is bn 9789526031101

7/18/2019 Is Bn 9789526031101

1/169

TKK Dissertations 220

Espoo 2010

EFFECT OF WELDING PARAMETERS OF PLASMA

TRANSFERRED ARC WELDING METHOD ON

ABRASIVE WEAR RESISTANCE OF 12V TOOL

STEEL DEPOSIT

Doctoral Dissertation

Marko Kernen

Aalto University

School of Science and Technology

Faculty of Engineering and Architecture

Department of Engineering Design and Production

7/18/2019 Is Bn 9789526031101

2/169

7/18/2019 Is Bn 9789526031101

3/169

TKK Dissertations 220

Espoo 2010

EFFECT OF WELDING PARAMETERS OF PLASMA

TRANSFERRED ARC WELDING METHOD ON

ABRASIVE WEAR RESISTANCE OF 12V TOOL

STEEL DEPOSIT

Doctoral Dissertation

Marko Kernen

Doctoral dissertation for the degree of Doctor of Science in Technology to be presented with due

permission of the Faculty of Engineering and Architecture for public examination and debate in

Auditorium K215 at the Aalto University School of Science and Technology (Espoo, Finland) on the

28th of April 2010 at 12 noon.

Aal to Universi ty




Aal to-yl iopisto

Teknillinen korkeakoulu

Insinriti eteiden ja arkkitehtuurin tiedekunta

Koneenrakennustekniikan laitos

7/18/2019 Is Bn 9789526031101

4/169

Distribution:

Aalto University




Engineering Materials

P.O. Box 14200 (Puumiehenkuja 3)

FI - 00076 AaltoFINLAND

URL: http://edp.tkk.fi/en/

Tel. +358-9-470 23538

Fax +358-9-470 23537

E-mail: [email protected]

2010 Marko Kernen

ISBN 978-952-60-3109-5

ISBN 978-952-60-3110-1 (PDF)

ISSN 1795-2239

ISSN 1795-4584 (PDF)

URL: http://lib.tkk.fi/Diss/2010/isbn9789526031101/

TKK-DISS-2743

Picaset Oy

Helsinki 2010

7/18/2019 Is Bn 9789526031101

5/169

ABSTRACT OF DOCTORAL DISSERTATION AALTO UNIVERSITYSCHOOL OF SCIENCE AND TECHNOLOGYP.O. BOX 11000, FI-00076 AALTO

http://www.aalto.fi

Author Marko Kernen

Name of the dissertation

Effect of welding parameters of plasma transferred arc welding method on abrasive wear resistance of 12V tool steeldeposit

Manuscript submitted 29.10.2009 Manuscript revised 15.03.2010

Date of the defence 28.04.2010

Monograph Article dissertation (summary + original articles)

Faculty Faculty of Engineering and Architecture

Department Department of Engineering Design and Production

Field of research Engineering Materials

Opponent(s) Prof. Leijun Li and Prof. Jyrki Vuorinen

Supervisor Prof. Hannu Hnninen

Instructor Prof. Hannu Hnninen

Abstract

In the plasma transferred arc, PTA, welding method the powder consumable makes it possible to weld wide variety ofalloys. The dilution of the deposit is typically 3-10 % and, thus, the properties of the deposit can be achieved with one-layer deposit. The studied alloy was an iron-based 12V tool steel reinforced with primarily precipitating vanadiumcarbides.

Wide deposits are welded by oscillating the plasma arc and overlapping the weld beads. The mobility of the molten pool of12V tool steel is good. The location of the plasma arc in relation with the molten pool during perpendicular movement andthe shape of the plasma arc had strong effects on both the dilution control and the microstructure of the 12V tool steeldeposit. The location of the plasma arc can be in the edge or on top of the molten pool.

Depending on the parameter window the mean sizes and the volume fractions of vanadium carbides were 0,5-1,8

micrometers and 12,2-17,5 vol.-%, respectively. When the plasma arc was precisely in the edge of the molten pool, thesize, the distance between vanadium carbides, and the shape of vanadium carbides were optimal. With the studied welding

parameters, the 12V tool steel deposit is extremely sensitive to small variations in the welding parameters. If the moltenpool is stirred slightly when oscillating the plasma arc, vanadium carbides cannot grow and the shapes of the carbides aremainly more needle-shaped instead of round-shape.

Wear tests were made by the rubber wheel abrasion test according to standard ASTM G 65-94. Round-shaped, 1,2-1,4micrometer-sized vanadium carbides were the most beneficial according to the wear surface examinations. The optimized

PTA welded 12V tool steel deposit had better abrasive wear resistance when compared to the 12V tool steel depositmanufactured by the hot isostatic pressing, HIP.

Keywords PTA, tool steel, powder consumable, dilution, vanadium carbide, abrasive wear resistance

ISBN (printed) 978-952-60-3109-5 ISSN (printed) 1795-2239

ISBN (pdf) 978-952-60-3110-1 ISSN (pdf) 1795-4584

Language English Number of pages 166 p.

Publisher Aalto University, Department of Engineering Design and Production

Print distribution Aalto University, Department of Engineering Design and Production, Engineering Materials

The dissertation can be read at http://lib.tkk.fi/Diss/2010/isbn9789526031101/

7/18/2019 Is Bn 9789526031101

6/169

7/18/2019 Is Bn 9789526031101

7/169

VITSKIRJAN TIIVISTELM AALTO-YLIOPISTOTEKNILLINEN KORKEAKOULUPL 11000, 00076 AALTO

http://www.aalto.fiTekij Marko Kernen

Vitskirjan nimiHitsausparametrien vaikutus jauheplasmamenetelmll hitsatun 12V tykaluterspinnoitteen abrasiiviseenkulumiskestvyyteen

Ksikirjoituksen pivmr 29.10.2009 Korjatun ksikirjoituksen pivmr 15.03.2010

Vitstilaisuuden ajankohta 28.04.2010

Monografia Yhdistelmvitskirja (yhteenveto + erillisartikkelit)

Tiedekunta Insinritieteiden ja arkkitehtuurin tiedekunta

Laitos Koneenrakennustekniikan laitos

Tutkimusala Materiaalitekniikka

Vastavittj(t) Prof. Leijun Li ja Prof. Jyrki Vuorinen

Tyn valvoja Prof. Hannu Hnninen

Tyn ohjaaja Prof. Hannu Hnninen

TiivistelmJauheplasmamenetelmll, PTA, hitsattaessa jauhemainen lisaine mahdollistaa laajan lisainevalikoiman kytn.Menetelmlle tyypillinen sekoittumisaste on 3-10 %, jolloin pinnoitteen halutut ominaisuudet saavutetaan yhdellpalkokerroksella. Koemateriaalina oli rautapohjainen 12V tykaluters, jossa on lujitteena primrisesti erkautuviavanadiinikarbideja.

Leve pinnoite hitsataan oskilloimalla plasmakaarta ja limittmll hitsipalkoja. 12V tykaluterksen hitsisulan

liikkuvuus on hyv. Plasmakaaren paikalla suhteessa hitsisulaan kaaren poikittaisen liikkeen aikana ja plasmakaarenmuodolla oli suuri vaikutus sek 12V tykaluterksen sekoittumisasteen kontrollointiin ett mikrorakenteeseen.Plasmakaaren paikka voi olla joko sulan reunassa tai sen pll.

Parametri-ikkunasta riippuen vanadiinikarbidien keskimrinen koko ja tilavuusosuus olivat 0,5-1,8 mikrometri ja12,2-17,5 til.-%, vastaavasti. Kun plasmakaari oli tarkasti sulan reunassa, vanadiinikarbidien koko, karbidien vlinen

etisyys ja muoto olivat optimaaliset. Tutkituilla hitsausparametreill 12V tykaluterspinnoite onerityisen herkkhitsausparametrien pienille vaihteluille. Jos hitsisulaa sekoitetaan hieman kun plasmakaartaoskilloidaan,vanadiinikarbidit eivt pse kasvamaan kokoa, ja niiden muoto on pasiallisesti enemmn neulasmainenpyrenmuodon sijasta.

Kulutuskokeet tehtiin kumipyrabraasiokulutuskoelaitteistolla ASTM G 65-94 standardin mukaisesti. Pyret,1,2-1,4 mikrometrin kokoiset vanadiinikarbidit olivat kulutuspinnan SEM tarkastelun perusteella kaikkeinhydyllisimpi. Optimoidulla jauheplasmalla hitsatulla 12V tykaluterspinnoitteella oli parempi kulumiskestvyys

kuin vastaavalla kuumaisostaattisella puristuksella, HIP, valmistetulla materiaalilla.

Asiasanat PTA, tykaluters, jauhemainen lisaine, sekoittumisaste, vanadiinikarbidi, abrasiivinen kulumiskestvyys

ISBN (painettu) 978-952-60-3109-5 ISSN (painettu) 1795-2239

ISBN (pdf) 978-952-60-3110-1 ISSN (pdf) 1795-4584

Kieli englanti Sivumr 166 s.

Julkaisija Aalto-yliopisto, Koneenrakennustekniikan laitos

Painetun vitskirjan jakelu Aalto-yliopisto, Koneenrakennustekniikan laitos, Materiaalitekniikka

Luettavissa verkossa osoitteessa http://lib.tkk.fi/Diss/2010/isbn9789526031101/

7/18/2019 Is Bn 9789526031101

8/169

7/18/2019 Is Bn 9789526031101

9/169

7

Preface

The study was started with the project "Surfacing with different methods", which wasconducted during the years 1999-2002. The project was funded by TEKES, MetsoPowdermet Oy, Metso Works Oy, Metso Paper Oy, Fortum Power and Heat Oy, FortumOil and Gas Oy, Sulzer Pumps Finland Oy, Plasma Modules Oy, and Esab Oy. Thedissertation was conducted in the Graduate school on Metallurgy and MetalsTechnology during the years 2002-2006 and it was funded by Academy of Finland andMetso Powdermet Oy.

I would like to thank the supervisor of the thesis Professor Hannu Hnninen for hisguidance and advices. Dr. Jari Liimatainen from Metso Powdermet Oy is acknowledgedfor his support and for additional funding for my studies during the years 2000-2006.Dr. Pekka Siitonen, Dr. Mikko Uusitalo, M.Sc. Mikko Luukas, Mr. Heikki Vestman,Mr. Samuli Laine, and Mr. Jari Hellgren are acknowledged for helping in the PTAwelding tests. M.Sc. Tapio Saukkonen is thanked for helping in the SEM studies. M.Sc.Timo Kiesi is thanked for helping with the X-ray diffraction examinations. Laboratoryof Machine Design is thanked for helping in the rubber wheel abrasion tests. Personneland colleagues in the Laboratory of Engineering Materials are thanked for the greatworking atmosphere.

I am very thankful to my dear wife Sirpa for her patience and support throughout thiswork.

Espoo, March 2010

Marko Kernen

7/18/2019 Is Bn 9789526031101

10/169

8

Contents

1

Introduction ............................................................................................................. 17

1.1 General ........................................................................................................................ 17

1.2

Problem description..................................................................................................... 182

PTA welding method .............................................................................................. 19

2.1 PTA equipments .......................................................................................................... 202.1.1 Plasma torch ........................................................................................................ 202.1.2 Plasma nozzle ...................................................................................................... 212.1.3 Process gas .......................................................................................................... 252.1.4 Dilution control ................................................................................................... 27

2.2 Welding parameters..................................................................................................... 282.2.1 Pre-set and adjustable welding parameters ......................................................... 282.2.2 Oscillation ........................................................................................................... 29

2.3 Consumables ............................................................................................................... 312.3.1 Gas atomized powder .......................................................................................... 312.3.2

Micropellets ......................................................................................................... 33

2.3.3 Rods and wires .................................................................................................... 372.3.4 Consumable efficiency ........................................................................................ 38

2.4 Modeling ..................................................................................................................... 392.4.1 Arc and melting efficiency .................................................................................. 392.4.2 Heat input ............................................................................................................ 412.4.3 Thermal changes of powder particles .................................................................. 432.4.4 Dilution ............................................................................................................... 462.4.5 Dissolution of carbides ........................................................................................ 472.4.6 Weld pool convection .......................................................................................... 482.4.7 Weld pool solidification ...................................................................................... 50

2.4.8

Interactions of welding parameters ..................................................................... 51

2.5

Effects of main welding parameters on dilution ......................................................... 53

2.5.1 Plasma gas flow rate and plasma arc current ...................................................... 532.5.2 Working distance ................................................................................................. 552.5.3 Powder feed rate .................................................................................................. 552.5.4 Powder fraction ................................................................................................... 572.5.5 Shielding and carrier gas flow rates .................................................................... 572.5.6 Preheating ............................................................................................................ 58

3

Degradation of materials ......................................................................................... 59

3.1 Wear mechanisms ....................................................................................................... 603.2 Abrasive wear .............................................................................................................. 613.3 Modeling abrasive wear .............................................................................................. 67

3.4

Strengthening mechanisms .......................................................................................... 743.5 Rubber wheel abrasion test ......................................................................................... 77

4 Objectives of the dissertation .................................................................................. 795

Experimental ........................................................................................................... 80

5.1 Materials ...................................................................................................................... 805.2 Welding equipment ..................................................................................................... 805.3 Welding experiments................................................................................................... 82

5.3.1 Welding parameters ............................................................................................. 825.4 Experimental methods ................................................................................................. 85

5.4.1 Powder consumable ............................................................................................. 855.4.2 Temperature of work piece.................................................................................. 855.4.3 Microstructure ..................................................................................................... 855.4.4

Rubber wheel abrasion test.................................................................................. 86

7/18/2019 Is Bn 9789526031101

11/169

9

6

Results ..................................................................................................................... 87

6.1 12V tool steel powder ................................................................................................. 876.1.1 Microstructure ..................................................................................................... 876.1.2 Powder fraction ................................................................................................... 87

6.2

Temperature of work piece during the welding process ............................................. 89

6.3 Dilution ....................................................................................................................... 906.3.1 Welding parameters A ......................................................................................... 906.3.2 Welding parameters B ......................................................................................... 916.3.3 Welding parameters C ......................................................................................... 926.3.4 Welding parameters D-G .................................................................................... 936.3.5 Hardness .............................................................................................................. 95

6.4 Hardness distribution................................................................................................... 966.5 Optical metallography ................................................................................................. 98

6.5.1 Microstructure ..................................................................................................... 986.5.2 Microstructural changes in weld bead cross-section ......................................... 101

6.6 Scanning electron microscopy .................................................................................. 103

6.6.1

Welding parameters A-C ................................................................................... 1036.6.2 Welding parameters D-F ................................................................................... 106

6.6.3 Welding parameters G ....................................................................................... 1106.6.4 Materials manufactured by the HIP process ...................................................... 1116.6.5 Microstructure of overlapping beads ................................................................. 1146.6.6 Microstructural changes in weld bead in welding direction .............................. 119

6.7 Abrasive wear resistance ........................................................................................... 1226.7.1 Surface roughness ............................................................................................. 1226.7.2 Applied load ...................................................................................................... 1236.7.3 Dilution ............................................................................................................. 1236.7.4 Hardness ............................................................................................................ 1266.7.5 Mean size of vanadium carbides ....................................................................... 128

6.7.6

Carbide morphology .......................................................................................... 129

6.7.7 Mean free path ................................................................................................... 1316.7.8 Volume fraction of vanadium carbides ............................................................. 1336.7.9 Martensite and retained austenite contents ........................................................ 134

6.8 Wear tracks ................................................................................................................ 1366.8.1 Low stress abrasive wear ................................................................................... 1366.8.2 High stress abrasive wear .................................................................................. 138

7

Discussion ............................................................................................................. 139

7.1 PTA equipment ......................................................................................................... 1397.2 Dilution ..................................................................................................................... 139

7.2.1 Temperature of work piece................................................................................ 1397.2.2 Plasma arc current ............................................................................................. 1407.2.3

Powder feed rate ................................................................................................ 140

7.2.4 Location of plasma arc ...................................................................................... 1417.2.5 Argon-hydrogen gas mixtures ........................................................................... 141

7.3 Microstructure ........................................................................................................... 1417.3.1 Volume fraction of vanadium carbides ............................................................. 1427.3.2 Size and morphology of vanadium carbides ..................................................... 1427.3.3 MFP ................................................................................................................... 1437.3.4 Cumulative volumes of vanadium carbide sizes and MFPs .............................. 1437.3.5 Retained austenite ............................................................................................. 1437.3.6 Microstructural changes in weld bead ............................................................... 1447.3.7 Oscillation mode and welding speed ................................................................. 144

7.3.8

Overlapping ....................................................................................................... 145

7/18/2019 Is Bn 9789526031101

12/169

10

7.4 Abrasive wear resistance ........................................................................................... 1457.4.1 Microstructure ................................................................................................... 1457.4.2 Dilution ............................................................................................................. 1467.4.3 Plasma arc location ............................................................................................ 146

7.4.4

Preheating .......................................................................................................... 148

7.4.5

Wear tracks ........................................................................................................ 148

7.4.6 Welding parameters G ....................................................................................... 1487.4.7 Reference material 15V-1 tool steel .................................................................. 1497.4.8 Modeling abrasive wear of weld deposits ......................................................... 1497.4.9 Optimized welding parameters .......................................................................... 150

8 Conclusions ........................................................................................................... 151References .................................................................................................................... 154

7/18/2019 Is Bn 9789526031101

13/169

11

LIST OF ABBREVIATIONS

Ar ArgonB4C Boron carbideCO2 Carbon dioxideEB Electron beam welding methodF.C.C. Face-centered cubicFZ Fusion zoneGMAW Gas metal arc welding methodGTA, TIG Gas tungsten arc welding method, tungsten inert gas welding methodHAZ Heat affected zoneHe HeliumHIP Hot isostatic pressingHPTA High power plasma transferred arc welding methodH

2 Hydrogen

LB Laser beam welding methodMFP Mean free pathMIG/MAG Metal inert/active gas welding methodMMC Metal matrix compositeMPTA Micro-plasma transferred arc welding methodO2 OxygenPA Plasma arc welding methodPH PreheatedPMZ Partially melted zonePTA Plasma transferred arc welding method

RT Room temperature, 20 CRWAT Rubber wheel abrasion test according to standard ASTM G 65-94SAW Submerged arc welding methodVC Vanadium carbide

7/18/2019 Is Bn 9789526031101

14/169

12

NOMENCLATURE

A Experimentally determined constant, maximum melting efficiencyAg Volume of a scratch groove [mm

3]Ar1, Ar2 Ridges formed during deformation by indented particle or asperity [mm3]A1 Area of the first weld bead layer without the first weld bead layer dilution

[mm2]A2 Area of the first weld bead layer dilution without the first weld bead layer

deposit area [mm2]A3 Area of the second weld bead layer without the dilution from the first weld

bead layer [mm2]A4 Area of the second weld bead layer dilution from the first weld bead layer

without the second weld bead layer deposit area [mm2]a1 Empirical constant, 60 [m]a

2 Empirical constant, 0,32

B Experimentally determined constant, slope on a plot of against( )-1

Bmf Magnetic flux vectorb Unit of slip, Burgers vector []br Width of the rubber wheel [mm]C Coefficient of arc concentration [m-2]CPL Specific heat rate of a powder particle at liquid stateCp Specific heat of a powder particleCR Cooling rateDsas Secondary arm spacing, 3-8 [ m]

Diffusivity of C in austeniteDilution of the first weld bead layer [%]Dilution of the second weld bead layer [%]

d Average diameter of the reinforcing carbides [ m]da Average diameter of the abrasives [ m]

dc Circle diameter on the base material surface, [mm]

d0 Initial diameter of a powder particle [ m]E Elastic modulus [GPa]Ebm Melting enthalpy of the base materialEm Melting enthalpy

Epc Melting enthalpy of the powder consumableEs Specific volume enthalpy of the deposit material including latent heat ofmelting

Farc Impinging force of the arc plasmaFb Buoyancy forceFem Electromagnetic forceF Surface tension forcef Resistance opposing motion of a dislocation (a force per unit length)fab Abrasive efficiencyg Gravitational acceleration [m/s2]Ha Hardness of the abrasive

Hbm Hardness of the base material or the wearing material

7/18/2019 Is Bn 9789526031101

15/169

13

Hdef Hardness measured on the wear debrish Average penetration depth of the abrasive particles [ m]hb Thickness of the weld bead [mm]hr Relative penetration depth of the abrasive particles, h/d

I Arc current [A]Jcd Current density vectorK Wear coefficientK Geometrical constantKc Fracture toughness of the wearing material [MN/m

3/2]K1, K2 Coefficients in relation with surface and abrasive hardnessk Modification factor in relation with interphase interactionLm Melting heat rate of a powder particlel Sliding distance [m]ld Displacement of the plasma arc, overlapping [mm]Ms Martensite start temperature [C]

mnet Abrasive net mass flow rate past the specimen [g/min]mp Mass of an abrasive particle [g]mpf Packing fractionNp Number of abrasive particles in contact zone [No.]n Number of deposited beads without preheating [No.]n Number of deposited beads with preheating [No.]P Applied load [N]Pp Dimensionless parameterq(0) Maximum heat input along arc central axisq(r) Specific heat input at a radial distance of r from arc axisqeff Effective thermal powerqeff, l Effective power of the plasma arc in local preheatingqeff, n Effective thermal power at n

thturnqnet Net powerRe Reynolds number of plasma particles system, ~1,06Rp Tip radius of an abrasiveRs Radius of the shaft to be deposited [mm]r Radial distance from the point source [mm]

Powder particle radius [ m]Sm Heat conducting potential of argon plasma fluid at powder particles

melting point

SP Heat conducting potential of argon plasma fluid at powder particlestemperatureS Heat conducting potential of argon plasma fluid at plasma temperatures Relative sliding speed between the rubber wheel and the specimen [m/s]TM Melting time for a particle [ s]TLH Liquid heating time for a particle [ s]TSH Solid heating time for a particle [ s]Tamb Ambient temperature [C]Tf, bm Temperature at front of the surfacing point [C]Tm, bm Surface temperature ensuring melting of the deposit and the base material

[C]

Tmp Temperature of the molten pool [C]

7/18/2019 Is Bn 9789526031101

16/169

14

Tm, p Temperature of a powder particle at melting point [C]Tp Temperature of a powder particle [C]Tr Temperature at radius r from the point source [C]Ts Surface temperature of the molten pool [C]

T0, bm Initial temperature of the base material [C]T0, mp Reference temperature of the molten pool [C]T0, p Initial temperature of a powder particle [C]tdiss Time to dissolve a particle [ s]t0 Powder particle transporting time in the plasma arc [ms]U Arc voltage [V]

Carbon content in austenite close to the carbide-matrix interface [%]Carbon content in austenite [%]Carbon content in cementite [%]

V Wear volume of the deposit [mm3]Vm Abrasive wear resistance of the matrix phaseVpc Volumetric powder consumable feed rateVpc, max Maximum volumetric powder consumable feed rateVr Abrasive wear rateVsp Abrasive wear resistance of the second phaseVvf Abrasive wear resistance of the composite materialvf Volume fraction of the second phase [vol.-%]vft(t) Volume fraction of carbides as a function of time

Initial volume fraction of carbides [vol.-%]W Wear resistancewb Width of the weld bead [mm]

xcl Contact length between the rubber wheel and the specimen [mm]x1, x2 Distance from the original deposit surface to the worn phase [mm]y Coordinate parallel to welding direction

T Temperature gradient at the weld pool surfaceThermal diffusivity

a Mean attack angle []300 K Thermal diffusivity at 300 K

Factor describing the decay of deformation with increasing depth belowthe wearing surface

e Coefficient of thermal expansion of liquid metalRetained austenite content [%]

s Surface tension of liquid metali Wear resistance of ith phase

Relative surfacing step, ld/Rs Arc efficiency

Melting efficiencyMelting efficiency for the welds having 2D dimensional heat flowMelting efficiency for the welds having 3D dimensional heat flowThermal efficiency

2 Angle of abrasive tip []Weighted average of the tan values of all the individual abrasives

MFP Mean free path, spacing, MFP [ m]Coefficient of friction

7/18/2019 Is Bn 9789526031101

17/169

15

c Specific volume heat capacity of the material of the componentDensity [g/mm3]

PL Mass density of a powder particle at liquid state [g/mm3]

Density of an abrasive particle [g/mm3]

lm Density of liquid metal [g/mm3]Mass density of a powder particle [g/mm3]

Stress in contact area [N/mm2]Driving forceShear stressDimensionless surfacing time of n turns, n

Dislocation yield strengthWelding speed [mm/s]Volume fraction of ith phase [vol.-%]

kin, mp Kinematic viscosity at melting pointThermal conductivityCapability of deformation of the wearing material during abrasive wearEffective deformation on the wearing surfaceFunction of heat distribution

1 Constantpo Initial powder injection speed [m/s]

7/18/2019 Is Bn 9789526031101

18/169

16

Original features

The studied material was an iron-based 12V tool steel alloy (C 2,9 % and V 11,5 %).The plasma transferred arc welding tests and experimental data of this thesis were firstlydirected to evaluate the effects of the dilution on the properties of the 12V tool steeldeposit. Secondly, they were directed to the evaluation of the effects of the weldingparameters on the microstructure and the abrasive wear resistance of the 12V tool steeldeposit.

The following features are believed to be original:

1. Optimization of the microstructure and the abrasive wear resistance of the PTAwelded 12V tool steel deposit are systematically studied by varying weldingparameters.

2. The effects of varying welding parameters on the vanadium carbide size and theMFP distributions of the 12V tool steel deposit were clearly detected by usingseven parameter windows.

3. The effects of varying welding parameters on the volume fraction of vanadiumcarbides and the retained austenite content of the 12V tool steel deposit weremeasured. The volume fraction of vanadium carbides of the PTA welded 12Vtool steel deposit was lower than in the materials manufactured by the HIPprocess, but the welding parameters for the volume fraction of vanadiumcarbides close to that of the HIP material 12V were found.

4. The effect of the location of the plasma arc in relation with the edge of themolten pool and the effect of the main welding parameters on the microstructurewere detected by measuring the vanadium carbide shape ratios from the 12Vtool steel deposits. The effect of the shape ratio on the abrasive wear resistancewas confirmed from the wear surfaces in the SEM examination.

5. The microstructures and the abrasive wear resistances of the PTA welded 12Vtool steel deposits were compared to those of the tool steel materialsmanufactured by the HIP process. The microstructural differences were definedand the welding parameters having better abrasive wear resistance for the used

abrasive were found.

7/18/2019 Is Bn 9789526031101

19/169

17

1 Introduction

1.1 General

The plasma transferred arc welding method, PTA, was developed in the 1960s from theplasma arc welding method, PA. The method has become common in the 1980s. It isalso possible to use the PTA welding method for joint welding. During last decades thefocus of the development has been towards the mechanized, the robotized, or theautomated PTA welding method. Therefore, it is possible to reduce the welding costs,enhance the process controllability, improve the deposit quality, and increase thedeposition rate. High amount of welding parameters, their interactions, and complicatedeffects of welding parameters on the properties of the various deposits have restrictedthe optimization of the properties of the deposits welded by the PTA welding method.

The main feature of the PTA welding method is the powder consumable. The propertiesof the deposit can be tailored for various applications, because different kinds ofpowders can be tailor-mixed (Saltzman et al., 1985). Low boiling point alloys, e.g.,zinc, cannot be welded because of high temperature of the plasma arc (Wahl et al.,1993). The powder consumable is fed with a flow of carrier gas from the powderconsumable hopper through the plasma torch straight into the plasma arc, where thepowder melts with the base material and forms the deposit. Gas atomized sphericallyshaped powders are preferred. The powder fraction is usually 50-150 m. The powderfraction depends on the deposition rate and, thus, the fraction is typically larger for thehigh power plasma transferred arc welding method, HPTA.

One of the main features of the PTA welding method is the low dilution of the deposit,which is from 3 to 10 %. The low dilution enables the use of one-layer deposit toachieve the intended deposition properties. Other features are strong metallurgical bondbetween the deposit and the base material, low heat input, low spattering, low distortion,less finishing, small heat affected zone (HAZ), no significant loss of alloying elements,no slag, high quality of the deposit, and excellent properties of the deposit.

The shape of the plasma arc is narrow and sharp, because of the plasma nozzle whichintensifies the plasma arc. The plasma arc is stable from low current levels to extremelyhigh current levels. The deposition rate for the PTA welding method with powderconsumable is in the range of 0,1-20 kg/h. Thickness of one bead layer is between 1 and6 mm. It is possible to weld multi-layer deposits. By oscillating the plasma torch, whichhelps to control the dilution of the deposit, wider passes up to 100 mm can be welded(Dilthey et al., 1994). Bigger surfaces can be welded by overlapping (distance from thebead centerline to the next bead centerline) the weld beads.

Less expensive material like low alloyed steel can be used as a base material in theapplications where wear resistance is only needed. In corrosive environments highlyalloyed base materials are preferred to withstand the requirements set to the deposit. It ispreferable that the base material has coefficient of thermal expansion close to that of the

weld metal. Thus, cracks induced in the fusion line by different coefficients of thermal

7/18/2019 Is Bn 9789526031101

20/169

18

expansion can be avoided during cooling of the component. Preheating, weld beadsequencing, and post-weld heat treating can be made if necessary (Raghu et al., 1996).

The equipments of the PTA welding method are nowadays relatively cheap and the

PTA welding method can be mechanized or automated economically, which helps tocontrol the welding process. Typical applications for the PTA welding method arevalves, valve seats, extrusion machine screws and components for cutting, wasterecycling, mining, crushing, energy industry, and pulp and paper industry.

1.2 Problem description

The PTA welding method is used to weld several millimeter thick wear and corrosionresistant deposits. Traditionally, the dilution of the deposit has been the most studiedparameter. It has been thought that it is the most important parameter of the weld bead,because low dilution means that the process is cost-effective and the properties of thedeposit are automatically typical to the deposit alloy. During the surfacing process thedilution of the deposit can be controlled and it is usually adjusted by plasma arc current.Welding parameters affecting most the dilution of the deposit are plasma arc current,temperature of the work piece, working distance, powder feed rate, process gas flowrates, and oscillation parameters like amplitude, frequency, and welding speed.

The 12V tool steel deposit is sensitive to the welding process variations. The abrasivewear resistance of the 12V tool steel deposit is based on the primarily precipitatedvanadium carbides (VC). The microstructure of the deposit, i.e., the volume fraction of

carbides, the mean free path (MFP), the size of carbides, the morphology, thedistribution of carbides, and the matrix phases determine the abrasive wear resistance ofthe 12V tool steel deposit. The dilution of the deposit has to be controlled during thesurfacing procedure to maintain basic features of the PTA welding method. A lot ofwelding parameters and their interactions have effects on both the dilution and themicrostructure of the deposit.

The size of the molten pool is large, when welding thick abrasive wear resistant depositby oscillating the plasma arc. The location of the plasma arc is normally in the edge ofthe molten pool (Hallen et al., 1992b). The risk of the weld pool stirring, i.e.,convection of the molten pool liquid by the changing location of the plasma arc

increases with the size of the molten pool. The location of the plasma arc in relationwith the molten pool affects the weld pool convection. The PTA welding method is notstable during the surfacing procedure, e.g., work piece temperature changes from roomtemperature (RT) up to approximately 300 C. This leads to a need to adjust weldingparameters. To maintain the optimized microstructure, both the dilution and the effectsof the plasma arc on the factors determining the abrasive wear resistance of the 12V toolsteel deposit have to be controlled and adjusted during the PTA welding method.

Some experimental tests and models are made during the years to determine the factors,which control the dilution and the abrasive wear resistance of the weld beads. Thedifferences between the welding parameters may be small, but they have major effects

7/18/2019 Is Bn 9789526031101

21/169

19

on the microstructure and the abrasive wear resistance of the 12V tool steel deposit,because of high temperature and velocity of the plasma arc.

2 PTA welding method

One of the most important features of the PTA welding method is its ability to producelow dilution deposits. This is related to the good plasma arc stability andreproducibility. Low dilution saves the powder consumable and production time. Thedeposits produced by thermal spray technologies are only mechanically bonded. Thebond between the PTA welded deposit and the base material is a metallurgical bondand, thus, the deposit is able to resist many kinds of wear modes including impact wear.The possibilities to use the PTA welded wear and corrosion resistant deposits inindustry are broad.

During the last decades the dilution of the PTA welded deposits has been between 5 and10 %. The development of the equipments has enhanced the ability to control it andnowadays the dilution is from 3 to 10 % for the majority of the PTA applications(Chattopadhyay, 1997; Harris et al., 1983). Because of the improved weldingequipments and knowledge about the PTA welding method, the dilution of the depositbelow 10 % is possible to maintain during the surfacing process. The dilution is one ofthe major parameters, which has to be controlled and adjusted during the weldingprocess. Low dilution is important, because highly alloyed consumables are expensiveand the base material and the deposit are usually dissimilar. Heat input to the basematerial is lower with low heat input.

The dilution of the first weld bead layer, , Figure 1, can be estimated from thecross-sectional weld bead sample by an image analysis program using equation 1. If thesecond weld bead layer is welded, equation 2 can be used to estimate the dilution of thesecond weld bead layer, . The dilution is mainly concentrated nearby thefusion line, thus, the estimated dilution is not an exact value. The dilutions of the firstand the second weld bead layers are:

[%] 100 %, (1)

[%] 100 % , (2)

where A1 is the area of the first weld bead layer without the first weld bead layerdilution, A2 is the area of the first weld bead layer dilution without the first weld beadlayer deposit area, A3is the area of the second weld bead layer without the dilution fromthe first weld bead layer, and A4is the area of the second weld bead layer dilution fromthe first weld bead layer without the second weld bead layer deposit area.

7/18/2019 Is Bn 9789526031101

22/169

20

Figure 1 Schematic presentation of the deposit cross-section, where A1 is thedeposit area without penetration and A2is the area of penetration

2.1 PTA equipments

The mechanized PTA welding unit consists of the plasma torch, pilot and plasma arcpower supplies, mechanized torch oscillation unit or robot, component manipulator orrotating table, consumable feed unit, torch cooling unit, and work piece preheating unit.The powder plasma torches have been manufactured also for manual hand welding(DuMola et al., 1988) and mechanized internal cladding for the pipes the inner diameterof which is 70 mm at minimum (Dilthey et al., 1994; Draugelates et al., 1993).

2.1.1 Plasma torch

A typical powder plasma torch is presented schematically in Figure 2. In the torch the

pilot arc (non-transferred arc) is maintained between the non-consumable tungstenelectrode (cathode) and the water-cooled plasma nozzle (anode). The pilot arc is neededfor initiating and stabilizing the plasma transferred arc. The transferred plasma arc ismaintained between the electrode and the work piece. Plasma gas is needed to form theplasma and to shield the electrode during the surfacing process. The transferred plasmaarc transmits heat to the base material more efficiently than the non-transferred plasmaarc does. Carrier gas is needed to transfer the powder consumable from the hopperthrough the plasma torch into the plasma arc, where the powder melts. Shielding gas isused to shield the molten pool against air atmosphere. Like in the conventional plasmawelding process, also in the PTA welding method the plasma arc is constricted by theplasma nozzle. The shape of the plasma arc can be controlled by the plasma nozzle,

which is available in different kinds of diameters.

7/18/2019 Is Bn 9789526031101

23/169

21

Figure 2 The plasma torch of the PTA welding unit (Raghu et al., 1996)

The basic phenomenon of the plasma arc of the PTA welding method is as follows.Plasma current passes air between the electrode and the work piece, while the gasmolecules accelerate and collide with each other. Binding forces between atoms andelectrons collapse and electrons are released from the nucleus. The gas, consisting ofneutral molecules, positively charged atoms, and negative electrons, is ionized and,thus, the gas is capable to conduct current (Craig, 1988).

The deposition rate of the mechanized PTA welding method with the powder

consumable is up to 20 kg/h. The thickness of one bead layer is between 1 and 6 mm.The deposit thickness is usually as much as the tolerance of the wear part. The plasmaarc is stable with low to high heat inputs. The plasma torch is designed to weld with acertain deposition rate, i.e., various welding torches have been developed to weld withdifferent deposition rates. For the micro-plasma transferred arc welding method, MPTA,the deposition rate is below 1 kg/h and the bead thickness 0,1-0,5 mm, for instance(Saltzman et al., 1992; Sun et al., 1998).

The pilot and the plasma arcs are maintained with direct current (electrode negative),variable polarity, or alternatively pulsed current, which makes it easier to control heatinput, penetration, and the shape of the molten pool (Craig, 1988; Deuis et al., 1997;

Fromowicz et al., 1993). The mass flow configuration allows the gas volume to becontrolled and allows variations in the gas flow to be made via computer controlledprogram (Shubert, 1992).

2.1.2 Plasma nozzle

Energy density and temperature of the plasma arc are much higher than those in the arcwithout the plasma nozzle (Elizagrate et al., 1981). Plasma arc temperatures in thedifferent regions of the plasma arc depend on plasma gas. Argon, Ar, hydrogen, H2, andhelium, He, can be used as plasma gas and temperatures of the plasma arc are on

7/18/2019 Is Bn 9789526031101

24/169

average 15000, 6000, anda plasma gas.

In Figure 3 are presented

plasma transferred arc, Phighest temperature is neawas 4,75 mm. Arc current14,5 and 29 V, respectivel

Figure 3 Differences bis plasma gas

Temperatures of the top suestimated to be about 270weld pool determined bymean temperature of the1,08 (Gladky et al.,1990).

Many kinds of powder plas

In Figure 4 is presented amanufactured a torch whicflow was introduced to thebenefits were reduced turdecreased deposition of tpresented plasma arc tempplasma arc currents 100 anbased on the experimentaplasma gas, plasma gastemperature range of theplasma nozzle diameters,al., 1990).

22

0000 C, respectively. The gas mixtures ca

differences between the gas tungsten arc

A, arc temperatures, when Ar is used asrby the tip of the electrode. Diameter of ths were 200 A. Voltages of the GTA and th(Elizagrate et al., 1981).

etween the GTA and the PTA arc temperat(Elizagrate et al., 1981)

rface and 4 mm below the top surface of theand 1700 C (Feng et al., 1999). Mean te

calorimeter, was around 1650-1700 C, in eld pool is equal to melting point temper

ma torches have been designed for the PTA

torch, which utilizes a separate external nozh utilizes a separate stabilizer gas. This se

plasma arc between the electrode and the plbulence, reduced thermal load to the plas

he powder particles on the plasma torch.erature vs. various external plasma arc nozzd 220 A. Plasma gas was Ar. The temperatl data. Temperature of the plasma arc de

flow rate, plasma arc current, and plas lasma arc column with low plasma strea

nd plasma arc currents is broad (Dresvin,

be also used as

, GTA, and the

lasma gas. Thee plasma nozzlePTA arcs were

res when argon

weld pool wereperature of the

dicating that theture times 1,01-

welding method.

zle. Babi (1993)arate cold gassma nozzle. Thema nozzle, andIn Figure 4 is

le diameters andre fields are setends mainly ona nozzle. Thespeed, various

1972; Gladky et

7/18/2019 Is Bn 9789526031101

25/169

Figure 4 Plasma arc tea) 100 A and1990)

In Figure 5 is presented radplasma nozzle. Plasma gasbeing moved across the arcolumn is 60-200 m/s. Plarate and small plasma ncolumnar and directional(Dresvin, 1972; Gladky etm/s for powder particle sihigh speed camera (Siaga

Figure 5 Radial distribplasma nozzl

23

peratures vs. various external plasma arc nb) 220 A plasma arc current (Dresvin, 197

ial distribution of plasma gas flow at 1 mmstream velocity has been estimated by a c

column. Plasma gas stream velocity range ima gas stream velocity is higher with highzzle radius. Higher plasma gas velocity

plasma flow from the plasma nozzle to t l., 1990; Wilden et al., 2006). Particle velozes of 63 and 200 m, respectively, wereet al., 2008). Plasma gas flow rate was 3,5 l

ution of plasma gas stream speed at 1 mm d(Dresvin, 1972; Gladky et al., 1990)

ozzle diameters:2; Gladky et al.,

istance from theylindrical probe,n the plasma arcplasma gas flowincreases moree base materialities of 20 and 7easured with a

/min.

istance from the

7/18/2019 Is Bn 9789526031101

26/169

In Figure 6 are presentedmethod. In the PTA welcolumnar. The electrode lenergy density of the plas

welding method. Heat inpuwith low heat input. The idistance when comparednozzle has also an effectreduces the intensity of threlatively long distance, win working distance does(Koivula et al., 1992).

Figure 6 Effect of arcand the plas1992)

A change in plasma arc voldistance or by resetting tincrease in working distan

position has an effect onelectrode position 1 mmeffect of working distanceal., 1984).

24

the differences between the GTA and thing method, the plasma arc is much nar

cates inside the plasma nozzle. The plasmama arc and the production rate are larger t

t concentrates to a small area, thus, producin tensity of the plasma arc reduces on abouto the GTA welding method. The diamete

n the intensity of the plasma arc. The biggplasma arc. The intensity of the plasma arich helps to control the dilution, because anot have a significant effect on the pla

shape on intensity of the arc in the gas tuna transferred arc, PTA, welding methods

tage may be achieved in two ways, i.e., by c he electrode, while the other parameters

e leads to an increase in plasma arc voltag

the dilution as well as working distance.hanges plasma arc voltage from 0,5 to 1 Von plasma arc voltage is presented (Hunt,

e PTA weldingower and morearc is hotter, thehan in the GTA

g a narrow weldten times longerr of the plasmar plasma nozzle

decreases on asmall alternationma arc density

gsten arc, GTA,(Koivula et al.,

anging workingre constant. Ane. The electrode

By resetting the. In Figure 7 the1988; Pfeiffer et

7/18/2019 Is Bn 9789526031101

27/169

Figure 7 Effect of wor140 A (Pfeiff

In the mechanized PTAsystem to control workingcontrol system measureswith plasma arc voltage.varies.

2.1.3 Process gas

Three gas lines are needshielding gas line. The gasline, and it can be changmechanical and the metallalloying elements and incrair atmosphere. Because cplasma arc, carrier gas miinput, concentration, and i

The process gas in the Pphysical properties of the gthe shape of the deposit, ashielding gas can lead to b

The properties of the prothermal conductivity. Den1,78, 0,18, 0,08 kg/m3 aThermal conductivity of tpool, molten pool degassin

conductivity of the process

25

king distance on plasma arc voltage with plr et al., 1984)

elding method it is possible to install an adistance between the plasma torch and thelasma arc voltage and adjusts working dislasma arc voltage changes linearly when

d for the PTA welding process, i.e., plascan be the same in all lines, it can be diffed during the welding process. The procesurgical processes in the molten pool througase of some components from both the pro

rrier gas and the powder consumable are fedxtures partly with plasma gas and its prop

nization potential, can change (Craig, 1988;

A welding method can be Ar, Ar-H2, Ar as have effects on the shape of the plasma ar

nd the deposit microstructure. The contentrn-out of, e.g., Si and Mn alloying elements

ess gas depend mainly on density, ionisasities and ionization energies of Ar, He, ad 15,8, 24,6, and 13,6 eV, respectively

he process gas has an effect on temperatug effect, welding speed, and weld shape. T

gas is the main feature when high welding

asma arc current

utomatic controlwork piece. Theance in relationorking distance

ma, carrier, andent gas in everygas affects the

h the loss of thecess gas and thestraight into the

erties, e.g., heatShubert, 1992).

He, or He. Thec, the heat input,of O2 or CO2 in(Menzel, 2003).

ion energy, andnd H2 gases are(Menzel, 2003).e of the moltenerefore, thermal

speed is needed.

7/18/2019 Is Bn 9789526031101

28/169

26

Thermal conductivity of Ar is around 0,015 W/cmC at 8000 C. Thermal conductivityof H2is multiple at 4000 C temperature when compared to thermal conductivity of Ar.Thermal conductivity of He is higher when compared to thermal conductivity of Ar(Menzel, 2003).

Ar is the most common process gas in the PTA welding method. It is inert and it doesnot oxidize or react with the weld pool. The properties of Ar as a process gas are goodshielding effect due to high gas density, relatively stable narrow plasma arc, goodelectrical conductivity, low ionization potential, high surface tension of the weld pool,and low burning of alloying elements. While Ar is used as plasma gas it ionizes easily,which determines the ease of the plasma arc ignition and its stability. It has low thermalconductivity and, thus, the plasma arc is narrow and concentrated (Craig, 1988;Yamamoto et al., 1997).

Ar-H2 gas mixture has excellent deoxidizing and focusing effects (Lugscheider et al.,

1992). Thus, Ar is mainly used as plasma and Ar-H2gas mixture (5-10 % H2) can beused as shielding gas. Ar-H2 compound can be used also as powder carrier gas. Lowamount of H2in the plasma arc increases thermal intensity, heat input, penetration, andwelding speed. Approximately at 3870 C temperature H2 molecules dissociate, andenergy is released when the molecules recombine or contact with the work piece (Craig,1988). H2reduces pores, surface tension of the weld pool, and bead surface oxidation. Itincreases fluidity and wetting of the melted powder, which is useful when the smoothdeposit surface is needed (Harris et al., 1983; Hunt, 1988; Oechsle et al., 2000).

H can be transported to the deposit with a contaminated powder, for instance.Dissolution of H into the iron melt is low, about 0,00075 % per 100 g of iron at melttemperature at 1 atm. Dissolution of H into iron melt depends on temperature. Solubilitydecreases at the solidification temperature of iron melt. When the temperature of themolten pool further decreases the solubility of H into the iron melt still decreases. At1371 C temperature the solubility of H increases temporarily. Below this temperaturethe problems caused by H arise. Diffusivity of H decreases with decreasing temperature.At RT iron is capable to retain only a small part of H soluble at high temperatures.Dislocation interactions can enhance the transportation of H by several orders ofmagnitude. H diffuses and forms molecular H2 and escapes as a gas. Under rapidcooling H is retained in the austenite and at low temperatures austenite transforms tomartensite or bainite. Lattice imperfections may trap H. Inclusions and pores may trap

H2. High internal stresses can develop and cracks may initiate at the discontinuities(Stout, 1987).

He has high thermal conductivity and high ionization potential, which offers increasedwelding speed, penetration, and improved weld quality when compared to Ar gas (Bolset al., 1997). Gas densities of He and H2are lower than air, which causes problems tothe shielding efficiency. He content in the process gas can be between 30 and 100 %,thus, flow rates have to be increased to maintain the shielding efficiency. Regular flowrates for He mixtures of 30, 50, 75, and 100 % are 18, 28, 35, and 40 l/min, respectively(Menzel, 2003; Oechsle et al., 2000).

7/18/2019 Is Bn 9789526031101

29/169

27

Welding tests welded with various gases without preheating gave results as follows.Welding tests with Ar-He-CO2 as a shielding gas resulted in a good bead appearanceonly for NiCrBSi 80 12 alloy, while Ar, Ar-H2, and Ar-He gas mixtures gave good beadappearances for CoCrWC 64 29 5, NiBSi + 60 % WC, NiCrBSiAl 90 4 1,

NiCrBSi 80 12, X 120 WCrMoV 6 5 4 4, and X 3 CrNiMo 18 13 3 alloys. The bestweld beads were obtained with Ar and Ar-H2gas mixture. The deposits welded with Arhad the highest hardnesses. Pores were typically found, when Ar-He gas mixture wasused (Oechsle et al., 2000).

2.1.4 Dilution control

Optical spectroscopy is one of the auxiliary methods for controlling and adjusting thedilution of the weld bead. Spectroscopic analysis is typically used to analyze materials.The significant wave length, 500-1000 nm, has to be set before the welding process and

after that the chemical composition of the molten pool is possible to control with thespectrometer. In Figures 8a, b is presented the local maximum and minimum wavelength measurements made by the spectrometer in relation with the dilution of thedeposit alloys Stellite 12 and Alloy 60. The location of the plasma arc was on themolten pool. The dilution control was more accurate with the bigger dilution ratios(Dilthey et al., 1996a; Pavlenko, 1996).

Figure 8 Relative intensities measured by an optical spectrometer as a function ofdilution of the deposits deposited onto the base materials GGG 70 and

St 52: a) Stellite 12 and b) Alloy 60 (Pavlenko, 1996)

7/18/2019 Is Bn 9789526031101

30/169

28

2.2 Welding parameters

The effect of an individual welding parameter on the PTA welding method has to beknown for optimizing welding parameters and to learn interactions between differentwelding parameters. In the mass production to obtain equal quality onto the depositedcomponents, the surfacing process has to be consistent enough (Herrstrm et al., 1993).

The main advantages of the PTA welding method are heat input and cooling rate, whichare in the range suitable for depositing a component with a broad range of alloys as wellas for surface alloying or surface hardening. Usually brittleness, cracking, and gasporosity can be avoided.

2.2.1 Pre-set and adjustable welding parameters

A number of process and material factors affect the dilution of the deposit. 37 variablesfor the PTA welding method are identified, which have effects on the welding process.18 variables were identified as pre-set variables and 19 variables as variables, whichneed to be adjusted. The settings preferred by welding operator and welding engineermay also vary which makes it difficult to find optimized welding parameters (Kapus,1980).

The main parameters have the biggest effect on the PTA welding method output quality.The selection of the test conditions is vast. Often a large set of welding tests are needed.Stable process conditions during the surfacing process cannot be ensured (Nefedov et

al., 1994). Even though straight passes without oscillation can be welded, the PTAwelding method is usually maintained by oscillating the plasma torch. The mostimportant factors are (Herrstrm et al., 1993; Hunt, 1988; Kapus, 1980; Sexton et al.,1994):

plasma arc current current upslope plasma gas flow rate composition of plasma, carrier, and shielding gas powder consumable feed rate moisture content of powder and gas pipes chemical composition of powder consumable overlapping preheating temperature interpass temperature working distance arc voltage arc polarity electrode diameter electrode type electrode setback substrate shape and size

7/18/2019 Is Bn 9789526031101

31/169

substrate and consu oscillation mode oscillation amplitud oscillation frequenc dwell time welding speed shielding gas flow r transport gas flow r deposit thickness cooling water flow torch angle in relati melting temperatur

2.2.2 Oscillation

The thickness of the depoobtained by oscillating thefrom 5 to 50 mm. Pilot adilution in the reverse poicommon oscillation modepresented in Figures 9b, dmany mechanical torch trwelding unit the adjustabletime, transversal welding

oscillation mode is used.oscillation width amplitudis used.

Figure 9 The most comethod: a) stc) sine waveSharples, 198

The oscillation wave modecomponent. For instance, rheat input and penetratiowelding a bead to the surfa

When comparing the we

noticed that the surfacing

29

mable materials

ey

ateate

rateon with base materials of powder and substrate.

it is typically between 1 and 6 mm. The wplasma torch, Figures 9a-d. The width of arc and plasma arc currents adjustment helts of the oscillation waves (Dilthey et al.,

s, i.e., zigzag and rectangular wave oscill. Sine wave is almost equal to the zigzagnsport units utilize it. In the mechanizedoscillation parameters are longitudinal weld

speed, and oscillation width amplitude,

The adjustable oscillation parameters are, and frequency, when zigzag or sine wave

mmon oscillation wave modes used in thraight pass without oscillation, b) triangular

and d) rectangular wave mode (Blunt, 205)

can be modified to face the properties of theverse points may have different kinds of d can be controlled more efficient way,e edge of the component (Rasche, 1973).

r resistance between the surfacing metho

arameters, e.g., the straight pass or the osci

ider pass can bebead is typicallys to control the

1994). The mosttion modes arewave mode andor the robotizeding speed, dwellhen rectangular

welding speed,oscillation mode

e PTA weldingor zigzag wave,00; Hunt, 1988;

deposit and theell times. Thus,specially, when

s, it has to be

llation mode are

7/18/2019 Is Bn 9789526031101

32/169

taken into account. In thmovement of the plasma tand, thus, the valve head c

Magnetic arc oscillation uoscillation unit. Oscillatioelectromagnets, which arethe plasma arc, the dwellcontrolled independently.the deposits welded by abead edges, which were irrand nozzle damage were sthe sweep time the penetrThe consumable efficiencyfrom the plasma arc. The

causes a transverse defleccontrolled by the field stredilution of the deposit bet

In Figure 10 weld pool shmm and frequency 50 Hz ithe arc, melt in the moldecreases. The result is thmolten pool. The unmelteWelding speed and oscillatthe molten pool as well a(Gladky et al., 1990).

Figure 10 Shape of thewidth 15 mm

Wider surfaces can be surbead and displacement of theights and valleys on t

30

small components, like the valve heads rch can cause too high heat input to the co n be overheated and distorted (Klimpel et al.

it can be also used for weld surfacing insteaand oscillation mode are therefore adjuste located parallel to the welding direction. T time, and the magnitude of the peak magn he quality of the deposit is not as good, w echanized oscillation unit. The defects wer

egularly shaped. Also spattering, arc instabilen during the surfacing process. By adjustition can be controlled in the edges of thewas low, because of excessive deflection ofinteraction between the plasma arc and th

ion of the arc, because of the Lorentz forcgth and polarity. The deposit width from 8 teen 4 and 16 % can be obtained (Blunt, 200

pe vs. plasma torch welding speed with osc s presented. The powder particles, which w

ten pool. Therefore, the temperature of tdecreasing effect on the size, mass, and ted powder particles melt in the molten poion parameters also affect the size, mass, an

s arc location in relation with the edge of

weld pool vs. plasma torch welding speedand frequency 50 Hz (Gladky et al., 1990)

aced by overlapping single weld beads. W he plasma arc ratio, wb/ld, is 1,1-1,2. The di

e deposit surface is usually 20 %. When

, the oscillationmponent surface, 2005).

d of mechanizedby one or twoe time to sweeptic field can been compared tofound from the

ity, powder loss,g the dwell andscillation wave.

droplets ejectingmagnetic field

e, which can beo 15 mm and the).

illating width 15re not melted inhe molten pool

perature of thel in 10-4-10-2 s.d temperature ofthe molten pool

with oscillating

idth of the weldference between

the wb/ld ratio

7/18/2019 Is Bn 9789526031101

33/169

31

increases, e.g., above wb/ld 2 the deposit may become more even (Nefedov et al.,1994).

2.3 Consumables

Term weldability is defined by the American Welding Society as the capacity of ametal to be welded under the fabrication conditions imposed, into a specific, suitablydesigned structure, and to perform satisfactorily in the intended service. The definitionof this term is a complex process. The evaluation of the whole successful process fromthe beginning of the first plan to the use of an application in its real environment ischallenging. Term weldability is mainly used to describe and to compare how weldablethe consumable is.

Fe-, Co-, and Ni-based alloys are mainly used as powder consumables. All metals canbe produced in the powder form. Gas or water atomization powder production methodsare the dominant and perhaps the most versatile methods that produce metal powders athigh production rates. Various powder production methods allow precise control of thechemical composition and the physical characteristics of the powders. Variation of thechemical composition of the powder and the powder fraction have effects on theproperties of the deposit, the viscosity of the molten pool, the consumable efficiency,and the size of the molten pool, even though the chemical composition is in theapproved range (Dilthey et al., 1996b; Dunkley, 1998; German, 2005).

2.3.1 Gas atomized powder

Water, nitrogen, and argon gas are used in the metal powder atomization processes. Thenitrogen and argon atomized powders are usually spherical. Nitrogen atomization canlead to a poor powder shape and to a high retained gas content, which may lead toirregular and porous weld deposits (Dunkley, 1998; German, 2005; Hunt, 1988).

The water atomized as well as crushed carbide powders tend to be irregular in shape,while the gas atomized powder is spherically shaped. In Figure 11 is presented theeffect of the powder manufacturing method on powder feed rate. The deposit can bethicker or thinner depending on the powder used, if welding parameters are not

changed. Therefore, the dilution can also change. Powder feed rate and plasma arccurrent can be varied so that the thickness and the dilution of the weld bead remainconstant during the PTA welding method (Dren et al., 1985; German, 2005).

7/18/2019 Is Bn 9789526031101

34/169

Figure 11 Powder feedi(Dren et al.,

The powder consumablesFigure 12. The shape of thshaped powder particles arm. In the PTA welding mfrom 50 to 150 m. Interprocess, and with the bigg2005; LEstrade et al., 198

32

ng characteristics of the water and the gas at1985)

are mainly manufactured by the gas atoe powder particles can be controlled and uspreferred. The range of the powder size is

ethod the range of the powder consumablesnal porosity exists always regardless of thr fractions the porosity increases (Dunkley).

omized powders

ization method,ually sphericallyfrom 10 to 1000used is typically

manufacturing, 1998; German,

7/18/2019 Is Bn 9789526031101

35/169

33

Figure 12 Gas atomized powder production method (German, 2005)

2.3.2 Micropellets

The wear resistance can be enhanced by adding ceramic carbides to the matrix alloy.The deposit is then so called metal matrix composite, MMC. MMCs have a combinationof mechanical and physical properties, the properties of which can be tailored byselecting a metal matrix alloy and a reinforcing phase or carbide. MMCs have highstiffness, strength, and hardness. The wear and the corrosion resistance can be tailoredaccording to the application. The optimal microstructure of the deposit depends on theapplication.

Weldability of the powder alloy determines partly the properties of the deposit. The sizeof ceramic carbides is typically around 100 m. High volume fraction or large-sized

7/18/2019 Is Bn 9789526031101

36/169

34

ceramic carbides decrease the weldability of the alloy. The maximum amount ofcarbides in the MMC is about 60 vol.-% (Tsubouchi et al., 1997). In the wear resistantMMC deposits, large-sized original carbides of the powder consumable have to remainat least partly unmelted during the surfacing procedure. The original carbides can be

introduced straight into the weld pool with the external powder feed nozzle, if necessary(Rthig et al., 1997). The melting of the original carbides in the plasma arc is, thus,prevented or minimized (Hunt, 1988; Herrstrm et al., 1993; Kammer et al., 1991).

The hard wear resistant carbides are commonly manufactured by crushing and,therefore, the shape of the carbides is irregular. The mixed powders, which are based onthe metal powder manufactured by the gas atomization and the crushed carbides, haveproneness to separate in the powder consumable hopper and the feeding cable due todifferences between the density, the size, and the morphology of the particles. Surfacinglayers tend also to exhibit segregation at the time of melting. The particles with a higherdensity tend to settle to the lower regions of the deposit and the particles with a lower

density tend to settle to the upper regions of the deposit (Lugscheider et al., 1990;Lugscheider et al., 1995).

The micropelletizing produces agglomerates of spherical form, irrespective of the initialparticle form, size, morphology, or density, which is the main reason to use them. Anykinds of mixtures can be adjusted. The density of the agglomerated powder is possibleto enhance by sintering. The density of the powder has also an effect on the powderfeeding properties (Lugscheider et al., 1990; Lugscheider et al., 1994).

By micropelletizing, it is possible to manufacture the micropellets from the sub-micrometer-sized particles. As micropellets, the fine particles are agglomerated and,thus, carried together in the powder feeding system. The dust-like fine particles mayinterfere the feeding of the powder, but when the fine particles are adhered to the largerparticles the problems do not appear in the powder feeding system. This increases theability of flow of the composite powders as compared with the mechanicallyhomogenized ones (Lugscheider et al., 1990; Lugscheider et al., 1992).

In Figures 13a-d the morphology of differently produced powders are presented. InFigure 14 the basic metal matrix alloys and the reinforcing carbides, which can beincluded into the micropellets, are schematically presented. The morphology of themicropellets is markedly enhanced when comparing to the morphology of the tailor-

mixed powders (Lugscheideret al

., 1990; Lugscheideret al

., 1994).

7/18/2019 Is Bn 9789526031101

37/169

Figure 13 MorphologyLugscheider

a) Fused tb) Cobalt(magni

c) Comptungsteis 5 m (m

d) Comptungstetungstealloy le

35

of differently produced powders (Ait-Mt al., 1990):

ungsten carbide, crushed powder (magnificachromium-tungsten-carbon alloy, gas-atfication 400:1)site powder of cobalt-chromium-tungsten-n carbide, agglomerated, grain size of fused

and cobalt-chromium-tungsten-carbon allgnification 1000:1)site powder of cobalt-chromium-tungsten-n carbide, agglomerated and sintered, gran carbide is 5 m and cobalt-chromium

ss than 45 m (magnification 1000:1)

kideche, 1989;

ion 400:1)mized powder

carbon + fusedtungsten carbideoy less than 15

carbon + fusedn size of fusedtungsten-carbon

7/18/2019 Is Bn 9789526031101

38/169

Figure 14 Metal matri(Lugscheider

In Figure 15 is presented

The initial components are(polyvinyl alcohol) and whot air, where the dropledependent on the selectedinitial components, the miin a reducing atmosphereagglomerated and sinteredbinder evaporates. The wesintering. Also heating duLugscheider et al., 1990; L

36

powder, reinforcing particles, and tet al., 1992)

spray dryer, which is used to manufacture

stirred to form a suspension together with ater. The suspension is sprayed through thes dry in free atmosphere. The size of the

arameters. To increase the strength of adheropellets can be sintered after agglomeratio

(H2) at 1050 to 1150 C with a dwell timmicropellets loose their binder content, beclding behavior of the micropellets is improvring free-fall evaporates the volatile agent

ugscheider et al., 1994).

e micropellets

he micropellets.

n organic bindertomizer into themicropellets is

ion between the. When sintered

e of 60 min theause the organiced as a result of(German, 2005;

7/18/2019 Is Bn 9789526031101

39/169

Figure 15 Spray dryer f

2.3.3 Rods and wires

Rod, external powder feedthe same time with the pCoarse carbides can be emetals and carbides are soperation increases efficiepool and limit the depositibe increased close to the mheating, which is introduce

The bead thickness and thfrom 4 to 10 mm and froup to 23 and 33 kg/h wrespectively. The dilutionsurfacing direction was lesThe optimized welding p

37

r manufacturing the micropellets (German,

ing torch for carbides, hot wire, or cold wirwder consumable, e.g., for increasing thecased into the steel wire tubes, but optimi

eldom achieved for rod and wire welders.tly the deposition rate, while cold wire mayn rate (Sexton et al., 1994). Temperature of

elting point by electrical resistance heating ato the wire.

e deposition rate for the plasma arc hot w18 to 31 kg/h, respectively (Meyer, 1976).ith one 3 and 4 mm diameter hot wire

as between 5 and 50 %. The length of thes than 6 mm. Surfacing speed was approxirameters were plasma arc current 300-31

005)

e can be used atdeposition rate.zed mixtures ofTwin hot wirecool the moltenthe hot wire can

nd additional arc

re surfacing areDeposition ratecan be welded,weld pool in the

ately 0,8 m2/h.A, arc current

7/18/2019 Is Bn 9789526031101

40/169

heating the wire 280-30026-28 V, welding speed 8mm, shielding gas Ar + 8flow rate 2-3 l/min (Steklo

2.3.4 Consumable efficie

Before the surfacing procseveral hours or over the npowder feeding nozzle is nthe plasma arc. The powddiameters of which are lowplasma arc column. Instabobtained, when the powder

In the PTA welding metobtained (Xibao, 2003). Tunmelted powder particlewelding method is betweAlloy 60 (X40CrSiMo 10of +95 %. Oberlnder et al.

In Figure 16 the best consarc intensity lowers the m

diameter causes a decreaschanged. Higher plasma abecause the increased entvolume of the powder consof the finer particles in thefficiency with high plasmal., 1986; Xibao et al., 199

Figure 16 Effect of pl

consumable e

38

A, current passing through the wire 180 m/h, frequency 60 Hz, amplitude 50 mm,

10 % H2with flow rate 5-6 l/min, and plaet al., 1989).

cy

ss the powder consumable is usually drie ight to eliminate the moisture from the pow eeded for feeding the powder consumable w

er range is normally 50-150 m. The power than 50 m, tend to blow away due to higility in the plasma arc column and the modiameter is above 150 m.

od the evaporation of the powder particle he loss of the powder is due to spattering

s. The best reported consumable efficienn 95-99 % (Dilthey et al., 1996b). The p

). Anderson et al. (2003) reported the consu(1992) reported the consumable efficiency

umable efficiency is 97 % and the lowest 6lting efficiency of the plasma arc. An incr

in the consumable efficiency, if plasma a rc current causes an increase in the consualpy of the plasma arc is then sufficient

umable. The finer fractions are too dust-likee plasma arc may occur. The reason for loa arc currents presented in Figure 16 is not).

sma arc current and powder fraction (

fficiency (Babiak et al., 1983; Babiak et al.,

200 A, voltagewire diameter 3ma gas Ar with

d at 120 C forer particles. Theith carrier gas toer particles, the

h pressure of theten pool can be

s is not usuallyand the loss of

cy in the PTAwder alloy was

mable efficiencyf 80-90 %.

8 %. Decreasingase of the grain

rc current is notable efficiency,to melt a givenand evaporationwer consumableclear (Babiak et

iCrBSiFeC) on

1986)

7/18/2019 Is Bn 9789526031101

41/169

39

The consumable efficiency can vary depending on how the powder is fed to the plasmaarc. According to Dilthey et al. (1998) the consumable efficiency is not optimized,when external coaxial annular powder feed nozzle is utilized. This nozzle allowsmelting of the powder particles up to 200 m. Also arc deflections can occur especially

if the powder is fed to the arc from several channels instead of an annular nozzle.Internal powder feed system, where the powder is fed with a carrier gas through thehollow electrode straight to the plasma arc, provides longer dwell time for a powderparticle. Thus, larger particles up to 300 m can be melted. It is also possible to feed thepowder straight to the molten pool, which is more common in the laser welding method,LB (Klimpel et al., 2005). A simple and useful system is to use gravitational force toassist the powder feed. Anyway, the powder traveling time in the plasma arc is veryshort, at least, when compared to the time in the molten pool.

With an increase in the consumable feed rate, more plasma arc heat is required to meltthe consumable and less heat remains to melt the base material and, thus, the dilutiondecreases. At the same time the thickness of the deposit increases, if the otherparameters are not changed. Larger deposition rate basically requires an increase inplasma arc current. However, it is obvious that finally, when the thickness of the beadincreases the plasma arc is no more able to form penetration. The plasma arc location isthen on the molten pool (DuPont, 1998; Marantz, 1980).

2.4 Modeling

The effect of the plasma arc on the PTA welding method can be modeled. The arc and

the melting efficiencies and their effects on heat input as well as physical properties ofmaterials and welding parameters have to be known to get exact estimation about thePTA welding method. The oscillation of the plasma torch is not included to theequations, thus, these equations describe more the conventional PA welding method.However, the physical properties and welding variables, which have an effect on the arcand the melting efficiencies, are determined.

2.4.1 Arc and melting efficiency

Term thermal efficiency, t, considers both the arc efficiency, a, and the melting

efficiency, m. The arc efficiency describes a fraction of total heat load which has beentransferred from the plasma arc to the component to be welded. A fraction of energywhich has been used to melt the powder and the base material is considered as meltingefficiency. The weld material has no effect on the arc efficiency. The melting efficiencydepends strongly on the process parameters (DuPont et al., 1996).

The plasma arc efficiency in the PA welding method is lower and more variable thanthe arc efficiency in the GTA welding method, because of the plasma nozzle. However,the penetration of the weld bead at the same heat input is higher in the deposits weldedby the PA welding method when compared to the deposits welded by the GTA weldingmethod. Increasing welding speed to a certain rate and pulsating plasma arc current

effectively increase the plasma arc efficiency. The effect of the pulsed plasma arc on the

7/18/2019 Is Bn 9789526031101

42/169

40

arc efficiency is not as stable in the PA welding method as in the GTA welding method.Decreasing pulsed current duty cycle decreases the arc efficiency in the weldingprocess, while decreasing pulsed current duty cycle increases the melting efficiency.One benefit of the pulsed current is high melting efficiency at low welding speeds. The

main difference between the GTA and the PA welding methods, i.e., the plasma nozzle,has an effect on both the arc and the melting efficiency. Thus, the plasma nozzlediameter depends on, e.g., powder feed rate, following the inability to get an exactcomparison between the methods and the surfacing situations. The plasma nozzle is alsowater-cooled, and the removed heat energy is not known. As a conclusion meltingefficiencies of the PA and the GTA welding methods are close to equal. The meltingefficiency was slightly better for the PA welding method (Fuerschbach et al., 1991).However, DuPont et al. (1996) measured that both calculated and measured meltingefficiencies were lower in the PA welding method when compared to the otherconventional welding methods.

The difference between measured and predicted results was observed to be within 10 % (Giedt et al., 1989). The plasma arc efficiency is assumed to be 0,470,03(DuPont et al., 1996), 0,50-0,75 (Fuerschbach et al., 1991), and 0,55-0,80 (Gladky etal.,1990). The measurements conducted by Seebeck Envelope arc welding calorimeterdisclose that the arc efficiency of the plasma arc decreases slowly from 0,50 to 0,44with increasing plasma arc current from 200 to 400 A (DuPont et al., 1996). SeebeckEnvelope arc welding calorimeter has been disclosed by Giedt et al. (1989). Thecombination of the thermopile and gradient layer forms a heat-rate meter based on theSeebeck thermoelectric effect, thus, the calorimeter is called as Seebeck EnvelopeCalorimeter. The arc efficiency measurements are usually based on measurements madewith calorimeters. The reproducibility of the measurements is excellent, the errors beingabout 2 % (Dilthey et al., 1993; Giedt et al., 1989; Kou et al., 1986). Equation 3estimates the arc efficiency, , and it is based on heat conduction and average constantthermal properties, even though fluid flow and convection heat transfer can be veryimportant affecting the weld penetration (Giedt et al., 1989):

= [( Tamb ) ] , (3)

where Tris temperature at radius r from the point source, Tambis ambient temperature,is thermal conductivity, r is radial distance from the point source, U is arc voltage, I isarc current, is welding speed, is thermal diffusivity, and y is coordinate parallel to

the welding direction.

A dimensionless parameter, Pp, which correlates with the melting efficiency, , ispresented in equation 4 (Fuerschbach et al., 1991):

Pp , (4)

where qnetis net power, Emis melting enthalpy, kin, mpis kinematic viscosity at meltingpoint, and 300 K is thermal diffusivity at 300 K.

7/18/2019 Is Bn 9789526031101

43/169

41

The melting efficiency depends about two- or three-dimensional heat flow and basematerial thermal diffusivity. Higher thermal diffusivity as well as thicker work piececause low melting efficiency, because heat is transported more rapidly away from thewelding area. The melting efficiency depends on oscillation width, i.e., wider pass

increases the melting efficiency (Rykalin, 1957). It is generally supposed that nospattering occurs. Equations 5 and 6, which are based on experimental data, disclosemelting efficiencies for the welds having 2D or 3D dimensional heat flows(Fuerschbach et al., 1991):

0,407 , (5)

0,346 . (6)

The melting efficiency, , can be also estimated by equation 7 which includesexperimentally determined constants A and B (DuPont et al., 1996; Fuerschbach et al.,1991):

A , (7)

where A is maximum melting efficiency, e.g. 0,5 and B is slope on a plot ofagainst ( )-1, e.g. 175.

2.4.2 Heat input

The temperature of the molten pool depends mainly on the physical properties of thematerial. The main thermal parameters are thermal conductivity, , [J/mmsK], specificheat capacity, c , [J/gK], density, , [g/mm3], and thermal diffusivity, , [mm2/s]. Theseparameters depend on the temperature. Thus, the constants are average values of theparameters which vary in a wide temperature range (Radaj, 1992).

The transferred plasma arc, i.e., the heat source can be assumed to be a spot or a circularheat source. Thermo-physical properties of the base material are constant andcorrespond to the mean process temperature. Heat transfer to the air is not considered.

Heat conduction from the surface of the deposited area is poor resulting in hot surface inrelation with interior parts of the base material if preheating is not used.

General presentation of Gaussian distribution of radial specific heat input to the surfaceof the base material is presented in equation 8. It is presumably designed for the openarc, but can be used to explain the phenomena of the plasma arc. Arc intensity, i.e.,plasma nozzle, plasma gas, plasma gas flow rate, and plasma arc current, have an effecton the coefficient of arc concentration. Gaussian distribution of the specific heat input tothe base material is (Pavelic et al., 1969; Rykalin, 1957):

q(r) q(0) , (8)

7/18/2019 Is Bn 9789526031101

44/169

42

where q(r) is specific heat input at a radial distance of r from arc axis, q(0) is maximumheat input along arc central axis, and C is coefficient of arc concentration.

If a uniform power density is distributed into a circle, its diameter on the base material

is dc . The energy input into this area is presented in equation 9 (Goldak et al.,1984). More concentrated arc has smaller circle diameter, dc, and larger value of C.Energy input is then (Goldak et al., 1984; Kou, 1987):

q(0) U I. (9)

Effective thermal power of the arc, which is needed for melting the powder consumableand a thin layer from the surface of the base material, is presented in equation 10. Theestimated and the measured data were in good agreement. The actual temperature of thework piece was slightly higher at the start and lower at the end when compared to the

estimated work piece temperatures. Effective thermal power of the arc is (Nefedov etal., 1995):

qeff= ( Es 1,07 c (Tm, bm Tf, bm)), (10)

where ld is displacement of the plasma arc, hb is thickness of the weld bead, Es isspecific volume enthalpy of the deposit material including latent heat of melting, c isspecific volume heat capacity of the material of the component, wbis width of the weldbead, Tm, bm is surface temperature ensuring melti

is bn 9789526031101

Documents

plasma arc

universi tyschool of

technology espoo

tool steeldepositmanuscript

molten pool of12v tool

tkk dissertations

layer deposit

hannu hnninenabstractin