 8]

11

Figure 2.1-1: Constant-potential power source illustration [8]

2.2 Wire Feed Speed

In conventional GMAW welding, as the electrode feed speed (wire feed speed) is varied, the welding current varies in a similar manner with the arc length. This occurs because the current output of the power source fluctuates dramatically with the slight changes in the arc voltage when alterations are made to the wire feed speed. If all other variables were held constant, an increase in welding current would result in the following: (1) an increase in the depth and width of the penetration, (2) an increase in the deposition rate, and (3) an increase in the size of the weld bead. [7 and 8] Unlike conventional GMAW units where current and voltage can be changed independently, these two key parameters are linked together in the CMT-GMAW via a digital process control, or an arc synergic line. An arc synergic line is a linear mathematical relationship (proprietary to Fronius International LLC) which incorporates the voltage and amperage process controls into the wire feed speed, and is dependent on the thermal and electrical resistivity properties of the material substrate/filler used. Therefore, each synergic line is uniquely different: that is, every synergic line consists of several points on a voltage current diagram (U-I diagram), which are formed from the connection of a series of certain current and pertinent voltage levels for any given wire composition and gas. Each point on the synergic line is recorded with the same arc

12

length, despite different performance, through the whole power range. An arc synergic line is experimentally determined via high-speed video techniques. [6, 7 and 8]

2.3 Traverse Speed

As with any GMAW process, including CMT-GMAW, the traverse or travel speed has a profound impact on the weld quality. Travel speed is the linear rate at which the arc is moved along the surface of the workpiece. The filler metal deposition rate per unit length increases when the travel speed is decreased. The welding arc impinges on the molten weld pool rather than the base metal at very slow speeds, thereby reducing the effective penetration and resulting in a wider bead. As travel speed is increased, the thermal energy per unit length of weld transmitted to the base metal from the arc is at first increased because the arc acts more directly on the base material. However, further increases in travel speed impart less thermal energy to the base metal. Melting of the base metal therefore first rises and then decreases with increasing travel speed. As the travel speed increases, there is a tendency for undercutting along the edges of the weld bead because of insufficient deposition of filler metal to fill the path melted by the arc. [7 and 8]

2.4 Electrode Orientation

Electrode orientation affects bead shape and penetration to a greater extent than arc voltage and travel speed. The electrode orientation is described in two ways: (1) by the relationship of the electrode axis with respect to the direction of travel and (2) the angle between the electrode axis and the adjacent workpiece surface. When the electrode points in a direction opposite to the travel direction, it results in a trail angle, which is known as the backhand method. Similarly, when the electrode points in the direction of travel, it results in a lead angle and is called the forehand method. The electrode orientation and its effect on the width and penetration of the weld bead are illustrated in figure 2.4-1. [7 and 8]

13

Figure 2.4-1: Effect of electrode position and welding technique [8]

When the electrode is changed from perpendicular to the lead angle technique with all other conditions unchanged, the penetration decreases and the weld bead exhibited is wider and flatter. Maximum penetration is obtained in the flat position with the drag technique, at a drag angle of about 25 degrees from perpendicular. The drag technique also produces a more convex, narrower bead, an increasingly stable arc, and less spatter on the workpiece. [7 and 8]

14

3. Method of Approach
The weld filler chosen for the weld trials performed in this research was 0.035inch diameter PMET 818 (AMS 5832) weld wire provided by Polymet Corporation (see table 3-1 for filler material specifications). The arc synergic line used during all trials was ER NiCrMo-3, and is based on 0.045-inch diameter weld filler AMS 5582, Inconel 718. The subsequent wire feed and traverse speeds were altered according to weld geometry and base material thickness, and will be discussed in further detail in the subsequent chapters.

Table 3-1: Material specification for PMET 818 (AMS 5832)

From previous weld experience using 0.025-inch thick wrought Inconel 718 plate, it was determined that the ideal torch angle for these trials would be between 5-10 degrees (from perpendicular) and using a backhand (pull) technique (see appendix A for trial details). A backhand technique is used to produce a weld with maximum penetration and a narrow, convex surface configuration. This angle also provides maximum shielding of the molten weld pool. The shielding gas for all weldments was 100% pure argon, with a flowrate of 30-35 CFH. A Plexiglas gas chamber was also used, as applicable, to ensure an inert atmosphere during welding. An illustration of the robotic weld set-up and listing of the equipment used during all automated CMT-GMAW weld trials is shown in figure 3-1 below.

15

Figure 3-1: Illustration of welding equipment used for robotic trials [4]

Several weldments with varying parameters were prepared in order to meet the parameter optimization, heat treatment, metallographic review and NDE (X-ray and FPI), objectives of this research. Each weldment represents the various weld profiles and geometries. These geometries include straight single-line beads, straight-line-weave pattern beads, circular weld build-ups, and material edge build-ups which are representative of the most common weld styles in aerospace applications. All specimens were heat treated accordingly, with a pre- and post-weld solution heat treatment and finally an age treatment. All weldments were then macro-evaluated using common NDE techniques - X-ray and FPI - to ensure weld quality (e.g. porosity, lack of fusion, and cracking).

16

Once NDE was complete, all specimens were micro-sectioned with a minimum of 21 cuts were made to examine the weld joint in the axial, transverse, and longitudinal planes of view. All weld microstructure and subsequent Heat Affected Zone (HAZ) were evaluated at up to 200x magnification and were compared to conventional GTAW. The micro-sections were evaluated for agreement of weld quality and microstructural equivalency, including phases, grain structure, grain size, and solidification patterns. Several micro hardness readings were taken in varying locations starting from the fusion zone and emanating toward the HAZ and finally into the parent material and weld material. These measurements were taken throughout each weldment and were compared to the AMS standards (Aerospace Material Specifications) to determine the mechanical properties and meet the objectives of the research. Additional weld plates were fabricated so as to obtain the tensile properties of the CMT-GMAW weldments. Tensile specimens were extracted through the welded region of the several weld plates. These tensile specimens were pulled perpendicular to the direction of the weld travel and tested at both room temperature (65F) and at the elevated engine service temperature (1100F). These results were compared to those obtained using the same testing scheme for the GTAW weldments.

3.1 Pre- and Post-Weld Thermal Treatments

Thermal (heat) treatment is any application of a temperature, for any amount of time, sufficiently high enough as to accomplish one of the following: (1) reduce stresses, (2) allow atom movements to redistribute existing alloy elements, (3) promote grain growth, (4) promote new recrystallization, (5) dissolve phases, (6) produce new phases owing to precipitation from solid-solutioning, or (7) support alloy surface chemistry. [3] Prior to welding, the following thermal treatment was performed in order to place the base material in a fully solutioned (weldable) condition and to obtain complete recrystallization and maximum softness. The pre-weld solution was performed in an vacuum atmosphere where (1) the base material specimens were heated to a temperature of 1750F 25F and held for a period of 60 minutes, then (2) cool to a temperature of 1100F at a rate of 35F/minute proceeded by (3) continued cooling to a temperature of 1000F at a rate of 15F /minute; all remaining subsequent cooling rates were 17

50F/minute. All cooling procedures were produced via back-filling with 100% argon gas. After welding, the following solution thermal treatment was performed to resolve any residual stresses and dissolve any secondary phases and prepare the alloy for subsequent aging. The aging is performed to obtain maximum ductility, this process also homogenizes the microstructure prior to aging. Like the initial solutioning treatment, the weldment specimens were once again (1) heated to a temperature of 1750F 25F and held for a period of 60 minutes, then (2) cooled to a temperature of 1100F at a rate of 35F/minute and proceeded by (3) continued cooling to a temperature of 1000F at a rate of 15F /minute; all remaining subsequent cooling rates were 50F/minute and cooling procedures were produced via back-filling with 100% argon gas. Immediately following the solution treatment, the weldment specimens were placed through the following precipitation (aging) thermal treatment to bring out the desirable strengthening precipitates and control other secondary phases, including carbides and detrimental topographically close packed phases. Some of the topographically close pack phases include , , and Laves which are variably detrimental when more than trace amounts are present. The weldment specimens were (1) heated to a temperature of 1325F 25F and held for a period of eight hours, then (2) cooled at a rate of 100F 25F/hour to a temperature of 1150F 25F and again held for a period of 18 hours; all remaining subsequent cooling rates were 50F/minute and all cooling procedures were produced via back-filling with 100% argon gas.

3.2 Weldment Specimen Fabrication

Nickel alloys are commonly susceptible to embrittlement by lead, sulfur, phosphorus, and other low-melting-point elements. These materials have been known to exist in grease, oil, paint, marking crayons, or ink and from lubricants, cutting fluids, shop dirt, and processing chemicals. Another important variable is the possibility of surface oxides which may have formed from previous processing, e.g. thermal treatments. Therefore, it is good practice to ensure that all specimens are completely free of foreign materials before they are welded. [7]

18

To ensure quality weldments, comprehensive cleaning procedures were performed prior to welding. The localized areas to be welded were wire brushed using handheld pneumatic stainless steel bristle wheels to remove any possible oxides and then followed by wiping the area with a clean acetone swab. Five weldments with varying wirefeed parameters (see table 3.2-1), were made utilizing standard 4-inch x 4-inch x 1/8-inch thick AMS 5596 shear cut sheet stock. Each parameter was chosen from previous trials on similar geometries. Each weldment exhibited several different weld profiles/geometries, including straight single-line beads, straight-line-weave pattern beads, circular weld build-ups (figure 3.2-1), and edge buildups (figure 3.2-2). These diverse geometries are representative of the most common weld styles in aerospace applications. A clear Plexiglas chamber was used during processing to note real-time observations for parameter optimization of the weld deposition during processing. The visual observations made during processing were in regards to overall arc stability, weld wetting, and solidification.

Table 3.2-1: Various parameters examined for representative CMT-GMAW geometry weldments

19

Figure 3.2-1: Illustration of straight single-line beads, straight-line-weave pattern beads, circular weld build-ups for CMT-GMAW representative geometry weld trials

Figure 3.2-2: Illustration of edge weld build-ups for CMT-GMAW representative geometry weld trials

Several manual GTAW (using a Lincoln Electric Square Wave TIG -355 welder) and CMT-GMAW weldments, with constant parameters (see table 3.2-2), were made utilizing shear cut AMS 5596 sheet stock with two 3/8 flat tip end mill machine trenches spanning the entire length of the specimen (see figure 3.2-3). After machining the base stock, a weld was placed along the trench length of the specimen, as shown in figure 3.24.

20

Table 3.2-2: Parameters used for tensile weldments

Figure 3.2-3: Stock geometry used for tensile weldment fabrication

Figure 3.2-4: Illustration of tensile weldment post-welding

3.3 Post-Weld Non-Destructive Evaluation

There are essentially four categories of flaws which can occur in GMAW and GTAW weldments: (1) gas porosity, figure 3.3-1; (2) Lack of Fusion (LOF); (3) Lack of Penetration (LOP), figure 3.3-2; and (4) cracks, figure 3.3-3. The form, location, and

21

orientation of each type are governed by several factors, including joint configuration, welding process, base metal, filler metal, and other variables associated with the joint. [9]

Figure 3.3-1: Types of gas porosity commonly found in weldments [9]

Figure 3.3-2: Lack of fusion in various weld joints [9]

Figure 3.3-3: Identification of cracks according to location in weld and base metal [9] 22

Although visual (white light) inspection is an invaluable and easy-to-perform assessment method, it should be mentioned that is also unreliable for detecting subsurface flaws and finite surface flaws. Consequently, it should not be the only inspection method performed when ensuring weld integrity. [9] All visual inspection were performed prior to Florescent Penetrant Inspection (FPI) and X-ray using a standard 14X eye loop. The primary purpose of this examination step was to verify the following: (1) conformity of welds with regards to surface roughness, weld spatter, and cleanliness; (2) presence of obvious surface flaws such as lack of fusion, undercuts, overlaps, and cracks. Florescent Penetrant Inspection is a non-destructive method of revealing discontinuities that are open to the surfaces of solid and essentially nonporous materials. Indications of a wide spectrum of flaw sizes can be found regardless of the configuration of the workpiece and flaw orientations. The florescent liquid penetrates deep into various types of minute surface openings by capillary action. [9] The florescent penetrant used for this research was ultra-high-sensitivity; simply meaning, it can penetrate into the smallest of indications. The primary purpose of this inspection step was to check for the presence of non-obvious surface flaws such as lack of fusion, undercuts, overlaps, and cracks which may not have been visible using white light inspection. The major limitation of visual inspection and FPI is that both can detect only imperfections that are open to the surface. As a result, an X-ray inspection was used for detecting subsurface flaws. X-rays can detect flaws that are completely internal and located well below the surface of the part. Although X-ray will reveal the interior of opaque objects, it cannot detect all types of discontinuities, such as small defects in thick objects and very fine cracks. Weldments were also metallographically sectioned and examined to ensure a complete analysis. [9] The primary purpose of this inspection step was to check for the presence of subsurface flaws such as (1) gas porosity; (2) Lack of Fusion (LOF); (3) Lack of Penetration (LOP); and (4) cracks that could not be detected via the previous inspection techniques. Secondly, any areas with indications would be marked as a location for micro sectioning.

23

3.4 Material Property Characterization

During this research it was imperative to classify the material properties of the weldments since welds (essentially a cast structure) generally behave differently from wrought structures (the base material). Wrought structures are generally more homogeneous and finer-grain than cast structures. The greater homogeneity enables more of the hardener elements to be taken into solution and to thus effectively converted to and phases. Moreover, the distribution of the and/or in wrought alloys is generally more uniform and finer in size. Additionally, wrought structures tend to have finer grain sizes than cast structures, and subsequently have enhanced tensile strengths. It is important to understand the metallurgy behind the solidified weld structure so as to provide the associated strength debit involved when placing a welded (cast structure) on a wrought base material. [3] Microstructural phase morphology can vary widely (e.g. script versus blocky carbides, cuboidal versus spheroidal , cellular versus uniform precipitation, and discrete versus envelopes). [3] The solidification patterns of welded Inconel 718 is governed by the laws of the phase diagram (as is the solidification of all metals). The kinetics of the solidification process determines the microstructure that is actually formed. However, what makes the solidification of superalloys like Inconel 718 exceptionally different from less sophisticated alloys is that the solute content of these alloys is very high. [3]

3.4.1 Micro-Analysis The purpose of the micro-analysis metallurgical review of this research was to classify the general microstructure of the solidified weld pool formed during the CMTGMAW process. This analysis was accomplished by examining the weld geometry in various planes of view (i.e. longitudinal, planar, and transverse) at magnification levels up to 200x using standard microscopy techniques. Additional micro-analysis was performed using electron microscopy to analyze the fractography of tensile specimens. The microstructure was compared to conventional GTAW for weld quality and microstructural equivalency, including phases, grain structure, solidification patterns, and

24

process induced defects including, porosity, Lack of Fusion (LOF), and Lack of Penetration (LOP) and cracks. After welding, all of the representative geometry weldments were sectioned according to figure 3.4.1-1 using handheld pneumatic silicon carbide cut off wheels. A total of 21 micro-sections were made, allowing for such visual orientations as, longitudinal, planar and transverse planes. Refer to figure 3.4.1-2 for view plane orientation definitions.

Figure 3.4.1-1: Illustration of CMT-GMAW representative geometry weldments with indicated cut planes

Figure 3.4.1-2: Illustration of micro section plane definitions [10]

After welding, one representative tensile weldment from each weld process (yielding two in total,) - manual GTAW and automated CMT-GMAW - was sectioned according to figure 3.4.1-3 using handheld pneumatic silicon-carbide cut-off wheels. 25

These sections were taken to confirm hardness and microstructural equivalence of the CMT-GMAW with GTAW prior to tensile testing.

Figure 3.4.1-3: Illustration of tensile weldments with indicated tensile extraction and cut planes

3.4.2 Specimen Mount Preparation Specimen preparation is crucial for the overall microscopic analysis of a materials microstructure. In order to be able to adequately analyze a material, the microspecimens surface must be properly prepared to maintain good edge retention and be free of scratches, smears, and contaminants. Therefore only one skilled in the field should prepare micro-sections. All cut sections were mounted in a two-piece castable clear epoxy resin. Castable epoxies consist of two or more liquid resins that are mixed in certain proportions. In a two-piece epoxy, one liquid is the resin and the other liquid is the hardener (also called activator or catalyst). After curing, the following procedures (table 3.4.2-1) were used to prepare the cut surface for metallographic examination. Each mount was ultrasonically cleaned prior to each subsequent step to ensure no grit cross-contamination. After all polishing procedures were performed; the micro-section mounts were submersed in a 10% oxalic acid and water bath and electrolytically etched using a 12v power source.

26

Table 3.4.2-1: Grinding and polishing procedure used for micro-sections

3.4.3 Mechanical (Hardness) Analysis The term hardness as commonly used, may be defined as the ability of a material to resist permanent indentation or deformation when in contact with an indenter under load. Hardness testing is perhaps the simplest and the least expensive method of mechanically characterizing a material since it does not require an elaborate specimen preparation, involves rather inexpensive testing equipment, and is relatively quick. The theoretical and empirical investigations have resulted in fairly accurate quantitative relationships between hardness and other mechanical properties of materials, such as ultimate tensile strength, yield strength, and strain hardening coefficient and fatigue strength and creep. [11] All micro-hardness readings were taken via the Vickers micro hardness method. The Vickers test method uses a diamond shaped indenter pressed into a material using a force of 500 gf. The resulting indentation diagonals are measured, and the hardness number is calculated, by dividing the force by the surface area of the indentation. [11] In order to fully characterize the weld processes, hardness readings were taken through three unique areas of interest including, the base material (this was used as a base-line), the HAZ (which is defined as the area between the fusion line and the unaffected parent material), and the weld material (see figure 3.4.3-1).

27

Figure 3.4.3-1: Illustration depicting the various locations for hardness testing

3.4.4 Mechanical (Tensile) Analysis Short-time properties are always under consideration for property-microstructure relationships. Short-time properties include tensile (or compressive) strengths, (yield and ultimate) produced by continuous fairly rapid application of a steady load to reach plastic deformation (proportional limit, 0.02% yield, 0.2% yield) or fracture. Elongations during and at the conclusion of a test are also measured, reductions in area at the conclusion of a test are also recorded, and finally elastic modulus is generally measured as well. It should be noted that tensile properties are usually at issue up to the region of about 1400 F. [3] Once welding was complete, standard (flat) tensile specimens (figure 3.4.4-1) were extracted from the weldments according to figure 3.4.4-2. All specimens were machined according to the procedure set-forth in table 3.4.4-1. It should be noted on figure 3.4.4-2 that the entire gage width and length is made of weld deposit. This was done in an effort to ensure fracture would be induced solely along the welded material. Extracting specimens in this manner also allotted a direct strength comparison of the GTAW deposit with CMT-GMAW.

28

Table 3.4.4-1: Machining procedure for tensile specimen extraction

Figure 3.4.4-1: Illustration of tensile specimen used for testing

Figure 3.4.4-2: Illustration of tensile specimen extraction

29

Tensile properties of the CMT-GMAW weldments were obtained using flat tensile specimens that were tested at both room temperature (65F) and at Inconel 718s serviceable elevated temperature (1100F). The tensile specimens were pulled perpendicular to the plane of the weld solidification pattern (figure 3.4.4-3). The

properties that were obtained via tensile testing include tensile strength, yield strength, elongation, reduction in area, strain and break, and modulus. After testing was complete, the fractured surfaces of both the GTAW and CMT-GMAW specimens were evaluated using Scanning Electron Microscopy (SEM) to obtain a general topographic image analysis of the fracture surfaces.

Figure 3.4.4-3: Illustration of tensile specimen loading scheme

30

4. Results and Discussion

4.1 Non-Destructive Analysis of CMT-GMAW Weldments
White light visual and subsequent ultraviolet light high-sensitivity Florescent Penetrant Inspection of all representative geometry weldments and tensile weldments using a standard 14X eye loop yielded no evidence of weld-induced surface defects. Of the seven representative geometry weldments evaluated, specimen numbers three (figure 4.1-1) and five (figure 4.1-2) exhibited the most consistent results in to overall weldment appearance. This was similar to the observations made during the welding process, table 4.1-1. The overall visual appearance of weldment with regard to surface roughness and bead shape appeared to be very uniform. This observation is consistent with automated weld processes. See appendix B for macro images of all representative geometry

weldment specimens.

Table 4.1-1: Visual observations made during processing of representative geometry CMT-GMAW weldments

31

Figure 4.1-1: White light macro image (post thermal treatment) of representative CMT-GMAW weldment specimen #3

Figure 4.1-2: White light macro image (post thermal treatment) of representative CMT-GMAW weldment specimen #5

Upon further evaluation of all weldments, it was noted that there was no evidence of process-induced weld spatter. This confirms the ability of the CMT-GMAW droplet

32

detachment method to produce wide bead weldments with virtually no weld spatter. Lastly, visual observation and FPI confirmed no presence of obvious surface flaws such as lack of fusion, undercuts, overlaps, and cracks. X-ray inspection of the representative geometry weldment specimens and tensile weldments using million-volt x-ray produced no evidence of subsurface weld-induced defects such as such as porosity, Lack of Fusion (LOF), Lack of Penetration (LOP), and cracks. This confirms the ability of CMT-GMAW process to produce a variety of weldment geometries with no weld-induced porosity. Figures 4.1-3 and 4.1-4 are images of the x-ray films taken of representative weldment specimens three and five verifying no subsurface indications.

Figure 4.1-3: X-ray image of representative geometry CMT-GMAW weldment specimen #3

Figure 4.1-4: X-ray image of representative geometry CMT-GMAW weldment specimen #5 33

4.2 Destructive Analysis

4.2.1 Representative Weldment Microstructure Destructive analysis is performed to view a material at a microscopic level. Destructive analysis is often only employed for critical tasks, as it is impractical to destructively analyze all components; hence the use of the previous mentioned NDE methods. To fully characterize the microstructure obtained using the CMT-GMAW weld process, of 21 micro-sections were taken from the representative geometry of specimens three and five. Sections were extracted through the following unique areas of interest, arc start (up slope), arc stop (down slope), and multi-pass build-up interfaces in three different orientations (longitudinal, planar, and transverse), reference figure 3.4.1-2. Upon initial analysis, it was found that the microstructure of each weldment was uniform within each group (weldment geometry) and cut orientation. Therefore, for the purposes of this discussion, only one representative micro-section from each orientation (independent of bead type) is shown in the following images below, figure 4.2.1(circular build-up), figure 4.2.1-2 (straight-line-weave pattern bead), figure 4.2.1-3 (single straight-line pattern), and figure 4.2.1-4 (edge weld build-up). Note: additional substantiating microphotographs can be found in appendix C.

Figure 4.2.1-1: Transverse micro-section location 3 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid 34

Figure 4.2.1-2: Planar micro-section location 8 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure 4.2.1-3: Longitudinal micro-section location 12 from single straight-line pattern, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

35

Figure 4.2.1-4:

Transverse micro-section location 4 from edge weld build-up,

specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid

In common with other weld processes, the major intermetallic phase found inside the CMT-GMAW weld and at the boundaries in the HAZ is the hexagonal Laves phase. Laves morphology in HAZ ranges from isolated particles to continuous sheets and massive dendritic walls as shown in the previous high magnification images above. Inside the weld liquidation zone, the recrystallized, dendritic network within weld grains is formed from discontinuous, cross-linked sheets of lamellar MC carbides, mixed with Laves precipitates. The precipitates in the weld or along the HAZ boundaries are Nbrich. Clusters of small carbides generally contain levels of Ti, Cr, and Ni (but not Fe) enhanced by a few percent, possibly reflecting local Nb depletion. Apart from occasional grown-in dislocations, the carbide crystals are generally well ordered. [12] In general, the structure of the CMT-GMAW weldments is parallel with the equiax casting. The fundamentals of the weld solidification are the same as those of a casting, but with different boundary conditions. As the weld filler solidifies, the primary dendrites reject solute into the interdendritic liquid. The low niobium primary dendrites grow into the solidifying metal in a direction perpendicular to the solidification front. This is best illustrated in figure 4.2.1-5 where the niobium dendrites are growing out of the page, which is perpendicular to the solidification front. [3] 36

Figure 4.2.1-5: Planar micro-section (post thermal treatment) showing primary dendrite growth, etchant: electrolytic 10% oxalic acid

Weldment heat extraction is not uniform, unlike conventional casting practices where the heat extraction is generally uniform and precisely controlled to enhance properties. Since the heat is extracted from the welded component in a uncontrolled manner, the subsequent primary dendrite growth is random and columnar in nature, as shown in figure 4.2.1-6 Although random in temperament, the dendrite growth is still perpendicular to the direction of heat extraction. The dendrites form a tightly packed columnar-grain (CG) structure.

37

Figure 4.2.1-6: Longitudinal micro-section (post thermal treatment) demonstrating the randomness of dendrite growth, etchant: electrolytic 10% oxalic acid

During solidification, solid dendrites begin to grow in what is referred to as a mushy zone. This region of solid dendrites and liquid interdendritic regions has a temperature gradient defined by the width of the mushy zone and the liquidus and solidus temperatures. The size of the dendrites in the mushy zone is directly related to the local solidification time (LST). LST is defined by equation 4.2.1-1, where TL is the liquidus temperature (F), Ts is the solidus temperature (F), G is the temperature gradient (F/in), and R is the solidification rate (in/min). Thus, with increasing heat extraction or decreasing heat input, (rate of molten metal introduction) G and R are increased. This intern decreases the LST, and, as is intuitively logical, increasing heat extraction or decreasing heat input decreases the dendrite size as will be seen in section 4.2.3. [3]

(Equation 4.2.1-1)

4.2.2 Macrostructural Comparison to GTAW At a macroscopic level, one can clearly observe the differences between the CMT-GMAW and GTAW processes. This can be seen in the bead shape and

distribution. The GTAW process produces a bead shape typical of a multi-pass manual

38

operation and appears to be less uniform than the single-pass automated CMT-GMAW (figure 4.2.2-1). From this figure one can also observe the massive amount of distortion in GTAW weldment base metal (see figure 4.2.2-1). Distortion is caused by residual (thermal) stresses induced during solidification and is generally prominent in a high heat input arc process. This cross-section clearly indicates that the GTAW process is much more heat invasive than the CMT-GMAW process.

Figure 4.2.2-1: Macro image analysis of tensile weldments (post thermal treatment) showing weld profile and distortion of manual GTAW and automated CMT-GMAW

4.2.3 Microstructural Comparison to GTAW At the microscopic level, the microstructure of the GTAW and CMT-GMAW appear to be equivalent in appearance. Both the GTAW and CMT-GMAW weld microstructures have a column grain dendrite structure that emanates in a direction perpendicular to the heat extraction, as previously described in section 4.2.1, figure 4.2.31. It can also be noted from this figure that the CMT-GMAW weld microstructure appears to be much more uniform with respect to randomness of the dendrite formation than the GTAW microstructural. This can be attributed to the automated processing and one pass as opposed to multiple passes, such as those performed with the manual GTAW (3 passes). A multi-pass weld has the tendency to create several locations of merging

39

dendrites and erratic growth patterns in which new grains nucleate and grow at the expense of original grains (the initial pass).

Figure 4.2.3-1: Longitudinal micro-sections of both CMT-GMAW and GTAW tensile weldments (post thermal treatment), etchant: electrolytic 10% oxalic acid

When an evaluation was performed on the transverse micro-sections, one can clearly distinguish the substantial amount of weld deposit delivered by 1-pass of the CMT-GMAW when compared to 3-passes of GTAW. In fact, with 1-pass of the CMTGMAW, there appears to be almost 4x as much weld deposit when compared to the 3passes of the GTAW (figure 4.2.3-2). This is enevitably the case because, as previous mentioned, the electrode is the filler in CMT-GMAW. Therefore, every time an arc is struck the filler is melted creating a larger amount of weld deposit. This is quite extraordinary given that the heat input transmitted from the CMT-GMAW process to the substrate appears to be substantially less than the GTAW process (see section 4.2.2).

40

Figure 4.2.3-2: Transverse micro-sections of both CMT-GMAW and GTAW tensile weldments (post thermal treatment) exhibiting deposit height, etchant: electrolytic 10% oxalic acid

When further evaluating the transverse micro-sections, it was also determined that the CMT-GMAW process consumes much less of the base material than the GTAW weld deposit. This is illustrated in figure 4.2.3-3, which is a transverse image of the weld joint taken at the base of the tensile weldment trench (refer to figures 3.2.2-1 and 3.2.2-2 for image orientation). The measurements shown were taken from the base of the stock material (which was used as the base line) and followed to the bottom of the lowest point of the HAZ from each weldment. From this image, it is clear that the GTAW weldment consumed much more of the base material during processing (approximately two times). This can be attributed to the fundamentals of GTAW processing which requires the heating of the substrate via the tungsten electrode prior to the addition of the weld filler, whereas the CMT-GMAW uses a consumable electrode. The consumable electrode in theory requires less heat input to produce the weld. This finding additionally confirms the observations made in section 4.2.2.

41

Figure 4.2.3-3: Transverse micro-sections of both CMT-GMAW and GTAW tensile weldments (post thermal treatment) exhibiting deposit depth, etchant: electrolytic 10% oxalic acid

4.3 Hardness Properties

To further characterize the CMT-GMAW weld process, 81 hardness readings were taken through three unique areas of interest. The areas included the base material (this was used as a base-line), the HAZ (which is defined as the area above and below the line of fusion), and the weld material. This analysis was performed in three different orientations (longitudinal, planar, and transverse) to confirm consistency and correspondence to the GTAW process. Overall, the results indicated that the CMTGMAW weldments material hardness was well within the current AMS specification and was also equivalent to the hardness obtained with the GTAW process as well (table 4.31). Refer to figures 3.4.1-2 and 3.4.3-1 for view plane orientation and weld location definitions.

42

Table 4.3-1: Material hardness results for GTAW and CMT-GMAW weldments

4.4 Tensile Properties

The mechanical properties of materials are usually temperature dependent and are subject to change with temperature. As a result, it is imperative to consider materials properties at different temperatures. Some of the typical material property concerns are: ultimate strength, yield strength, modulus, and ductility. The most common technique in characterizing the temperature dependence of mechanical properties is to conduct tensile tests at varying temperatures. As previously discussed in section 3.4.4, tensile properties of the CMT-GMAW and GTAW weldments were obtained using a total 12 flat tensile specimens (figure 4.4-1) and were tested at both room temperature (65F) and at elevated temperature (1100F). [11]

43

Figure 4.4-1: Macro image of flat specimen used for testing, specimen CMT-1
The tensile strength or ultimate tensile strength (su) is the maximum load divided

by the original cross-sectional area of the specimen (see equation 4.4-1). The tensile strength is the value most frequently cited in the results of a tension test. The tensile strength, however, is a value of little fundamental significance with regard to the strength of a metal. For ductile metals, the tensile strength should be considered as a measure of the maximum load that a metal can withstand under the very restrictive conditions of uniaxial loading. This value bears little relation to the useful strength of the metal under the more complex conditions of stress that are usually encountered during service. [11]

Equation 4.4-1 Yield strength is the stress at which plastic deformation or yielding is first observed. With most materials, there is a gradual transition from elastic to plastic behavior and identifying the point at which plastic deformation begins is difficult to define with precision. [11] Ductility specifically refers to a material's ability to deform under tensile stress and is generally broken into two major categories: reduction in area and elongation, where Reduction in Area (RA) is defined by equation 4.4-2 and elongation (e) is defined by equation 4.4-3. [11]

Equation 4.4-2

Equation 4.4-3

44

The slope of the initial linear portion of the stress-strain curve is the elastic modulus (E), or Young's modulus. Elastic modulus is a measure of the stiffness of the material; the greater the modulus, the smaller the elastic strain resulting from the application of a given stress. The binding forces between atoms determine the elastic modulus. Because these forces cannot be changed without altering the basic nature of the material, the elastic modulus is one of the most structure-insensitive of the mechanical properties. [11]

4.4.1 Room Temperature Properties At room temperature, it was established that the overall tensile strength of the CMT-GMAW was very comparable to its GTAW counterpart (see table 4.4.1-1). The average ultimate tensile strength of the CMT-GMAW was found to be within 8 ksi of the GTAW, whereas the yield strength was within 13 ksi. One statistic which can be clearly observed from the data is that the CMTGMAW exhibited an average of 2x increase in elongation %, Reduction in Area % (RA), and strain at break when compared to the GTAW. The specifics of this phenomenon will be later discussed in sections 4.4.3 and 4.4.4. Finally, it is also worth noting that the modulus of both CMT-GMAW and GTAW appear to be equivalent as well.

Table 4.4.1-1: Room temperature (65F) tensile properties

After testing was complete, the fractured surfaces of both processes were macroscopically examined and further microscopically examined using electron microscopy (see figures 4.4.1-1 and 4.4.1-2). At a macroscopic level, as expected with 45

room temperature fracture, both the GTAW and CMT-GMAW specimens appeared to have fractured in a less ductile manner than what would be expected at elevated temperatures which is evident through the rough appearance of the fractured surface. Other substantiating macroscopic images can be seen in appendix D. In figure 4.4.1-2, it is important to point out that the surface features of both the GTAW and CMT-GMAW concur with those findings at a macroscopic level. At lower magnification, it can be noted that surface of the CMT-GMAW appears to show more cleavage than that of the GTAW specimen as evidenced by the smoother surface of the CMT-GMAW fracture. At higher magnification of the GTAW specimen, one can observe the striation lines which intrinsically form as the specimen begins to stretch and deform under continual tension.

Figure 4.4.1-1: Representative macro photographs, room temperature specimens GTAW-1 and CMT-2

46

Figure 4.4.1-2: SEM photographs of fractured surfaces, room temperature specimens GTAW-1 and CMT-2

4.4.2 Elevated Temperature Properties At elevated temperatures, (1100F) it was similarly established that the overall tensile strength of the CMT-GMAW was very comparable to GTAW (see table 4.4.2-1). The average ultimate tensile strength and yield strength of the CMT-GMAW was found to be within 6 ksi of the GTAW processed specimens. Additionally, it was also observed that the CMT-GMAW once again exhibited an average increase in elongation %, Reduction in Area % (RA), and strain at break when compared to the GTAW, though the increase was not as extreme as those exhibited at room temperature. This will again be later discussed in detail in sections 4.4.3 and 4.4.4. The modulus of both CMT-GMAW and GTAW appear to be equivalent as well.

47

Table 4.4.2-1: Elevated temperature (1100F) tensile properties

Similarly to the room temperature analysis, after testing the fractured surfaces of the elevated temperature, specimens were macroscopically and microscopically examined using electron microscopy (see figures 4.4.2-1 and 4.4.2-2). At a macroscopic level, both the GTAW and CMT-GMAW specimens appeared to have fractured in a ductile manner as seen from the extensive cleaving (sliding) appearance of the fractured surface. This surface appearance is commonly observed at elevated temperatures since the grain boundaries and successive dislocations tend to slip with continued tension. Other substantiating macroscopic images can be seen in appendix D. In figure 4.4.2-2, the surface features of both the GTAW and CMT-GMAW at a microscopic level correspond with those findings at a macroscopic level, though the CMT-GMAW appears to have more cleaving which also coincides with the equivalent ductility values obtained. This is particularly evident in the high magnification image of the CMT-GMAW sample, since the surface features appear to be smooth and continuous. The higher magnification image of the GTAW sample focuses on the rapid tensile overload (or final fracture) region of the fracture surface given that the surface features appear to be much coarser.

48

Figure 4.4.2-1: Representative macro photographs, elevated temperature specimens GTAW-6 and CMT-4

Figure 4.4.2-2: SEM photographs of fractured surfaces, elevated temperature specimens GTAW-6 and CMT-4

49

4.4.3 General Comments and Property Comparison In general, strength is reduced at high temperatures and materials become softer and more ductile as temperature increases. However, the rate and direction of property changes can vary widely. For example, the yield strength and elongation of various alloys are functions of the temperature. The majority of these changes are due to various metallurgical factors. Other features that often cannot be easily predicted, can also affect mechanical behavior at high temperatures. For example, re-solutioning of phases, precipitation, and aging (diffusion-controlled particle growth) can occur, both during heating prior to testing and during the testing itself. These processes have also been known to produce a wide variety of responses in mechanical behavior depending on the material, as previously discussed in section 4.2. As temperature increases, the strength of a material usually decreases and the ductility increases. The general reduction in strength and increase in ductility of metals at high temperatures is related to the effect of temperature on deformation of the material. At room temperature, plastic deformation occurs when there are dislocations in the material slip. The dislocations also intersect and build up in the material as they slip. This build-up of dislocations restricts the slip, and, thus, increases the forces necessary for continued deformation. This practice of continued deformation is known as strain hardening or work hardening. [11] At elevated temperatures, dislocation climb comes into play as another deformation mechanism. The build-up of strain energy from strain hardening can be relieved at high temperatures when crystal imperfections are rearranged or eliminated into new configurations, also know as recovery. A much more rapid restoration process is recrystallization in which new, dislocation-free crystals nucleate and grow at the expense of original grains. The restoration processes can be greatly enhanced by the increase in the thermal activity and mobility of atoms at higher temperatures. As a result, lower stress is required for deformation, as shown in the stress-strain diagrams in appendix E. [11] The crucial difference between the properties of the GTAW and CMT-GMAW weldments is the ductility. In general, it has been shown that with larger grain size, there is a corresponding decrease in ductility. Therefore, as previously mentioned in section

50

4.2.1, the temperature exhibited during the solidification process governs grain size and the subsequent dendrites that form (the size of the dendrites is defined by equation 4.2.11). From what has been observed through visual observations, the GTAW process inherently induces a greater amount of heat input during welding; therefore, upon solidification, the solidification rate (R) will be small since it will take longer for the filler to solidify. With a smaller R value, the local solidification temperature will be larger (equation 4.2.1-1). When the solidification conditions become sufficiently slow, the dendrites and the separation between them become large, in addition to the grain size. This will decrease the ductility of the weldment (as with the GTAW specimens), whereas the CMT-GMAW has been shown to have a lower thermal input. The lower thermal input will create a smaller local solidification temperature and will result in smaller, more finely, precipitated grains that will increase the ductility. The above is represented in the tensile data obtained during this research.

51

5. Conclusions
During parameter optimization, it was observed that the key attribute of the CMTGMAW process is its electronically controlled short circuit droplet detachment method. This precision process control is optimized by the push-pull servomotor located within the torch head and is controlled by the synergic line. The synergic line is a linear mathematical relationship that incorporates the voltage and amperage process controls into the wire feed speed. The synergic line helps with the detachment of the molten droplet during the short-circuiting, drastically reduces weld spatter, and provides a low thermal gradient during arcing. During parameter optimization, it was also found that the traverse and wire feed speeds had a profound impact on the weld quality and overall bead profile shape. As with any arcing welding process, when all other controls were held constant, varying the traverse speed could change the bead shape and wire feed speed. In contrast, the other process controls parameters had little affect on the overall weld quality, since the synergic line dictates the majority of the process controls which simplified the overall weld process and parameter optimization. However, it was noted that fine-tuning parameters could greatly affect the initial wetting of the filler material. After the material was subjected to welding, it was placed through conventional post weld heat treatment cycles. The heat treatments included a solution heat treatment at 1865F for one hour, and a precipitation treatment at 1400F for 16 hours. All non-destructive analyses of the CMT-GMAW weldments showed little evidence of spatter, porosity, surface and sub-surface cracking, Lack of Fusion, and Lack of Penetration. The absences of the above can be attributed to the precision droplet detachment method of the CMT-GMAW process. The CMT-GMAW process reduced porosity and spatter because it creates less turbulence within the weld pool, as the wire feed is not a constant forward moving motion. Surface and sub-surface cracking was reduced by the lower thermal input properties of the CMT-GMAW process. Visual white macro analysis of the CMT-GMAW weldments exhibited a reduced amount of substrate distortion when compared its GTAW counterpart. At the microscopic level, the microstructure of the GTAW and CMT-GMAW appears to be equivalent. Both the GTAW and CMT-GMAW weld microstructures have 52

a column grain dendrite structure that emanates in a direction perpendicular to the heat extraction. The microstructure of the CMT-GMAW was equivalent to that of GTAW with the exception of a reduced substrate consumption level and subsequent Heat Affected Zone (HAZ) with an increased amount weld filler deposition. With the reduced thermal input of the CMT-GMAW, it was also noted that the consequent grain size was reduced, and the dendrites were finer than those seen in the GTAW weldments. Mechanical hardness properties of the CMT-GMAW were evaluated in 81 different locations including the substrate, HAZ, and weld filler locations; all of which confirmed equivalency to the current AMS standards and those obtained from the GTAW weldments. Tensile properties of the CMT-GMAW weldments obtained were tested at both room temperature (65F) and at Inconel 718 serviceable elevated temperature (1100F). The results obtained at room and elevated temperature indicated equivalency between the CMT-GMAW and GTAW with the exception that the CMT-GMAW specimens exhibited approximately 2x increase in the percentage of Reduction in Area (RA) at both room and elevated temperatures. This increase can be attributed to the finer grain and dendrite sizes obtained when using the CMT-GMAW. When evaluating the fractured surface, it was noted that the CMT-GMAW exhibit more cleaving during fracture, which, of course, is a byproduct of greater ductility. On the basis of overall evaluation, it was concluded that CMT-GMAW provides the following benefits: excellent weld quality on wrought Inconel 718; comparable metallographic structure; increased deposition rates when compared to GTAW; a reduction in overall thermal input by achieving almost current-free metal transfer; virtually spatter-free metal transfer (as compared to conventional GMAW) by controlling the short circuiting; comparable material hardness; comparable tensile strength with increased RA when compared to GTAW.

53

6. References
[1] The CMT Process A Revolution in Materials-Joining Technology. Brighton: Fronius USA LLC, 2004. [2] Prof. Charles Albright. "Arc Welding Processes: Short Course, Pratt and Whitney" Edison Joining Technology Center, Ohio State University. 23 Jan. 2001. [3] American Society for Metals. Superalloys: A Technical Guide, Second Edition, Matthew J. Donachie and Stephen J. Donachie. Ohio: ASM International, 2003 [4] CMT: Cold Metal Transfer, MIG/MAG dip-transfer arc process. Brighton: Fronius USA LLC, 2007. [5] CMT: Cold Metal Transfer,MIG/MAG dip-transfer process for automated applications. Brighton: Fronius USA LLC, 2004. [6] Bruckner, Jergen. "Cold Metal Transfer Has a Future Joining Steel to Aluminum." American Welding Society. 2005 <http://www.aws.org />. [7] American Society for Metals. Metals Handbook: Welding, Brazing, and Soldering, 10th Edition, Volume 6. Ed. David L. Olson. Ohio: ASM International, 2007. [8] American Welding Society. Welding Handbook, 8th Edition, Volume 2. Ed. R. L. O Brien. Miami: U.S.A American Welding Society, 1991. [9] American Society for Metals. Metals Handbook: Non-Destructive Inspection and Quality Control, 9th Edition, Volume 11. Ed. Howard E. Boyer. Ohio: ASM International, 1992. [10] American Society for Metals. Metals Handbook: Metallography and Microstructures, Volume 9. Ed. George F. Vander Voort. Ohio: ASM International, 2004. [11] American Society for Metals. Metals Handbook: Mechanical Testing and Evaluation, Volume 8. Howard Kuhn and Dana Medlin. Ohio: ASM International, 2000.

54

[12] Thompson, R.G. "Microfisuring of Alloy 718 in the Weld Heat-Affected Zone." Journal of Metals. July 1988: 44-48.

55

Appendix A
All weld trials were performed on the edge of 0.025-inch thick AMS 5596 base material using the ER NiCrMo-3 arc synergic line. The wire feed speed, current, and voltage held constant while varying the torch angle and technique. The shielding gas for all weldments was 100% pure argon, with a flowrate of 30-35 CFH; a gas chamber was also used to ensure an inert atmosphere during welding. All appropriate heat treatments were performed pre- and post-weld.

Table A-1: Parameters used for weld trials

A-1

Figure A-1: Specimen 25-1 (20, Pull Angle)

Figure A-2: Specimen 25-3 (0, Neutral Angle)

Figure A-3: Specimen 25-6 (20, Push Angle)

A-2

Appendix B
The following appendix pertains to the visual inspection of all representative geometry weldment specimens. Reference table 4.1-1 for parameters used and observations made during weld trials.

B-1

Figure B-1: White light macro image (post thermal treatment) of representative weldment specimen #1 (as welded)

Figure B-2: White light macro image (post thermal treatment) of representative weldment specimen #2 (as welded) B-2

Figure B-3: White light macro image (post thermal treatment) of representative weldment specimen #3 (as welded)

Figure B-4: White light macro image (post thermal treatment) of representative weldment specimen #4 (as welded)

B-3

Figure B-5: White light macro image (post thermal treatment) of representative weldment specimen #5 (as welded)

B-4

Appendix C
The following appendix pertains to the micro-section analysis of representative geometry weldment specimens 3 and 5. The following photographs are of the additional substantiating micro-sections. Reference table 4.1-1 for parameters used and observations made during weld trials.

C-1

Figure C-1: Planar micro-section location 1 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-2: Longitudinal micro-section location 2 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-2

Figure C-3: Transverse micro-section location 3 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-4: Transverse micro-section location 4 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-3

Figure C-5: Transverse micro-section location 5 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-6: Longitudinal micro-section location 6 from the circular weld build-up, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-4

Figure C-7: Longitudinal micro-section location 7 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-8: Planar micro-section location 8 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-5

Figure C-9: Transverse micro-section location 9 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-10: Transverse micro-section location 10 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-6

Figure C-11: Transverse micro-section location 11 from the straight-line-weave pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-12: Longitudinal micro-section location 12 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-7

Figure C-13: Planar micro-section location 13 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-14: Transverse micro-section location 14 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-8

Figure C-15: Transverse micro-section location 15 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-16: Transverse micro-section location 16 from single straight-line pattern bead, specimen #3 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-9

Figure C-17: Longitudinal micro-section location 1 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-18: Transverse micro-section location 2 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-10

Figure C-19: Transverse micro-section location 3 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid

Figure C-20: Transverse micro-section location 4 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid C-11

Figure C-21: Longitudinal micro-section location 5 from edge weld build-up, specimen #5 (post thermal treatment), etchant: electrolytic 10% oxalic acid

C-12

Appendix D
The following appendix pertains to the macro analysis of the fractured surfaces of all tensile specimens evaluated. The following photographs are of the additional substantiating macro-sections. Reference tables 4.4-1 and 4.4-2 for individual tensile specimen testing results.

D-1

Figure D-1: Post-test macro image of room temperature specimen CMT-1

Figure D-2: Macro image of fractured surface, specimen CMT-1

D-2

Figure D-3: Post-test macro image of room temperature specimen CMT-2

Figure D-4: Macro image of fractured surface, specimen CMT-2

D-3

Figure D-5: Post-test macro image of room temperature specimen CMT-3

Figure D-6: Macro image of fractured surface, specimen CMT-3

D-4

Figure D-7: Post-test macro image of elevated temperature specimen CMT-4

Figure D-8: Macro image of fractured surface, specimen CMT-4

D-5

Figure D-9: Post-test macro image of elevated temperature specimen CMT-5

Figure D-10: Macro image of fractured surface, specimen CMT-5

D-6

Figure D-11: Post-test macro image of elevated temperature specimen CMT-6

Figure D-12: Macro image of fractured surface, specimen CMT-6

D-7

Figure D-13: Post-test macro image of room temperature specimen GTAW-1

Figure D-14: Macro image of fractured surface, specimen GTAW-1

D-8

Figure D-15: Post-test macro image of room temperature specimen GTAW-2

Figure D-16: Macro image of fractured surface, specimen GTAW-2

D-9

Figure D-17: Post-test macro image of room temperature specimen GTAW-3

Figure D-18: Macro image of fractured surface, specimen GTAW-3

D-10

Figure D-19: Post-test macro image of elevated temperature specimen GTAW-4

Figure D-20: Macro image of fractured surface, specimen GTAW-4

D-11

Figure D-21: Post-test macro image of elevated temperature specimen GTAW-5

Figure D-22: Macro image of fractured surface, specimen GTAW-5

D-12

Figure D-23: Post-test macro image of elevated temperature specimen GTAW-6

Figure D-24: Macro image of fractured surface, specimen GTAW-6

D-13

Appendix E
The following appendix pertains to stress-strain curves generated during tensile testing. Reference tables 4.4.1-1 and 4.4.2-1 for individual tensile specimen testing results.

E-1

Figure E-1: Stress-Strain curve of room temperature CMT specimens

E-2

Figure E-2: Stress-Strain curve of elevated temperature CMT specimens

E-3

Figure E-3: Stress-Strain curve of room temperature GTAW specimens

E-4

Figure E-4: Stress-Strain curve of elevated temperature GTAW specimens

E-5