Lecture Notes
Table of Contents
Chapter 0. 1. 2. 3. 4. 5. 6. Subject Introduction Gas Welding Manual Metal Arc Welding Submerged Arc Welding TIG Welding and Plasma Arc Welding Gas Shielded Metal Arc Welding Narrow Gap Welding, Electrogas - and Electroslag Welding 7. 8. Pressure Welding Resistance Spot Welding, Resistance Projection Welding and Resistance Seam Welding 9. 10. 11. 12. 13. 14. 15. 16. Electron Beam Welding Laser Beam Welding Surfacing and Shape Welding Thermal Cutting Special Processes Mechanisation and Welding Fixtures Welding Robots Sensors Literature 101 115 129 146 160 175 187 200 208 218 73 85 43 56 Page 1 3 13 26
0. Introduction
2003
0. Introduction
Welding fabrication processes are classified in accordance with the German Standards DIN 8580 and DIN 8595 in main group 4 Joining, group 4.6 Joining by Welding, Figure 0.1.
1 Casting
2 Forming
3 Cutting
4 Joining
5 Coating
6
Changing of materials properties
Figure 0.1
Welding: permanent, positive joining method. The course of the strain lines is almost ideal. Welded joints show therefore higher strength properties than the joint types depicted in Figure 0.2. This is of advantage, especially in the case of dynamic stress, as the notch effects are lower.
Adhesive bonding Riveting Screwing
Soldering
Welding
br-er0-02.cdr
ISF 2002
Connection Types
Figure 0.2
0. Introduction
Figures 0.3 and 0.4 show the further subdivision of the different welding methods according to DIN 1910.
Flow welding
br-er0-03.cdr
Figure 0.3
Cast welding
Gas welding
Arc welding
Beam welding
Resistance welding
br-er0-04.cdr
Figure 0.4
1. Gas Welding
2003
1. Gas Welding
has been introduced long time ago it is still applied for its flexibility and mo6
bility. Equipment for oxyacetylene welding consists of just a few elements, the energy necessary for welding can be transported in cylinders, Figure 1.1.
7 1
2 1 2 3 4 5 6 7 8 9 oxygen cylinder with pressure reducer acetylene cylinder with pressure reducer oxygen hose acetylene hose welding torch welding rod workpiece welding nozzle welding flame 9
br-er1-01.cdr
Figure 1.1
density in normal state [kg/m ]
3
Process energy is obtained from the exothermal chemical reaction between oxygen and a combustible gas, Figure 1.2. Suitable combustible gases are C2H2, lighting gas, H2, C3H8 and natural gas; here C3H8 has the highest calorific value. The highest flame intensity from point of view of calorific value and flame propagation speed is, however, obtained with C2H2.
propane
1.43 oxygen
ISF 2002
0.9
645
air
300
490 335
510
KW k
/cm2
cm
/s
Figure 1.2
1. Gas Welding
loading funnel
formation of calcium carbide with water, Figure 1.3. Carbide is obtained from the reaction of lime and carbon in the arc furnace.
commercial quantities can be stored when C2H2 is dissolved in acetone (1 l of acetone dissolves approx. 24 l of C2H2 at 0.1 MPa), Figure 1.4.
to sludge pit
br-er1-03.cdr ISF 2002
Acetylene Generator
Figure 1.3 Acetone disintegrates at a pressure of more than 1.8 MPa, i.e., with a filling pressure of 1.5 MPa the storage of 6m of C2H2 is possible in a standard cylinder (40 l). For gas exchange (storage and drawing of quantities up to 700 l/h) a larger surface is necessary, therefore the gas cylinders are filled with a porous mass (diatomite). Gas consumption during welding can be observed from the weight reduction of the gas cylinder.
br-er1-04.cdr
acetone
acetylene
porous mass
acetylene cylinder
acetone quantity : acetylene quantity : cylinder pressure : ~13 l 6000 l 15 bar
ISF 2002
Storage of Acetylene
Figure 1.4
5 is by profrac-
duced
cylinder
tional distillation of liquid air and stored in cylinders with a filling pressure of up to 20 MPa, Figure 1.5. For higher oxygen
liquid air
oxygen
pipeline liquid
compressor
separation
supply
ISF 2002
Figure 1.5
The standard cylinder (40 l) contains, at a filling pressure of 15 MPa, 6m of O2 (pressureless state), Figure 1.6. Moreover, cylinders with contents of 10 or 20 l (15 MPa) as well as 50 l at 20 MPa are common. Gas consump-
50 l oxygen cylinder
protective cap cylinder valve take-off connection
gaseous
N
p = cylinder pressure : 200 bar V = volume of cylinder : 50 l Q = volume of oxygen : 10 000 l content control
tion can be calculated from the pressure difference by means of the general gas equation.
manometer foot ring
Q=pV
safety valve
liquid
vaporizer
user gaseous
Storage of Oxygen
Figure 1.6
1. Gas Welding
In order to prevent mistakes, the gas cylinders are colour-coded. Figure 1.7 shows a survey of the present colour code and the future colour code which is in accordance with DIN EN 1089. The cylinder valves are also of show a thread right-hand union nut.
actual condition
blue
actual condition
grey
DIN EN 1089
brown grey
Acetylene
cylinder
oxygen techn.
yellow
helium
brown red
valves are equipped with screw clamp retentions. Cylinder valves for other
acetylene
grey dark green grey
hydrogen
grey vivid green grey
argon
darkgreen black darkgreen
argon-carbon-dioxide mixture
grey grey
thread-connection
br-er1-07.cdr
nitrogen
carbon-dioxide
ISF 2002
Pressure regulators reduce the cylinder pressure to the requested working pressure, Figures 1.8 and 1.9.
cylinder pressure working pressure
br-er1-08.cdr
ISF 2002
Figure 1.8
1. Gas Welding
At a low cylinder pressure (e.g. acetylene cylinder) and low pressure fluctuations, single-stage regulators are applied; at higher cylinder pressures normally two-stage pressure regulators are
discharge pressure locking pressure
the
throttle the
closes
increased pressure
Single Pressure Reducing Valve, Shut Down
onto brane.
the
mem-
Figure 1.9
The
injector-type
welding torch injector or blowpipe
coupling nut mixer nozzle oxygen valve hose connection for oxygen A6x1/4" right
injector pressure nozzle suction nozzle welding nozzle fuel gas valve
torch body
ISF 2002
Welding Torch
welding
plate thicknesses.
1. Gas Welding
The special form of the mixing chamber guarantees highest possible safety against flashback, Figure 1.11. The high outlet speed of the escaping O2 generates a negative pressure in the acetylene gas line, in consequence C2H2 is sucked and drawn-in. C2H2 is therefore available with a very low pressure of 0.02 up to 0.05 MPa compared with O2 (0.2 up to 0.3 MPa).
pressure nozzle
torch body
br-er1-11.cdr
ISF 2002
Injector-Area of Torch
Figure 1.11 A neutral flame adjustment allows the differentiation of three zones of a chemical reaction, Figure 1.12:
escaping gas mixture acetylene decomposition C2H2 -> 2C+H2 1st stage of combustion 2C + H2 + O2 (cylinder) -> 2CO + H2 2nd stage of combustion 4CO + 2H2 + 3O2 (air) -> 4CO2 + 2H2O
2. welding zone:
3. outer flame:
complete reaction:
1. Gas Welding
3200C
2500C
1800C
1100C
400C
br-er1-12.cdr
ISF 2002
Figure 1.12
welding flame ratio of mixture
By changing the mixture ratio of the volumes O2:C2H2 the weld pool can greatly be influenced, Figure 1.13. At a neutral flame adjustment the mixture ratio is O2:C2H2 = 1:1. By reason of the higher flame temperature, an excess oxygen flame might allow faster welding of steel, however, there is the risk of oxidizing (flame cutting). Area of application: brass
excess of acetylene
normal (neutral)
excess of oxygen
The excess acetylene causes the carburising of steel materials. Area of application: cast iron
Figure 1.13
1. Gas Welding
10 By changing the gas mixture outlet speed the flame can be adjusted to the heat requirements of the welding job, for example when welding plates (thickness: 2 to 4 mm) with the weldsoft flame
welding flame
balanced (neutral) flame nozzle size: for plate thickness of 2-4 mm discharging velocity and weld heat-input rate: low 2
ing chamber size 3: 2 to 4 mm, Figure 1.14. The gas mixture outlet speed is 100 to 130 m/s when using a medium or normal flame, applied to
moderate flame
discharging velocity and weld head-input rate: high 4
at, for example, a 3 mm plate. Using a soft flame, the gas outlet speed is lower (80 to 100 m/s) for the 2 mm plate, with a hard flame it is higher
hard flame
br-er1-14.cdr
Figure 1.14 Depending on the plate thickness are the working methods leftward welding and rightward welding applied, Figure 1.15. A decisive factor for the designation of the working method is the sequence of flame and welding rod as well as the manipulation of flame and welding rod. The welding direction itself is of no importance. In leftward welding the flame is pointed at the open gap and wets the molten pool; the heat input to the molten pool can be well controlled by a slight movebr-er1-15e.cdr
Leftward welding is applied to a plate thickness of up to 3 mm. The weld-rod dips into the molten pool from time to time, but remains calm otherwise. The torch swings a little. Advantages: easy to handle on thin plates
welding-rod
flame
welding bead
Rightward welding ist applied to a plate thickness of 3mm upwards. The wire circles, the torch remains calm. Advantages: - the molten pool and the weld keyhole are easy to observe - good root fusion - the bath and the melting weld-rod are permanently protected from the air - narrow welding seam - low gas consumption
weld-rod
flame
ISF 2002
Flame Welding
Figure 1.15
1. Gas Welding In rightward welding the flame is directed onto the molten pool; a weld keyhole is formed (s = 3 mm). Flanged welds and plain butt welds can be applied to a plate thickness of approx. 1.5 mm without filler material, but this does not apply to any other
1,0
11
gap preparations
~ s+1 ~
denotation
symbol
r=
flange weld
12,0
1-2
V - weld
1,0
8,0
1-2
corner weld
By the specific heat input of the different welding methods all welding positions can be carried out using the oxyacetylene welding method, Figures 1.17 and 1.18
br-er1-16.cdr
1,0
8,0
lap seam
1,0
8,0
fillet weld
ISF 2002
Figure 1.16
butt-welded seams in gravity position
When working in tanks and confined spaces, the welder (and all other persons present!) have to be protected against the welding heat, the gases produced during welding and lack of oxygen ((1.5 % (vol.) O2 per 2 % (vol.) C2H2 are taken out from the ambient atmosphere)), Figure 1.19. The addition of pure oxygen is unsuitable (explosion hazard!).
s f
PA
gravity fillet welds
PB PF PG PC
horizontal fillet welds vertical fillet and butt welds vertical-upwelding position vertical-down position horizontal on vertical wall
PE
overhead position
PD
br-er1-17.cdr
ISF 2002
Welding Positions I
Figure 1.17
1. Gas Welding
12 A special type of autogene method is flame-straightening, where specific locally applied flame heating allows for
PA PB
shape correction of workpieces, Figure 1.20. Much experience is needed to carry out flame straightening processes. The basic principle of flame straightening
PC
PF
depends on locally applied heating in connection with prevention of expansion. This process causes the appearance of a
PG
PD PE
heated zone. During cooling, shrinking forces are generated in the heated zone and lead to the desired shape correction.
br-er1-18.cdr
ISF 2002
Welding Positions II
Figure 1.18
Flame straightening
welded parts
Hazards through gas, fumes, explosive mixtures, electric current protective measures / safety precautions 1. requirement for a permission to enter 2. extraction unit, ventilation 3. second person for safety reasons 4. illumination and electric machines: max 42volt 5. after welding: Removing the equipment from the tank
br-er1-19e.cdr
ISF 2002
br-er1-20.cdr
ISF 2002
Flame Straightening
Figure 1.19
Figure 1.20
2003
13 Figure 2.1 describes the burn-off of a covered stick electrode. The stick electrode consists of a core wire with a mineral covering. The welding arc between the electrode and the workpiece melts core wire and covering. Droplets of the liquefied core wire mix with the molten base material forming weld metal while the molten covering is forming slag which, due to its lower density, solidifies on the weld pool. The slag layer and gases which are generated inside the arc protect the metal during transfer and also the
c ISF 2002
br-er2-01.cdr
Weld Point
Figure 2.1
Covered stick electrodes have re1. Conductivity of the arc plasma is improved by a) ease of ignition b) increase of arc stability 2. Constitution of slag, to a) influence the transferred metal droplet b) shield the droplet and the weld pool against atmosphere c) form weld bead 3. Constitution of gas shielding atmosphere of a) organic components b) carbides 4. Desoxidation and alloying of the weld metal 5. Additional input of metallic particles
placed the initially applied metal arc and carbon arc The
electrodes.
ISF 2002
Figure 2.2
14
The covering of the stick electrode consists of a multitude of components which are mainly mineral, Figure 2.3.
coating raw material quartz - SiO2 rutile -TiO2 magnetite - Fe3O4 calcareous spar -CaCO3 fluorspar - CaF2 calcareous- fluorspar K2O Al2O3 6SiO2 ferro-manganese / ferro-silicon cellulose kaolin Al2O3 2SiO2 2H2O potassium water glass K2SiO3 / Na2SiO3
br-er2-03.cdr
effect on the welding characteristics to raise current-carrying capacity to increase slag viscosity, good re-striking to refine transfer of droplets through the arc to reduce arc voltage, shielding gas emitter and slag formation to increase slag viscosity of basic electrodes, decrease ionization easy to ionize, to improve arc stability deoxidant shielding gas emitter lubricant bonding agent
ISF 2002
Figure 2.3 For the stick electrode manufacturing mixed ground and screened covering materials are used as protection for the core wire which has been drawn to finished diameter and subsequently cut to size, Figure 2.4.
raw material storage for flux production raw wire storage jaw crusher descaling magnetic separation cone crusher for pulverisation sieving to further treatment like milling, sieving, cleaning and weighing sieving system
drawing plate
6 mm
5,5 mm
4 mm
3,25 mm
wet mixer
inspection
ISF 2002
br-er2-04.cdr
Figure 2.4
15
inspection electrode compound core wire magazine electrodepress compound packing inspection
inspection
TO DELIVERY
Figure 2.5
The core wires are coated with the covering material which contains binding agents in electrode extrusion presses. The defect-free electrodes then pass through a drying oven and are, after a final inspection, automatically packed, Figure 2.5.
pressing mass core rod guide pressing cylinder core rod coating pressing nozzle pressing cylinder
Figure 2.6 shows how the moist extruded covering is deposited onto the core wire inside an electrode extrusion press.
br-er2-06.cdr
Figure 2.6
16
Stick electrodes are, according to their covering compositions, categorized into four different types, Figure 2.7. with concern to burn-off characteristics and achievable weld metal toughness these types show fundamental differences.
cellulosic type cellulose 40 rutile TiO2 20 quartz SiO2 25 Fe - Mn 15 potassium water glass almost no slag droplet transfer : medium- sized droplets toughness value: good
br-er2-07.cdr
acid type magnetite Fe3O4 50 SiO2 20 quartz CaCO3 10 calcite Fe - Mn 20 potassium water glass slag solidification time: long droplet transfer : fine droplets to sprinkle toughness value: normal
rutile type rutile TiO2 45 magnetite Fe3O4 10 SiO2 quartz 20 CaCO3 10 calcite Fe - Mn 15 potassium water glass slag solidification time: medium droplet transfer : medium- sized to fine droplets toughness value: good
basic typ fluorspar CaF2 45 CaCO3 40 calcite SiO2 10 quartz 5 Fe - Mn potassium water glass slag solidification time: short droplet transfer : medium- sized to big droplets toughness value: very good
ISF 2002
Figure 2.7
The melting characteristics of the different coverings and the slag properties result in further properties; these determine the areas of application, Figure 2.8.
coating type symbol current type/polarity gap bridging ability welding positions sensitivity of cold cracking weld appearance slag detachability characteristic features
br-er2-08.cdr
cellulosic type C ~/+ very good PG,(PA,PB, PC,PE,PF) low moderate good spatter, little slag, intensive fume formation
acid type A ~/+ moderate PA,PB,PC, PE,PF,PG high good very good
rutile type R ~/+ good PA,PB,PC, PE,PF,(PG) low good very good
basic type B =/+ good PA,PB,PC, PE,PF,PG very low moderate moderate low burn-out losses hygroscopic predrying!!
ISF 2002
universal application
Figure 2.8
17
The dependence on temperature of the slags electrical conductivity determines the reignition behaviour of a stick electrode, Figure 2.9. The electrical conductivity for a rutile stick electrode lies, also at
conductivity
slag taining tile-con high ru r nducto semico
ence
in
the
Conductivity of Slags
details
hydrogen content < 5 cm /100 g welding deposit butt weld: gravity position fillet weld: gravity position suitable for direct and alternating current recovery between 125% and 160% basic thick-coated electrode chemical composition 1,4% Mn and approx. 1% Ni o minimum impact 47 J in -30 C 2 minimum weld metal deposit yield strength: 460 N/mm distinguishing letter for manual electrode stick welding
br-er2-10.cdr
ISF 2002
manual electrode welding flux cored arc welding submerged arc welding
G W
18
The identification numbers give information about yield point, tensile strength and elongation of the weld metal where the tenfold of the identification number is the minimum yield point in N/mm, Figure 2.11.
key number
minimum elongation*) % 22 20 20 20 18
35 38 42 46 50 *) L0 = 5 D0
br-er2-11.cdr
ISF 2002
Figure 2.11
The identification figures for the minimum impact energy value of 47 J a parameter for the weld metal toughness are shown in Figure 2.12.
characteristic figure Z A 0 2 3 4 5 6 7 8
minimum impact energy 47 J [ C] no demands +20 0 -20 -30 -40 -50 -60 -70 -80
The minimum value of the impact energy allocated to the characteristic figures is the average value of three ISO-V-Specimen, the lowest value of whitch amounts to 32 Joule.
br-er2-12.cdr
Figure 2.12
19 chemical of
composition
br-er2-13.cdr
ISF 2002
Alloy Symbols for Weld Metals Minimum Yield Strength up to 500 N/mm2
Figure 2.13
The properties of a stick electrode are characterised by the covering thickness and the covering type. Both details are determined by the identification letter for the electrode covering, Figure 2.14.
A B C R RR RA RB RC acid coating basic coating cellulose coating rutile coated (medium thick) rutile coated (thick) rutile acid coating rutile basic coating rutile cellulose coating key letter type of coating
br-er2-14.cdr
ISF 2002
Figure 2.14
20
plains the additional identification figure for electrode recovery and applicable type The of current.
subsequent
identification figure determines the application ties for possibilidifferent Figure 2.15 12345all positions all positions, except vertical down position flat position butt weld, flat position fillet weld, horizontal-, vertical up position flat position butt and fillet weld as 3; and recommended for vertical down position
br-er2-15.cdr ISF 2002
welding positions:
The last detail of the European Standard designation determines the maximum hydrogen content of the weld metal in cm per 100 g weld metal. Welding amperage current and
lengths to
are each
assigned
diameter,
Figure 2.16
Figure 2.16.
21
metal arc welding. Polarity and type of current depend on the trode known applied types. elecAll
power
work piece
br-er2-17.cdr ISF 2002
Since in manual metal arc welding the arc length cannot always be kept constant, a steeply descending power
power source characteristic U
A2 A1
source is used. Different arc lengths lead therefore to just minimally altered weld current intensities, Figure 2.18. Penetration remains basically unal-
2 1
A2
tered.
A1
ISF 2002
Figure 2.18
used for a.c. welding. For d.c. welding mainly converters, rectifiers and series regulator transistorised power sources (inverters) are applied. Con-
transformer
verters are specifically suitable for site welding and are mains-
tages of inverters are their small size and low weight, however, a more complicated electronic design is necinverter type
br-er2-19.cdr
ISF 2002
Figure 2.19
45 RA73
Figure 2.20 shows the standard welding parameters of different stick electrode diameters and stick electrode
medium weld voltage
V 40 RR73
types.
35
The rate of deposition of a stick electrode is, besides the used current intensity, dependent on the so-called electrode recovery, Figure 2.21. This describes the mass of deposited weld metal / mass of core wire ratio in percent. Electrode recovery can reach values of up to 220% with metal
30
25
= = = =
B15
3,25 4 5 6
20
100
200
300
400
23
A survey of the material spectrum which is suitable for manual metal arc welding is given in Figure 2.22. The survey comprises almost all metals known for technical applications and also explains the wide application range of the method.
7
eff ic ien cy
kg/h 6
de po s it ion
steel:
de po s it io n
ef fic
th
ed at co ick
16 0%
constructional steels shipbuilding steels high-strength constructional steels boiler and pressure vessel steels austenitic steels creep resistant steels austenitic-ferritic steels (duplex) scale resistant steels wear resistant steels hydrogen resistant steels high-speed steels cast steels combinations of materials (ferritic/ austenitic) cast iron with lamella graphite cast iron with globular graphite pure nickel Ni-Cu-alloys Ni-Cr-Fe-alloys Ni-Cr-Mo-alloys electrical grade copper (ETP copper) bronzes (CuSn, CuAl) gunmetal (CuSnZnPb) Cu-Ni-alloys pure aluminium AlMg-alloys AlSi -alloys
ISF 2002
22 0%
ien
cy
2
th in-
ed at co
1
X=
copper:
100
a = A- and R- coated electrodes, recovery 105% b = basic-coated electrodes, recovery <125% c = high-performance electrodes
br-er2-21.cdr ISF 2002
aluminium:
br-er2-22.cdr
Figure 2.21
producing forces can lead to the deflection of the arc from power supply point on the side of the workpiece, Figure 2.23. The mabr-er2-23e.cdr
terial
Arc Blow Effect through Concentration of Magnetic Fields
transfer
also
Figure 2.23
although, in that case, in the direction of the respective mass, Figure 2.24.
close to current-connection
Figures 2.25 and 2.26 show how by various measures the magnetic arc blow can be compensated or even
avoided.
opposite direction brings about the correct placement of the weld metal. Numerous strong tacks close the
br-er2-24.cdr
ISF 2002
tilting of electrode
tacks
through using a welding transformer alternating current (not applicable for all types of electrodes)
br-er2-25.cdr
ISF 2002
br-er2-26.cdr
ISF 2002
Figure 2.25
Figure 2.26
25 of the power supply point the various reasons for arc deflection can be eliminated. The fast magnetic reversal
4,0 %
20C / 70% RF
when a.c. is used minimises the influence of the magnetic arc blow.
3,0
the water absorption of a stick electrode may vary strongly during stor-
1,0
age, Figure 2.27. The absorbed humidity leads during subsequent weld-
10
Tage
100
ing frequently to an increased hydrogen content in the weld metal and, thus, increases cold cracking suscep-
Time of storage
br-er2-27.cdr
ISF 2002
tibility.
Figure 2.27 Stick electrodes, particularly those with a basic, rutile or cellulosic cover have to be baked before welding to keep the water content of the cover during welding below the permissible values in order to avoid hydrogen-induced cracks, Figure 2.28. The baking temperature and time are specified by the manufacturer. Baking is carried out in special ovens; in damp working conditions
Water content of the coating 1,0 % 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 30
br-er2-28.cdr
basic-coated electrode (having been stored at 18 - 20C for one year) 0,74
and only just before welding are electrodes from taken out
electrically
2003
26
In submerged arc welding a mineral weld flux layer protects the welding point and the freezing weld from the influence of the surrounding atmosphere, Figure 3.1. The arc burns in a cavity filled with ionised gases and vapours where the droplets from the
flux hopper electrode contact piece
continuously-
fed wire electrode are transferred into the weld pool. Unfused flux can be extracted from behind the welding head and subsequently recycled.
br-er3-01e.cdr
Figure 3.1
Main components of a submerged arc welding unit are: the wire electrode reel, the wire feed motor equipped with grooved wire feed rolls which are suitable for the demanded wire diameters, a wire straigthener as well as a torch head for current transmission, Figure 3.2. Flux supply is carried out via a hose from the flux container to the feeding hopper mounted torch which on is the DeAC or DC current supply wire straightener wire feed rolls flux supply indicators power source wire reel welding machine holder
head.
stall a flux excess pickup behind the torch. Submerged Figure 3.2
Assembly of a SA Welding Equipment
alloy type Mn
commercial wire electrodes S1 S2 S3 S4 S2Mo S3Mo S4Mo S2Ni1 S2Ni2 S2NiMo1 S3NiMo1 S3NiV1 S1NiCrMo2,5 S2NiCrMo1 S3NiCrMo2,5
main alloying elements Mn Ni Mo Cr V 0,5 1,0 1,5 2,0 1,0 1,5 2,0 1,0 1,0 1,0 1,5 1,5 0,5 1,0 1,5 1,0 2,0 1,0 1,0 1,0 2,5 1,0 2,5 0,6 0,8 0,6 0,5 0,6 0,8 0,5 0,5 0,15 0,5 0,5 0,5
either an a.c. power source or a d.c. power source where the electrode is normally connected to the positive terminal. Welding advance is provided by the welding machine or by workpiece movement.
MnMo
From a diameter of 3 mm upwards all wire electrodes have to be marked with the following symbols: S1 Si Mo S6: I : _ : IIIIII Example: S2Si: II _ S3Mo: III
ISF 2002
the average Mn-content and is carried out in steps of 0.5%, Figure 3.3. Standardisation for welding filler materials for unalloyed steels as well as for fine-grain structural steels is contained in DIN EN 756, for creep resis-
br-er3-03e.cdr
Figure 3.3 tant steels in DIN pr EN 12070 (previously DIN 8575) and for stainless and heat resistant steels in DIN pr EN 12072 (previously DIN 8556-10).
S3 1.5064 S2Si 1.5034 DIN EN 756 Reference analysis mat.-no. approx. weight % S1 C = 0,08 1.0351 Si = 0,09 Mn = 0,50 S2 1.5035 C = 0,10 Si = 0,10 Mn = 1,00 C = 0,11 Si = 0,15 Mn = 1,50 C = 0,10 Si = 0,30 Mn = 1,00 C = 0,10 Si = 0,15 Mn = 1,00 Mo = 0,50 C = 0,09 Si = 0,12 Mn = 1,00 Ni = 1,20 C = 0,10 Si = 0,12 Mn = 1,00 Ni = 2,20 C = 0,12 Si = 0,15 Mn = 1,00 Mo = 0,50 Ni = 1,00
For lower welding joint quality requirements;in: boiler and tank construction, pipe production, structural steel engineering, shipbuilding For higher welding joint quality requirements; in: pipe production, boiler and tank construction, sructural steel engineering, shipbuilding. Fine-grain structural steels up to StE 380. For high-quality welds with medium wall-thicknesses. Fine-grain structural steels up to StE 420. Especially suitable for welding of pipe steels, no tendency to porosity of unkilled steels. Fine-grain structural steels up to StE 420. For welding in boiler and tank construction and pipeline production with creep-resistant steels. Working temperatures of up 500 C. Suitable for higher-strength fine-grain structural steels. For welding low-temperature fine-grain structural steels. Non-ageing. Especially suitable for low-temperature welds. Non-ageing.
The proportions of additional alloying elements are dependent on the materials to be welded and on the mechanical-technological demands which emerge from the prevailing operating conditions, Figure 3.4. Connected to this, most important alloying elements are manganese for strength, molybdenum for high-temperature
S2Mo 1.5425
S2Ni1
S2Ni2
S3NiMo1
For quenched and tempered fine-grain structural steels. Suitable for normalising and/or re-quenching and tempering.
ISF 2002
br-er3-04e.cdr
28 The identification
Symbols of the chemical composition: S0, S1...S4, S1Si, S2Si, S2Si2, S3Si, S4Si, S1Mo,..., S4Mo, S2Ni1, S2Ni1.5, S2Ni2, S2Ni3, S2Ni1Mo, S3Ni1.5, S3Ni1Mo, S3Ni1.5Mo
br-er3-05e.cdr
Figure 3.5
During manufacture of fused welding fluxes the individual mineral constituents are, with regard of their future composition, weighed and subsequently
lime quarz rutile bauxite magnesite
fused in a cupola or electric furnace, Figure 3.6. In the dry granulation process, the melt is poured stresses break the crust into large fragments. During water granulation the melt hardens to form small grains with a diameter of
tapping raw material molten metal air coke roasting kiln silos balance coke
approximately 5 mm.
coal-burning stove
As a third variant, compressed air is additionally blown into the water tank resulting in finely blistered grains with low bulk weight. The fragments or grains are subsequently ground and screened thus bringing about the desired grain size. Figure 3.6
br-er3-06e.cdr
air
drying oven
balance
ISF 2002
rutile
Mn - ore
fluorspar
magnesite
alloys
ated weld fluxes the raw materials are very finely ground, Figure 3.7.
After weighing and with the aid of a suitable binding agent (waterglass) a pre-stage granulate is produced in the mixer.
gas
dish granulator
drying oven
Manufacture of the granulate is finished on a rotary dish granulator where the individual grains are rolled
cooling pipe
nealing furnace the remaining water is subsequently removed at temperatures of between 500C and 900C, depending on the type of flux.
Figure 3.7
Properties Fused fluxes
1)
The fused welding fluxes are characterised by high homogeneity, low sensitivity to moisture, good storing properties and high abrasion resistance. An important advantage of the agglomerated fluxes is the relatively low manufacturing temperature, Figure
Agglomerated 1) fluxes
uniformity of grain size distribution grain strength homogeneity susceptibility to moisture storing properties resistance to dirt current carrying capacity slag removability high-speed welding properties multiple-wire weldability flux consumption
1) 2)
+/++ +/++ +/++ +/++ +/++ --/+ +/++ -/+ +/++ -/++ -/+
-/++ -/++ -- /++ -/+ -/++ -/++ +/++ +/++ +/++ +/++ +/++
2)
3.8. The technological properties of the welded joint can be improved by the addition of temperature-sensitive deoxidation and alloying constituents to the flux. Agglomerated fluxes have, in general, a lower bulk weight (lower consumption) which allows the use of components which are reacting among
br-er3-08e.cdr
Figure 3.8
30 themselves during
MS CS ZS RS AR AB
MnO + SiO2 CaO CaO + MgO + SiO2 CaO + MgO ZrO2 + SiO2 + MnO ZrO2 TiO2 + SiO2 TiO2 Al2O3 + TiO2 Al2O3 + CaO + MgO Al2O3 CaF2 Al2O3 + SiO 2 + ZrO2 CaF2 + MgO ZrO2 Al2O3 + CaF2 CaO + MgO + CaF2 + Mo SiO2 CaF2 other compositions
AS AF FB Z
min. 50% max. 15% min. 55% min.15% min. 45% min. 15% min. 50% min. 20% min. 40% min. 40% min. 20% max. 22% min. 40% min. 30% min. 5% min. 70% min. 50% max. 20% min. 15%
the melting process. However, the higher susceptibility to moisture during storage andprocessing has to be taken intoconsideration.
br-er3-09e.cdr
Figure 3.9
The SA welding fluxes are, in accordance with their mineralogical constituents, classified into nine groups, Figure 3.9. The composition of the individual flux groups is to be considered as in principle, as fluxes which belong to the same group may differ substantially with regards to their
MS - high manganese and silicon pickup - restricted toughness values - high current carrying capacity/ high weld speed - unsusceptible to pores and undercuts - unsuitable for thick parts - suitable for high-speed welding and fillet welds CS acidic types - highest current carrying capacity of all fluxes - high silicon pickup - suitable for welding by the pass/ capping method of thick parts with low requirements basic types - low silicon pickup - suitable for multiple pass welding - current carrying capacity decreases with increaseing basicity - high-speed welding of single-pass welds - high manganese pickup/ high silicon pickup - restricted toughness values of the weld metal - suitable for single and multi wire welding - typical: welding by the pass/ capping pass method - average manganese and silicon pickup - suitable for a.c. and d.c. - single and multi wire welding - application fields: thin-walled tanks, fillet welds for structural steel construction and shipbuilding
ISF 2002
welding or weld metal properties. In addition to the groups mentioned above there is also the Z-group which allows free compositions of the flux.
The calcium silicate fluxes are recognized by their effective silicon pickup. A low Si pickup has low cracking tendency and liability to rust, on the other hand the lower current carrying capacity of these fluxes has to
ZS RS
AR
be accepted. A high Si pickup leads to a high current currying capacity up to 2500 A and a deep penetration. Aluminate-basic fluxes have, due to the higher Mn pickup, good mechanical
br-er3-10ae.cdr
Figure 3.10a
AB
- medium manganese pickup - good weldability - good toughness values in welding by the pass/ capping pass method - application field:unalloyed and low alloyed structural steels - suitable for a.c. and d.c. - applicable for multilayer welding or welding by the pass/ capping pass method - mainly neutral metallurgical behavior - manganese burnoff possible - good weld appearance and slag removability - to some degree suitable for d.c. - recommended for multi layer welds for high toughness requirements - application field: high-tensile fine grain structural steels, pressure vessels, nuclear- and offshore components - suitable for welding stainless steels and nickel-base alloys - neutral behaviour as regards Mn, Si and other constituents - mainly neutral metallurgical behaviour - however, manganese burnoff possible - highest toughness values right down to very low temperatures - limited current carrying capacity and welding speed - recommended for multi layer welds - application field: high-tensile fine-grain structural steeels - all other compositions
ISF 2002
AS
Fluoride-basic fluxes are characterised by good weld metal impact values and high cracking insensitivity. Figures 3.10a and 3.10b show typical properties and application areas for the different flux types.
AF
FB
Figure 3.11 shows the identification of a welding flux according to DIN EN 760 by the example of a fused calcium silicate flux. This type of flux is suitable for the welding of joints as well as for overlap welds. The flux can
br-er3-10be.cdr
Figure 3.10b
and low-alloy steels, as, e.g. general structural steels, as well as for welding hightensile and creep resistant steels. The silicon pickup is 0.1 0.3% (6), while the manganese pickup is expected to be 0.3 0.5% (7). Either d.c. or a.c. can be used, as, in principle, a.c. weldability allows
w e l d i n g f l u x D I N EN 760-SF CS 1 67 AC H10
also for d.c. power source. The hydrogen content in the clean weld metal is lower than the weld
hydrogen content
(table 4)
10 ml/100 g metal.
flux type
(figure 3.9)
br-er3-11e.cdr
Figure 3.11
32
The flux classes 1-3 (table 1) explain the suitability of a flux for welding certain material groups, for welding of joints and for overlap welding. The flux classes also characterise the metallurgical material behaviour. In table 2 defines the identification figure
table 2 table 1
unalloyed and low-alloyed steel general structural steel high-tensile & creep resistant steels stainless and heat resistant steels Cr- & CrNi steels welding of joints hardfacing pickup of elements as C, Cr, Mo
for
the
flux class 1 2 3
metallurgial behaviour
identification proportion flux in all-weld metal figure % 1 2 3 4 5 6 7 8 9 over 0,7 0,5 up to 0,7 0,3 up to 0,5 0,1 up to 0,3 0 up to 0,1 0,1 up to 0,3 0,3 up to 0,5 0,5 up to 0,7 over 0,7
5 10 15
br-er3-12e.cdr
Figure 3.12
Figure 3.13 shows the identification of a wire-flux combination and the resultant weld metal. It is a case of a combination for multipass SA welding where the weld metal
wire-flux combination D I N E N 7 5 6 - S 4 6 3 AB S2
standard no. wire electrode and/or wire-flux combination for submerged arc welding strength and fracture strain
(table1 and 2)
br-er 3-13e.cdr
shows
minimum
yield
mum metal impact value of 47 J at 30C (3). The flux type is aluminate-
impact energy
(table 3)
Figure 3.13
33
The tables for the identification of the tensile properties as well as of the impact energy are combined in Figure 3.14. The chemical composition of the weld
table 1
identification Identification for strength properties of multipass weld joints minimum yield point n/mm2 tensile strength minimum fracture strain N/mm2 %
metal and the structural constitution are dependent on the different metallurgical reactions during the welding process as well as on the used materials, Figure 3.15. The welding flux
35 38 42 46 50
440 up to 570 470 up to 600 500 up to 640 530 up to 680 560 up to 720
22 20 20 20 18
table 2 identification 2T 3T 4T 5T
Identification for strength properties of welding by the pass/ capping pass method welded joints
minimum base metal yield strength N/mm2 275 355 420 500
influences the slag viscosity, the pool motion and the bead surface. The different combinations of filler material and welding flux cause, in direct dependence on the weld parameters
table 3
Identification for the impact energy of clean all-weld metal or of welding by the pass/ capping pass method welded joints
Z
no demands
A +20
0 0
(current, voltage), a different melting behaviour and also different chemical reactions. The dilution with the base
br-er3-14e.cdr
ISF 2002
metal leads to various strong weld pool reactions, this being dependent on the weld parameters.
Figure 3.14
burn-off of the
behaviour element
br-er 3-15e.cdr
weld metal
manganese, Figure 3.16. For example: A welding flux with Figure 3.15
Mn-pickup
teristic and when a wire electrode S3 is used, has a neu1,0% S1 S2 S3 2,0% S4 S5 3,0% Mn in wire S6
tral
point
where
Mn-burnoff
br-er 3-16e.cdr
Figure 3.16
The pickup and burn-off behaviour is, besides the filler material and the welding flux, also directly dependent on the welding amperage and welding voltage, Figure 3.17. By the example of the selected flux a higher welding voltage causes a more steeply descending manganese characteristic at a constant neutral point. Silicon pickup increases with the increased voltage. The influence of current and voltage on the carbon content is, as a rule, negligible.
pickup/ burnoff rX in weight %
% Mn wire
neutral point
Inversely proportional to the voltage is the rising characteristic as regards manganese in dependence on the welding current, Figure 3.18. Higher currents cause the characteristic curve to flatten. As the welding voltage, the welding current also has practically no influence on the location of the neutral point. Silicon pickup decreases with increasing current intensity.
% Si wire
% C wire
br-er3-17e.cdr
ISF 2002
Figure 3.17
neutral point
In this example, the two points on the axis which determine the flux characteristic are defined for the parameters 600A
% Mn wire
450 A
welding current and 29V welding voltage, with the aid of the auxiliary straight line
% Si wire
and the neutral point curve (MnNP). In this case, the two points are positioned at
% C wire
0.6% Mn and 1.25% MnSZ. Dependent on the manganese content of the used filler material, the pickup or burn-off con-
br-er3-18e.cdr
ISF 2002
tents can be recognized by the reflection with respect to the characteristic line Figure 3.18 (0.38% Mn-pickup with a wire containing 0.5%Mn, 0.2% Mn-burnoff with a wire containing 1.75%Mn).
flux diagramm LW 280, manganese
wire electrode 4 mm acc. to Prof. Thier example: I = 580 A U = 29 V MnSZ1 = 0.48 % Mn MnSZ2 = 1.69 % Mn
The structure of the characteristic line for the determination of the silicon pickup content, is, in principle, exactly the same as described above, Figure 3.20. As silicon has only pickup properties and therefore no neutral point exists, the second auxiliary straight line must be considered for the determination of the second characteristic line point.
br-er3-19e.cdr
ISF 2002
Figure 3.19
36
Weld preparations for multipass fabrication are dependent on the thickness of the plates to be welded, Figure 3.21. If no root is planned during weld preparaflux diagramm LW 280, silicon
wire electrode 4 mm acc. to Prof. Thier example:
tion and also no support of the weld pool is made, the root pass must be welded using low energy input.
When welding very thick plates which are accessible from both sides, the double-U butt weld may be applied, Figure 3.22. Before the opposite side is welded, the root must be milled out (gouging/sanding). This type of weld cannot be produced by flame cutting
preparation and correct selection of the welding parameters lead to a high Figure 3.20 weld quality.
joints is the so-called steep single-V butt weld, Figure 3.23. The very steep edges keep the welding volume at a very low level. This technique, however, requires the application of special narrow-gap torches. Figure 3.21
br-er 3-21e.cdr
Welding Procedure Sheets for Single-V Butt Welds, Single-Y Butt Welds with Broad Root Faces and Double-V Butt Welds
preparation geometry
weld buildup
ing
weld-related
manual metal arc welding turning and sanding manual metal arc welding
SA SA
problems.
Here,
turn
SA SA
turn
side 2
SA SA
turn
SA SA
Special narrow-gap
ISF 2002
br-er3-22e.cdr
Figure 3.22
The most important welding parameters as regards weld bead formation are welding current, voltage and speed, Figure 3.24. A higher welding current causes higher deposition rates and energy input, which leads to reinforced beads and a deeper penetration. The weld width remains roughly constant. The increased welding voltage leads to a longer arc which also causes the bead to be wider. The change in welding speed causes - on both sides of an optimum - a decrease of the penetration depth. At lower weld speeds, the weld pool running ahead of the welding arc acts as a buffer between arc and base metal. At high speeds, per the unit
GMA welding
GMA welding
SA welding
lower
SA welding oscillated
br-er3-23e.cdr ISF 2002
Figure 3.23
38
constant:
w tp
plate thickness: wire electrode: flux:
penetration depth tp in mm
I w tp
weld width b in mm
2,4 2,2 2,0 1,8 1,6 1,4 1,2 1,0 0,8 0,6 0,4 0,2 0
agglomerated fluxes
400
500
600
700
w te
constant:
agglomerated fluxes
500
600
700
Figure 3.24
Figure 3.25
Weld flux consumption is dependent on the selected weld type, Figure 3.25. Due to geometrical shape, the flux consumption of a fillet weld is significantly lower than that of a butt weld. Because of their lower bulk weight, the specific consumption of agglomerated fluxes is
direction of welding
L1
L2
L3
br-er3-26e.cdr ISF 2002
Figure 3.26
U U0 US A U A I IS IK
external regulation ( U-regulation)
I I
The external regulation of the arc length by the control of the wire feed speed requires a power source with a steeply descending characteristic,
Figure 3.27. In this case, the shortening of the arc caused by some process disturbance, entails a strong
U U0 US U IS I A A I I
ISF 2002
voltage drop at a low current rise. As a regulated quantity, this voltage drop reduces the wire feed speed. Thus, the initial arc length can be regulated at an almost constant deposition rate. In contrast, the internal regulation effects, when the arc is reduced, a strong current rise at a low voltage
Figure 3.27 drop (slightly descending characteristic). At a constant wire feed speed the initial arc length is independently regulated by the increased burn-off rate which again is a consequence of the high current.
ceramic backing bar backing flux
The reaction of the internal regulation to process disturbance is very fast. This process is self regulating and does not require any machine expenditure.
flux copper backing
In submerged arc welding of butt joints, it is, depending on the weld preparation, necessary to support the
br-er3-28e.cdr
Figure 3.28
40
liquid weld pool with a backing, Figure 3.28. This is normally done with either a ceramic or copper backing with a flux layer or by a backing flux. Dependent on the shape of the backing bar, direct formation of the underside seam can be achieved. When welding circumferential tubes,
0 30
the inclination angle of the electrode has a direct influence onto the formation of the weld bead, Figure 3.29. For external as well as for internal tube welds, the best weld shapes may be obtained with an adjusted an-
1 = 0
gular position of the torch. If the advance is too low, the molten bath runs ahead and produces a narrow weld with a medium-sized ridge, too high
b1
b2
b3
the molten bath and a wide seam with a formed trough in the centre. The
t1
t2 inclusion
br-er3-29e.cdr
ISF 2002
processes described here for external tube welds are, the other way round, also applicable to internal tube welds.
efficiency of submerged arc welding, different process variations are applied, Figure
parallel twin wire tandem, twin wire
3.30. In multiwire welding, where up to 6 wires are used, each welding torch
br-er3-30e.cdr ISF 2002
from
one
power
arranged
ISF 2002
in
parallel or in a tanbr-er3-31e.cdr
dem.
Figure 3.31
In submerged arc welding with iron powder addition can the deposition rate be substantially increased at constant electrical parameters, Figure 3.31. The increased deposition rate is realised by either the addition of a currentless wire (cold wire) or of a preheated filler wire (hot wire). The use of a rectangular strip instead of a wire electrode allows a higher current carrying capacity and opens the SA method also for the wide application range of surfacing.
three-wire welding
= 35 1. WH HW = 1. WH 2. WH ~ 65 12..16
tandem welding
1. WH
2. WH ~
3. WH ~ 65
12..16 2. WH 3. WH ~ ~
However,
the
mentioned
process
=
variations can be combined over wide ranges, where the electrode distances and positions have to be appropriately optimised, Figure 3.32. Current type, polarity, geometrical coordination of the individual weld heads and the selected weld parameters also have substantial influence on the weld result. Figure 3.32
four-wire welding
br-er3-32e.cdr
15
10 10 35 12..16
~ 80
~ 15 18
~ 75 12
ISF 2002
100 kg/h
welding shows that this method can be applied sensibly and economically
single wire+ metal powder single wire+ hot wire double wire three-wire tandem single wire 0 500 1000 1500 2000 2500 A 3500 four-wire
deposition rate
80 70 60 50 40 30 20 10
0
over a very wide operating range, Figure 3.33. It is a high-efficiency welding process with a deposition rate of up to 100 kg/h. Due to large molten pools and flux application positional welding is not possible.
current intensity
12 kg/h
weld metal
9 6
5,0 mm 4,0 mm
800
current intensity
br-er3-33e.cdr ISF 2002
Figure 3.33
When more than one wire is used in order to obtain a high deposition rate, arc interactions occur due to magnetic arc blow, Figure 3.34. Therefore, selection of the the
(_) +
(_) +
elektrode
(+) _
_( ) +
arc
2003
43
TIG welding and plasma welding belong to the group of the gas-shielded tungsten arc welding processes, Figure 4.1. In all processes mentioned in Figure 4.1, the arc burns between a
Gas-shielded arc welding
trode
and or,
workpiece
narrow-gap gas-shielded arc welding electrogas welding Tungsten inert- Tungsten plasma gas welding welding with TIG electrode Tungsten hydrogen welding
Plasma arc Plasma arc Plasma welding with arc welding welding with semi-transferred non-transferred with transferred arc arc arc
ISF 2002
br-er4-01e.cdr
the torch. Exclusively inert gases (Ar, He) are used as shielding gases.
Figure 4.1
The potential curve of the ideal arc, as shown in Figure 4.2, can be divided into three characteristic sectors: 1.cathode- drop region 2.arc 3. anode-drop region
l
In
the
cathode-
K L
+
A
50% voltage
US
10 0
voltage
10
br-er4-02e.cdr
-4
4 mm 5
0,5
ISF 2002
however,
Arc Potential Curve
over
Figure 4.2
20 Argon 60 A V UK = 6,5 V
length is comparatively low. Main energy conversion occurs accordingly in the anode-drop and cathode-drop re-
UARC
gion.
Figure 4.3 shows the potential distribution by the example of a real TIG arc under the influence of different shielding gases. UA and UK have different values, the potential curve in the arc is not exactly linear. There is no discernible expansion of the cath-
UARC
20 anode UA = 6,1 V 0 1 2 3 4 mm 6
ISF 2002
10
cathode
XARC
br-er4--03e.cdr
Figure 4.3 The electrical characteristics of the arc differ, depending on the selected shielding gas, Figure 4.4. As the ionisation potential of helium in comparison with argon is higher, arc voltage must
arc voltage V 20
he lium
25 4 mm 2
15
4 2
n argo
10
50
100
150
200
250
350
weld current
br-er4-04e.cdr
ISF 2002
Figure 4.4
arc length
necessarily be higher.
45 temperature
mm
x
Figure 4.5
In TIG welding just approximately 30% of the input electrical energy may be used for melting the base metal, Figure 4.6. Losses result from the arc raradiation P = U .I welding direction
diation and heat dissipation in the workpiece and also from the heat conversion in the tungsten electrode.
R.I2 melting of wire
Figure 4.6
br-er4-06e.cdr
Figure 4.5.
mm
ISF 2002
4. TIG Welding and Plasma Arc Welding Figure 4.7 describes the process principle of TIG welding.
46
Figure 4.8 explains by an example the code for a TIG welding wire, as stipulated in the drafts of the European Standardisations.
tungsten electrode electric contact shielding gas shielding gas nozzle welding power source
workpiece arc
br-er4-07e.cdr isf 2002
Figure 4.7
A table with the chemical compositions of the filler materials is shown in Figure 4.9.
W 46 3 W2
chemical composition table rods and wires for tig-welding minimum impact energy value 47 J at -30C minimum weld metal yield point: 460 N/mm2 identification letter for TIG-welding
Figure 4.8
47
According to Figure 4.10, a conventional TIG welding installation consists of a transformer, a set of rectifiers and a torch. For most applications an electrode with a negative polarity is used. However, for welding of aluminium, current alternating must be
frequency
voltage is superimposed
br-er4-09e.cdr ISF 2002
and ionisation
causes
Figure 4.9
The central part of the torch for TIG welding is the tungsten electrode which is held in a collet inside the torch body, Figure 4.11. The hose package contains the supply lines for shielding gas and welding current. The shielding gas nozzle is more often than not made of
L1 L2 L3 N PE
mains
torches welding
TIG
_ O
~
St
well
as
rectifier transformer SC: scattering core for adjusting the characteristic curve
br-er4-10e.cdr
torches
+ O
= ~
duty
ISF 2002
cycles
selector switch
water-cooled.
Figure 4.10
48
longer arc
R and U rise
torch cap with seal handle of the torch control switch control cable
I drops
I rises
voltage
shielding gas supply cooling water supply cooling water return with welding current cable
increasing
U
decreasing
current intensity
br-er4-11e.cdr ISF 2002
br-er4-12e.cdr
decreasing
increasing
i
isf 2002
Figure 4.11
Figure 4.12
In order to keep the influence of torch distance variations on the current intensity and thus on the penetration depth as low as possible, power sources used for TIG welding always have a steeply drooping characteristic, Figure 4.12.
The
voltage reignition of the arc by voltage impulses
non-contact
A.C.
+ -
+ -
frequent
br-er4-13e.cdr
ISF 2002
voltage
Figure 4.13.
Figure 4.13
49
When argon is used as a shielding gas, metals as, for example, aluminium and magnesium, which besides having low melting points and also simultaneously forming tight and hard-to-melt-off oxide skins, cannot be welded with a negative polarity electrode. With a positive polarity, however, a cleaning effect takes place which is caused by the impact of the positive charged ions from the shielding gas atmosphere on the negative charged work surface, thus destroying the oxide skin due to their large cross-section. However, as a positive polarity would cause thermal overload of the electrode, these materials are welded with alternating current.
However, this has a disturbing side-effect. The electron emission and, consequently, the current flow are dependent on the temperature of the cathode. During the negative phase on the work surface the emission is, due to the lower melting temperature substantially lower than during the negative phase on the tungsten electrode. As a consequence, a positively connected electrode leads to lower welding current flows than this would be the case with a negatively connected electrode, Figure 4.14. A filter capacitor in the welding current circuit filters out the d.c. component which results in equal half-wave components. With modern transistorised power sources which use
without filter capacitor
current a current a
alternating current
+ + time
+ +
electrode polarity
time
+ + +
lower
time
- time
+ -
and height of the phase components adjustable. electrode stress The thermal and the
time
- time
stronger increasing
smaller
br-er4-14e.cdr
isf 2002
enced.
Figure 4.14
50 Figure 4.15 shows that the thermal electrode load can be recognized from the shape of the electrode tip. While the normal-load negative connected electrode end has the shape of a pointed cone (point angle approx.
10), a flattened electrode tip is the result from a.c. welding (higher thermal load by positive half-waves).The tip of a thermally overloaded electrode is hemispherical and leads to a stronger spread of the arc and thus to
overloaded electrode
br-er4-15e.cdr
ISF 2002
Figure 4.15
All fusion weldable materials can be joined using the TIG process; from the economical point of view this applies especially to plate thickness of less than 5 mm. The method is, moreover, predestined for welding root passes without backing support, Figure 4.16.
materials: - steels, especially high-alloy steel - aluminium and aluminium alloys - copper and copper alloys - nickel and nickel alloys - titanium - circonium - tantalum workpiece thickness: - 0,5 - 5,0 mm weld types: - plain butt weld, V-type welds, flanged weld, fillet weld - all positions - surfacing application examples: - tube to tube sheet welding - orbital welding - root welding
br-er4-16e.cdr ISF 2002
Figure 4.16
51
For circumferential welding of fixed pipes, the TIG orbital welding method is applied. The welding torch moves orbitrally around the pipe, i.e., the pipe is welded in the positions flat, vertical down, overhead, vertical-up and also interpass welding is applied.
preflow of the shielding gas postflow of the shielding gas movement in switch-on position
Moreover,
de-
shielding gas
achieved. welding
Orbital installa-
tions are equipped with process operational controls determine appropriate parame-
rise of current
br-er4-17e.cdr
preheating
pulsing
current decay
overlap
ISF 2002
which the
process
In plasma arc welding burns the arc between the tungsten electrode (- pole) and the plasma gas nozzle (+ pole) and is called the non-transferred arc, Figure 4.18. The non-transferred arc is mainly used for metal-spraying and for the welding of metal-foil strips.
In
contact tube shielding gas tungsten electrode shielding gas nozzle
plasma
welding transferred
burns the arc beplasma gas nozzle plasma gas filler material surface weld non-transferred arc workpiece welding power source Ignition device
br-er4-18e.cdr
isf 2002
Figure 4.18
contact tube shielding gas nozzle shielding gas plasma gas nozzle plasma gas
tungsten electrode
concentrated
heat input than in TIG welding and allows thus the exploitation of the keyhole effect. Plasma arc
work piece
welding arc
with is for
transferred arc
br-er4-19e.cdr isf 2002
transferred mainly
used
welding of joints.
Figure 4.19 Plasma arc welding with semi-transferred arc is a combination of the two methods mentioned above. This process variant is used for microplasma welding, plasma-arc powder surfacing and weld-joining of aluminium, Figure 4.20
The plasma welding equipment includes, besides the water-cooled welding torch, a gas supply for plasma gas (Ar) and shielding gas (ArH2-mixture, Ar/He mixture or Ar); the gas supply is, in most cases, separated, Figure 4.21. The copper anode and the additional focusing gas flow constrict the plasma arc which leads, in comparison with TIG welding, to a more concentrated heat input and thus to deeper
contact tube shielding gas nozzle shielding gas conveying gas and welding filler (powder) plasma gas plasma gas nozzle surface weld non-transferred arc transferred arc
br-er4-20e.cdr
tungsten electrode
burns more stable and is not easy to deflect, as, for example, at work-
workpiece
ISF 2002
Figure 4.20
53
br-er4-21e.cdr
ISF 2002
br-er4-22e.cdr
ISF 2002
Figure 4.21
Figure 4.22
The TIG arc is cone shaped or bell shaped, respectively, and has an aperture angle of 45. The plasma arc, in comparison, burns highly concentrated with almost parallel flanks, Figure 4.22. The shielding gas
arc length
configuration.
br-er4-23e.cdr
ISF 2002
Figure 4.23
plasma torch
than 2.5mm the so-called keyhole effect is utilised, Figure 4.24. The plasma jet penetrates the material,
welding direction
forming a weld keyhole. During welding the plasma jet with the keyhole
weld (seam)
moves along the joint edges. Behind the plasma jet as result of the surface tension and the vapour pressure in
weld surface
the keyhole, the liquid metal flows back together and the weld bead is
keyhole
created.
root
br-er4-24e.cdr
ISF 2002
Figure 4.24
Very
thin
sheets
0.05 and 50 A.
br-er4-25e.cdr
ISF 2002
Figure 4.25
55
Figures 4.25 and 4.26 show these application examples: The circumferential weld in a diaphragm disk with a wall thickness of 0.15mm and the joining of fine metal grids made of Cr-Ni steel.
br-er4-26e.cdr
ISF 2002
Figure 4.26
2003
56
The difference between gas-shielded metal arc welding (GMA) and the gas tungsten arc welding process is the consumable electrode. Essentially the process is classified as metal inert gas welding (MIG) and metal active gas welding (MAG). Besides, there are
gas-shielded arc welding (SG) gas-shielded metal-arc welding (GMAW)
metal inert gas welding (MIG) electrogas welding (MSGG) Narrow-gap gasshielded arc welding (MSGE) metal active gas welding (MAG) plasma gas metal arc welding (MSGP)
variants, the electrogas and the narrow gap welding and also the gasshielded plasma
hydrogen tungsten arc welding (WHG) plasma jet plasma arc welding (WPSL)
gas mixture gas metalarc CO2 metal-arc welding welding (GMMA) (MAGC)
metal arc welding, a combination of both plasma welding and MIG welding, Figure 5.1.
consumable electrode
br-er5-01e.cdr
Figure 5.1 In contrast to TIG welding, where the electrode is normally negative in order to avoid the melting of the tungsten electrode, this effect is exploited in MIG welding, as the positive pole is strongly heated than the negative pole, thus improving the melting characteristics of the feed wire. Figure 5.2 shows the principle of a GMA welding installation. The welding power source is assembled using the following assembly groups: The transformer converts the mains voltage to low voltage which is subsebr-er5-02e.cdr
water cooling
shielding gas control device
control switch
cooling water control
rectifier transformer
ISF 2002
quently rectified.
GMA Welding Installation
Figure 5.2
57 Apart from the torch cooling and the shielding gas control, the process
compact device
universal device
5, 10 or 20m 3 to 5m 3 to 5m
control is the most important installation component. The process control ensures that once set welding data are adhered to.
mini-spool device
push-pull device
A selection of common welding installation variants is depicted in Figure 5.3, where the universal device
10, 20 or 30m
5 to 10m
with a separate wire feed housing is the most frequently the industry. used variant in
br-er5-03e.cdr
ISF 2002
Figure 5.4 shows in detail a manually operated inert-gas shielded torch with the common swan-neck shape. A
Figure 5.3 machine torch has no handle and its shape is straight or swan-necked. The hose package contains the wire core and also supply lines for shielding gas, current and cooling water, the latter for contact tube cooling. The current is transferred to the wire electrode over the contact tube. The shielding gas nozzle is shaped to ensure a steady gas flow in the arc space, thus protecting arc and molten pool against the atmosphere.
1 torch handle 2 torch neck 3 torch trigger 4 hose package 5 shielding gas nozzle 6 contact tube 7 contact tube fixture 8 insulator 9 wire core 10 wire guide tube 11 wire electrode 12 shielding gas supply 13 welding current supply
br-er5-04e.cdr
ISF 2002
Figure 5.4
1 4 2 2
feed system is of the most simple design, as shown in Figure 5.5. The wire is pulled off a wire reel and fed into the hose package. The wire transport
roller, which shows different grooves depending on the used material, is driven by an electric motor. The counterpressure roller generates the frictional force which is needed for wire
5
1 wire reel 2 wire guide tube
6
3 wire transport roll 4 counter pressure roll
feeding.
More complicated but following the same operation principle is the Four ISF 2002
5 wire feed roll with a V-groove for steel electrodes 6 wire feed roll with a rounded groove for aluminium
br-er5-05e.cdr
Wire-Drive, Figure 5.6. Here, the second pair of rollers guarantees higher feeding reliability by reducing
Figure 5.5
4-roller drive
the risk of wheel slip. Another design among the wire feed drive systems is the planetary drive, where the wire is fed in axial direction by the motor. A rectilinear rotation-free wire feed motion is the outcome of the motor rotation and the angular offset of the drive rollers which are firmly connected to the motor shaft.
planetary drive
1
1 wire guide tube 2 drive rollers 3 counter pressure rollers 4 wire guide tube
direction of rotation
Figure 5.7 depicts the metal transfer in the short arc range. During the burning phase of the arc, material is molten
1 2 3 1 wire guide tube 2 roller holding device 3 drive rollers
ISF 2002
and accumulates at the electrode end. The voltage drops slowly while the arc shortens. Electrode and workpiece
br-er5-06e.cdr
Wire Drives
Figure 5.6
59
make contact and a short-circuit occurs. In the short-circuit phase is the liquid electrode material drawn as result of surface tension into the molten pool. The narrowing liquid root and the rising current lead
welding voltage
density
that
welding current
time
1 ms 1 mm
br-er5-07e.cdr
particularly suitable
Short-Circuiting Arc Metal Transfer
for and
Figure 5.7
welding.
The limitation of the rate of the current rise during the short-circuit phase with a choke leads to a pointed burn-off process which is smoother and clearly shows less spatter formation, Figures 5.8
proportion a long arc is formed in the upper power range, Figure 5.9. Material
time
time
transfer
is
unde-
choke effect
medium
ISF 2002
Choke Effect
Figure 5.8
5. Gas-Shielded Metal Arc Welding caused by the formation of very large droplets at the electrode end.
60
welding current
welding current
time
time
br-er5-09e.cdr ISF 2002 br-er5-10e.cdr
time
ISF 2002
Long Arc
Spray Arc
Figure 5.9
Figure 5.10
If the inert gas content of the shielding gas exceeds 80%, a spray arc forms in the upper power range, Figure 5.10. The spray arc is characterised by a non-shortcircuiting
35
and
C1 shielding gas composition: C1: CO2 M21: 82% Ar, 18% CO2 M23: 92% Ar, 8% O2
long arc
M21 M23
welding voltage
deposition
25
15
filler
and
cover
Welding Parameters in Dependence on the Shielding Gas Mixture (SG 2, 1,2 mm)
Connections tween
be-
welding
Figure 5.11
61 parameters, shielding gas and arc type are shown in Figure 5.11. When
helium
the shielding gas M23 is used, the spray arc may already be produced
thermal conductivity
hydrogen
decreasing argon proportion the amperage has to be increased in order to remain in the spray arc range. When
argon
pure carbon dioxide is applied, the spray arc cannot be produced. Figure 5.11 shows, moreover, that with the
temperature
argon 82%Ar+18%CO2
CO2
helium
increasing CO2 content the welding voltage must also be increased in order to achieve the same deposition
br-er2-12e.cdr
ISF 2002
rate.
The different thermal conductivity of Figure 5.12 the shielding gases has a considerable influence on the arc configuration and weld geometry, Figure 5.12. Caused by the low thermal conductivity of the argon the arc core becomes very hot this results in a deep penetration in the weld centre, the so-called argon finger-type penetration. Weld reinforcement is strongly pronounced. Application of CO2 and helium leads, due to the better thermal conductivity of these shielding gases, to a wide and deep penetration.
argon
br-er5-13e.cdr
temperature
argon
carbon dioxide
Fa
F Fr
Fr F
Fa
carbon dioxide
ISF 2002
A recombination (endothermic break of the linkage in the arc space exoFigure 5.13
wire elektrodes
this effect when CO2 is used. In argon, the current-carrying arc core is wider and envelops the wire electrode end, Figure 5.13. This generates electromagnetic forces which
argon carbon dioxide
bring about the detachment of the liquid electrode material. This socalled pinch effect causes a metal transfer in small drops, Figure 5.14.
until gravity has overcome that force component and material transfer in the form of very coarse drops appear.
Besides the pinch effect, the inertia and the gravitational force, other
surface tension S viscosity droplets necking down
forces, shown in Figure 5.15, are active inside the arc space; however these forces are of less importance.
backlash forces fr of the evaporating material suction forces, plasma flow induced
work piece
br-er5-15e.cdr ISF 2002
Figure 5.15
63
If the welding voltage and the wire feed speed are further increased, a rotating arc occurs after an undefined transition zone, Figure 5.16. High-efficiency MAG welding has been applied since the beginning of the nineties; the deposition rate, when this process is used, is twice the size as, in comparison, to spray arc welding. Apart from a multicomponent gas with a helium proportion,
also a high-rating power source and a precisely controlled wire feed system for high wire feed
Rotating Arc
Figure 5.16
Figure 5.17 depicts the deposition rates over the wire feed speed, as achievable with modern high-efficiency MAG welding processes.
During
25
kg/h
1,2 mm
the
transi-
deposition rate
20 15 10 5 0 0 5 10
conventional GMA
0,8 mm
degree.
15
20
25
30
35
40
45 m/min
critical
Deposition Rate
Figure 5.17
64
and is, with inert gas constituents of lower than 80%, hardly achievable thereafter. This effect facilitates the pulsed-arc welding technique, Figure 5.18.
300 number of droplets 1/s 200 critical current range 100 100 300 35 10 cm 200
-4 3
UEff
V arc voltage 25 20 15 10 5 Um
drop volume
500 A 400 welding current 350 300 250 200 150 100 50 0 0
ISF 2002 br-er5-19e.cdr
Setting parameters: - background current IG - pulse voltage UP - impulse time tP - background time tG or frequency f with f = 1 / ( tG + tP), resp. - wire feed speed vD
IEff Im
time
Ikrit
tG
IG
Im
10
15 time
20
ms
30
ISF 2002
br-er5-18e.cdr
Pulsed Arc
Figure 5.18
Figure 5.19
In pulsed-arc welding, a change-over occurs between a low, subcritical background current and a high, supercritical pulsed current. During the background phase which corresponds with the
welding current
pulsed current intensity Non-short-circuiting metal tranfer range
short arc range, the arc length is ionised and wire electrode
pulsed material
phase is
the
molten
superseded magnetic
by
Pulsed Metal Transfer
the
65
Figure 5.19 shows an example of pulsed arc real current path and voltage time curve. The formula for mean current is:
Im =
1T idt T 0
1T 2 i dt T 0
Ieff =
50 working range welding current / arc voltage 45 40 35 voltage [v] 30 transition arc 25 short arc 20 15 10 50 75 100 125 150 175 200 225 250 welding current 275 300 325 350 375 400 shielding gas: 82%Ar, 18%CO2 wire diameter: 1,2 mm wire type: SG 2 optimal setting lower limit upper limit spray arc
welding
technique allows a selection of different arc types which are by distinguished their metal
br-er5-21e.cdr
ISF 2002
transfer way. Figure 5.21 shows the setting range for a good welding
Figure 5.21
filler metal: SG2 -1,2 mm shielding gas: Ar/He/CO2/O2-65/26,5/8/0,5
of
conventional
GMA welding.
voltage
Figure 5.22 shows the extended setting range for the high-efficiency MAGM
100
br-er5-22e.cdr
20
high-efficiency short arc
10
short arc
welding with a
200
400
A
Quelle: Linde, ISF2002
600
process
rotating arc.
Figure 5.22
arc types
spray arc welding methods MAGC MAGM MIG aluminium copper steel unalloyed, lowalloy, high-alloy long arc short arc aluminium (s < 1,5 mm) pulsed arc aluminium copper
plications of the different arc types are depicted in Figure 5.23. The
steel unalloyed, low-alloy steel unalloyed, low-alloy fillet welds or inner passes and cover passes of butt welds at medium-thick or thick components in position PA, PB
steel unalloyed, low-alloy, steel low-alloy, high-alloy high-alloy steel unalloyed, low-alloy
applications
fillet welds or inner passes and cover passes of butt welds at medium-thick or thick components in position PA, PB welding of root layers in position PA
fillet welds or butt welds fillet welds or inner at thin sheets, all positions passes and cover passes of thin and root layers of butt welds medium-thick at medium-thick or thick components, all components, all positions positions inner passes and cover passes of fillet or butt welds in position PC, PD, PE, PF, PG (out-of-position) root layer welds only conditionally possible
rotating arc, (not mentioned in the figure), is applied in just the same way as the spray
br-er5-23e.cdr
ISF 2002
Figure 5.23
The arc length within the working range is linearly dependent on the set welding voltage, Figure 5.24. The weld seam shape is considerably influenced by the arc length. A long arc produces a wide flat weld seam and, in the case of fillet welds, generally undercuts. A short arc produces a narrow, banked weld bead.
operating point: wire feed speed: arc length: welding current: deposition efficiency: AL low long low low AM medium medium medium medium vD, I AL AM AK U arc length: long medium short
On the other hand, the arc length is inversely proportional to the wire feed speed, Figure 5.25. This has influence on the current over the internal adjustment with a slightly dropping power source characteristic. This
weld appearance:
br-er5-24e.cdr
ISF 2002
again is of considerable importance for the deposition rate, i.e., a low wire feed speed leads to a low deposition Figure 5.24
U AL AM AK
low base metal fusion. At a constant weld speed and a high wire feed speed a deep penetration can be obtained.
vD, I
AL high long
AM medium medium
AK low short
At equal arc lengths, the current intensity is dependent on the contact tube distance, Figure 5.26. With a
large contact tube distance, the wire stickout is longer and is therefore characterised by a higher ohmic resis-
br-er5-25e.cdr
ISF 2002
the contact tube distance, as a thumb rule, ten to twelve times the size of
Welding Voltage
influence on weld formation and welding process, Figure 5.27. When welding with the torch pointed in forward direction of the weld, a part of the weld pool is moved in front of the arc. This results in process instability. However, it ha s the advantage of a
30 mm 20
operating rule: lk = 10 to 12 dD
1
10
0 200
250
300 A
current
wire electrode: shielding gas: arc voltage: wire feed speed: welding speed:
br-er5-26e.cdr
flat smooth weld surface with good gap bridging. When welding with the torch pointed in reversing direction of the weld, the weld process is more stable and the penetration deeper, as
Figure 5.26
advance direction
Figure 5.28 shows a selection of different application areas for the GMA technique and the appropriate shieldpenetration: gap bridging: arc stability: shallow average deep
ing gases.
good bad
bad
narrow
assemblies, Figure 5.29. An additional ripple-filter choke suppresses the residual ripple of the rectified current and has also a process-stabilising
br-er5-27e.cdr
ISF 2002
Torch Position
effect.
83% Ar + 15% He + 2% CO2 90% Ar + 5% O2 + 5% CO2 80% Ar + 5% O2 + 15% CO2 92% Ar + 8% O2 92% Ar + 8% CO2 forming gas (N2-H2-mixture)
Ar/He-mixture Ar + 5% H2 or 7,5% H2
ment
of
efficient
shielding gases
Argon 4.8 Helium 4.6 Argon 4.6
application examples
autoclaves, vessels, mixers, cylinders panelling, window frames, gates, grids stainless steel pipes, flanges, bends spherical holders, bridges, vehicles, dump bodies reactors, fuel rods, control devices rocket, launch platforms, satellites valves, sliders, control systems stator packages, transformer boxes passenger cars, trucks radiators, shock absorbers, exhausts cranes, conveyor roads, excavators (crawlers) shelves (chains), switch boxes braces, railings, stock boxes mud guards, side parts, tops, engine bonnets attachments to flame nozzles, blast pipes, rollers vessels, tanks, containers, pipe lines stanchions, stands, frames, cages beams, bracings, craneways harvester-threshers, tractors, narrows, ploughs waggons, locomotives, lorries
ISF 2002
power source follows the principle of an audio frequency amplifier which amFigure 5.28
plifies a low-level to a high level input signal, possibly distortion-free. The transistor power source is, as conventional power sources, also equipped with a three-phase
69
transformer, with generally only one secondary tap. The secondary voltage is rectified by silicon diodes into full wave operation, smoothed by capacitors and fed to the arc through a transistor cascade. The welding voltage is steplessly adjustable until no-load voltage is reached. The difference between source voltage and welding voltage reduces at the transistor cascade and produces a comparatively high stray power which, in general, makes water-cooling necessary. The efficiency factor is between 50 and 75%. This disadvantage is, however, accepted as those power sources are characterised by very short reaction times (30 to 50 s). Along with the development of transistor analogue power sources, the consequent separation of the power section (transformer and rectifier) and electronic control took place. The analogue or digital control sets the reference values and also controls the welding process. The power section operates exclusively as an amplifier for the signals coming from the control.
The output stage may also be carried out by clocked cycle. A secondary clocked transistor power source features just as the analogue power sources, a transformer and a rectifier, Figure 5.30. The transistor unit functions as an on-off switch. By varying the on-off period, i.e., of the pulse duty factor, the average voltage at the output of the transistor stage may be varied. The arc voltage achieves small ripples, which are of a limited amplitude, in the switching frequency of, in general, 20 kHz; whereas the welding current shows to be strongly smoothed during the high pulse frequencies caused by inductivities. As the transistor unit has only a switching function, the stray power is lower than that
three-phase transformer mains supply fully-controlled three-phase bridge rectifier energy store transistor power section welding current
of
analogue
of
current pickup
these
clocked
u1 . . un
iist
Figure 5.29
70
Series regulator power sources, the so-called inverter power sources, differ widely from the afore-mentioned welding machines, Figure 5.31. The alternating voltage coming from the mains (50 Hz) is initially rectified, smoothed and converted into a medium frequency alternating voltage (approx. 25-50 kHz) with the help of controllable transistor and thyristor switches. The alternating voltage is then transformer reduced to welding voltage levels and fed into the welding process through a secondary rectifier, where the alternating voltage also shows switching frequency related ripples. The advantage of inverter power sources is their low weight. A transformer that voltage transforms with fre3-phase transformer
mains supply
energy store
transistor switch
former,
consideraUist U1 . . Un Iist
bly lower magnetic losses, that is to say, its size may accordingly smaller and be
br-er5-30e.cdr
current pickup
ISF 2002
its
filter
energy storage
transistor inverter
current pickup
br-er5-31e.cdr
ISF 2002
Figure 5.31
71
All welding power sources are fitted with a rating plate, Figure 5.32. Here the performance capability and the properties of the power source are listed. The S in capital letter (former K) in
manufacturer rotary current welding rectifier
insulations class
cooling type
~
type welding MIG/MAG
VDE 0542
production number
DIN 40 050
that
power range power capacity in dependence of current flow power supply
the
power
S
U1 220 V U1 380 V U1 U1 V V
situations,
Figure 5.32
Besides the familiar solid wires also filler wires are used for gas-shielded
br-er5-33e.cdr
sectional shapes.
Figure 5.33
72
Filler wires contain arc stabilisators, slag-forming and also alloying elements which support a stable welding process, help to protect the solidifying weld from the atmosphere and, more often than not, guarantee very good mechanical properties. An important distinctive criteria is the type of the filling. The influence of the filling is
symbol R P B M V W Y S slag characteristics rutile base, slowly soldifying slag rutile base, rapidly soldifying slag basic filling: metal powder rutile- or fluoride-basic fluoride basic, slowly soldifying slag fluoride basic, slowly soldifying slag other types customary application* S and M S and M S and M S and M S S and M S and M
of
the
electrode
chapter
*) S: single pass welding - M: multi pass welding **) C: CO2 - M2: mixed gas M2 according to DIN EN 439
br-er5-34e.cdr ISF 2002
Figure 5.34
2003
73
Up to this day, there is no universal agreement about the definition of the term Narrow Gap Welding although the term is actually self-explanatory. In the international technical literature, the process characteristics mentioned in the upper part of Figure 6.1 are frequently connected with the definition for narrow gap welding. In spite of these
Process characteristics: - narrow, almost parallel weld edges. The small preparation angle has the function to compensate the distortion of the joining members - multipass technique where the weld build-up is a constant 1 or 2 beads per pass - usually very small heat affected zone (HAZ) caused by low energy input
universally advantages
Advantages: - profitable through low consumption quantities of filler material, gas and/ or powder due to the narrow gaps - excellent quality values of the weld metal and the HAZ due to low heat input - decreased tendency to shrink
Disadvantages - higher apparatus expenditure, espacially for the control of the weld head and the wire feed device - increased risk of imperfections at large wall thicknesses due to more difficult accessibility during process control - repair possibilities more difficult
br-er6-01e.cdr
ISF 2002
Figure 6.1 The numerous variations of narrow gap welding are, in general, a further development of the conventional welding technologies. Figure 6.2 shows a classification with emphasis on several important process characteristics. Narrow gap TIG welding with cold or hot wire addition is mainly applied as an orbital process method or for the joining
submerged arc electroslag narrow narrow gap welding gap welding process with straightened wire electrode (1R/L, 2R/L, 3R/L) process with oscillating wire electrode (1R/L) process with twin electrode (1R/L, 2R/L) process with lengthwise positioned strip electrode (2R/L) flat position
br-er6-02e.cdr
of
high-
electrogas process with linearly oscillating wire electrode electrogas process with bent, longitudinally positioned strip electrode
process with hot wire addition (1R/L, 2R/L) MIG/MAGprocesses (1R/L,2R/L,3R/L) process with cold wire addition (1R/L, 2R/L)
als. This method is, however, hardly applied in the practice. The other are widely and are detail
processes more
vertical up position
spread
explainedin
Survey of Narrow Gap Welding Techniques Based on Conventional Technologies
in the following.
Figure 6.2
74
dance with this, the fundamental distinguishing feature of the methods is whether the process is carried out
with or without wire deformation. Overlaps in the structure result from the application of methods where a single or several additional wires are
used. While most methods are suitable for single layer per pass welding, other methods require a weld build-up with at least two layers per pass. A
A: method without forced arc movement B: method with rotating arc movement C: method with oscillating arc movement D: method with two or more filler wires
ISF 2002
further subdivision is made in accordance with the different types of arc movement.
Figure 6.3 In the following, some of the GMA narrow gap technologies are explained: Using the turning tube method, Figure 6.4, side wall fusion is achieved by the turning of the contact tube; the contact tip angles are set in degrees of between 3 and 15 towards the torch axis. With an electronic stepper motor control, arbitrary transversearc oscillating mocorrugated wire method with mech. oscillator
tions with defined dwell periods of oscillation and oscillation frequencies can be realised - independent of the filler wire properties. In contrast, when the corrugated method with wire me1 - wire reel 2 - drive rollers 3 - wire mechanism for wire guidance 4 - inert gas shroud 5 - wire guide tube and shielding gas tube 6 - contact tip
br-er 6-04e.cdr
2 3 4 5 6
2 3 4 5 6
12 - 14
1 - wire reel 2 - mechanical oscillator for wire deformation 3 - drive rollers 4 - inert gas shroud 5 - wire feed nozzle and shielding gas tube 6 - contact tip
8 - 10
75
wire electrode. The deformation is obtained by a continuously swinging oscillator which is fixed above the wire feed rollers. Amplitude and frequency of the wave motion can be varied over the total amplitude of oscillation and the speed of the mechanical oscillator or, also, over the wire feed speed. As the contact tube remains stationary, very narrow gaps with widths from 9 to 12 mm with plate thicknesses of up to 300 mm can be welded.
br-er6-05e.cdr
ISF 2002
Figure 6.5 Figure 6.5 shows the macro section of a GMA narrow gap welded joint with plates (thickness: 300 mm) which has been produced by the mechanical oscillator method in approx. 70 passes. A highly regular weld build-up and an almost straight fusion line with an extremely narrow heat affected zone can be noticed. Thanks to the correct setting of the oscillation parameters and the precise, centred torch manipulation no
2 3 4 5 rotation method 1 spiral wire method 1
2 3 4
13 - 14
1 - wire reel 2 - drive rollers 3 - mechanism for nozzle rotation 4 - inert gas shroud 5 - shielding gas nozzle 6 - wire guiding tube
br-er 6-06e.cdr
1 - wire reel 2 - wire mechanism for wire deformation 3 - drive rollers 4 - wire feed nozzle and shielding gas supply 5 - contact piece
weave-bead
Figure 6.6
9 - 12
spite
of
the
low
76
nique is the high crystal restructuring rate in the weld metal and in the basemetal adjacent to the fusion line an advantage that gains good toughness properties.
Two narrow-gap welding variations with a rotating arc movement are shown in Figure 6.6. When the rotation method is applied, the arc movement is produced by an eccentrically protruding wire electrode (1.2 mm) from a contact tube nozzle which is rotating with frequencies between 100 and 150 Hz. When the wave wire method is used, the 1.2 mm solid wire is being spiralwise deformed. This happens before it enters the rotating 3 roll wire feed device. With a turning speed of 120 to 150 revs per minute the welding wire is deformed. The effect of this is such that after leaving the contact piece the deformed wire creates a spiral diameter of 2.5 to 3.0 mm in the gap adequate enough to weld plates with thicknesses of up to 200 mm at gap widths between 9 and 12 mm with a good sidewall fusion.
Figure 6.7 explains two GMA narrow gap welding methods which are operated without forced arc movement, where a reliable sidewall fusion is obtained either by the wire deflection through constant deformation (tandem wire method) or by forced wire deflection with the contact tip (twin-wire method). In both cases, two wire electrodes with thicknesses between 0.8 and 1.2 mm are used. When the tandem technique is applied, these electrodes are transported to the two weld heads which are arranged inside the gap in tandem and which are indeFigure pendently selectable.
When
tandem method 1
350
the
twin-
twin-wire method 1
2 3 4 5 6 2 3 4 5
1 - wire reel 2 - deflection rollers 3 - drive rollers 4 - inert gas shroud 5 - shielding gas nozzle 6 - wire feed nozzle and contact tip
br-er 6-07e.cdr
9 - 12
1 - wire reel 2 - drive rollers 3 - inert gas shroud 4 - wire feed nozzle and shielding gas supply 5 - contact tips
15 - 18
sword-
Figure 6.7
77
strip electrode
SO stick out
s
weld edges at small spaces behind each other (approx. 8 mm) and mol-
s a x h w
gap width electrode deflection distance of strip tip to flank twisting angle bead hight bead width
ten.
so
x
In place of the SA narrow gap welding methods, mentioned in Figure 6.2, the method with a lengthwise po-
twin-wire electrode
vw
vw a H z s h w p
weld speed electrode deflection stick out distance torch - flank gap width bead height bead width penetration depth
more detail, Figure 6.8. SA narrow gap welding with strip electrode is carried out in the multipass layer technique, where the strip electrode is deflected at an angle of approx. 5
H
z
a h
p
br-er6-08e.cdr
Figure 6.8
10 7
cored-strip electrodes with widths between 10 and 25 mm are used. The gap width is, depending on the number of passes per layer, between 20 and 25 mm. SA twin-wire welding is, in general, carried out using two elec6 double-U butt weld SA-DU weld preparation (8UP DIN 8551) 8
16
trodes (1.2 to 1.6 mm) where one electrode is deflected towards one weld edge and the other towards the bottom of the groove or towards the opposite weld edge. Either a single pass layer or a two pass layer technique are applied. Dependent on the electrode di-
double-U butt weld GMA-DU weld preparation (Indexno. 2.7.7 DIN EN 29692)
br-er6-09e.cdr
Figure 6.9
78
ameter and also on the type of welding powder, is the gap width between 12 and 13 mm.
Figure 6.9 shows a comparison of groove shapes in relation to plate thickness for SA welding (DIN 8551 part 4) with those for GMA welding (EN 29692) and the unstandardised, mainly used, narrow gap welding. Depending on the plate thickness, significant differences in the weld crosssectional dimensions occur which may lead to substantial saving of mabr-er6-10e_sw.cdr ISF 2002
terial and energy during welding. For example, when welding thicknesses of 120 mm to 200 mm with the narrow
Figure 6.10 gap welding technique, 66% up to 75% of the weld metal weight are saved, compared to the SA square edge weld.
electrode shielding gas arc +
workpiece
wire guide
The practical application of SA narrow gap welding for the production of a flanged calotte joint for a reactor pressure vessel cover is depicted in Figure 6.10. The inner diameter of the pressure vessel is more than
designation: gas-shielded metal arc welding (GMAW acc. DIN 1910 T.4) position: vertical (width deviations of up to 45) plate thickness: 6 - 30 mm square-butt joint or V weld seam 30 mm double-V weld seam materials: unalloyed, lowalloy and highalloy steels gap width: 8 - 20 mm electrodes: only 1 (flux-cored wire, for slag formation between copper shoe and weld surface) 1.6 - 3.2 mm amperage: 350 - 650 A voltage: 28 - 45 V weld speed: 2 - 12 m/h shielding gas: unalloyed and lowalloy steels CO2 or mixed gas (Ar 60% and 40% Co ) 2 highalloy steels: argon or helium
br-er6-11e.cdr
40,000 mm.
The
total
3,000 tons. The weld depth at the joint was 670 mm, so it had been necesFigure 6.11
Electrogas Welding
79
sary to select a gap width of at least 35 mm and to work in the three pass layer technique.
Electrogas welding (EG) is characterised by a vertical groove which is bound by two water-cooled copper shoes. In the groove, a filler wire electrode which is fed through a copper nozzle, is melted by a shielded arc, Figure 6.11. During this process, are groove edges fused. In relation with the ascending rate of the weld pool volume, the welding nozzle and the copper shoes are pulled upwards. In order to avoid poor fusion at the beginning of the welding, the process has to be started on a run-up plate which closes the bottom end of the groove. The shrinkholes forming at the weld end are transferred onto the run-off plate. If possible, any interruptions of the welding process should be avoided. Suitable power sources are rectifiers with a slightly dropping static characteristic. The electrode has a positive polarity.
The application of electrogas welding for low-alloyed steels is more often than not limited, as the toughness of the heat affected zone with the complex coarse grain formation does not meet sophisticated demands. Long-time exposure to temperatures of more than 1500C and low crystallisation rates are responsible for this. The same applies to the weld metal. For a more wide-spread application of electrogas welding, the High-Speed Electrogas Welding
6. copper shoe 7. water cooling 8. weld seam 9. Run-up plate 1 2 3 4 5 6 7 8 9 1. base metal 2. welding boom 3. filler metal 4. slag pool 5. metal pool
Method has been developed in the ISF. In this process, the gap crosssection is reduced and additional metal powder is added to increase the deposition rate. By the increase of the welding speed, the dwell times of weld-adjacent regions above critical temperatures and thus the brittleness effects are significantly reduced. Figure 6.12
designation: position: plate thickness: gap width: materials: electrodes:
resistance fusion welding vertical (and deviation of up to 45) 30 mm (up to 2,000 mm) 24 - 28 mm unalloyed, lowalloy and highalloy steels 1 or more solid or cored wires 2.0 - 3.2 mm plate thickness range per electrode: fixed 30 - 50 mm oscillated: up to 150 mm amperage: 550 - 800 A per electrode voltage: 35 - 52 V welding speed: 0.5 - 2 m/h slag hight: 30 - 50 mm
br-er6-12e.cdr
Electroslag Welding
80
Figure 6.12 shows the process principle of Electroslag Welding. Heating and melting of the groove faces occurs in a slag bath. The temperature of the slag bath must always exceed the melting temperature of the metal. The Joule effect, produced when the current is transferred through the conducting bath, keeps the slag bath temperature constant. The welding current is fed over one or more endless wire electrodes which melt in the highly heated slag bath. Molten pool and slag bath which both form the weld pool are, sideways retained by the groove faces and, in general, by water-cooled copper shoes which are, with the complete welding unit, and in relation with the welding speed, moved progressively upwards. To avoid the inevitable welding defects at the
~
powder
beginning
of
sion of unmolten powder) and at the end of the welding (shrinkholes, slag
start of welding
br-er6-13e.cdr
welding
end of welding
ISF 2002
Figure 6.13 The electroslag welding process can be divided into four process phases, Figure 6.13. At the beginning of the welding process, in the so-called ignition phase, the arc is ignited for a short period and liquefies the non-conductive welding flux powder into conductive slag. The arc is extinguished as the electrical conductivity of the arc length exceeds that of the conductive slag. When the desired slag bath level is reached, the lower ignition parameters (current and voltage) are, during the so-called Data-Increase-Phase, raised to the values of the stationary welding process. This occurs on the run-up plate. The subsequent actual welding process starts, the process phase. At the end of the weld, the switch-off phase is initiated in the run-off plate. The solidifying slag bath is located on the run-off plate which is subsequently removed.
81
The electroslag welding with consumable feed wire (channel-slot welding) proved to be very useful for shorter welds.
The copper sliding shoes are replaced by fixed Cu cooling bars and the welding nozzle by a steel tube, Figure 6.14. The length of the sheathed steel tube, in general, corresponds with the weld seam length (mainly shorter than 2.500 mm) and the steel tube melts during welding in the ascending slag bath. Dependent on the plate thickness, welding can be carried out with one single or with several wire and strip electrodes. A feature of this process variation is the handiness of the welding device and the easier welding
drive motor welding cable run-off plate workpiece = ~ workpiece wire or strip electrode
Electroslag fusing nozzle process (channel welding) position: vertical plate thickness: 15 mm materials: unalloyed, lowalloy and highalloy steels welding consumables: wire electrodes: 2.5 - 4 mm strip electrodes: 60 x 0.5 mm plate electrodes: 80 x60 up to 10 x 120 mm fusing feed nozzle: 10 - 15 mm welding powder: slag formation with high electrical conductivity
area
preparation.
Also curved seams can be welded with a bent consumable electrode. As the groove width can be significantly when with
fusing feed nozzle workpiece cable workpiece run-up plate copper shoes workpiece
reduced comparing
copper shoes
br-er 6-14e.cdr
conventional processes, and a strip shaped filler material with a consumable guide
Figure 6.14
technological measures
post weld heat treatment decrease of peak temperature and dwell times at high temperatures increase of welding speed reduction of energy per unit length continuous normalisation furnace normalisation increase of deposit rate application of several wire electrodes, metal powder addition decrease of groove volume V, double-V butt joints, multi-pass technique
metallurgical measures
increase of purity application of suitable base and filler metals
piece is used, this welding process is rightly placed under the group of narrow gap weld-
reduction of S-, P-, H2-, N2 and O2 - contents and other unfavourable trace elements
C-content limits Mn, Si, Mo, Cr, Ni, Cu, Nb, V, Zr, Ti
ing techniques.
br-er 6-15e.cdr
Figure 6.15
82
process is also characterised by a large molten pool with a simultaneously - low heating and cooling rate. Due to the low cooling rate good degassing and thus almost porefree hardening of the slag bath is possible. Disadvantageous, however, is the formation of a coarse-grain structure. There are, however, possibilities to improve the weld properties, Figure 6.15.
To avoid postweld heat treatment the electroslag welding process with local continuous normalisation has been developed for plate thicknesses of up to approx. 60 mm, Figure 6.16. The welding temperature in the weld region drops below the Ar1temperature and is subsequently heated to the normalising temperature (>Ac3). The specially designed torches follow the
temperature C
1. filler wire 2. copper shoes 3. slag pool 4. metal pool 5. water cooling 6. slag layer 7. weld seam 8. distance plate 9. postheating torch 10. side plate 11. heat treated zone
copper
1 2 3 4 5 6 7 8 9 10
along
seam. By reason of the residual heat in the workpiece the necessary perature can tembe
reached in a short
br-er 6-16e.cdr
time.
Figure 6.16
In order to circumvent an expensive postheat weld treatment which is often unrealistic for use on-site, the electroslag high-speed welding process with multilayer technique has been developed. Similar to electrogas welding, the weld cross-section is reduced and, by application of a twin-wire electrode in tandem arrangement and addition of metal powder, the weld speed is increased, as in contrast to the conventional technique. In the heat affected zones toughness values are determined which correspond with those of the unaffected base metal. The slag bath and the molten pool of the first layer are retained by a sliding shoe, Figure 6.17. The weld preparation is a double-V butt weld with a gap of approx. 15 mm, so the carried along sliding shoe seals the slag and the metal bath. Plate preparation is, as in conventional elec-
83
troslag welding, exclusively done by flame cutting. Thus, the advantage of easier weld preparation can be main1 2 3 4 9 5 6 7 8 4
1 magnetic screening 2 metal powder addition 3 tandem electrode 4 water cooling 5 copper shoe (water cooled) 6 slag pool 7 molten pool 8 solidified slag 9 welding powder addition 10 weld seam
tained.
For larger plate thicknesses (70 to 100 mm), the three passes layer technique has been developed.
width: 20 to 30 mm; gap width: approx. 15 mm) two sliding shoes which are adjusted to the weld groove are used. The first layer is welded using the conventional technique, with one wire electrode without metal powder addition.
10
br-er6-17e.cdr
ISF 2002
Figure 6.17
12
When welding the outer passes flat Cu shoes are again used, Figure 6.18. Three wire electrodes, arranged in a triangular formation, are used. Thus, one electrode is positioned close to the root and on the plate outer sides two electrodes in parallel arrangement are fed into the bath. The single as well as the parallel wire electrodes are fed with different metal powder quantities which as outcome permit a welding speed 5 times higher than the
11
1 2 3 4 9 5 6 7 8 4
1 magnetic screening 2 metal powder supply 3 three-wire electrode 4 water cooling 5 copper shoe (water cooled) 6 slag pool 7 molten pool 8 solidified slag 9 welding powder supply 10 weld seam 11 first pass 12 second pass
ISF 2002
10
speed of the single layer conventional technique and also leads to strong increase of toughness in all zones of the welded joint. Figure 6.18
br-er6-18e.cdr
84
If wall thicknesses of more than 100 mm are to be welded, several twin-wire electrodes with metal powder addition have to be used to reach deposition rates of approx. 200 kg/h. The electroslag welding process is limited by the possible crack formation in the centre of the weld metal. Reasons for this are the concentration of elements such as sulphur and phosphor in the weld centre as well as too fast a cooling of the molten pool in the proximity of the weld seam edges.
7. Pressure Welding
2003
7. Pressure Welding
85
Figure 7.1 shows a survey of the pressure welding processes for joining of metals in accordance with
welding
DIN 1910.
pressure welding
fusion welding
In
friction welding
gas
pressure
projection welding
ure 7.2.
Figure 7.1 Both methods use efficient gas torches to bring the workpiece ends up to the welding temperature. When the welding temperature is reached, both joining members are butt-welded by the application of axial force when a flash edge forms. The long preheating time leads to the formation of a coarse-grained structure in the joining area, therefore the welds are of low toughness values. As the process is operated mainsindependently and
initial state: gap closed initial state: gap opened
(for special cases)
and
main
application
Figure 7.2
7. Pressure Welding
86
In pressure butt welding, the input of the necessary welding heat is produced by resistance heating. The necessary axial force is applied by copper clamping jaws which also receive the current supply, Figure 7.3. The current circuit is closed over the abutting surfaces of the two joining members where, by the increased interface resistance, the highest heat generation is obtained. After the welding temperature which is lower than the melting temperature of the weld metal is reached, upset pressure is applied and the current circuit is opened. This produces a thick flash-free upset seam which is typical for this method. In order to guarantee the uniform heating of the abutting
before upset force has been applied upset force
faces, they must be conformable in their cross-sectional sizes and shapes with each other and they must have
parallel faces.
_ ~
As no molten metal
br-er7-03e.cdr
develops
during
pressure upset butt welding, the joining surfaces must be free from contaminations and from Figure 7.3
a+b b 2 a
oxides. Suitable for welding are unalloyed and low-alloy steels. The welding of aluminium and copper material is, because of the tendency towards oxidation and good Figure 7.4
primary side secondary side
copper shoe
welding transformer
br-er7-04e.cdr
conductivity, possi-
7. Pressure Welding
87
ble only up to a point. For the most part, smaller cross-sections with surfaces of up to 100 mm are welded. Areas of applications are chain manufacturing and also extensions of wires in a wire drawing shop.
A flash butt welding equipment is, in its principal structure, similar to the pressure butt welding device, Figure 7.4.
While in pressure upset butt welding the joining members are always
br-er7-05e.cdr ISF 2002
strongly pressed together, in flash butt welding only fusing contact is made during the heating phase. During the welding process, the workpiece ends
Figure 7.5
are progressively advanced towards each other until they make contact at several points and the current circuit is over these contact bridges closed. As the local current density at these points is high, the heating also develops very fast. The metal is liquified and, partly, evaporated. The metal vapour pressure causes the liquified metal to be thrown out of the gap. At the same time, the metal vapour is generating a shielding gas atmosphere; that is to say, with the exception of pipe welds, welding can be carried out without the use of shielding gas. The constant creation and destruction of the contact bridges causes the abutting faces to burn, starting from the contact points, with heavy spray-type ejection. Along with the occurrence of material loss, the parts are progressively advanced towards each other again. New contact bridges are created again and again. When the entire abutting face is uniformly fused, the two workpiece ends are, through a high axial force, abruptly pressed together and the welding current is switched off. This way, a narrow, sharp and, in contrast to friction welding, irregular weld edge is produced during the upsetting progress, which, if necessary, can be easy mechanically removed while the weld is still warm, Figure 7.5.
7. Pressure Welding
88
In flash butt welding, a fundamental distinction is made between two different working techniques. During hot flash butt welding a preheating operation precedes the actual flashing process, Figure 7.6. The preceding resistance heating is carried out by reversing, i.e., by the changing short-circuiting and pressing of the joining surfaces and by the mechanical separation in the reversed motion. When the joint ends are sufficiently heated, is the flashing process is initialised automatically and the following process is similar to cold flash butt welding. In contrast to cold flash butt welding, the advantage of hot flash butt welding is that, on one hand, sections of 20 times the size can be welded with the same machine efficiency and, on the other hand, a smaller temperature drop and with that a lower cooling rate in the workpiece can be obtained. This is of importance, especially with steels which because of their chemical composition have a tendency to harden. The cooling rate may also be reduced by conductive reheating inside the machine. A smooth and clean surface is not necessary with hot flash butt welding. If the abutting faces differ greatly from the desired
upset force upset travel flashing travel
preheating
flashing
flashing
amperage
time
time
Flashing Travel, Upset Travel, Upset Force and Welding Current in Timely Order
The welding area of the structure of a flash butt weld shows a zone which is reduced in carbon and other alloying elements, Figure 7.7. Moreover, all flash butt welded joints have a pronounced coarse grain zone, whereby the toughness properties of the welded joint are lower than of the base metal. By the impact normalizing effect in the machine successive to the actual welding process, can the toughness properties be considerably increased. By one or several current impulses the weld
7. Pressure Welding
89
temperatures are increased by up to approximately 50 over the austeniting temperature of the metal. Steels, aluminium, nickel and copper alloys can be welded economically with the flash butt welding process. Supported
heat affected zone
10 mm
material: C60 E
only very low residual stresses occur. It is, therefore, posweld coarse grain zone fine grain zone soft-annealing zone base metal
Profiles of all kind are butt welded with this method. The method is used
n
for large-scale manufacture and with components of equal dimensions. The weldable cross-sections may reach dimensions of up to 120,000 mm. Areas of application are the welding of
n
F1 friction force
rails, the manufacture of car axles, wheel rims and shafts, the welding of chain links and also the manufacture of tools and endless strips for pipe
F2 upset force
production. Friction welding is a pressure welding method where the necessary heat
br-er7-08e.cdr
ISF 2002
for joining is produced by mechanical friction. The friction is, as a rule, generated by a relative motion between a
Figure 7.8
7. Pressure Welding
90
rotating and a stationary workpiece while axial force is being applied at the same time, Figure 7.8.
After the joint surfaces are adequately heated, the relative motion is discontinued and the friction force is increased to upsetting force. An even, lip-shaped bead is produced which may be removed in the welding machine by an additional accessory unit. The bead is often considered as the first quality criterion.
br-er7-09e_sw.cdr
friction welding process. In most cases this method is used for rota-
Figure 7.9 tion-symmetrical parts. It is, nowabrake clamping tool workpiece clamping tool pressure element for axial pressure
days, also possible to accurately join rectangular sections. and polygonal cross-
clutch
The most common variant of friction welding is friction welding with a continuous drive and friction welding with a flywheel drive, Figure 7.10. In friction welding with continuous drive, the clamped-on part to be joined is kept at a constant nominal speed by a drive, while the workpiece in the sliding chuck is pressed with a defined friction force. The nominal speed is maintained until the demanded temFigure 7.10
br-er7-10e.cdr
flywheel
clamping tool
ISF 2002
7. Pressure Welding
91
perature profile has been achieved. Then the motor is declutched and the relative motion is neutralised by external braking. In general, the friction force is raised to upsetting force after the rotation movement has been discontinued. During flywheel friction welding, the inertia mass is raised to nominal speed, the drive motor is declutched and the stationary workpiece is, with a defined axial force, pressed against the rotating workpiece. Welding is finished when the total kinetic energy - stored in the flywheel has been consumed by the friction processes. This is the so-called self-breaking effect of the system. The method is used when, based on metallurgical processes, extremely short welding times may be realised. Further process variants are radial friction welding, orbital friction welding, oscillation friction welding and friction However, process surfacing. these variants
number of revolutions
friction welding time 1...100s braking 0,1...0,5s friction welding time 0,125...2s
900... 5400min
-1
time
40...280
-2
40...280 Nmm
-2
Nmm
-2
Comparison of the Welding Processes for Conventional and Flywheel Friction Welding
Figure 7.11
Figure 7.11 depicts the variation in time of the most important process parameters in friction welding with continuous drive and flywheel friction welding. The occuring moments maxima may be interpreted as follows: The first maximum, at the start of the frictional contact, is explained by the formation of local fusion zones and their shearing off in the lower temperature range. The torque decreases as a result of the risen temperature - which again is a consequence of the increased plasticity - and of the lowered deformation resistance. The second maximum is generated during the braking phase which precedes the spindle standstill. The second maximum is explained by the increased deformation resistance at dropping temperatures. The temperature drop in the joining zone is ex-
7. Pressure Welding
92
plained by the lowered energy input due to the rotation-speed decrease and also by the augmented radial displacement of the highly heated material into the weld upset.
In friction welding
number of revolutions upset force
friction force
welding
upsetting
ments, Fig. 7.12. Figure 7.12 In this case, the rotation-energy which has been stored in the drive motor, the spindle and also in the clamping chuck, may be totally or partially converted by selfbreaking. Here, the breaking phase matches the breaking phase in flywheel welding. The use of this process variant allows the welding structures to influence each other in a positive way when many welding tasks are to be carried out. Moreover, the torque range may
a) b)
c)
d)
e)
f)
br-er7-13e.cdr
Figure 7.13
7. Pressure Welding
93
chine is in its structure similar to a turning lathe, however, for the transmission of the high axial forces, the machine structure must be considerably more rigid. Basically, there are three types of friction welding: a) friction welding with a rotating workpiece and a translational motion of the other workpiece; b) friction welding with rotation and translational motion of one workpiece facing a stationary other workpiece, c) rotation and translation of two workpieces against a stationary intermediate piece. The remaining variations, shown in Figure 7.13, also find applications when both workpieces have to rotate in opposite direction to each other. For example, when a low diameter and, in connection with this, the low relative speeds demand the necessary heat quantity.
A survey of possible joint shapes achievable with friction welding is given in Figure 7.14. The specimen preparation of the joining members should, if possible, be carried out in a way that the heat input and the heat dissipation is equal for both members. Depending on the combination of materials can this provision facilitate the joining task considerably. abutting The
1. a)round stock with round stock
before welding
after welding
4. pipe with pipe
before welding
after welding
1..2
d 0,6D
surfaces
2. a)round stock with round stock (different cross-sections, partially machined) b) round stock with round stock (different cross-sections, bevelled) 3. round stock with pipe
D
0,75d
7. round material with plate, without preparation 8. pipe with plate, without preparation
(1/6)d
d=0,75D
dimensions.
br-er7-14e.cdr
ISF 2002
A simple saw cut is, for many applications, sufficient. Figure 7.14
The method of heat generation causes a comparatively low joining temperature lower than the melting temperature of the metals. This is the main reason why friction welding is the suitable method for metals and material combinations which are difficult to weld. It is also possible to weld material combinations (e.g. Cu/Al or Al/steel) which cannot be joined using other welding processes otherwise only with increased expenditure. Figure 7.15 shows a survey of possible material combinations. Many
7. Pressure Welding
94
combinations have, however, not yet been tested on their suitability to friction welding. Metallurgical reasons which may reduce the friction weldability are:
non-metal inclusions, 2. formation of low-melting or intermetallic phases, 3. embrittlement by gas absorption (as a rule, the costly, inert gas shielding can be dispensed with, even when welding titanium), 4. softening of hardened or pre-
aluminium aluminium alloys aluminium, sintered lead cast iron (GGG, GT) hard metal, sintered cobalt copper magnesium brass molybdenum nickel nickel alloys niobium silver steel, unalloyed steel, alloyed steel, high alloyed steel, austentic cast steel free cutting steel stellite tantalum titanium vanadium tungsten cirkon
friction weldable restricted friction weldable not friction weldable not tested
cipitataly-hardened and
materials
a cooling rate.
By the adjustment of the welding paFigure 7.15 rameters in respect toweld joints, can
in most cases joints with good mechano-technological properties be obtained. The secondary structure along the friction-welded joint is depicted in Figure 7.16. An extremely grained (forge finestructure structure)
metal: S235JR
10 mm
weld metal
10 m
Figure 7.16
7. Pressure Welding
95
Figure 7.17 shows a comparison between a flash butt-welded and a frictionwelded cardan shaft. The two welds are distinguished by the size of their heat affected zone and the development of the weld upset. While in friction welding a regular, lip-shaped upset is produced, the weld flash formation in flash butt welding is narrower and sharper and also considerably more irregular. Besides, the heat affected zone during friction welding is substantially smaller than during flash butt welding. Friction welding machines are fully mechanized and may well be integrated into production lines. Loading and unloading equipment, turning attachments for the preparation of the abutting surfaces and for upset removal and also storage units for complete welding programs make these machines well adaptable to automation. The machines may furthermore be
equipped with parameter supervisory systems. During welding are parameters: welding path, pressure, rotational speed, and time are governed by the desired value/actual value compariflash butt welding
son. This allows an indirect quality control. A further complement to the retension of parameters is the torque control, however this method is costly and it cannot be used for all applications because of its structural dimen-
friction welding
sions.
br-er7-17e.cdr
ISF 2002
Friction welding machines are mainly used in the series production and industrial mass production.
Figure 7.17 Nevertheless, these machines are also always applied when metals and material combinations which are difficult to weld have to be joined in a reliable and costeffective way. With the machines that are presently used in Germany, it is possible to weld massive workpieces in the diameter range of 0.6 up to 250 mm For steel pipes, the maximum weldable diameter is at present approximately 900 mm, the wall thicknesses are approx. 6 mm.
7. Pressure Welding
96
3
1,2 joint ring material combination: Cf53/ Ck45
br-er7-18e_sw.cdr
4
3 loading device 4 unloading grippers
1 cardan shaft, AIZn 4,5 Mg 1 2 cardan shaft, retracted tube 3 cardan shaft, flattening test specimen 4 crown wheel, 16MnCr5/ 42Cr4 5 bimetal valve, X45CrSi9-3/ NiCr20 TiAl
ISF 2002
br-er7-19e_sw.cdr
ISF 2002
Figure 7.18
Figure 7.19
Figures 7.18 to 7.20 show a selection of examples for the application of friction welding.
Figure 7.21 shows a comparison of the cost expenditure for the manufacture of a cardan shaft, carried out by forging and by friction welding, respectively.
1 pump shaft 2 shaft C22E/ E295 3 press cylinder S185/9S 20K 4 hydraulic cylinder S235J3G2/ C60E or S235JR/ C15 5 cylinder case S235JR/ S355J2G3 6 piston rod 42Cr4 7 connecting rod 100Cr6/ S235JR 8 stud S235J2G3/ X5CrNi18-10 9 knotter hook 15CrNi6
br-er7-20e_sw.cdr ISF 2002
It shows that the application of the friction welding method may save approx. 20% of the production costs. This comparison is, however, not an universally valid statement as for each component a profitability evaluation must be carried out
Figure 7.20
7. Pressure Welding
97
separately. The comparison is just to show that, in many applications, considerable savings can be made if the matter of the joining technology by friction welding could be circulated to a wider audience of design and production engineers.
tant and
advantages disadvan-
tages of friction
940 mm
welding in comforged piece motor shaft 20,friction-welded piece flange,forged material costs shaft 30 und 40 mm 2x friction welds incl. upset removal 7,50 4,25 3,14,75
parison
with
the
20,br-er7-21e.cdr
butt welding.
Cost Comparison of Forging/ Friction Welding in a Case of a Cardan Shaft
Figure 7.21 Pressure welding with magnetically impelled arc, Magnetarc Welding, is an arc pressure welding method for the joining of closed structural tubular shapes, Figure 7.23. The weldable wall thickness range is between 0.7 and 5 mm, the weldable diameter range between 5 and 300 mm. In Magnetarc Welding an arc burns between the joining surfaces and is rotated by external magnetic forces. This is achieved by a magnet coil system that produces a magnetic field.
Advantages and disadvantages of friction welding in comparison with the competitive flash butt welding
advantages: - clean and well controllable bulging - low heat influence on joining members - better control of heat input - low phase seperation phenomena in the bond zone - hot forming causes permanent recovery and recrystallisation processes in the welding area thus forming a very fine-grained structure with good toughness and strength properties (forged structure) - low susceptibility to defects, extremely good reproducibility within a wide parameter range - frequently shorter welding times - more choice in the selection of weldable materials and material combinations disadvantages: - torque-safe clamping necessary - machine-determined smaller maximum weldable cross-sections - susceptibility to non-metal inclusions - high expenditure requested because of high manufacturing tolerances - high capital investment for the machine
br-er7-22e.cdr ISF 2002
The combined action of this magnetic field and the arcs own magnetic field
Figure 7.22
7. Pressure Welding
98
effects a tangential force to act upon the arc. The rotation of the arc heats and melts the joint surfaces. After an adequate heating operation, the two workpiece members are pressed and fused together. A regular weld upset develops which is normally not removed. The welding operation takes place under shielding gas (mainly CO2).
1. starting position
a) both workpieces are brought into contact b) welding current and magnetic field are switched on
2. starting of welding
a) both workpieces are seperated until a defined gap width is reached (retracting movement) - the arc ignites
3. heating
a) the arc rotates b) the joint surfaces are melting
4. completion of welding
a) both workpieces are broght into contact again and upset b) welding current and magnetic field are switched off
br-er7-23e.cdr
Figure 7.23 motion behaviour and the regularity of the weld bead are therefore improved.
materials
free cutting steel steel, lowalloyed steel, unalloyed malleable
ISF 2002
electrical conductivity and melting behaviour. Figure 7.24 gives a survey of the material combinations which are nowadays already weldable under industrial conditions. As reason is the symmetric heat input, the subsequent upsetting of the liquid phase and the cooling off under pressure. The cracking sensitivity of the
br-er7-24e.cdr
not tested
welds is, in general, relatively low. This has a positive effect, particulary Figure 7.24
cast steel
7. Pressure Welding
99
when steels with a high carbon content or machining steels are welded. The joining faces of the workpieces must be free from contamination, such as rust or scale. To obtain a defect-free weld, normally a simple saw cut is a sufficient preparation of the abutting surfaces. If special
friction welding.
Applications for Magnetic Arc Welding
Figure 7.25
Figure 7.27 shows a summary of the most important advantages and disadvantages of this method in comparison with the competitive methods of friction welding and flash butt welding.
In friction-stir welding a cylindrical, mandrel-like tool carries out rotating self-movements between two plates which are knocked and clamped onto a fixed backing. The resulting friction heat softens the base metal, although Figure 7.26
br-er7-26e_sw.cdr ISF 2002
7. Pressure Welding
100
the melting point is not reached. The plastified material is displaced by the
Advantages and disadvantages of magnetic arc welding in comparison with flash butt welding and/ or friction welding
mandrel and transported behind the tool where a longitudinal seam devel-
advantages: - lower energy demands - material savings through lower loss of length - better dimensional accuracy in joining especially for small wall thicknesses - in comparison with friction welding less moving parts (only axial movement of one joining member during upsetting) - no restrictions to the free clamping length - smaller and more regular welding edge - no spatter formation disadvantages: - suitable for small wall thicknesses only (maximum wall thickness: 4 - 5 mm) - welding parameters must be kept within narrow limits - only magnetizing steels are weldable without any difficulties
ops.
The advantages of this method which is mainly used for welding of aluminium alloys is the low thermal stress of the component which allows joining with a minimum of distortion and shrinkage. Welding fumes do not develop and the addition of filler metal or shielding gases is not required.
ISF 2002
br-er7-27e.cdr
Figure 7.27
fixed backing
contoured pin
br-er7-28e.cdr
Friction-Stir Welding
Figure 7.28
8. Resistance Spot Welding, Resistance Projection Welding and Resistance Seam Welding
2003
101
Figure 8.1 shows an extract from the classification of the welding methods according to DIN 1910 with a detailed account of the conductive resistance pressure welding.
In the case of resistance pressure welding, the heating occurs at the welding point as a consequence of Joule resistance heating caused by current flow through an electrical conductor, Figure 8.2. In spot and projection welding, the plates to be welded in overlap. Current supply is carried out through spherical or flat electrodes, respectively. In roller seam welding, two driven roller electrodes are applied. The plates to be
welding
welded
are
pressure welding
friction welding
projection welding
br-er8-01e.cdr
ISF 2002
tity Qeff is instrumental in the formation of the weld nugget. Qeff is com-
Figure 8.1
spot welding l workpieces overlap l electrode l weld nugget roller seam welding l workpiece usually in general overlap l driven roller electrode l spot rows (stitch weld, roller spots) projection welding l workpieces with elevations (concentration of electicity) l workpieces overlap l pad electrode l several joints in a single weld l weld nugget joint
posed of the input heat minus the dissipation heat. The heat loss the arises heat
2 3
from
4 loaded area
5 projection
Figure 8.2
102
The resistance during resistance heating is composed of the contact resistances on the two plates and of their material resistance. The reduction of the electrode force down to 90% inelectrode force effective heat total heat input current (time dependence) heat losses losses into the electrodes losses into the sheet metal losses by heat radiation total resistance material resistance contact resistance Qeff = Qinput - Q1l
t=tS
input
Q4
rate
by
Q4
Q2
Q3
Qeff
Q3
current 90%
down
to
Q4 Q2
Q4
decreases
Qinput = C
t=0
the heat rate to 80% and a welding time reduction to 90% decreases
The time progression of the resistance is shown in Figure 8.4. The contact resistance is composed of the interface resistances between the electrode and the plate (electrode/plate) and between the plates (plate/plate). The resistance height is greatly dependent on the applied electrode force. The higher this force is set, the larger are the conductive
theoretical contact area 100% metallic conduction contact
mOhm
crossat the
sections
resistance
low electrode force high resistance high electrode force low resistance proportion after first milliseconds welding time
periods
tact
surfaces,
10
welding time
br-er8-04e.cdr
Figure 8.4
103
With the formation of the weld nugget the interface resistances between the plates disappear. During the progress of the weld the material resistance increases from a low value (surrounding temperatures) to a maximum value above the melting temperature.
Figure 8.5 shows diagrammatically the different resistances during the spot welding process with acting electrode force, but without welding current. Weld nugget formation must therefore start in the joining zone because of the existing high contact resistance there.
electrode force resistance rate
R3 R6 R5 R7 R4 0 100 200
tance welding. The coolant is normally water. In the cooling tube, the cooling
R [Ohm]
base. The diagram shows the temperature distribution in the electrodes and
cooling tube
Figure 8.5
in the plates. The maximum temperature is reached in the centre of the weld nugget and
6-8
cooling drill-hole
slope
10 - 20 2-5
Electrode Cooling
Figure 8.6
104
Sequence of a resistance spot welding process, Figure 8.7: 1 ->2 Lowering of the top electrode 2->3 Application of the adjusted electrode force Set-up time tpre, sequence 3->4 Switching-on of the adjusted welding current for the period of the welding time tw. Formation of the weld nugget in the joining zone of both workpieces. An example shows the macrosection of a weld nugget after the welding time has ended. 4->5 Maintaining the electrode force for the period of the set post-weld holding time th. 5->6 Switching-off the force generating system and lifting the electrodes off the workpiece.
The functions of the set-up time and the post-weld holding time are listed in Figure 8.8. Dependent on the welding task different force and current programs can be set in the welding machines, Figure 8.9. In the top the simplest possible welding program sequence is shown: The application of the electrode force, the set-up time sequence tpre, the switching-on of the welding current and the sequence of the weldFel Iw
electrode force Fel
set-up time - compressing the workpiece - build-up of electrode force to preset value - setting-up of reproducible resistance before welding - electrode resting after bounce - preventing resting of electrode on workpiece under electricle voltage
welding current Iw
time t
tpre
tw
th
top electrode
postweld-holding time - holding time of workpiece during cooling of molten metal - prevention of pore formation in the welding nugget - prevention of lifting the electrode under voltage The postweld-holding time has influence on the weld point hardening within certain limits.
br-er8-08e.cdr ISF 2002
Figure 8.7
Figure 8.8
105
ing time tw, the sequence of the postwelding current electrode force
Fel 5 Iw
tpre = pre-weld time tw = welding time th = holding time tpres = pressure time
weld holding time th and the switchingoff of the force generating system. The diagram in the centre is almost identi-
tpre
tw tpres
th
time
welding current
electrode force
Fel 5 7 Iw 8
Merely in the welding current range, welding is carried out using an adjustable current rise (7) and current decay
time 1 - initial force 2 - welding pressure force 3 - post pressure force 4 - preheating current 5 - welding current 6 - postheating current 7 - ascending current 8 - descending current time
(8). The diagram below depicts a more sophisticated current program. In addition, welding is carried out with a variable electrode force (2) and with preheating (4) and post-heating current (6). Dependent on the control system,
welding current
electrode force
1 2 5 7 4 6
3 Iw 8
Fel
br-er8-09e.cdr
stance, the electrode path, the resistance progress, the welding current or the welding voltage.
11 12 4 9 10 6 2 3
Figure 8.10 shows the principle structure of a resistance spot welding machine. The main components are: the machine frame, the welding transformer with secondary lines, the electrode pressure system and the control system. This principle design applies to spot, projection and roller seam welding machines. Differences are to be found merely in the type of electrode fittings and in the electrode shapes.
7 5 8
1 electrode force cylinder 2 pneumatic equipment 3 machine tool frame 4 welding transformer 5 power control unit 6 current conductor 7 lower arm 8 foot switch 9 top arm 10 electrical power supply cable 11 water cooled electrode holder 12 electrode
br-er8-10e.cdr ISF 2002
Figure 8.10
106
Figure 8.11 depicts the possible process variations of resistance spot welding. These are distinguished by the number of plates to be welded and by the arrangement of the electrodes or, respectively, of the transformers. It has to be noted that with a corresponding arrangement also plates can be welded where one of the two plates has a non-conductive surface (as, for example, plastics). Figure 8.12 shows the current types which are normally used for resistance welding. Alternating current has the simplest structure (Figure 8.13) and is most price effective, unavoidable are, however, the disadvantages of current zeros and weld nugget cooling. In relation to the average current values, peak
~ ~ ~
loads
occur
and,
two-sided single-shear single-spot welding
~
+
+
two-sided duplex spot welding
br-er8-11e.cdr
The structural design of a d.c. supply unit more and, more is, however, Figur 8.11
alternating current
20
complicated
[kA]
10 5 0 0.00 0.02 0.04 0.07 0.09 0.11 0.13 0.16 -5 -10 -15 -20
[kA]
therefore, expensive
15
current
current
[kA]
machines
current
operate with a 50
current
[kA]
Figur 8.12
welding time
[s]
0.14 0.16
welding time
[s]
welding time
[s]
ISF 2002
Current Types
107
units
single-phase alternating current static-inverter direct current
(1
period).
frequency
tech-
period
and a more precise control of the welding current is posFigure 8.13 sible.
In order to realise higher currents and shorter welding times, the impulse capacitor resistance welding technique is applied. The rectified primary current is stored in capacitors and, through a high-voltage transformer, converted to high welding currents. The advantages of this technique are low heat input and high reproducibility. Because of the high energy density, materials with good conductivity can be welded and also multiple-projection welds can be carried out. A disadvantage of this method is, apart from the high equipment costs, the difficult regulation of the welding current.
electrode caps
form E form F form G form A
electrodes
form B form C form D
Electrodes for spot resistance welding have the property of transferring the electrode force and the welding current. They are wearing parts and, therefore, easily replaceable. Depending on the shape and type of electrode, solid electrodes or electrode Figure 8.14
electrode holders
br-er8-14e.cdr ISF 2002
108
recycled. Figure 8.14 depicts various types of electrodes, electrode caps and holders.
Key
Key
Cu - ETP Cu Cdl Cu Crl Cu Crl Zr Cu CO2 Be Cu Ni2 Si Cu Ni1 P Cu Be2 Co Ni Cu Ag6 CuAl10NiFe5Ni5
10 11 12 B 13 14 15
added alloying elements influence the red hardness, the tempering resistance, the conductivity, the fusion temperature, the electrode alloying tendency, and, finally, the machinability of the electrode material. When beryllium is used as an alloying ele-
br-er8-15e.cdr
Electrode Materials
Figure 8.15
poor good
Already during the design phase of the components to be welded, importance must be attached to a good accessibility of the welding point. Moreover, the electrode force which is imperative to the process must be applied in a way that no damage is done to the workpiece. In the ideal case, the welding point is accessible from the top and from below, Figure 8.16.
br-er8-16e.cdr ISF 2002
Figure 8.16
109
poor
good
In order to avoid the displacement of the electrodes, the electrode working surface must be flat. Also during the design phase space must be provided for an adequately large clearing zone around the working point, in order to guarantee the unimpeded electrode approach to the working point, Figure 8.17.
Dependent on the joining job, the process variation, or the resistance welding method, a so-called shunt current/effect may be noticed. This
br-er8-17e.cdr ISF 2002
Figure 8.17 weld nugget; under certain circumstances it might even prevent a reliable welding process. In the example, shown in Figure 8.18, the shunt current leads to undesired fusing contacts and, because of the lacking electrode force at this point, also to damages to the plate surface.
current path copper
spot welding
A
If unsuitable welding parameters have been set, weld spatter formation may occur, Figure 8.19. Liquid molten metal forms on the plate surface or in
the joining zone. The reasons for this kind of process disturbance are, for
br-er8-18e.cdr
Shunting
Figure 8.18
110
with regard to the set welding current or welding time, too high an energy input with regard to the plate thickness or too small an edge distance of the welding point.
fig. 1
Reason here is high welding current, (fig. 1) or too-small edge distance (fig. 2)
fig. 2
Figure 8.20 shows a list of a large number of possible disturbances in resistance spot welding. Welding current changes are caused by: shunt, electrode wear, cable wear, mains voltage variations, secondary
impedance.
Welding Spatter
Figure 8.19
Different welding conditions are caused by welding machine wear, different heat dissipation. Non-uniform conditions by alterations to components are: different plate thicknesses, plate
welding current changes
shunt connection wear of electrodes wear of cable mains voltage fluctuation secondary electrical impedance
Electrode
wear
plate
diversion heat
fluctuations
plate thickness quality of plates number of plates plate surface edge distance
Figure 8.20
welding equipment
111
materials. A list of the various materials is shown in Figure 8.21. The alloying elements which are used for steels have a varying influence on the suitability for resistance spot welding. The values which are indicated in the table are valid only when the stated element is the sole alloying constituent of the steel material.
aluminium
weldable materials
br-er8-21e.cdr
Figure 8.22 shows a comparison between resistance spot and resistance projection welding. The fun-
Weldable Materials
damental difference between the two methods lies in the definition of the
Figure 8.21
The differences between both methods are illustrated in Figure 8.23. The short life of the electrodes used for resistance spot welding is explained by the higher thermal load and the larger pressing area caused by the smaller electrode contact areas. The term electrode life stands for the number of welds that can be carried out with one pair of
follow-up distance
before welding
after welding
electrodes
without
elektrode
projection
weld.
Figure 8.22
112
spot welding
elektrodes: diameter tip face electrode life up to 20 mm convex less
projection welding
elektrodes
projections
one
several
Circular
follow-up distance
small
big
weld nut
crossed wires
Longitudinal
cut
wire-plate
Annular
br-er8-23e.cdr ISF 2002
br-er8-24e.cdr
pushed
bolt-pipe
Figure 8.23
Figure 8.24
die
Depending on the demands on the joint strength or on the projection rigidity, different projection shapes are applied. These are annular, circular or longitudi-
d1
mould plate
d1
nal projections. The welding projections are, according to their size, adapted to
embossed projection
ring projection
the used plate thickness and may, therefore, appear as different types in the
die
l
mould plate
b
plate
workpiece: embossed projections, solid projections and natural projections. The survey is shown in Figure 8.24.
longitudinal projection
br-er8-25e.cdr ISF 2002
In Figure 8.25 the production of embossed projections in different shapes is shown. The shape is embossed onto the
Figure 8.25
113
plate surface by appropriate die plates, dies and, if necessary, counter dies.
alternating current distribution intensity of current increases from the center to the outer area caused by current displacement
force distribution of a C-frame projection press welder during bending of machine tool frame
direct current distribution intensity of current decreases from the center to the outer area caused by the longer current path
br-er8-26e.cdr ISF 2002 br-er8-27e.cdr
force distribution of a C-frame projection press welder with non-parallel positioning tables
ISF 2002
Figure 8.26
Figure 8.27
Various problems occur in projection welding caused by the welding of several joints in a single working cycle. Due to different current paths - when using direct current - and a current displacement - when
lap joint lap joint with wire electrode lap joint with foil squash seam weld butt weld with foil
using alternating current -, welding nuggets with differing are probefore welding
qualities
duced when no preventive remedies are taken, Figure 8.26. A varying force distribr-er8-28e.cdr ISF 2002
after welding
114
application using projection welding are depicted. In this example, the shapes are of the embossed type.
process variations of roller seam welding. Seam welding is actually a continuous spot welding process, but with the application of roller elec-
trodes. In contrast to resistance spot welding the electrodes remain in contact and turn continuously after the first weld spot has been produced. At
Figure 8.29 be produced again current flow is initiated. Dependent on the electrode feed rate and on the
welding current frequency, spot welds or seal welds with overlapping nuggets are weld probr-er8-30e_sw.cdr ISF 2002
Figure 8.30
2003
115
The application of highly accelerated electrons as a tool for material processing in the fusion, drilling and welding process
high voltage supply
and also for surface treatment has been known since the Fifties. Ever
beam generation
since, the electron beam welding process has been developed from the laboratory stage for particular applications. In this cases, this materials could not have been joined by any industrially applied high-production
viewing optics
focussing coil
defelction coil
joining method.
workpiece
workpiece handling
working chamber
chamber door
br-er9-01e.cdr ISF 2002
and forming and working chamber. These components may also have separate vacuum systems, Figure 9.1.
Figure 9.1
A tungsten cathode which has been heated under vacuum emits electrons by thermal emission. The heating of the tungsten cathode may be carried out directly - by filament current - or indirectly - as, for example, by coiled filaments. The electrons are accelerated by high voltage between the cathode and the pierced anode. A modulating electrode, the so-called Wehnelt cylinder, which is positioned between anode and cathode, regulates the electron flow. Dependent on the height of the cut-off voltage beFigure 9.2
br-er9-02e_f.cdr
control panel
ISF 2002
116
tween the cathode and the modulating electrode, is a barrier field which may pass only a certain quantity of electrons. This happens during an electron excess in front of the cathode where it culminates in form of an electron cloud. Due to its particular shape which can be compared to a concave mirror as used in light optic, the Wehnelt cylinder also effects, besides the beam current adjustment, the electrostatic focussing of the electron beam. The electron beam which diverges after having passed the pierced anode, however, obtains the power density which is necessary for welding only after having passed the adjacent alignment and focussing system. One or several electromagnetic focussing lenses bundle the beam onto the workpiece inside the vacuum chamber. A deflection coil assists in maintaining the electron beam oscillating motion. An additional stigmator coil may help to correct aberrations of the lenses. A viewing optic or a video system allows the exact positioning of the electron beam onto the weld groove.
The core piece of the electron beam welding machine is the electron beam gun where the electron beam is generated under high vacuum. The tightly focussed electron beam diverges rapidly under atmospheric pressure caused by scattering and ionisation development with air. As it would, here, loose power density and efficiency, the welding process is, as a rule, carried out under medium or high vacuum. The necessary vacuum is generated in separate vacuum pumps for working chamber and beam gun. A shut-off valve which is positioned between electron gun and working chamber serves to maintain the gun vacuum while the working chamber is flooded. In universal machines, Figure 9.2, the
back-scattered electrons x-ray
heat conduction
z
br-er9-03e.cdr ISF 2002
117
the vacuum chamber onto the workpiece platform. A distinction is made between electron beam machines with vertical and horizontal beam manipulation systems.
The energy conversion in the workpiece, which is schematically shown in Figure 9.3, indicates that the kinetic energy of the highly accelerated electrons is, at the operational point, not only converted into the heat necessary for welding, but is also released by heat radiation and heat dissipation. Furthermore, a part of the incident electrons (primary electrons) is subject to backscatter and by secondary processes the secondary electrons are emitted from the workpiece thus generating X-rays.
The impact of the electrons, which are tightly focussed into a corpuscular beam, onto the workpiece surface stops the electrons; their penetration depth into the workpiece is very low, just a few m. Most of the kinetic energy is released in the form of heat. The high energy density at the impact point causes the metal to evaporate thus allowing the following electrons a deeper penetration. This finally leads to a metal pour cavity which is vapour cavity which is surrounded by a shell of fluid metal, covering the entire weld surrounded by a
shell of fluid metal, covering the entire weld depth, Figure 9.4. This deep-weld effect allows nowadays depths penetration into steel
a) b) c) d)
The diameter of the cavity corresponds approximately with the beam diameter. By a relative motion in the direction of the weld groove between workpiece and electron beam the cavity penetrates through the material, Figure 9.5. At the front side of the cavity new material is molten which, to some extent, evaporates, but for the most part
118
flows around the cavity and rapidly solidifies at the backside. In order to maintain the welding cavity open, the vapour pressure must press the molten metal round the vapour column against the cavity walls, by counteracting its hydrostatic pressure and the surface tension.
However, this equilibrium of forces is unstable. The transient pressure and temperature conditions inside the cavity as well as their respective, momentary diameters are subject to dynamic changes. Under the influence of the resulting, dynamically changing geometry of the vapour cavity and
motion of the molten metal groove melting pool welding direction keyhole molten zone vapour capillary electron beam groove front side
F1
rameters,
F2 F3 F1
F1 : force resulting from vapour pressure F2 : force resulting from surface tension F3 : force resulting from hydrostatic pressure
br-er9-05e.cdr
Condition in Capillary
Figure 9.5
molten backside of the vapour cavity to strong and irregular changes in shape (case II). Pressure variations interfere
with
the
regular
II
workpiece movement
III
br-er9-06e.cdr
ISF 2002
Figure 9.6
119 molten
end crater
g len
width of seam
weld penetration depth
weld thickness
Nahtdicke
root reinforcement
root weld
ISF 2002
Basic Definitions
collapse
sporadi-
solidification (case Ill). The angle (case I) increases with the rising weld speed and this is defined as a turbulent process. Flaws such as a constantly open vapour cavity and subsequent continuous weld solidification could be avoided by selection of jobsuitable welding parameter combination and in particular of beam oscillation characteristics, it has to be seen to a constantly of the molten metal, in order to avoid the above-mentioned defects. Customary beam oscillation types are: circular, sine, double parabola or triangular functions.
by pressure: l high vacuum machine l fine vacuum machine l atmospheric machine (NV-EB welding) by accelerating voltage: l high voltage machine (UB=150 kV) l low voltage machine (UB=60 kV)
deep-weld effect and, with that, the possibility to join in a single working cycle with high weld speed and low heat input quantity. A comparison with the submerged-arc and the gas metalarc welding processes illustrates the depth-to-width ratio which is obtain-
Figure 9.8
120
able with the electron beam technology, Figure 9.7. Electron beam welding of thick plates offers thereby decisive advantages. With modern equipment, wall thicknesses of up to 300 mm with length-to-width ratios of up to 50 : 1 and consisting of low and high-alloy materials can be welded fast and precisely in one pass and without adding any filler metal. A corresponding quantification shows the advantage in regard of the applied filler metal and of the primary energy demand. Compared with the gas-shielded narrow gap welding process, the production time can be reduced by the factor of approx. 20 to 50.
Numerous specific advantages speak in favour of the increased application of this high productivity process in the manufacturing practice, Figure 9.8. Pointing to series production, the high profitability of this process is dominant. This process depends on highly energetic efficiency together with a sparing use of resources during fabrication. Considering the above-mentioned advantages, there are also disadvantages which emerge from the process. These are, in particular, the high cooling rate, the high equipment costs and the size of the chamber, Figure 9.9.
< 5 x 10-4 mbar < 1 x 10 mbar
-6
In accordance with DIN 32511 (terms for methods and equipment applied in electron and laser beam welding), the specific designations, shown in Figure 9.10, have been standardised for electron beam welding. Figure 9.9
br-er9-09e_f.cdr
Electron beam units are not only distinguished by their working vacuum quality or the unit concept but also by the acceleration voltage level, Figure 9.11. The latter exerts a considerable influence onto the obtainable welding results. With the increasing acceleration voltage, the achievable weld depth and the depth-to-width ratio of the weld
121 geometry are also increasing. A disadvantage of the increasing accelerating voltage is, however, the exponential increase of X-rays and, also, the
< 5 x 10 mbar
-2
the size of the workpiece to be welded and the size of the chamber volume, high-voltage beam generators (150 200 kV) with powers of up to 200 kW are applied in industrial production, while the low-voltage technology
(max. 60 kV) is a good alternative for smaller units and weld thicknesses.
br-er9-10e_f.cdr
Figure 9.10 to the lower acceleration voltage, a separate complete lead covering of the unit is not necessary. While during the beam generation, the vacuum (p = 10-5 mbar) for the insulation of the beam generation compartment and the prevention of cathode oxidation is imperative, the possible working pressures inside the vacuum chamber vary between a high vacuum (p = 10-4 mbar) and atmospheric pressure. A collision of the electrodes with the residual gas molecules and the scattering of the electron beam which is connected to this is, naturally, lowest in high vacuum. Figure 9.11
br-er9-11e_f.cdr
< 1x 10 mbar
-4
~ 10 mbar ~ 1 mbar
-1
122 The beam diameter is minimal in high vacuum and the beam power density
in vacuum
r r r
thin and thick plate welding (0,1 mm bis 300 mm) extremely narrow seams (t:b = 50:1) low overall heat input => low distortion => welding of completely processed components high welding speed possible no shielding gas required high process and plant efficiency material dependence, often the only welding method
is maximum in high vacuum, Figure 9.12. The reasons for the application of a high vacuum unit are, among others, special demands on the weld (narrow, deep welds with a minimum energy input) or the choice of the materials to be welded (materials with a high oxygen affinity). The application of the electron beam welding process also entails advantages as far as the structural design of the components is concerned.
r r r r
at atmosphere
r r r
very high welding velocity good gap bridging no problems with reflection during energy entry into workpiece
br-er9-12e.cdr
ISF 2002
Advantages of EBW
With a low risk of oxidation and reduced demands on the welds, the so-
in vacuum r r r r r r r electrical conductivity of materials is required high cooling rates => hardening => cracks high precision of seam preparation beam may be deflected by magnetism X-ray formation size of workpiece limited by chamber size high investment
cause of economic considerations, as, for instance, the reduction of cycle times, Figure 9.13. Areas of application are in the automotive industry (pistons, valves, torque converters, gear parts) and also in the metal-working industry (fittings, gauge heads, accumulators).
at atmosphere r r r X-ray formation limited sheet thickness (max. 10 mm) high investment
Under extreme demands on the welding time, reduced requirements to the weld geometry, distortion and in case of full material compatibility with air or shielding gas, out-of-vacuum welding
br-er9-13e.cdr
ISF 2002
Disadvantages of EBW
Figure 9.13
123 units are applied, Figure 9.14. Their advantages are the continuous welding time and/or short cycle times. Areas of application are in the metalworking industry (precision tubes, bimetal strips) and in the automotive industry (converters, pinion cages, socket joints and module holders).
A further distinction criterion is the adjustment of the vacuum chambers to the different joining tasks. Universal machines are characterised by their simply designed working chamber,
br-er9-14e_f.cdr
Figure 9.14 generators. Here, several workpieces can be welded in subsequence during an evacuation cycle. The largest, presently existing working chamber has a volume of 265 m.
Clock system machines, in contrast, are equipped with several small vacuum chambers which are adapted to the workpiece shape and they are, therefore, characterised by short
evacuation times, Figure 9.16. Just immediately before the welding starts, is the beam gun coupled to the vacuum chamber which has been evacuated during the preceding evacuation Figure 9.15
br-er9-15e_f.cdr
124 cycle, while, at the same time, the next vacuum chamber may be flooded and charged/loaded.
Conveyor machines allow the continuous production of welded joints, as, for example, bimetal semi finished products such as saw blades or thermostatic bimetals, Figure 9.17. In the main chamber of these units is a gradually raising pressure system as partial vacsemi-finished material
uum pre and post activated, to serve as a vacuum lock. Systems which are operating with a
endproduct
br-er9-16e_f.cdr
Figure 9.16 uum by decreasing the pumping volume. In the local vacuum systems, with the use of suitable sealing, is the vacuum produced only in the welding area. In mobile vacuum systems welding is carried out in a small vacuum chamber which is restricted to the welding area but is travelling along the welded seam. In this case, a sufficient sealing between workpiece and vacuum chamber is more difficult. With these types of machine design, electron beam welding may be carried out with components which, due to their sizes, can not be loaded into a stationary vacuum chamber (e.g. ves-
butt weld
a)
b)
lap weld
br-er9-17e.cdr
ISF 2002
Figure 9.17
125 sel skins, components for particle accelerators and nuclear fusion plants).
In general the workpiece is moved during electron beam welding, while the beam remains stationary and is directed onto the workpiece in the horizontal or the vertical position. Depending on the control systems of the working table and similar to conventional welding are different welding positions possible. The weld type preferred in electron beam welding is the plain butt weld. Frequently, also cenbr-er9-18e.cdr ISF 2002
tring allowance for centralising tasks and machining is made. For the execution of axial welds, slightly over-
Figure 9.18 sized parts (press fit) should be selected during weld preparation, as a transverse shrinkage sets in at the beginning of the weld and may lead to a considerable increase of the gap width in the opposite groove area. In some cases also T-welds may be carried out; the T-joint with a plain butt weld should, however, be chosen only when the demands on the strength of the joints are low, Figure 9.18. As the beam spread is large under atmosphere, odd seam formations have to be considered during Non-Vacuum Electron Beam Welding, Figure 9.19. Figure 9.19
br-er9-19e.cdr ISF 2002
UP (conventional)
UP (narrow gap) 650 A 30 V
2
UP (conventional) 510 A 28 V
4905 mm 81
150
5966 mm 143
2
0 7,7 kg/h
6410 kJ
3
23 kg 5 kg/h
12810 kJ
3
54 kg 13 kg/h
29310 kJ
3
66 kg 9 kg/h
37710 kJ
3
27 min
4 h 35 min
4 h 11 min
7 h 20 min
126
In order to receive uniform and reproducible results with electron beam welding, an exact knowledge about the beam geometry is necessary and a prerequisite for:
- tests on the interactions between beam and substance - applicability of welding parameters to other welding machines - development of beam generation systems.
The objective of many tests is therefore the exact measurement of the beam and the investigation of the effects of different beam geometries on the welding result. For the exact measurement of the electron beam, a microprocessor-controlled measuring system has been developed in the ISF. The electron beam is linearly scanned at a high speed by means of a point probe, which, with a diameter of 20 m is much smaller than the beam diameter in the focus, Figure 9.20. When the electron beam is deflected through the aperture diaphragm located inside the sensor, the electrons flowing through the diaphragm are picked up by a Faraday shield and
industrial areas: l automotive industries l aircraft and space industries l mechanical engineering l tool construction l nuclear power industries l power plants l fine mechanics and electrical industries l job shop material: l almost all steels l aluminium and its alloys l magnesium alloys l copper and its alloys l titanium l tungsten l gold l material combinations (e.g. Cu-steel, bronze-steel) l ceramics (electrically conductive)
br-er9-20e.cdr
diverted over a precision resistor. The time progression of the signal, intercepted at the resistor, corresponds with the intensity distribution of the electron beam in the scanning path. In order to receive an overall picture of the power density distribution inside the electron beam, the beam is line scanned over the slit sensor (60 lines). An evaluation program creates a perspective view of the power density distribution in the beam and also a two-dimensional representation of
Figure 9.20
hole sensor
hole with aperture diaphragm Faraday cup (20 m) cross section of the beam
be seen clearly that the cathode had not been heated up sufficiently.
Therefore, the electrons are sucked off directly from the cathode surface during saturation and unsaturated
slit sensor
beams, which may lead to impaired welding results, develop. During the
slit with Faraday cup cross section of the beam
space charge mode of a generator, the electron cloud is sufficiently large, i.e., there are always enough elec-
voltage
trons which can be sucked off. In the ideal case, the developed power denbeam deflection
br-er9-21e.cdr
Figure 9.21 The electron signals are used for the automatic seam tracking. These may be either primary or secondary electrons or passing-through current or the developing X-rays. When backscattered primary electrons are used, the electron beam is scanned transversely to the groove. A computer may determine the position of the groove relative to the beam by the signals from the reflected electrons. In correspondence with the deflection the beam is guided by electromagnetic deflection coils or by moving the working table. This kind of seam tracking system may be used either on-line or off-line. Figure 9.22
br-er9-22e_f.cdr
FILENAME: R I N G S T R Accel. voltage: 150 kV Beam current: 600 mA Prefocus current: 700 mA Main focus current: 1500 mA Cath. heat current: 500 mm Max. Density: 26,456 kW/mm2 2 Ref. Density: 26,456 kW/mm
ISF 2002
128
The broad variation range of the weldable materials and also material thicknesses offer this joining method a large range of application, Figure 9.22. Besides the fine and micro welding carried out by the electronics industry where in particular the low heat input and the precisely programmable control is of importance, electron beam welding is also particularly suited for the joining of large cross-sections.
2003
1917 postulate of stimulated emission by Einstein 1950 work out of physical basics and realisation of a maser (Microwave Amplification by Stimulated Emission of Radiation) by Towens, Prokhorov, Basov 1954 construction of the first maser 1960 construction of the first ruby laser (Light Amplification by Stimulated Emission of Radiation) 1961 manufacturing of the first HeNe lasers and Nd: glass lasers 1962 development of the first semiconductor lasers 1964 nobel price for Towens, Prokhorov and Basov for their works in the field of masers construction of the first Nd:YAG solid state lasers and CO2 gas lasers 1966 established laser emission on organic dyes since increased application of CO2 and solid state laser 1970 technologies in industry 1975 first applications of laser beam cutting in sheet fabrication industry 1983 introduction into the market of 1-kW-CO2 lasers 1984 first applications of laser beam welding in industrial serial production
br-er10-01e.cdr
,,Light Amplification by Stimulated Emission of Radiation. The laser is the further development of the maser (m=microwave), Figure 10.1. Al-
mechanical fundamentals have already been postulated by Einstein in the beginning of the 20th century, the first laser - a ruby laser - was not implemented until 1960 in the Hughes Research Laboratories. Until then numerous tests on materials had to be carried out in order to gain a more
precise knowledge about the atomic structure. The following years had
Figure 10.1
opment of the laser technology. Already since the beginning of the Seventies and, increasingly since the Eighties when the first high-performance lasers were available, CO2 and solid state lasers have been used for production metal working. The number of the annual sales of laser beam sources has constantly increased in the
9
10
1.5
0.5
1986
1988
1990
1992
1994
1996
1998
2000
130
in 1994 are shown in Figure 10.3. The main application areas of the laser in the field of production metal working are joining and cutting jobs.
beam
sources
financial
con-
Figure 10.5 shows the characteristic properties of the laser beam. By reason of the induced or stimu40 kW 20 10 laser power 5 4 3 2 1 diode laser 1970 1975 1980 1985 1990 1995 2000 0 Nd:YAG CO2
br-er10-04e.cdr
Figure 10.4
131
Inside the resonator, Figure 10.6, the laser-active medium (gas molecules, ions) is excited to a higher energy level (pumping) by energy input (electrical gas discharge, flash lamps).
During retreat to a lower level, the energy is released in the form of a light quantum (photon). The wave length depends on the energy difference between both excited states and is thus a characteristic for the respective laser-active medium. A distinction is made between spontaneous and induced transition. While the spontaneous emission is non-directional and in coherent (e.g. in fluorescent tubes) is a laser beam generated by induced when with a a
0,46"
E2
light bulb
Laser
induced emission
emission particle
higher energy level is hit by a photon. The resulting photon has the same properties quency, (fredirection,
br-er10-05e.cdr
E1
exited state
ground state
0,9 4"
polychromatic
(multiple wave length)
monochromatic coherent
(in phase)
incoherent
(not in phase)
large divergence
small divergence
ISF 2002
ing photon (coherence). In order to maintain the ratio of the desired induced emission I spontaneous emission as high as possible,
active laser medium laser beam partially reflecting mirror R < 100%
ISF 2002
Figure 10.5
resonator energy source
the upper energy level must be constantly overfully reflecting mirror R = 100%
br-er10-06e.cdr
energy source
132
inversion. As result, a stationary light wave is formed between the mirrors of the resonator (one of which is semi-reflecting) causing parts of the excited laser-active medium to emit light. In the field of production metal working, and particularly in welding, especially CO2 and Nd:YAG lasers are applied for their high power outputs. At present, the development of diode lasers is so far advanced that their sporadic use in the field of material processing is also possible. The industrial standard powers for CO2 lasers are, nowadays, approximately 5 - 20 kW, lasers with powers of up to 40 kW are available. In the field of solid state lasers average output powers of up to 4 kW are nowadays obtainable.
eV
transmission of vibration energy
where
the
energy
0,4
resonator is filled
0,3 0,288 eV 0,2 0,1
thrust of first type 1 E = 0,002 eV 001
LASER = 10,6 m
with a N2-C02-He gas mixture, pumping is carried out over the vibrational
ISF 2002
0
br-er10-07e.cdr
000
N2
CO2
Figure 10.7
which again, with thrusts of the secradio frequency high voltage exitaion
vibrational
cooling water
cooling water
laser gas: CO2: 5 l/h He: 100 l/h N2 : 45 l/h gas circulation pump
br-er10-08e.cdr
vakuum pump
emit
Figure 10.8
laser beam
The efficiency of up
mirror
(partially reflecting)
gas discharge
br-er10-09e.cdr
cooling water
relatively
Figure 10.9
is the heat which must be discharged from the resonator. This is achieved by means of the constant gas mixture circulation and cooling by heat exchangers. In dependence of the type of gas transport, laser systems are classified into longitudinal-flow and transverse-flow laser systems, Figures 10.8 and 10.9.
With transverse-flow laser systems of a compact design can the multiple folding ability of the beam reach higher output powers than those achievable with longitudinal-flow systems, the beam quality, however, is worse. In d.c.-excited systems (high voltage), the
f2,57" d0
unfocussed beam
focussed beam
molecules
0<K<1
causes
Figure 10.10
134
also entails a contamination of the laser gas. Parts of the gas mixture must be therefore exchanged permanently. In high-frequency a.c.-excited systems the electrodes are positioned outside the gas discharge tube where the electrical energy is capacitively coupled. High electrode lives and high achievable pulse frequencies characterise kind of this
resonator partially reflecting mirror absorber shutter beam divergence mirror
excitation
end mirror
LASER
ever necessary. Figure 10.11 The intensity distribution is not constant across the laser beam. The intensity distribution in the case of the ideal beam is described by TEM modes (transversal electronic-magnetic). In the Gaussian or basic mode TEM00 is the peak energy in the centre of the beam weakening towards its periphery, similar to the Gaussian normal distribution. In practice, the quality of a laser beam is, in accordance with DIN EN 11146, distinguished by the nonbeam (or
dimensional quality
factor
10.10.
distance of
field a
divergence
CO2 Laser Beam Welding Station
Figure 10.12
beam may be
objec-
taively.
Figure 10.13 The CO2 laser beam is guided from the resonator over a beam reflection mirror system to one or several processing stations, Figures 10.11 and 10.12. The low divergence allows long transmission paths. At the processing station is the beam, with the help of the focussing optics, formed according to the working task. The relative motion between beam and workpiece may be realised in different ways: moving workpiece, fixed optics moving (flying) optics moving workpiece and moving optics (two handling facilities).
laser,
beam
contaminations.
As
Figure 10.14
136
sipated only over the holders, there is a risk of deformation (alteration of the focal length) or destruction through thermal overloading.
In the case of solid state laser, the normally cylindrical rod serves only the purpose to pick up the laser-active ions (in the case of the Nd:YAG laser with yttriumaluminium-garnet crystals dosed with Nd3+ ions), Figure 10.14. The excitation is, for the most part, carried out using flash or arc lamps, which for the optimal utilisation of the excitation energy are arranged as a double ellipsoid; the rod is positioned in their common focal point. The achieved efficiency is below 4%. In the meantime, also diode-pumped solid
possibility to guide the solid-state laser beam over flexible fibre optics makes these systems destined for the robot
br-er10-15e_f.cdr ISF 2002
application, whereas the CO2 laser application is restricted, as its Figure 10.15
Nd:YAG Laser Beam Welding Station
necessary complex mirror systems may cause losses, 10.15. Some types of optical fibres allow, with fibre diameters of 1 mm bending radii
br-er10-16e_f.cdr ISF 2002
radiation Figure
of up to 100 mm.
Diode Laser
With
137
switches a multiple utilisation of the solid state laser source is possible; with beam splitters (mostly
0,30 Nd:YAGlaser CO2laser
0,25
neous
welding
at
0,20
absorption A
Cu Al Ag Stahl
0,15 Fe 0,10
projection impaired
is
the beam
0
br-er10-17e.cdr
0,1
0,2 0,3
0,5 0,8 1
wave lenght
8 10
m 20
The semiconductor or diode lasers are characterised by their mechanical robustness, high efficiency and compact design, Figure 10.16. High performance diode lasers allow the welding of metals, although no deep penetration effect is achieved. In material processing they are therefore particularly suitable for welding thin sheets.
Energy input into the workpiece is carried out over the absorption of the laser beam. The absorption coefficient is, apart from the surface quality, also dependent on the wave length
1010
en 106 erg y
shock hardening
4
W/cm
10 power density
10
10
10
glaz e
0
drilling
10
10
remelting coating
welding cutting
105
steel
10
which is exposed
har den ing
0
103 10
-8
10
-6
br-er10-18e.cdr
10 acting time
-4
-2
10
10
138
tion, Figure 10.17. As copper is a highly reflective metal with also a good heat conductivity, it is frequently used as mirror material.
Intensity
heat conducting welding deep penetration welding metal vapour blowing away laser-induced plasma soldified weld metal keyhole (vapour-/ plasma cavity) molten pool soldified weld metal
adjust-
molten pool
variation working
speed
make the laser a flexible and contactless tool, Figbr-er10-19e.cdr ISF 2002
ure
Principle of Laser Beam Welding
10.18.
The
Figure 10.19
mand high intensities in the focal point, which means the distance between focussing optics and workpiece surface must be maintained
reflection R penetration depth t 100 % 60 40 20 0
within close tolerances. At the same time, highest accuracy and quality demands are set on all machine components (handling, optics, resonator, beam manipulation, etc.).
Steel materials with treated surfaces reflect the laser beam to a degree of up to 95%, Figures 10.19 - 10.22.
mm 2 1 0 5 10
When metals are welded with a lowintensity laser beam (I 105 W/cm2), just the workpiece surfaces and/or edges are melted and thus thermalbr-er10-20e.cdr
10 laser intesity I
W/cm
10
ISF 2002
Figure 10.20
1010
steel rF = 100 m = 10.6 m
penetration effect is possible. Above the threshold intensity value (I 106 W/cm2) a phase transition occurs and laser-induced plasma develops. The
W/cm
10 laser intensity I
and the vapour density, absorbs an increased quantity of radiation. A vapour cavity forms and allows the laser beam to penetrate deep into the material (energy input deep beyond
A = 0,1
10
A=1 10
-10
10
-8
10
-6
radiation time t
br-er10-21e.cdr ISF 2002
called the deep penetration effect. The cavity which is moved though the joining zone and is prevented to close due to the vapour pressure is sur-
Figure 10.21
ten metal. The residual material vaporises and condenses either on the cavity side walls or flows off in an ionised form. With suitable parameter selection, an almost complete energy input into the workpiece can be obtained.
However, pendence
in of
dethe
Transformation of electromagnetic energy into thermal energy within nm range at the surface of the work piece by stimulation of atoms to resonant oscillations
"normal" absorption: "abormal" absorption:
electron density in the plasma and of the radiated beam intensity, may plasma from
detach
the workpiece surface and screen off the working zone. The plasma is
140
high degree that only a fraction of the beam radiation reaches the workpiece. This is the reason why, in laser beam welding, gases are applied for plasma control. The gases ionisation potential should be as high as possible, since also the formation of shielding gas plasmas is possible
beam energy
0-2,5% 2,5-12,5% 85-95% diagnostics beam transmission focussing system
Only a part of the beam energy from the resonator is used up for the actual
work piece
ca. 5%
reflection
welding process, Figure 10.23. Another part is absorbed by the optics in the beam manipulation system, another
ca. 10%
transmission
ca. 40%
part is lost by reflection or transmission (beam penetration through the vapour cavity). Other parts flow over thermal
ca. 30%
10-15%
recombination
fusion energy
br-er10-23e.cdr
Figure 10.23 Penetration depths in dependence of the beam power and welding speed which are
work piece
- minimum thermal stress - little distortion - completely processed components - welding at positions difficult to access - different materials weldable - expensive edge preparation - exact positioning required - danger of increased hardness - danger of cracks - Al, Cu difficult to weld - expensive beam transmission and forming - power losses at optical devices - laser radiation protection required - high investment cost - low efficency (CO2-Laser: < 20%, Nd:YAG: < 5%)
ISF 2002
advantages
- high power density - small beam diameter - high welding speed - non-contact tool - atmosphere welding possible
disadvantages
process
achievable in laser beam welding are depicted in Figure 10.25. Further relevant influential
installation
- short cycle times - operation at several stations possible - installation availability > 90% - well suitable to automatic function
br-er10-24e.cdr
28 mm
0,2% C-steel CO2-laser
(cross flow)
the resonator (beam quality), the focal position and the applied optics (focal length; focus diameter).
penetration depth
20 16 12
10 kW laser power: 15 kW
8 4
8 kW 6 kW 4 kW 1,5 kW
Figure 10.26 shows several joint shapes which are typical for car body production and which can be welded
welding speed
15
X 5 CrNi 18 10 CO2-laser
(axial flow)
penetration depth
mm
5
4 kW
laser power:
2 kW
6 kW
0
br-er10-25e.cdr
1 kW
m/min
welding speed
Penetration Depths
Figure 10.25 The high cooling rate during laser beam welding leads, when transforming steel materials are used, to significantly increased hardness values in comparison with other methods, welding Figure
butt weld fillet weld at overlap joint
Figure 10.26
142
The small beam diameter demands the very precise manipulation and positioning of the workpiece or of
500 HV 0,4
WMA MAZ MAZ
hardness
ration, 10.28.
MAZ MAZ
Figure Otherwise,
as result, lack of
0 distance from the weld centre 12
fusion,
sagged
weld
Figure 10.27
Caused by the high cooling rate and, in connection with this, the insufficient degassing of the molten metal, pore formation may occur during laser beam welding of, in particular, thick plates (very deep welds) or while carrying out welding-in works (insufficient degassing over the root), Figure 10.29. However, too low a weld speed may also cause pore formation when the molten metal picks up gases from the root side. The materials that
edge preparation misalignment
may
be
welded
gap
beam
mispositioning
nium
and
nickel
br-er10-28e.cdr
Figure 10.28
143
high cooling rate, to be considered a critical influential factor where contents of C > 0.22% may be stipulated as the limiting reference value. Aluminium and copper properties cause problems during energy input and process stability. Highly reactive materials demand, also during laser beam welding, sufficient gas shielding beyond the solidification of the weld seam. The sole application of working gases is, as a rule, not adequate.
vw = 0,7 m/min
br-er10-29e.cdr
vw = 0,9 m/min
vw = 1,5 m/min
ISF 2002
Porosity
Figure 10.29
The application of laser beam welding may be extended by process variants. One is laser beam welding with filler wire, Figures 10.30 and 10.31 which offers the following advantages: - influence on the mechanic-technological properties of the weld and fusion zone (e.g. strength, toughness, corrosion, wear resistance) over the metallurgical composition of the filler wire - reduction of the demands on the accuracy of the weld preparation in regard to edge misalignment, edge preparation and beam misalignment, due to larger molten pools - filling of non-ideal, for example, V-shaped groove geometries - a realisation of a defined weld reinforcement on the beam entry and beam exit side.
144
The exact positioning of the filler wire is a prerequisite for a high weld quality and a sufficient dilution of the molten pool through which filler wire of different composition as the base can reach right to the root. Therefore, the use of sensor systems is indispensable for industrial application, Figure 10.32. The sensor systems are to take over the tasks of - process control, - weld quality as surance - beam positioning and joint tracking, respectively.
welding direction
laser beam
laser beam
plasma
plasma
weld metal
work piece
weld metal
work piece
molten pool
keyhole
molten pool
keyhole
Figure 10.30
weld zone
weld zone
br-er10-31e.cdr
Figure 10.31
145
The present state-of-the-art is the further development of systems for industrial applications which until now have been tested in the laboratory. Welding by means of solid state lasers has, in the past, mainly been applied by manufacturers from the fields of precision mechanics and microelectronics. Ever since solid state lasers with higher powers are available on the market, they are applied in the car industry to an ever increasing degree. This is due to their more variable beam manipulation possibilities when comparing with CO2 lasers. The CO2 laser is mostly used by the car industry and
with sensing device; fill factor 120 %
by
their
ancillary
Probe MS1-6C Probe MS1-5A Probe MS1-4C Probe MS1-3A Probe MS1-2B Probe MS1-1C
0.1 mm 0.2 mm 0.3 mm 0.4 mm 0.5 mm 0.6 mm
Probe OS1-6A Probe OS1-5C Probe OS1-4C Probe OS1-3B Probe OS1-2B Probe OS1-1B
without sensing device; wire speed vD = 4 m/min constant
br-er10-32e.cdr
1 mm
steel industry
- pipe production - vehicle superstructures - continuous metal strips - tins
- body-making
(bottom plates, skins)
- engine components
(tappet housings, diesel engine precombustion chambers)
br-er10-33e.cdr
ISF 2002
Figure 10.33
2003
welding according
process to its
dissimilar l hardfacing (wear protection) l cladding (corrosion prevention) l buffering (production of an appropriate-to-the-type-of-duty joint of dissimilar materials)
applications: welding of joints and surfacing. According to DIN 1910 surfacing is the
coating of a workbr-er11-01e.cdr
applied filler material a further classification may be made: deposition repair welding and surfacing for the production of a composite material with certain functions. Surfacing carried out with wear-resistant materials in preference to the base metal material is called hardfacing; but when mainly chemically stable filler materials are used, the method is called cladding. In the case of buffering, surfacing layers are produced which allow the appropriate-to-the-type-of-duty joining of dissimilar materials and/or of materials with differing properties, Figure 11.1.
m
wear caused by very high impact and compressive stress wear by friction (metal against metal) during high impact and compression stress strong sanding or grinding wear very strong wear caused by grinding during low impact stress cold forming tools hot forming tools cavitation wear parts (plastics industry) corrosion temperature stresses
m m
diate layer made from a relatively tough material between two layers with strongly differing thermal expansion coefficients. Figure 11.2 shows different kinds of stresses which demand the surfacing of components. Furthermore surfacing may be used for primary forming as well as for joining by primary forming.
br-er11-02e.cdr
m m m m m
Figure 11.2
147 In case of surfacing - as for all fabrication processes - certain limiting con-
component (material)
ditions have to be observed. For example, hard and wear-resistant weld filler metals cannot be drawn into solid wires. Here, another form has to be
stress compatibility
coating
rods, powder). Process materials, as for example SA welding flux demand a certain welding position which in terms limits the method of welding.
consumable
surfacing method
The coating material must be selected with view to the type of duty and,
br-er11-03e.cdr
Figure 11.3 For all surfacing tasks a large product line of welding filler metals is available. In dependence on the welding method as well as on the selected materials, filler metals in the form of wires, filler wires, strips, cored strips, rods or powder are applied, Figure 11.4. The filler/base metal dilution is rather important, as the desired high-quality properties of the surfacing layer deteriorate with the increasing degree of dilution.
corrosion prevention q ferritic to martensitic chromium steel alloys q soft martensitic chromium-nickel steel alloys q austenitic-ferritic chromium-nickel steel alloys q austenitic chromium-nickel steel alloys wearing protection (armouring) hard facing on q cobalt base q nickel base q iron base
objective to optimise the degree of dilution in order to guarantee a sufficient Figure 11.4
148
adherence of the layer with the minimum metal dissimilation. A planimetric determination of the surfacing and penetration areas will roughly assess the proportion of filler to base metal. When the analysis of base and filler metal is known, a more precise calculation is possible by the determination of the content of a certain element in the surfacing layer as well as in the base
Definition of Dilution
AD = AD= FP FP + FB x 100% base metal surface built up by welding FB
penetration area FP
x 100%
ISF 2002
metal, Figure 11.5. Figure 11.5 Figure 11.6 shows record charts of an electron beam microprobe analysis for the elements nickel and chromium. It is
30 Cr percentages by mass % 20
evident that - after passing a narrow transition zone between base metal and layer the analysis inside the
10
arc welding methods are not only suitable for joining but also for surfacing.
In the case of the strip-electrode submerged-arc surfacing process normally strips (widths: 20 - 120mm)
10
0
br-er11-06e.cdr
are used. These strips allow high clad0 100 200 distance 300 m 500
ding rates. Solid wire electrodes as well as flux-cored strip electrodes are used. The flux-cored strip electrodes contain
Microprobe Analyses
Figure 11.6
149
certain alloying elements. The strip is continuously fed into the process via feed rollers. Current contact is normally carried out via copper contact jaws which in some cases are protected against wear by
metal-arc welding - stick electrode - filler wire arc welding with self-shielded cored wire electrode - filler wire inert gas-shielded arc welding - MIG / MAG - MIG cold wire - filler wire TIG welding - TIG cold wire submerged arc welding - wire electrode - strip electrode
Should the slag developed on top of the welding bead not detach itself, it will have to be removed mechanically in order to avoid slag inclusions during overwelding. The arc wanders along the lower edge of the strip. Thus the strip is melted consistently, Figure 11.8.
+
filler metal
flux support
br-er11-08e.cdr
Figure 11.8
150
Figure 11.9 shows the cladding of a roll barrel. The coating is deposited helically while the workpiece is rotating. The weld head is moved axially over the workpiece.
br-er11-09e.cdr
Figure 11.9
The macro-section and possible weld defects of a strip-electrode submerged-arc surfacing process are depicted in Figure 11.10.
lack of fusion
sagged weld
base metal
gusset
undercuts
br-er11-10e.cdr
Figure 11.10
151
Electroslag surfacing using a strip electrode is similar to strip-electrode SA surfacing, Figure 11.11. The difference is that the weld filler metal is not melted in the arc but in liquefied welding flux the liquid slag as a result of Joule resistance heating. The slag is held by a slight inclination of the
method
work) with normally relatively low deposition rates. The process principle has already been shown when the TIG joint welding process was explained, Figure 11.12. The arc is
arc rod/ wire-shaped filler metal shielding gas nozzle
burning between a gas-backed nonconsumable tungsten electrode and the workpiece. The arc melts the base metal and the wire or rod-shaped weld filler metal which is fed either continuously or intermittently. Thus a fusion welded joint develops between base metal and surfacing bead.
base metal (+ / ~) tungsten electrode (- / ~) surfacing bead
In the case of MIG/MAG surfacing processes the arc burns between a consumable wire electrode and the
br-er11-12e.cdr
ISF 2002
Figure 11.12
higher
+ -
power source
used.
The
wire a
electrode
has
to has
be a
oscillation
feed direction
br-er11-13e.cdr
surfaced negative
polarity,
A further development of the TIG welding process is plasma welding. While the TIG arc develops freely, the plasma welding arc is mechanically and thermally constricted by a water-cooled copper nozzle. Thus the arc obtains a higher energy density.
In the case of plasma arc powder surfacing this constricting nozzle has a positive, the tungsten electrode has a negative polarity, Figure 11.14. Through a pilot arc power supply a non-transferred arc (pilot arc) develops inside the torch. A second, separate power source feeds the transferred arc between electrode and workpiece. The non-transferred arc ionises the centrally gas fed (inert plasma gases,
tungsten electrode filler metal
plasma gas
HIG
UNTA
conveying gas power sources shielding gas pilot arc welding arc surfacing bead
UTA
plasma jet of high energy to emerge from This the nozzle. jet
br-er11-14e.cdr
plasma
workpiece
oscillation
Figure 11.14
153
arc striking ability of the transferred arc gap. The surfacing filler metal powder added by a feeding gas flow is melted in the plasma jet. The partly liquefied weld filler metal meets the by transferred arc molten base metal and forms the surfacing bead. A third gas flow, the shielding gas, protects
section A
ZW
high-temperature zone from the surrounding influence. The applied gases are mainly inert
br-er11-15e.cdr
GW
gases, as, for example, Ar and He and/or Ar-/He mixtures. Figure 11.15
The method is applied for surfacing small and medium-sized parts (car exhaust valves, extruder spirals). Figure 11.15 shows a cross-section of armour plating of a car exhaust valve seat. The fusion line, i.e., the region between surfacing and base metal, is shown enlarged on the right side of Figure 11.15 (blow-up). It shows hardfacing with cobalt which is high-temperature and hot gas corrosion resistant.
plasma gas
=
wires from spool
surfacing bead
~
workpiece
weld pool
br-er11-16e.cdr
154
pendently. The arc between the tips of the two parallel wires is generated through the application of a separate power source. The plasma arc with a length of approx. 20 mm is oscillating (oscillation width between 20 to 50 mm). The two wires are fed in a V-formation at an angle of approx. 30 and melt in the high-temperature region in the trailing zone of the plasma torch.
For surfacing purposes, besides the arc-welding methods, the beam welding methods laser beam and electron beam welding may also be applied. Figure 11.17 shows the process principle of laser surfacing. The powder filler metal is added to the laser beam via a powder nozzle and the powlaser beam shielding gas nozzle powder nozzle direction of the oscillation powder flow surfacing bead shielding gas
Friction
surfacing
Werkstck
br-er11-17e.cdr
for the production of joints which due to Figure 11.17 the different materials are difficult to produce with fusion
electron beam surface layer base material metal foil metal foil feeding
workpiece
welding, 11.18.
Figure
The filler metal is advanced over the workpiece with high pressure and rotation. By the pressure and the relative frictional
ISF 1998
feed direction
ka11-18.cdr
movement
Figure 11.18
155
puts the weld filler end into a pasty condition. The advance motion causes an adherent, spreaded layer on the base metal. This method is not applied frequently and is mainly used for materials which show strong differences in their melting and oxidation behaviours.
A comparison of the different surfacing methods shows that the application fields are limited - dependent on the welding method. A specific method, for example, is the low filler/base metal dilution. These methods are applied where high-quality filler metals are welded. Another criterion for the selection of a surfacing method is the deposition rate. In the case of cladding large surfaces a method with a high deposition rate is chosen, this with regard to profitability.
In thermal spraying the filler metal is melted inside the torch and then, with a high kinetic energy, discharged onto the unfused but preheated workpiece surface.
There is no fusion of base and filler metal but rather adhesive binding and mechanical interlocking of the spray deposit with the base material. These mechanisms are effective only when the workpiece surface is coarse (pre-treatment by sandblasting) and free of oxides. The filler and base materials are metallic and non-metallic. Plastics may be sprayed as well. The utilisation of filler metals in thermal spraying is relatively low. The most important methods of thermal spraying are: plasma arc spraying, flame spraying and arc spraying.
force filler metal rotation advance
base metal
br-er11-18e.cdr
156
the acceleration and atomisation of the filler metal is produced by compressed gas (air).
compressed air
spraying material
workpiece
In contrast to powder flame spraying, is for flame spraying a wire filler metal fed mechanically into the centre cone, melted, at-
flame cone
spray deposit
Figure 11.20. Figure 11.20 In plasma arc spraying an internal, high-energy arc is ignited between the tungsten cathode and the anode, Figure 11.21. This arc ionises the plasma gas (argon, 50 100 l/min). The plasma emerges from the torch with a high kinetic and thermal energy and carries the side-fed powder along with it which then meets the workpiece surface in a semi-fluid state with the necessary kinetic energy. In the case of shape welding, steel shapes with larger dimensions and higher weights are produced from molten weld metal only. In comparison to cast parts this method about more brings essentially favourable
adjustable wire feed device spraying wire
br-er11-20e.cdr
compressed air
spraying jet
gas mixture
spray deposit
toughness
157
for this lies mainly in the high purity and the homogeneity of the steel which is helped by the repeated melting process and the resulting slag reactions. These properties are also put down to the favourable fine-grained structure formation which is achieved by the repeated subsequent thermal treatment with the multi-pass technique. Also in contrast with the shapes produced by forging, the workpieces produced by shape welding show quality advantages, especially in the isotropy and the regularity of their toughness and strength properties as far as larger workpiece thicknesses are conpowder injector
back frame isolation ring gas middle distributor frame anode carrier
cerned. In Europe, due to the lack of expensive equipment, high forging very
copper anode
individual
jet of particles
Therefore,
shape
br-er11-21e.cdr
welding is, for certain applications, a sensible logical nomical technoand ecoFigure 11.22
alternative
shape welding
Figure 11.23 shows an early application which is related to the field of arts.
br-er11-22e.cdr
joining (welding)
ISF 2002
Figure 11.23
158
Baumkuchenmethode
+ several weld heads possible + no interruption during weld head failure - core made of foreign material necessary applications: shafts, large boiler shell rings, flanges
Tpfermethode
+ free rotationally-symmetrical shapes + several weld heads possible + weld head manipulation not necessary + each head capable to weld a specific layer + small diameters possible - component movement must correspond with the contour - number of weld heads limited when smaller diameters are welded applications: spherical caps, pipe bends, braces
Klammeraffe
br-er11-23e.cdr
br-er11-25e.cdr
ISF 2002
Figure 11.24
Figure 11.25
The higher tooling costs in forging make the shape welding method less expensive; this applies to parts with certain increasing complexity. This comparison is, however, related to relatively low numbers of pieces, where the tooling costs per part are accordingly Figure 11.24. higher,
Figure 11.25 shows the principal procedure for the production of typical
phase 7 phase 6 phase 4 phase 2
traction mechanism
joist
turntable
duced
with
the
BaumkuchenmeFigure 11.26
159 thode method: the filler metal is welded by submerged-arc with helical
1. welding of the half-torus 2. stress relief annealing 3. mechanical treatment 4. seperating/ halving 5. folding 6. welding togehter 7. stress relief annealing 8. testing
movement in multiple passes into a tube which has the function of a traction mechanism (for the most part mechanically removed later). This brings about the possibility to produce seamless containers with bottom and flange in one working cycle.
ti tes ng
Elbows are mainly manufactured with the Tpfer method. On the traction mechanism a rotationally symmetrical part with a semicircle cross-section is
br-er11-27e.cdr
Figure 11.27
connection pieces onto pipes. A portable unit which is connected with the pipe welds the connection pipe in a similar manner to the Tpfer method.
shafts
braces
Figure 11.28
pipe bends
11.28.
2003
160
Thermal cutting processes are applied in different fields of mechanical engineering, shipbuilding and
of components and for the preparation of welding edges. The thermal cutting processes
br-er12-01e.cdr
Figure 12.1 Figure 12.2 shows the classification according to the physics of the cutting process: - flame cutting the material is mainly oxidised (burnt) - fusion cutting the material is mainly fused - sublimation cutting the material is mainly evaporat The gas jet and/or evaporation expansion is in all processes responsible for the ejection of molten material or emerging reaction products such as slag.
The different enFlame cutting The material is mainly oxidised;the products are blown out by an oxygen jet. Fusion cutting The material is mainly fused and blown out by a high-speed gas jet. Sublimation cutting The material is mainly evaporated. It is transported out of the cutting groove by the created expansion or by additional gas.
ergy carriers for the thermal cutting are depicted in Figure 12.3: - gas, electrical gas
br-er12-02e.cdr
Electron
beams
Figure 12.2
161
are listed in the DIN-Standard, they produce, however, only very small boreholes. Cutting is impossible.
Figure 12.4 depicts the different methods of thermal cutting with gas according to DIN 8580. These are: - flame cutting - metal powder flame cutting - metal powder fusion cutting - flame planing -oxygen-lance cutting - flame gouging or scarfing
br-er12-03e.cdr
- gas - electrical gas discharge - sparks - arc - plasma - beams - laser beam (light) - electron beam - ion beam
-flame cleaning
Figure 1.3 In flame cutting (principle is depicted in Figure 12.5) the material is brought to the ignition temperature by a heating flame and is then burnt in the oxygen stream. During the process the ignition temperature is maintained on the plate top side by the heating flame and below the plate top
thermal cutting processes using gas:
side
by
thermal and
l
conduction convection.
oxygen cutting
metal powder
fusion cutting
flame planing
l l
flame cleaning
automation and is, also easy to apply on site. Figure 12.6. shows a commerFigure 1.4
br-er12-04e.cdr
162
cial torch which combines a welding with a cutting torch. By means of different nozzle shapes the process may be adapted to varying materials and plate thicknesses. Hand-held or torches
cutting oxygen heating oxygen gas fuel
machine-type are
heating flame
torches
equipped with different cutting nozzles: Standard or block-type nozzles (cutting-oxygen pressure 5 bar) are used for hand-held torches and for
cutting jet
br-er12-05e.cdr
workpiece
The high-speed cutting nozzle (cutting-oxygen pressure 8 bar) allows higher cutting speeds with increased cutting-oxygen pressure. The heavy-duty cutting nozzle (cutting-oxygen pressure 11 bar) is mainly applied for economic cutting with flamecutting machines. A further development of the heavy-duty nozzle is the oxygenshrouded nozzle which allows even faster and more economic cutting of plates within thickness
cutting oxygen heating oxygen gas fuel mixing chamber
certain ranges.
Gas mixing is either carried out in the torch handle, the cutting attachment, the torch
(gas
mixing
Figure 12.6
163
of the nozzle. As the design of cutting torches is not yet subject to standardisation, many types and systems exist on the market.
The selection of a
heating and cutting nozzle nozzle-to-work distance torch cutting jet kerf width
torch
kerf start
or
nozzles
cut thickness
cutting
quality,
cutting le
br-er12-07e.cdr
Figure 12.7
flame-cutting.
In flame cutting, the thermal conductivity of the material must be low enough to constantly maintain the ignition temperature, Figure 12.8. Moreover, the material must neither melt during the oxidation nor form high-melting oxides, as these would produce difficult cutting surfaces. In accordance, only steel or titanium materials fulfill the conditions for oxygen cutting., Figure 12.9
The heating flame has to perform the following tasks: - rapid heating of the material (about 1200C) - substitution of losses due to heat conduction in order to maintain a positive heat balance - preheating of cutting oxygen - stabilisation of the cutting oxygen jet; formation of a cylindrical geometry over a extensive length and protection against nitrogen of the surrounding air
Steel materials with a C-content of up to approx. 0.45% may be flame-cut without preheating, with a C-content of approx. flame-cutting 1.6% is with be-
br-er12-08e.cdr
carried
Function of the Flame During Flame Cutting
out
preheating,
164
C-content demands more heat. Carbon accumulates at the cutting surface, so a very high degree of hardness is to be expected. Should the carbon content exceed 0.45% and should the material not have been subject to prior heat treatment, hardening cracks on the cutting
The material has to fulfill the following requirements: - the ignition temperature has to be lower than the melting temperature - the melting temperature of the oxides has to be lower than the melting temperature of the material itself - the ignition temperature has to be permanently maintained; i. e. the sum of the supplied energy and heat losses due to heat conduction has to result in a positive heat balance
surface
are
ides which impair the slag expulsion and influence the thermal conductiv-
br-er12-09e.cdr
ity.
Figure 12.9
The iron-carbon equilibrium diagram illustrates the carbon content-temperature interrelation, Figure 12.10. As the carbon content increases, the melting temperature is lowered. That means: from a certain carbon content upwards, the ignition temperature is higher than the melting temperature, i.e., this would be the first violation to the basic requirement in flame cutting.
steel
cast iron
liquid
pasty
solid
Liquidus
Solidus
solid
2,0
12. Thermal Cutting ents are therefore only reference values for the evaluation of the flame cuttability of steels, as the cutting quality is substantially deteriorating, as a rule lower tents. Figure 12.11 already alloy with conbr-er12-11e.cdr
165
Maximum allowable contents of alloy-elements: carbon: silicon: manganese: chromium: tungsten: nickel: copper: up to 1,6 % up to 2,5 % with max. 0,2 %C up to 13 % and 1,3 % C up to 1,5 % up to 10 % and 5 % Cr, 0,2 % Ni, 0,8 % C up to 7,0 % and/or up to 35 % with min. 0,3 % C up to 0,7 % not suitable for cutting
nozzles already during the cutting phase a weld preparation may be carried out and certain welding grooves be produced. Figure 12.12 shows torch arrangements for - the square butt weld, - the single V butt weld, - the single V butt weld with root face, - the double V butt weld and the double V butt weld with root face.
It has to be considered that, particularly in cases where flame cutting is applied for weld
square butt weld single-V butt weld single-V butt weld with rootface
cratering: sporadic craterings connected craterings cratering areas adherent slag: slag adhearing to bottom cut edge
in
order
to
cut face defects: kerf constriction or extension angular deviation step at lower edge of the cut excessive depth of cutting grooves
subsequent processes.
br-er12-13e.cdr
Figure
Possible Flame Cutting Defects
12.13
Figure 12.13
defects
flame cutting.
In order to improve the flame-cutting capacity and/or cutting of materials which are normally not to be flame-cut the powder flame cutting process may be applied. Here, in addition to the cutting oxygen, iron powder is blown into the cutting gap. In the flame, the iron powder oxidises very fast and adds further energy to the process. Through the additional energy input the high-melting
oxygen water seperator compressed air acetylene
powder dispenser
167
methods are suited for the weld preparation; material is removed but not
Figure 12.15
Figure 12.16 shows the methods of thermal cutting processes by electrical gas discharge: plasma cutting with non-transferred arc plasma cutting with transferred arc plasma cutting with transferred arc and secondary gas flow plasma cutting with transferred arc and water injection arc air gouging (represented diagrammatically) arc oxygen cutting (represented diagrammatically) In plasma cutting
Thermal cutting processes by electrical gas discharge:
plasma cutting
- with non-transferred arc - with transferred arc -with secondary gas flow -with water injection
carbon electrode
compressed air
cutting oxygen
jet. The nozzle forms the plasma jet only in a restricted way and limits thus the cutting ability of
br-er2-16e.cdr
Figure 12.16
168
Figure 12.17. Characteristic for the plasma cut are the cone-shaped formation of the kerf and the rounded edges in the plasma jet entry zone which were caused by the hot gas shield that envelops the plasma jet. These process-specific disadvantages may be significantly reduced or limited to just one side of the plate (high quality or scrap side), respectively, by the inclination of the torch and/or water addition. With the plasma cutting process, all electrically conductive
plasma gas cooling water electrode
HF R +
power source
nozzle
workpiece
br-er12-17e.cdr
Plasma Cutting
In order to cool and to reduce the emissions, plasma torches may be surrounded by additional gas or water curtains which also serve as arc constriction, Figure 12.18. In dry plasma cutting where Ar/H2, N2, or air are used, harmful substances always develop which not
plasma gas electrode
plasma
water bath workpiece
br-er12-18e.cdr
Figure 12.18
169
the UV radiation are, for the most part, held back by the water. A further, positive effect is the cooling of the cutting surface, Figure 12.18. Careful disposal of the residues inevitable. is here
cutting with water bath water injection plasma cutting with water curtain
Figure 12.19 gives a survey of the different cutting methods using a water bath. Figure 12.19 Figure 12.20 shows a torch which is equipped with an additional gas supply, the socalled secondary gas. The secondary gas shields the plasma jet and increases the transition resistance at the nozzle front. The so-called double and/or parasite arcs are avoided and nozzle life is increased. Thanks to new electrode materials, compressed air and even pure oxygen may be applied as plasma gas therefore, in flame cutting, the burning of unalloyed steel may be used for increased capacity and quality. of The the
plasma gas electrode
plasma cutting with workpiece on water surface
br-er12-19e.cdr
selection plasma
forming
secondary gas nozzle
workpiece
170
The advantage of the use of oxygen as plasma gas is in the achievable cutting speeds within the plate thickness range of approx. 3 12 mm (400 A, WIPC). In the steel plate thickness range of approx. 1 10 mm the application of 40 A-compressed air units is recommended. In comcutting speed [m/min]
8 1 6 2 4 2 3 4 5
br-er12-21e.cdr
machine type and plasma medium 1 WIPC, 400 A, O2 2 WIPC, 400 A, N2 3 200 A, s < 8 mm: N2 s > 8 mm: Ar/H2 4 40 A, compressed air
narrower
kerfs, but with lower cutting speeds. Figure 12.21 different shows cutting Figure 12.21
10
plate thickness [mm]
15
20
the laser is used as the jet generator for cutting, 12.22. Variations of the
br-er12-22e.cdr
- laser beam combustion cutting - laser beam fusion cutting - laser beam sublimation cutting
Figure
laser beam combustion cutting, Figure 12.25 laser beam fusion cutting, Figure 12.26 laser beam sublimation cutting, Figure 12.27.
171
The process sequence in laser beam combustion cutting is comparable to oxygen cutting. The material is heated to the ignition temperature and subsequently burnt in the oxygen stream, Figure 12.23. Due to the concentrated energy input almost all metals in the plate thickness range of up to approx. 2 mm may be cut. In addition, it is possible to achieve very good bur-free cutting qualities for stainless steels (thickness of up to approx. 8 mm) and for structural steels (thickness of up to 12 mm). Very narrow and parallel cutting kerfs are characteristic for laser beam cutting of structural steels.
contribution for oxidising materials) or an inactive cutting gas may be applied depending on the cutting job. Besides,
br-er12-23e.cdr
slag jet
the very high beam powers (pulsed/superpulse d mode of operation) allow a direct evaporation of the material (sublimaabsorption factor
Figure 12.23
80
cutlaser
40
20
melting point Tm
temperature
Figure 12.24
melting
60
boiling point Tb
evaporating
heating-up
172
creases unevenly when the process starts. In laser beam fusion cutting remains the reflexion on the molten material, however, at more than 90%! Figure 12.24 shows the absorption factor of the laser light in dependence on the temperature. This factor mainly depends on the wave length of
laser cutting (with oxygen jet) - the laser beam is focused on the workpiece surface and the material burns in the oxygen jet starting from the heated surface materials: - steel aluminium alloys, titanium alloys cutting gas: - O2, N2, Ar criteria: - high cutting speed, cut faces with oxide skin
br-er12-25e.cdr
the light.
used When
laser the
melting point of the material has been reached, the absorption increases evenly factor unand
Figure 12.25
During laser beam combustion cutting of structural steel high cutting speeds are achieved due to the exothermal energy input and the low laser beam powers, Figure 12.25. In the above-mentioned case (dependent on beam quality, focussing, etc.), above a beam power of approx. 3,3 kW, spontaneous evaporation of the material takes place and allows sublimation cutting. Significantly higher laser powers are necessary to fuse the
laser fusion cutting: - the laser beam melts the entire plate thickness (optimum focus point 1/3 below plate surface) - high reflection losses (>90%) materials: - metals, glasses, polymers cutting gas: - N2, Ar, He criterions: - cutting speed is only 10-15% in comparison to cutting with oxygen jet, characteristics melting drag lines
br-er12-26e.cdr
material and blow it out with an inert gas, as the reflexion loss remains constant.
influ-
Figure 12.26
173
ting are the focus intensity, the position of the focus point in relation to the plate surface and the formation of the cutting gas flow. A prerequisite for a high intensity in the focus is the high beam quality (Gaussian intensity distribution in the beam) with a high beam power and suitable focussing optics. Laser beam cutting of contours, especially of pointed corners and narrow root faces, requires adaptation of the beam power in order to avoid heat accumulation and burning of the material. In such a case the beam power might be reduced in the continuous wave (CW) operating mode. With a decreasing beam efficiency decreases the cuttable plate thickness as well. Better suited is the switching of the laser to pulse mode (stanlaser evaporation cutting: - spontaneous evaporation of the material starting from 105 W/cm2 with high absorption rate and deep-penetration effect - metallic vapour is pressed from the cavity by own vapour pressure and by a supporting gas flow materials: - metals, wood, paper, ceramic, polymer cutting gas: - N2, Ar, He (lens protection) criteria: - low cutting speed, smooth cut edges, minimum heat input
br-er12-27e.cdr
dard equipment of HF-excited lasers) where pulse height can right be up selected to the
pulse (in-
Figure 12.27
laser 600 W 1500 W 600 W 1500 W 1500 W plasma 50 A 5 kW 250 A 25 kW 500 A 150 kW oxy-flame Stahl Cr-NiStahl 1 10
operation
and
laser
beam
sublimation cutting.
br-er12-28e.cdr
Figure 12.28
174
Laser beam cutting of aluminium plates thicker than appx. 2 mm does not produce bur-free results due to a high reflexion property, high heat conductivity and large temperature difCO2-laser (1500 W) 10 cuttig speeds [m/min] plasma cutting (WIPC, 300-600 A)
ferences between Al and Al2O3. The addition of iron powder allows the flame cutting of
stainless
steels
100
Figure 12.28 shows a comparison of the different plate thicknesses which were cut using different processes. For the plate thickness range of up to 12 mm (steel plate), laser beam cutting is the approved precision cutting process. Plasma cutting of plates > 3 mm allows higher cutting speeds, in comparison to laser beam cutting, the cutting quality, however, is
6
costs [DM/m cut length] total costs machine costs
significantly
5 4
laser
cutting
plasma flame cutting with 3 torches
plates
3 2 1 5 10 15 20
25
30
35
40
plasma
cutting,
increasing
Figure 12.30
175
ference in the cutting speed is reduced. Plates with a thickness of more than 40 mm may be cut even faster using the flame cutting process.
Figure 12.29 shows the cutting speeds of some thermal cutting processes.
Apart from technological aspects, financial considerations as well determine the application of a certain cutting method. Figures 12.30 and 12.31 show a comparison of the costs of flame cutting, plasma arc and laser beam cutting the costs per m/cutting
extract from a costing acc. to VDI 3258
flame cutting (6-8 torches) investment total (replacement value) calculation for a 6-yearaccounting depreciation maintenance costs energy costs production cost unit rate costs/1 operating hour plasma cutting (plasma 300A) laser beam cutting (laser 1500W)
length
170,000.00
220,000.00
500,000.00
be a deterrent to
/h 65.00 75.00 130.00
exploit cutting
the
high
qualities
Figure 12.31
2003
175
Apart from the welding processes explained earlier there is also a multitude of special welding processes. One of them is stud welding. Figure 13.1 depicts different stud shapes. Depending on the application, the studs are
Figure 13.1 In arc stud welding, a distinction is basically made between three process variations. Figure 13.2. depicts the three variations the differences lie in the kind of arc ignition and in the cycle of motions during the welding process.
Figure 13.2
176
The switching arrangement of an arc stud welding unit is shown in Figure 13.3. Besides a power source which produces high currents for a short-time, a control as well as a lifting device are necessary.
Figure 13.3
In drawn-arc stud welding the stud is first mounted onto the plate, Figure 13.4. The arc is ignited by lifting the stud and melts the entire stud diameter in a short time. When base stud plate and are
molten weld pool while the ceramic ferrule is forming the weld. After the solidification of the liquid weld pool the ceramic ferrule is knocked off. Figure 13.4
177
Figure 13.5 illustrates tip ignition stud welding. The tip melts away immediately after touching the plate and allows the arc to be ignited. The lifting of the stud is dispensed with. When the stud base is molten, the stud is positioned onto the partly molten workpiece.
Studs with diameters of up to 22 mm can be used. Welding currents of more than 1000 A are necessary.
The arc stud welding process allows to join different materials, see Fig-
ure 13.6. Problematic are the different melting points and the heat dissipation of the individual materials. Aluminium studs, for example, may not be welded onto steel.
The relatively high welding currents in the arc stud welding process cause the somewhat troublesome sideeffects of the arc blow. Figure 13.7 depicts different of Figure 13.5
arrangements current
contact
points and cable runs and illustrates the developing arc deflection (B,C,E). A, D and F show possible measures. Figure 13.6 counter-
178
In high-frequency welding of pipes the energy input into the workpiece may be carried out via sliding contacts, as shown in Figure 13.8, or via rollers, as shown in Figure 13.9. Only the high-frequency technique allows a safe current transfer in spite of the scale or oxide layers.
Through the skin effect the current flows only conditionally at the surface. Therefore no thorough fusion of thick-wall pipes
may be achieved.
Figure 13.7
Figure 13.8
Figure 13.9
13. Special Processes Only welding of small wall thicknesses is profitable as the weld speed must be greatly reduced with increasing wall thicknesses, Figure 13.10.
179
In induction welding a process which is used frequently nowadays the energy input is received contactless, Figure 13.11. Varying magnetic fields produce eddy currents inside the workpiece, which again cause resistance heating in the slotted tube. A distinction is made between coil inductors (left) and line inductors (right).
Figure 13.10
Also in case of induction welding flows the current flows only close to the surface areas of the pipe. Only the current part which reaches the joining zone and causes to fill the gap may be utilised. ure 13.12 Figillus-
trates two current paths. On the left side: current the useful on
path,
path which does not contribute to the fusion of the Figure 13.11 edges.
13. Special Processes Figure 13.13 shows the effective depth during the inductive heating for different materials, pendence in on dethe
180
reached, the effective depth for ferritic steels increases. Figure 13.12
Figure 13.13
Figure 13.14
181 The application of the induction welding method allows high speeds than of welding more
100m/min,
Figure 13.14.
Aluminothermic fusion welding or cast mainly Figure 13.15 joining welding used is for
railway
tracks on site. A crucible is filled with a mixture consisting of aluminium powder and iron oxide. An exothermal reaction is initiated by an igniter the aluminium oxidises and the iron oxide is reduced to iron, Figure 13.15. The molten iron flows into a ceramic mould which matches the contour track. of After the the
process
sembly.
Figure 13.16
182
quently used for joining dissimilar materials, as, for example, unal-
loyed steel/alloyed steel, per/aluminium steel/aluminium. The materials Figure 13.17 copor
Figure 13.18
The determined cladding speed must be strictly adhered to during the welding process. If the welding speed is too low, lack of fusion is the result. If the welding speed is exceeded, the development of the waves in the joining zone is erratic. Figure 13.19 shows the critical cladding speeds for different material combinations.
183
Figure 13.19
Figure 13.20
Figure 13.20 shows a diagrammatic representation of a diffusion welding unit. Diffusion welding, like ultrasonic welding, is welding in the solid state. The surfaces which are to be joined are cleaned, polished and then joined in a vacuum with pressure and temperature. After a certain time (minutes, right up to several days) joining is achieved by diffusion processes.
The advantage of this costly welding method lies in the possibility of joining dissimilar materials without taking the risk of structural transformation due to the Figure 13.21
184
tions. The joining of two extremely different materials, as, e.g. austenite and a zirconium alloy,
ate layers.
Figure 13.22 shows the structure of a joint where nickel, copper and vanadium had been used as intermediate layers. As the diffusion of the individual components takes place only in the region close to the surface, very thin layers may be realised.
In cold pressure welding - in contrast to diffusion welding - a deformation is produced by the high contact pressure in the bonding plane, Figure 13.23. The joint surfaces are
adhesion
185
Ultrasonic welding is used as a microwelding method. The process principle is shown in Figure 13.24. The surface layers of overlap arranged plates are destroyed by applying mechanical vibrator energy. At this instance are joining surfaces deformed by very short localised warming up and point-
interspersed connected. The joining members are welded under pressure, where one part small amplitudes (up to 50 m) relative to the other is moved with with ultrasonic frequency. As far as metals are concerned, the vibratory vector is
in the joining zone, in contrast to ultrasonic welding of plastics. The ultrasonics which have been produced by a magnetostrictive transducer and transmitted by a sonotrode lie in the Figure 13.24 frequency range of 20 up to 60 Hz.
weld-
Further microwelding processes are methods which are also called heated element methods, example, welding as, for
186
bonding. These methods are applied in the electronics industry for joining very fine wires, as, for example, gold wires from microchips with aluminium strip conductors.
point via a feeding nozzle. The welding wedge is lowered and the wire with is the
welded
In nailhead bonding, the wire which emerges from the feeding nozzle may have diameters from 12 to 100 m. By a reducing hydrogen flame its end is molten to a globule, Figure 13.27. The nozzle then presses this globule onto the part aimed at and shapes it into a nail head.
Figure 13.28
de-
soldering
187
The individual soldering methods are classified into different mechanisms depending on the type of heating, Figure 13.30. There are two basic distinctions: soft soldering (melting temperature of the solder is approx. up to 450C) and brazing (melting temperature of the brazing solder is approx. up to
ture is approx. up to 1200C) are used. This process is frequently subject to automation. Figure 13.28
Figure 13.29
Figure 13.30
2003
188
As the production costs of the metal-working industry are nowadays mainly determined by the costs of labour, many factories are compelled to rationalise their manufacturing methods
Designation examples gas-shielded arc welding TIG GMAW movement/ working cycles
torch-/ workpiece control filler wire feeding workpiece handling
by fully
partially
and
mechanised proc-
production
manually
of
welding
engia
mechanically mechanically
manually
neering
where
consistently
mechanically mechanically mechanically
good
Figure 14.1
The levels of mechanisation in welding are stipulated in DIN 1910, part 1. Distinctions are made with regard to the type of torch control and to filler addition and to the type of process sequence, as, e.g., the transport of parts to the welding point. Figure 14.1 explains the four levels of mechanisation.
Figure 14.2. shows manual welding, in this case: manual electrode welding. The control of the electrode and/or the arc is carried out manu-ally. The filler metal (the consumable elecbr-er14-02e.cdr ISF 2002
189
shielded welding,
manipulation is carried out manually, the filler metal addition, however, is executed mechanibr-er14-03e.cdr
Figure 14.3.
In fully mechanised welding, Figure 14.4, an automatic equipment mechanism carries out the welding advance and thus the torch control. Wire feeding is realised by
must
positioned
br-er14-04e.cdr
manu-
ally in accordance
Fully Mechanised Welding (Gas-Shielded Metal-Arc Welding)
Figure 14.4
chine support.
In automatic welding, besides the process sequences described above, the workpieces are mechanically positioned at the welding point and, after welding, automatically trans-ported to the next working station. Figure14. 5 shows an example of automatic welding (assembly line in the car industry).
devices which
available
facilitate or make Figure 14.5 the welding process at all possible. Figure 14.6 shows
assembly line welding robot machine carrier linear travelling mechanism track-mounted welding robots spindle / sliding head turntable turn-/ tilt table dollies assembly devices
assisting devices.
Before
welding,
br-er14-06e.cdr
welding
lower part of the device has the shape of a prism. This allows to clamp pipes with different diameters.
Devices, however, may be significantly more complex. Figure 14.8 shows an example of an assembly equipment used in car body manufacturing. This type of device allows to fix complex parts at several points. Thus a defined position of any weld seam is reproducible.
Figure 14.7
middle: the rollers automatically adapt to the tank diameter; Figure bottom: the roller spacing may be varied by a scissor-like arrange-
br-er14-08e.cdr
Figure 17.8
ment.
In general, dollies are motor-driven. This provides also an effortless movement of heavy components, Figure 14.10.
192
set of rollers 1
br-er14-09e.cdr
set of rollers 2
br-er14-10e.cdr
Turning Rolls
Turning Rolls
Figure 14.9
Figure 14.10
A work piece positioner, e.g. a turn-tilt-table, is part of the standard equipment of a robot working station. Figure 14.11 shows a diagrammatic representation of a turntilt-table. Rotations
table top rotational axis gear segment table support tilting axis support
135 are possible while the turn-table can be turned by 365. Those types of turn-tables are designed for working parts with
br-er14-11e.cdr
weights of just a few kilograms right Figure 14.11 up to several hundred tons.
193
A turn-tilt table with hydraulic adjustment of the tilting and vertical motion as well as chucking grooves for the part fixture is depicted ure 14.12. in Fig-
br-er14-12e.cdr
Figure 14.12
In robot technology the types of turn-tilt-tables - as shown in Figure 14.13 - are gaining importance. Positioners with orbital design have a decisive advantage because the component, when turning around the tilting axis, remains approx. equally distant to the welding robot.
single-column turn-tilt-table
table top table support tilting axis support rotational axis
orbital turn-tilt-table
table top table support tilting axis support rotational axis
br-er14-13e.cdr
ISF 2002
Turn-Tilt-Tables
Figure 14.13
194
Other types of workpiece positioners are shown in Figure 14.14 the double column turn-tilt-table and the spindle and sliding holder turn-tilt-table. Those types of positioners are used for special component geometries and allow welding of any seam in the flat and in the horizontal position.
tilting axis
support
table support
br-er14-14e.cdr
ISF 2002
Double-Column Turn-Tilt-Table
Figure 14.14
table tops
bed way
br-er14-15e.cdr
ISF 2002
Figure 14.15
In the field of welding, special units are designed for special tasks. Figure 14.16 shows a pipe-flange-welding machine. This machine allows the welding of flanges to a pipe. The weld head has to be guided to follow the seam contour.
195
br-er14-16e.cdr
Figure 14.16
Plain plates or rounded tanks are clamped by means of longitudinal jigs for the welding of a longitudinal seam, Figure 14.17. The design and the gripping power are very dependent of the thickness of the plates to be welded.
br-er14-17e.cdr
Figure 14.17
A simple example of a special welding machine is the tractor travelling carriage for submerged-arc welding, Figure 14.18. This device is designed for the application
196 on-site and provides, besides the supply of the filler metal, also the welding speed as well as the feeding and suction of the welding flux.
For the guidance of a welding head and/or welding device, machine supports may be used. Figure 14.19 shows different types of machine supports for welding and cutting. Apart from the translatory and rotary principal axes they are often also equipped with additional axes to allow precise positioning.
br-er14-18e.cdr
Robots are handling devices which are equipped with more than three userd
197
designation
cartesian robot
arrangement
R z y z C B C
A R C z C D B
kinematic schedule
operating space
br-er14-20e.cdr
ISF 2002
Kinematic Chains
Figure 14.20
The most common design of a trackmounted welding robot is shown in Figure 14.21. The robot depicted here is a hinged-arm robot with six axes. The axes are divided into three principal and three additional axes or hand axes. The wire feed unit and the spool carriers for the wire electrodes are often fixed on the robot. This allows a compact welding design.
br-er14-21e.cdr
Robot Motions
Figure 14.21
198
Varying lever lengths permit the design of robots with different operating ranges. Figure 14.22 shows the operating range of a robot. In the unrestricted operating range the component may be reached with the torch in any position. The restricted operating range
allows the torch to reach the component only certain In a the sus-
positions. case of
arrangerobot is
br-er14-22e.cdr
device
For the completion of a robot welding station workpiece positioners are necessary. Figure 14.23 shows positioner devices where also several axes may be combined. These axes may either turn to certain defined positions or be guided by the robot control and moved synchronically with the internal axes. The complexity and versatility of the
Figure 14.23
199
Movement by means of a linear travelling mechanism increases the operating range of the robot, Figure 14.24. This may be done in ease of stationary as well as suspended arrangement, where there is a possibility to move to fixed end positions or to stay in a synchronised motion with the other movement axes.
br-er14-24e.cdr
Figure 14.24
2003
200
Increased quality requirements for products and the trend to automate production processes along with increased profitability result in the use of industrial robots in modern manufac-
introduced
The definition says that an industrial robot for gas welding is an universal movement automaton with more than three axes which are user-programmable and may be sensor-controlled. It is equipped with a welding torch and carries out welding jobs.
Core of a modern robot welding cell are one or more seam welding robots of swan neck type. Normally, they have six user-programmable axes; so they can access any point within the
working range at any orientation of the welding torch. To extend their range,
working
201
installation of the robot onto a linear carriage with Cartesian axes. Such 'external' axes are also user-programmable, Figure 15.3.
To turn the workpiece in the welding-favourable downhand position and to ensure accessibility to any joints, workpiece positioners are used as external axes which are steered by the robot control. Multistation cycle tables are often used to increase profitability of the complete system installation. The operator feeds and removes the welded workpiece on one side, while the robot is welding Figure 15.3 on the other side.
The robot control is the centre of an industrial robot system for arc welding, Figure 15.4. It provides and processes all information for robot mechanics, positioner, welding unit, safety equipment, and external sensors. The robot program transforms information into signals for control of robotand posi-
Figure 15.4
202
Modern industrial robot controls are build as multi-processor controls due to the multitude of parallel calculations and control functions. Figure 15.5 shows the internal structure of such a control. Individual assemblies which are designed for special jobs and equipped with an own micro-processor are linked with the host computer via the system bus. The host controls and coordinates the actions of the components based on the operating system and the robot program. Examples of such assemblies, which are mostly installed on individual printed boards, are e.g. the axes computers. They are responsible for calculation of
control the drive motors, two interconnected control loops per axis are available which
Further
assemblies
control the display screen, the manual programming (PHG); semblies sponsible these are unit asrefor
203
and analogue in- and outputs and field bus systems. Or they complete the data transmission with external control systems. To reduce downtimes in the case of malfunction, some robot controls can be connected via internet with telediagnosis systems of the robot manufacturer to support service personnel during troubleshooting and commissioning.
Programming of welding robots can be carried out in different ways which are distinguished in On-Line (programming at the robot) and Off-Line (programming out of the robot cell), Figure 15.6.
The robot is manually guided along the later track with decoupled drives during PlayBack programming. The path of the track is recorded and transformed into a corresponding robot control program. This procedure is preferably used for painting jobs.
A common technique to program a robot is the Teach-In procedure. During Teach-In programming, with the help of the manual programming unit, the welding torch is moved to notable points of the groove to be welded which are stored with information about position and orientation. In addition, track parameters must be entered, like e.g. type of movement and speed or welding parameter sets.
During sensor supported Teach-In programming, the path progress through some typical points is only roughly indicated. Then the accurate path is picked-up by sensors and autocalcu-
matically
gram is supplemented by
204
Textual programming belongs to mixed procedures. The sequence program in form of a text file is created on an external computer and is then transmitted to the robot steering control, Figure 15.7. The recording of the position of points is carried out in the same way as with Teach-In programming: moving into position and recording.
Macro-programming is also regarded as a mixed method which shortens programming time at the robot, Figure 15.8. Macros are structured processing sequences which are created online to fulfil working functions and which can be repeated for further similar working functions. Geometry macros contain information about torch guidance to produce certain joints or joint sections. Welding technol-
ogy parameters for individual situations summarised welding This torch torch Figure 15.8 welding are in
macros. for
applies
positioning, inclination,
online or offline) of such macros, the programming time can be shortened for workpieces with often Figure 15.9 repeated
205
welding jobs, e.g. steel construction when welding stiffeners and head plates Using offline programming practice, the programming work is shifted out from the producing robot cell. This avoids unproductive stoppages and allows for economicviable, limited number of pieces to be reduced. During textual programming, the 3-dimensional point coordinates and torch orientations are entered into an external computer in a manufacturer-specific program language. To achieve a complete program sequence, each instruction must be entered individually.
The graphical offline programming uses CAD data for modelling the complete robot working cell and parts to be welded. Planning of the path is carried out with CAD functions directly at the workpiece which is displayed on a screen. In most cases, the programming systems provide a graphical simulation of the movement, e.g. to check for collisions between torch and
workpiece,
Figure
15.9. For the following transformation of the program into the robot control, a calibration between model
In the case of knowledge-based offline programming, the operator is supported by integrated expert systems when it comes to creation of robot welding programs, e.g. for determination of job-specific welding parameters. However, checking and adapting the program must be carried out by the operator. Modern robot controls provide the programmer with some functions for movement control and for modification of program sequence, Figure 15.10. PTP movement (point to point) serves to move the robot in the space. All axes are controlled in such
206
a way that they reach their set-point at the same time. Thereby the actual path of the torch depends on kinematics of the robot and on current position of the axes.
A linear interpolation (CP procedure, continuous Path), Figure 15.11, is used for accurate movement along a straight line, e.g. movement to weld start point or welding. The active point of the tool 'arc' (ToolCentre-Point, TCP) is moved along a straight tween line two bepropoints,
grammed
adapting torch angle and torch inclination between the two points. Figure 15.11
Circles and graduated circles are entered by means of circle interpolation programs, Figure 15.12. Then the orientation of the torch can be adapted through turning the knuckle axis or 6th axis of the robot and the value of spill-weld at the end of the seam can be indicated.
control the program run, commands are available for: reFigure 15.12
207
ditional and unconditional program jumps, waiting periods, waiting for inputs, and working with sub-programs. The software of modern seam welding robots contains as special functions 3dimansional transfor-mations and mirroring of programs and partial programs, palletising functions,
(Master/Slave
16. Sensors
2003
16. Sensors
208
The welding process is exposed to disturbances like misalignment of workpiece, inaccurate preparation, machine and device tolerances, and proess disturbances, Figure 16.1.
The manual welder notices them by eyesight and corrects them manually according to strategies learned and gained by experience. To record process irregularities and path deviations, a fully mechanised welding plant requires sensors providing control signals which are then used in accordance with implemented rules. Using corresponding control elements, the control loop is closed for the welding process.
Scopes of duty of the sensors is finding the weld start point and seam
online and offline. The ideal sensor for a robot application should measure the welding point Figure 16.1
16. Sensors
209
as possible (no restriction in accessibility). The ideal sensor which combines all three requirements, does not yet exist, therefore one must select a sensor which is suitable for the individual job. Figshows sensor
The
frequently used systems in practice are tactile, optical, and arc based sensor systems with mechanical adjustment. arc Figure 16.3
With tactile scanning systems, the simplest type of scanning is a mechanical sensor. Pins, rollers, balls, or similar devices may be used as sensors.
Such scanning systems show a long distance between sensor and torch, the application range is limited. Only grooves with large dimensions and relatively straight seam path can be scanned with these systems. Figure 16.3 shows some examples of different groove geometries.
Tactile sensors can recognise 3dimensional offsets of the workpiece. Figure 16.4
16. Sensors
210
Through scanning of three levels the 3-dimensional point of intersection can be calculated and the robot program for correcting the deviation can be shifted accordingly thus finding the start point of the weld. In this case, the gas nozzle of the torch serves as a sensor, Figure 16.4, which is charged with electrical tension. As soon as the torch touches the workpiece, a current flows, which is then taken by the robot control as a signal for obtaining the level to be scanned.
Inductive sensors are graded as non-contact measurement systems. Due to their function principle, they can be applied for metallic and electrically conductive materials. The simplest type is a ring coil. If alternating current flows though the coil, ,a magnetic field is generated close to the workpiece. When the coil approaches the workpiece surface, the magnetic field weakens. Figure 16.5 shows the distancedependent electrical signal. Such
simple sensors are used to recognise the workpiece position. several sensors, welding Using distance also a
groove
can be scanned. Figure 16.5 With multi-coil arrangements in one sensor, the position of the welding groove, the angle between sensor and workpiece surface and the distance can be recorded. Figure 16.6 shows a principle arrangement. A transmitter coil generates an magnetically alternating field which induces
alternating currents in the two receiver coils. In the undisturbed case, these currents are phase-shifted by 180 and neutralise each other. If the sensor is moved crosswise to the groove, magnetical asymmetries will occur in the scanning area, which
16. Sensors
211
will show in the presented signal shape. The output signal will be zero, if the coils are positioned exactly above the centre of the groove. The radar sensor in Figure 16.6 uses Doppler's effect to generate a signal. Here the phase difference between transmitter signal and receiving signal is evaluated. A mathematical process transforms such signals into distance values. To record the position and the depth of the groove, the sensor must be continuously moved along the seam. Radar sensors form a so called radar baton, which is focussed onto a measurement spot of about 0,7 mm diameter for this application. Figure 16.6 shows the sensor signal, which represents
can be determined with a resolution in the range of 1/10 mm. Figure 16.6 Arc sensors evaluate the continuous change welding of the
current
with a change of the contact tip-towork distance, Figure 16.7. A signal for side control of the torch is determined by
16. Sensors
212
subtraction of the currents on the flanks of a groove. A comparison between actual welding current and programmed rated current provides a signal for distance control of the welding torch. To let this sensor method work, a divergence of the arc or the use of a second arc is required.
To realise this principle, there are numerous possibili-ties. Figure 16.8 shows some variants of signal recording. The most frequently used method is a mechanical oscillation of the welding torch, which is carried out by a rotor movement with an oscillation frequency up to 5 Hz. The second method is mainly used with submerged arc
direction
Magnetic fields can diverge only the arc itself. The advantage of this method is a high divergence frequency of about 15 Hz. A disadvantage is the size of the electromagnets and the limited accessibility to the workpiece. The last variant of an arc sensor incorporates a mechanical rotation of the welding wire. In this case, the divergence frequency of the arc can reach up to 30 Hz.
The signal recording is continuous during the movement. In this way, information about orientation of the torch and groove width is also provided. The arc sensor principle is limited to groove shapes with clear flanks. Together with the tactile torch gas nozzle sensor, it provides a frequently used combination for seam finding and seam tracking during robot welding.
16. Sensors
213
Optical sensors can be used for a great number of jobs. The easiest method is the recognition radiation which is of the
intensity, reflected
during welding. E.g. with laser beam welding, this is carried out through the relaser
recording flected
The procedure is based on the line-up between the degree of reflection and shaft relation (penetration depth/focus position) of the capillary. The amount of backreflection of the laser beam power is measured, which due to multi-reflection is not absorbed by the workpiece. Changes of penetration depth due to modified laser power or a shifted focus position can be identified by the signal of reflected laser power and can be used for control of the penetration depth. However, optical sensors can also be used for measuring geometrical values. Such information may be used for finding the start point of a seam, for seam tracking, and for identification of groove profile. The two last mentioned functions provide the possibility to use the information for filling rate control and/or quality control.
Geometry-measuring optical sensors are normally external systems, which are positioned in front of the torch as a leading element. It is practical to equip the sensor with additional axes, because both, torch and sensor, must be moved along the groove. Without additional axes, a robot would be limited in its accessibility to the workpiece and in its working range. Another problem is the tremendous effort to introduce the control-technical integration into the robot control. Among other things, information must be exchanged in real time.
16. Sensors
214
Most of geometry-measuring sensors use the triangulation principle or a variant of this measurement procedure. The triangulation measurement procedure provides information about the distance to the workpiece surface. A light spot is projected onto the workpiece surface and displayed to a line-type receiver element under a certain angle. With distance changes emerge corresponding positions on the receiver element, Figure 16.10. Sensors which use this triangulation principle are applied for recognition of workpiece position and for offline seam finding. Figure 16.10 Both, the laser scanner and the light-section procedure are based on the triangulation measurement principle. With the laser scanner, Figure 16.11, this principle is complemen-ted by an oscillating axis in parallel to the groove axis. The measurement of a sequence of distances along a line becomes possible and provides a 2-dimensional re-
procedure,
16. Sensors
215
provide information about the 2-dimensional position of the groove. As a function of this system, one or more light lines are projected onto the workpiece surface and displayed to a CCD matrix under a certain angle, Figure 16.12. In contrast to scanning, information about the groove profile is provided by taking a picture scene. Using sensors, it is pssible to obtain additional 3-dimensional information through evaluation of more, in succession taken, while the camera moves over the grooves. Systems, which generate their information through a projection of several light lines, provide additional information about the path of the seam and the orientation of the sensor related to the workpiece surface. Both, scanning
systems and sensors based on the light section procedure, can be used for recognition of the welded seam to make automised an quality
Another measurement
optical prin-
stereo procedure to record information the Two optics the weld geometry across groove.
16. Sensors
216
groove area and displays them onto two image converter elements (CCD-lines or CCD-matrix). Based on the corresponding image points in both picture scenes, the 3dimensional position of object points is evaluated. Figure 16.13 shows the measurement principle, which uses CCD lines as image converter elements, and idealised signals for generating information. The grey scale drop in the signal is ideally used as corresponding image area, which occurs with butt welds due to different reflection intensity between workpiece surface and gap. Both, the lateral position of the groove and the distance to the sensor can be determined by evaluating the centre positions of both signal drops. The width of the groove is taken from the width of the signal drop.
Optical sensors may also be used for geometrical recognition of the weld pool, to adapt process parame-ters in the case of possible deviations. Figure 16.14 depicts such a system for use with laser beam welding. The welding process is monitored by a CCD camera through a filter system. An optical filter allows to observe the weld pool surface without disturbing effects of the plasma in the near infrared spectrum. Picture data are transferred to an image processing computer which measures the geometry of the weld pool. Geometry data contain information which is used online for control of the welding Among penetration process. others, depth
provides the recognition of protrusionwelded joints and welding defects like e.g. molten pool Figure 16.14
ejections.
16. Sensors
217
During electron beam welding, the beam is in combination with a detector used for both, to carry out a seam tracking and a monitoring of the welded seam. For this, the beam can be diverged as well as bent, Figure 16.15. Backscattered electrons are recognised by a special detector and converted into grey values. The line or area surface scanning by the spotted electron beam provides a progressive series of greys across the scanned line or area. During electron beam welding, these signals can be used for seam tracking by scanning an edge which is parallel to the groove. The
area-type scanning provides the possibility observing welded seam for the or
Literature
Literature
Altmann, E., J. Derse u. A. Farwer Sauerstoff-Plasmaschneiden von unleg. Stahl - ein wirtschaftlicher und technologischer Vergleich DVS-Berichte, Bd. 131, 1990 Baum, L. u. V. Fichter Der Schutzgasschweier, Teil 2: MIG/MAG-Schweien DVS-Verlag GmbH, Dsseldorf 1982 Behnisch, H. Das thermische Schneiden Technica 29, 1980, Heft 7 Beyer, E. Einflu des laserinduzierten Plasmas beim Schweien mit CO2-Lasern Schweitechnische Forschungsberichte Bd. 2 DVS-Verlag GmbH, Dsseldorf 1985 Beyer, E. u. L. Cleemann Schweien mit CO2-Hochleistungslasern Technologie Aktuell 4, VDI-Verlag 1987 Blasig, K., U. Lttmann u. H. Nies Unterpulver-Engspaltschweien mit dnnen Doppeldrahtelektroden Adaptives Nahtfhrungssystem Industrie Anzeiger 109, 1987, Nr. 82, S. 30-32 Bhme, D., R. Killing u. R. Helwig Beitrag zur Frage der gnstigsten Stromart und Energieeinbringung beim Unterpulvertandemschweien Schweien und Schneiden 34, 1982, Heft 10 Cloos Romat Roboter ProgrammieranIeitung C. Cloos Schweitechnik, Haiger Derse, J. Wasser-Injektions-Plasmaschneiden ein neues Qualittsverfahren Trennen u. Fgen 17, 1986 Dickmann, K. Lasertechnologie fr die Materialbearbeitung Technica 10/1990
219 Dilthey, U. Programmieren von Industrierobotern DVS-Berichte Bd. 118 DIN 1910 Teil 1, Mechanisierungsgrade in der Schweitechnik, Juli 1983 DIN 1910 Blatt 5 Schweien, Widerstandsschweien, Verfahren, Nov. 1972
Literature
DIN 1913 Stabelektroden fr das Verbindungsschweien von Stahl, un- und niedriglegiert, Jan. 1976 DIN 1732 Schweizusatzwerkstoffe fr Aluminium, Apr. 1975 DIN 2310 Teil 6 Thermisches Schneiden, Einteilung, Verfahren, Feb. 1991 DIN 8555 Schweizusatzwerkstoffe zum Auftragschweien, Jan. 1978 DIN 8556 Schweizusatzwerkstoffe fr das Schweien nichtrostender und hitzebestndiger Sthle, Mrz 1976 DIN 8573 Schweizusatzwerkstoffe zum Schweien von Gueisen, Jan. 1978 DIN 8575 Teil 1 Schweizusatzwerkstoffe zum Lichtbogenschwei8en warmfester Sthle, Dez. 1983 DIN 8593 Teil 6 Fertigungsverfahren Fgen, Fgen durch Schweien, Einordnung, Unterteilung, Sept. 1985 DIN 32511 Elektronen- und Laserstrahlverfahren zur Materialbearbeitung, Juni 1996 DIN EN 440 Schweizustze - Drahtelektroden und Schweigut zum MetallSchutzgasschweien von unlegierten Sthlen und Feinkornsthlen, Nov. 1994 DIN EN 756 Schweizustze - Drahtelektroden und Draht-Pulver-Kombinationen zum Unterpulverschweien von unlegierten Sthlen und Feinkornsthlen, Dez. 1995 DIN EN 758 Schweizustze - Flldrahtelektroden zum Metall Lichtbogenschweien mit und ohne Schutzgas von unlegierten Sthlen und Feinkornbausthlen, Mai 1997 DIN EN 760 Schweizustze - Pulver zum Unterpulverschweien, Mai 1996 DIN EN 1089 Ortsbewegliche Gasflaschen - Gasflaschen-Kennzeichnung, Apr. 1998 DIN ISO 857 Einteilung der Schutzgasverfahren, Juni. 1996
220
Literature
DIN EN 12070 Schweizustze - Drahtelektrode, Drhte und Stbe zum Lichtbogenschweien von warmfesten Sthlen, Jan. 2000 DIN EN 12072 Schweizustze - Drahtelektrode, Drhte und Stbe zum Lichtbogenschweien von nichtrostenden und hitzebestndigen Sthlen, Jan. 2000 DIN EN ISO 9692 Teil 2 Schweien und verwandte Verfahren - Schweinahtvorbereitung Unterpulverschweien von Stahl, Sept. 1999 DIN EN ISO 11146 Laser und Laseranlagen - Prfverfahren fr Laserstrahlparameter, Sept. 1999 EN ISO 9692 Teil 2 Schweien und verwandte Verfahren, Schweinahtvorbereitung Unterpulverschweien von Stahl, Apr. 1998 Dorn, L. u. P. Rippl Prozeanalyse beim Unterpulverschweien Lichtbogenstabilitt und Momentanwertverlauf der elektrischen Gren bei nderung der Verfahrensparameter Schweien und Schneiden 37, 1985, Heft 2 Draugelates, U. u. J. Krohn Plasma-Heidraht-Auftragschwei8en von Hartlegierungen DVS-Bericht Band 81 DVS-Merkbltter, Widerstandsschweitechnik, Fachbuchreihe Schwei8technik Band 68/III DVS-Verlag GmbH, Dsseldorf 1979 DVS-Merkblatt 0902, Lichtbogenbolzenschweien mit Hubzndung DVS-Verlag GmbH, Dsseldorf 1988 DVS-Merkblatt 0903, Lichtbogenbolzenschweien mit Spitzenzndung DVS-Verlag GmbH, Dsseldorf 1989 DVS-Merkblatt 0941, Flldrahtelektroden fr das Verbindungs- und Auftragschweien, Grundlagen und Begriffsbestimmung DVS-Verlag GmbH, Dsseldorf Mai 1991 DVS-Merkblatt 2901, Teil 1, Abbrennstumpfschweien DVS-Verlag GmbH, Dsseldorf 1988 DVS-Merkblatt 2909, Teil 1, Reibschweien von metallischen Werkstoffen DVS-Verlag GmbH, Dsseldorf 1989 DVS Merkblatt, Rohrlngsnahtschweien mit Rolltransformator DVS-Verlag GmbH, Dsseldorf 1974 DVS-Merkblatt 2934, Preschweien mit magnetisch bewegtem Lichtbogen
221 DVS-Verlag GmbH, Dsseldorf 1987 Dynamit Nobel Sprengplattierte Verbundwerkstoffe Eichhorn, F. Schweitechnische Fertigungsverfahren, Bd. 1, Schwei- und Schneidtechnologien VDI-Verlag GmbH, Dsseldorf 1983 Eichhorn, F., K. Blasig u. H. Nies Entwicklung eines Unterpulver-Engspaltschweikopfes fr Bandelektroden DVS-Berichte Bd. 100, 1985, S. 51-55
Literature
Eichhorn, F., E. Engindeniz, D. Pyrasch und J. Remmel Einsatzmglichkeiten des Elektrogas- und Elektroschlackeschweiens von Kehlnhten DVS-Berichte Bd. 90, 1984, S. 130-135 Eichhorn, F. u. H.W. Langenbahn Spritzerfreies MAGM-Impulslichtbogenschweien Schweien und Schneiden 37, 1985 Eichhorn, F. u. J. Remmel Leistungssteigerung des Elektroschlackeschweiverfahrens bei Verbindungen an niedriglegierten Sthlen im Blechdickenbereich von 100 bis 250 mm Schweien und Schneiden 37, 1985, Heft 11, S. 573-579 Ellis, D.J. Submerged-Arc Welding An Update Welding and Metal Fabrication, Okt. 1990 ESAB Firmenprospekt Eversheim, W. u. G. Luscell Stand und Entwicklungstendenzen der Programmierung von Robotern zum Bahnschweien Schweien und Schneiden 42, 1990, Heft 2, S. S.2ff Fischer, Baum Der Schutzgasschweier, Teil 1: WIG-, Plasmaschweien DVS-Schweitechnische Praxis Band 11 Grix, H. Tips fr den Werkstoffpraktiker zum Gasschweien und Flammlten Praktiker 32, 1980, Heft 2, S. 42-44 Grger, P. Engspaltschweien fgt dicke Bleche
222 VDI-Nachrichten Nr. 13, 1984, S. 32 Grger, P., G. Groten, D. Pyrasch u. H. Wietrzniok Neue Entwicklungen auf dem Gebiet des Schutzgas-Engspaltschweiens DVS-Bericht 127, 1989, S. 112-119 Grger, P. u. J. Koivula Metall-Schutzgasschweien Verfahrensvarianten des Engspaltschweiens Industrie-Anzeiger 106, 1984, Nr. 39, S. 28-33 Grnauer, H. Reibschweien von Metallen Expert-Verlag, Ehningen 1987 HAANE Firmenprospekt Hase, C. u. W. Reitze Lehrbuch des Gasschweiers und verwandte Autogenverfahren Verlag W. Girardet, Essen 1980 Hirschherg, H. Thermisches Schneiden, Stand der Entwicklung und Anwendung Technica 38, 1989, Heft 13, S. 67-73 Hrmann, E. Hochfrequenz-Widerstandsschweien mit Kontaktelektroden Schweien und Schneiden 12, 1960, Heft 10, S. 431-438 Hoult, A. P. Neuartige Erkenntnisse bei der Materialbearbeitung mit gepulsten Nd:YAG Hochleistungslasern im Kilowatt-Bereich Laser-Praxis, Juni 1989 Industrial Laser Review 1988 International Institute of Welding The Physics of Welding Pergamon Press, Frankfurt 1986 ISO 5182, Materials for resistance welding elektrodes and ancillary equipment, 1978 ISO 5184, Staight resistance spot welding electrodes, 1979 ISO 5821, Resistance spot welding electrode caps, 1979 Kessel, A.
Literature
223 Wirtschaftliches Schneiden von Baustahlblechen mit Luft-Plasma von 10A bis 70A Metallhandwerk und Technik, 1987, Heft 9, S. 743-744 Killing, R. Handbuch der Schweiverfahren, Teil 1: Lichtbogenschweiverfahren DVS-Fachbuchreihe Bd.76, DVS-Verlag GmbH, Dsseldorf 1984
Literature
King, F.J. Erhhung des Mechanisierungsgrades beim maschinellen Lichtbogenschweien durch Schweikopf positionierung und Fugengeometrieerfassung Dissertation RWTH Aachen, 1977 Kosfeld, G. Schweiverfahren DVS-Bericht Band 105 KUKA Firmenprospekt Laser Focus Annual Economic Survey 1989 Mair, M. Einflu der Sauerstoffreinheit auf die Schneidgeschwindigkeit und die Schneidkosten beim Laserstrahlbrennschneiden DVS Berichte Bd. 123, 1989 Marfels, W. Der Gasschweier Schweitechnische Praxis Bd.I, DVS-Verlag GmbH, Dsseldorf 1982 Marfels, W. Der Lichtbogenschweier Schweitechnische Praxis Bd.II, DVS-Verlag GmbH, Dsseldorf Marfels, W. u. A. Schneider Vorrichtungen in der Schweitechnik, Manahmen zur Rationalisierung der Fertigung DVS-Verlag GmbH, Dsseldorf 1989 Matzner, H.R. Qualittssteigerung beim spritzerarmen MAGM-Impulslichtbogenschweien durch Regelung der Prozegren Schweitechnische Forschungsberichte Bd. 40 DVS-Verlag GmbH, Dsseldorf 1991
Meleka, A.H. Electron Beam Welding Published for the Welding Institute by McGraw-Hill, 1971
Literature
Metzbower, E.A., D.W. Moon u. F.W. Fraser Laser welding of structural alloys Proceedings of International Conference on Welding Technology for Energy Applications, Gatlinburg, Tn, USA, May 1982 Meyer, C., A. Rosenthal u. V. Bdecker Festkrperlaser im kW-Betrieb Industrie Anzeiger 51/1988 Mller, P. u. L. Wolff Handbuch des Unterpulverschweiens Fachbuchreihe Schweitechnik Bd.63 Neff, F., P. Scherl, K. Winter u. H. Ornig Neue Verfahren zum Schweiplattieren dickwandiger Stahlbleche und -behlter Schweitechnik Berlin 7/74 Nies, H. u. H. Krebs UP-Formschweien mit Bandelektrode Oerlikon-Schweimitteilungen Mrz 1988 N.N. Laserstrahltechnologien in der Schweitechnik Fachbuchreihe Schweitechnik, Bd. 86, DVS-Verlag GmbH, 1989 Ortmann, R. Werkstoffe zum Verschleischutz DVS-Bericht Band 105 Pfeifer, L. Fachkunde des Widerstandsschweiens Girardet-Verlag, Essen 1969 Plasma-Technik AG Plasma Spraying Technique Wohlen (Schweiz) 1974
225 Neuere Entwicklungen beim Plasmaschneiden Trennen und Fgen, 1985, Heft 15, S. 55-58 Ruckdeschel, W. Plasmaheidraht-Auftragschweien Ein neues Plattierungsverfahren DVS-Bericht Band 23/1972 Ruge, J. Handbuch der Schweitechnik, Bd. II, Verfahren und Fertigung Springer-Verlag, Berlin Heidelberg New York 1980 Schfer, P. Industrielle Anwendungen von Festkrperlasern Laser und Optoelektronik, 2/1988 Schellhase, M. Der Schweilichtbogen ein technologisches Werkzeug VEB Verlag Technik, Berlin 1985 Schiller, S. et al. Elektronenstrahltechnologie Wissenschaftliche Verlagsgesellschaft mbH, Stuttgart, 1977 Schmidt, H. u. K. Ludewig Hochleistungs-Festkrperlaser Laser und Optoelektronik, 2/1988 Schultz, H. Elektronenstrahlschweien DVS-Verlag, Dsseldorf, 1989 Seiler, P. Schweien mit YAG-Laser Feinwerktechnik & Messtechnik, 96 (1988) 7-8 SOUDOMETAL Firmenprospekt Taylor D.S. u. C.E. Thornton High Deposition Rate Submerged-Arc Welding Welding Review, Aug. 1989 Tong S. u. Z. Ding Effect Of Plasma Spraywelding Technology On Dilution Wuhan (China) 1985 Tradowsky, Klaus Laser: Grundlagen, Technik, Basisanwendungen, Kamprath-Reihe Technik
Literature
Literature
Wahl, W. Auftragschweien Standzeitverlngerung durch gezielten Werkstoffeinsatz und optimale Schweiverfahren Schweien und Schneiden 6/79 Yamamoto, H. Recent Trends in Low Current Airplasma Cutting Welding International 55, 1987, S. 35-43
Lecture Notes
Table of Contents
Chapter 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
Subject Weldability of Metals TTT - Diagrams Residual Stresses Heat Treatment and its Function During Welding Welding Plain and Low Alloy Steels Welding High Alloy Steels Welding of Cast Materials Welding of Aluminium Welding Defects Testing of Welded Joints
1. Weldability of Metals
1. Weldability of Metals DIN 8580 and DIN 8595 classify welding into production technique main group 4 "Joining, group 3.6 "Joining by welding, Figure 1.1.
Figure 1.1
Weldability of a component is determined by three outer features according to DIN 8528, Part 1. This also indicates whether a given joining job can be done by welding, Figure 1.2.
Figure 1.2
1. Weldability of Metals
5 Material influence on weldability, i.e. welding suitability, can be detailed for a better understanding in three subdefinitions, Figure 1.3.
The chemical composition of a material and also its metallurgical properties are mainly set during its production, Figure 1.4. They have a very strong influence on the physical characteristics of the material. Process steps on steel manufacturing, shown in Figure 1.4, are the essential steps on the way to a processible and usable material. During manufacture, the requested chemical composition (e.g. by alloying) and metallurgical properties (e.g. type of teeming) of the steel are obtained. Figure 1.4
Another modification of the material beha viour takes place during subsequent treatment, where the raw material is rolled to processible semi-finished goods, e.g. like strips, plates, bars, profiles, etc.. With the rolling process, materialtypical transformation processes, hardening and precipitation processes are used to adjust an optimised material characteristics Figure 1.3
1. Weldability of Metals
A survey from quality point of view about the influence of the most important alloy elements to some mechanical and metallurgical properties is shown in Figure 1.5.
Figure 1.5
Figure 1.6 depicts the decisive importance of the carbon content to suitability of fusion welding of mild steels. A guide number of flawless fusion weldability is a carbon content of C < 0,22 %. with higher C contents, there is a danger of hardening, and welding becomes only possible by observing special precautions (e.g. pre- and post-weld heat treatment). Figure 1.6
1. Weldability of Metals
In addition to material beha viour, weldability is also essentially determined through the design of a component. The influence of the design is designated as welding safety, Figure 1.7.
Figure 1.7
The influence of the manufacturing process to weldability is called welding possibility, Figure 1.8. For example, a preand post-weld heat
treatment is not always possible, or grinding the weld surface before welding the
subsequent pass cannot be carried out (na rrow gap welding). Figure 1.8
2. TTT - Diagrams
2. TTT Diagrams
An essential feature of low alloyed ferrous materials is the crystallographic transformation of the body-centred cubic lattice which is stable at room temperature (-iron, ferritic structure) to the facecentred cubic lattice (-iron, austenitic structure), Figure 2.1. The temperature, where this transformation occurs, is not constant but depends on factors like alloy content, crystalline structure, tensional status, heating and cooling rate, dwell times, etc.. In order to be able to understand the basic processes it is necessary to have a look at the basic processes occuring in an idealized binary system. Figure 2.2 shows the state of a binary system with complete solubility in the liquid and solid state. If the melting of the L1 alloy is cooling down, the first crystals of the composition c1 are formed with reaching the temperature T1. These crystals are depicted as mixed crystal , since they consist of a compound of the components A (80%) and of B (20%). Further, a melting with the composition c0 is present at the temperature T1. With dropping temperature, the remaining melt is enriched with component B,
Temperature T TsA T1 T2 1 2 5 So L1 S L1
-Iron body-centered
-Iron face-centered
Figure 2.1
3 4 Li TsB Temperature T
following the course of line Li (liquidus line, up to point 4). In parallel, always new and B richer -mixed crystals are forming along the connection line So (solidus line, points 1, 2, 5). The dis-
- ss
A (Ni)
br-eI-02-02.cdr
c1
c2
c0
c3
c4
Concentration c
B (Cu)
Time t
Figure 2.2
2. TTT Diagrams
10
tribution of the components A and B in the solidified structure is homogeneous since concentration differences of the precipitated mixed crystals are balanced by diffusion processes. The other basic case of complete solubility of two components in the liquid state and of complete insolubility in the solid state shows Figure 2.3 If two components are completely insoluble in the solid state, no mixed crystal will be formed of A and B. The two liquidus lines Li cut in point e which is also designated as the eutectic point. The isotherm Te is the eutectic line. If an alloy of free composition solidifies according to Figure 2.3, the eutectic line must be cut. This is the temperature (Te) of the eutectic transformation: S A+B (T = Te = const.). This means that the melt at a constant temperature Te dissociates in A and B. If an alloy of the composition L2 solidifies, a purely eutectic structure results. On account of the eutectic reaction, the temperature of the alloy remains constant up to the completed transformation (critical point) (Figure 2.2). Eutectic
L1 TsA S 2 1 TsB Li Li L2 L1 L2
structures a
are
Temperature T
2 S+A
constituents. The alloy L1 will consist of a compound of alloy A and eutectic alloy E in the solid state.
So Te
S+B
3 4
A+E
B+E
A
br-eI-02-03.cdr
c1
ce
Concentration c
Time t
You can find further information on transformation behaviour in relevant specialist literature.
Figure 2.3
The definite use of the principles occurs in the iron-iron carbide diagram. Transformation behaviour of carbon containing iron in the equilibrium condition is described by the stable phase diagram iron-graphite (Fe-C). In addition to the stable system Fe-C which is specific for an equilibrium-close cooling, there is a metastable phase diagram iron cementite (Fe-Fe3C). During a slow cooling, carbon precipitates as graphite in accord with the stable system Fe-C,
2. TTT Diagrams
11
while during accelerated cooling, what corresponds to technical conditions, carbon precipitates as cementite in agreement with the metastable system (Fe-Fe3C). Per definition, iron carbide is designated as a structure constituent with cementite although its stoichiometric composition is identical (Fe3C). By definition, cementite and graphite can be present in steel together or the cementite can decompose to iron
melt
melt + graphite
and graphite during heat treatment of carbon rich alloys. However, it is fundamentally valid that the formation of cementite is encouraged with increasing cooling rate and decreasing carbon content. In a double diagram, the stable system is shown by a dashed, the metastable by a solid line, Figure 2.4.
solid sol.
+
solid sol.
Fe3C (cementite)
melt + austenite
Temperature C
austenite
melt + cementite
austenite + ferrite
ferrite
perlite
Mass % of Carbon
Figure 2.4
The metastable phase diagram is limited by the formation of cementite with a carbon content of 6,67 mass%. The strict stoichiometry of the formed carbide phase can be read off at the top X-coordinate of the molar carbon content. In accordance with the carbon content of Fe3C, cementite is formed at a molar content of 25%. The solid solutions in the phase fields are designated by Greek characters. According to convention, the transition points of pure iron are marked with the character A - arrt (stop point) and distinguished by subjacent indexes. If the transition points are determined by cooling curves, the character r = refroidissement is additionally used. Heat-up curves get the supplement c - chauffage. Important transition points of the commercially more important metastable phase diagram are:
1536 solidification temperature (melting point) -iron, C: 1392 A4- point - iron, C: 911 A3- point non-magnetic - iron, C: 723 A1- point (perlite point). C:
ledeburite
2. TTT Diagrams
12
The corners of the phase fields are designated by continuous roman capital letters. As mentioned before, the system iron-iron carbide is a more important phase diagram for technical use and also for welding techniques. The binary system iron-graphite can be stabilized by an addition of silicon so that a precipitation of graphite also occurs with increased solidification velocity. Especially iron cast materials solidify due to their increased silicon contents according to the stable system. In the following, the most important terms and transformations should be explained more closely as a case of the metastable system. The transformation mechanisms explained in the previous sections can be found in the binary system iron-iron carbide almost without exception. There is an eutectic transformation in point C, a peritectic one in point I, and an eutectoidic transformation in point S. With a temperature of 1147 and a carbon concentration of 4.3 mass%, the eutectic phase called LeC deburite precipitates from cementite with 6,67% C and saturated -solid solutions with 2,06% C. Alloys with less than 4,3 mass% C coming from primary austenite and Ledeburite are called hypoeutectic, with more than 4,3 mass% C coming from primary austenite and Ledeburite are called hypereutectic.
If an alloy solidifies with less than 0,51 mass percent of carbon, a -solid solution is formed below the solidus line A-B (-ferrite). In accordance with the peritectic transformation at 1493 melt (0,51% C) and -ferrite (0,10% C) decompose to a -solid solution (austenite). C,
The transformation of the -solid solution takes place at lower temperatures. From -iron with C-contents below 0.8% (hypoeutectoidic alloys), a low-carbon -iron (pre-eutectoidic ferrite) and a fine-lamellar solid solution (perlite) precipitate with falling temperature, which consists of -solid solution and cementite. With carbon contents above 0,8% (hypereutectoidic alloys) secondary cementite and perlite are formed out of austenite. Below 723 tertiary cementite C, precipitates out of the -iron because of falling carbon solubility.
The most important distinguished feature of the three described phases is their lattice structure. - and -phases are cubic body-centered (CBC lattice) and -phase is cubic facecentered (CFC lattice), Figure 2.1.
2. TTT Diagrams
13
Different carbon solubility of solid solutions also results from lattice structures. The three above mentioned phases dissolve carbon interstitially, i.e. carbon is embedded between the iron atoms. Therefore, this types of solid solutions are also named interstitial solid solution. Although the cubic face-centred lattice of austenite has a higher packing density than the cubic body-centred lattice, the void is bigger to disperse the carbon atom. Hence, an about 100 times higher carbon solubility of austenite (max. 2,06% C) in comparison with the ferritic phase (max. 0,02% C for -iron) is the result. However, diffusion speed in -iron is always at least 100 times slower than in -iron because of the tighter packing of the -lattice.
Although - and -iron show the same lattice structure and properties, there is also a difference between these phases. While -iron develops of a direct decomposition of the melt (S ), -iron forms in the solid phase through an eutectoidic transformation of austenite ( + Fe3C). For the transformation of non- and low-alloyed steels, is the transformation of ferrite of lower importance, although this -phase has a special importance for weldability of high alloyed steels. Unalloyed steels used in industry are multi-component systems of iron and carbon with alloying elements as manganese, chromium, nickel and silicon. Principally the equilibrium diagram Fe-C applies also to such multi-component systems. Figure 2.5 shows a schematic cut through the three phase system Fe-M-C. During precipitation, mixed carbides of the general composition M3C develop. In contrast to the binary system Fe-C, is the three phase system
Ac3
three-phase field + + M3C. The beginning of the transformation of + M3C to is marked by Aclb, the end by Acle. The indices b
br-eI-02-05.cdr
Figure 2.5
2. TTT Diagrams
14
The described equilibrium diagrams apply only to low heating and cooling rates. However, higher heating and cooling rates are present during welding, consequently other structure types develop in the heat affected zone (HAZ) and in the weld metal. The struc C
ture transformations during heating and cooling are described by transformation diagrams, where a temperature change is not carried out close to the equilibrium,
s
but A
at
different
heating
br-eI-02-06.cdr
Figure 2.6
during isothermal austenitizing shows Figure
ASTM4; L=80m
ASTM11; L=7m
2.6. This figure must be read exclusively along the time axis! It can be recognised that several transformations during isothermal austenitizing occur with e.g. 800 Inhomogeneous austenC. ite means both, low carbon containing austenite is formed in areas, where ferrite was present before transformation, and carbon-rich austenite is formed in areas during transformation, where carbon was present before transformation. During sufficiently long annealing times, the concentration differences are balanced by diffusion, the border to a homogeneous austenite is passed. A growing of the
20m
20m
Temperature
Time
br-er02-07.cdr ISF 2002
austenite grain size (to ASTM and/or in m) can here simultaneously be observed with longer annealing times.
Figure 2.7
2. TTT Diagrams
15
The influence of heating rate on austenitizing is shown in Figure 2.7. This diagram must only be read along the sloping lines of the same heating rate. For better readability, a time pattern was added to the pattern of the heating curves. To elucidate the grain coarsening during austenitizing, two microstructure photographs are shown, both with different grain size classes to ASTM. Figure 2.8 shows the relation between the TTA and the Fe-C diagram. It's obvious that the Fe-C diagram is only valid for infinite long dwell times and that the TTA diagram applies only for one individual alloy. Figure 2.9 shows the different
Ac1e Ac1b
Ac3
passes during austenitizing and down. The heating period is composed of a continuous and an isothermal section.
br-eI-02-08.cdr
During cooling down, two different ways of heat control can be distinguished: 1. : During continuous a
Figure 2.8
Ac3
temperature
continuous
Ac1e
control
cooling is carried out with a constant cooling rate out of the area of the homogene-
Ac1b
isothermal
ous and stable austenite down to room temperature. 2. : During isothermal control a
temperature
br-eI-02-09.cdr
Figure 2.9
2. TTT Diagrams the area of the homogeneous and stable austenite down to room temperature.
16
2. : During isothermal temperature control a quenching out of the area of the austenite is carried out into the area of the metastable austenite (and/or into the area of martensite), followed by an isothermal holding until all transformation processes are completed. After transformation will be cooled down to room temperature.
Figure
2.10
shows
the
time-temperature diagram of a isothermal transformation of the mild steel Ck 45. Read such diagrams only along the time-axis! Below the Ac1b line in this figure, there is the area of the metastable austenite, marked with an A. The areas
marked with F, P, B, und M represent areas where ferFigure 2.10 rite, perlite, Bainite and martensite are formed. The lines which limit the area to the left mark the beginning of the formation of the respective structure. The lines which limit the area to the right mark the completion of the formation of the respective structure. Because the ferrite formation is followed by the perlite formation, the completion of the ferrite formation is not determined, but the start of the perlite formation. Transformations to ferrite and perlite, which are diffusion controlled, take place with elevated temperatures, as diffusion is easier. Such structures have a lower hardness and strength, but an increased toughness.
Diffusion is impeded under lower temperature, resulting in formation of bainitic and martensitic structures with hardness and strength values which are much higher than those of ferrite and perlite. The proportion of the formed martensite does not depend on time. During quenching to holding temperature, the corresponding share of martensite is spontanically formed. The present rest austenite transforms to Bainite with sufficient holding time. The right
2. TTT Diagrams
17
detail of the figure shows the present structure components after completed transformation and the resulting hardness at room temperature. Figure 2.11 depicts the graphic representation of the TTT diagram, which is more important for welding techniques. This is the TTT diagram for continuous cooling of the steel Ck 15. The diagram must be read along the drawn cooling passes. The lines, which are limiting the individual areas, also depict the beginning and the end of the respective transformation. Close to the cooling curves, the amount of the formed structure is indicated in per cent, at the end of each curve, there is the hardness value of the structure at room temperature.
Figure 2.12 shows the TTT diagram of an alloyed steel containing approximately
the same content of carbon as the steel Ck 15. Here you can see that all transformation processes are
strongly postponed in relation to the mild steel. A completely transformation Figure 2.11 marte nsitic is carried
out up to a cooling time of about 1.5 seconds, compared with 0.4 seconds of Ck 15. In addition, the completely diffusion controlled transformation processes of the perlite area are postponed to clearly longer times.
different, Figure 2.13. With Figure 2.12 this carbon content, a pre-
2. TTT Diagrams
18 eutectoid ferrite formation cannot still be carried out (see also Figure 2.3). The term of the figures 2.9 to 2.11 "austeniti zing temperature means the temperature, where the workpiece transforms to an auste nitic microstructure in the course of a heat treatment. Dont mix up this temperature with the AC3 temperature, where above it there is only pure auste nite. In addition you can see that only martensite is formed from the austenite, provided that the cooling rate is sufficiently high, a formation of any other
microstructure is completely depressed. With this type of transformation, the steel gains the highest hardness and strength, but loses its toughness, it embrittles. The slowest cooling rate where such a transformation happens, is Figure 2.13 called critical cooling rate.
Figure 2.14
Figure 2.15
2. TTT Diagrams
19
Figure 2.14 shows schematically how the TTT diagram is modified by the chemical composition of the steel. The influence of an increased austenitizing temperature on transformation beha viour shows Figure 2.15. Due to the higher hardening temperature, the grain size of the austenite is higher (see Figure 2.6 and 2.7).
This grain growth leads to an extension of the diffusion lengths which must be passed during the transformation. As a result, the "noses" in the TTT diagram are shifted to longer times. The lower part of the figure shows the proportion of formed martensite and
Bainite depending on cooling time. You can see that Figure 2.16 with higher austenitizing
temperature the start of Bainite formation together with the drop of the martensite proportion is clearly shifted to longer times. As Bainite formation is not so much impeded by the coarse austenite grain as with the completely diffusion controlled processes of ferrite and perlite formation, the maximum Bainite Figure 2.17 proportion is increased
2. TTT Diagrams
20
Due to the strong influence of the austenitizing temperature to the transformation behaviour of steel, the welding technique uses special diagrams, the so called Welding -TTT-diagrams.
They are recorded following the welding temperature cycle with both, higher austenitizing temperatures (basically between 950 and 1350C) and shorter a usteniti zing times. You find two examples in Figures 2.16 and 2.17.
Figure 2.18 proves that the iron-carbon diagram was developed as an equilibrium diagram for infinite long cooling time and that a TTT diagram applies always only for one alloy.
Figure 2.18
3. Residual Stresses
3. Residual Stresses
22
The emergence of residual stresses can be of very different nature, see three
tension
Figure
3.2
details
the
pressure
tension
duced workpiece, material, production-, and wearcaused residual stresses are overlaying in such a way that a certain condition of residual stresses is created. Such a workpiece shows in service more or
weld
br-eI-03-01e.cdr
ISF 2002
Figure 3.1
less residual stresses, and it will never be stress-free! Figure 3.3 defines residual stresses of 1., 2., and 3. type. This grading is independent from the origin of the residual stresses. It is rather based on the three-dimensional extension of the stress conditions. Based on this definition, FigAnalysis of Residual Stress Development
ure 3.4 shows a typical distribution of residual stresses. Residual stresses, which build-up around dislocations and other lattice imperfections (III), superimpose within a
forming
deforming residual stresses due to inhomogenuous deformationanisotropy
relevant material
production
wear
mechanical e.g. partial-plastic deformation of notched bars or close to inclusions, fatigue strain
residual
Figure 3.2
stresses in a transition-free
3. Residual Stresses
23
steel cylinder is shown in Figures 3.5. and 3.6. During water quenching of the homogeneous heated cylinder, the edge of the cylinder cools down faster than the core. Not before 100 seconds have elapsed is the temperature across the cylinder's cross section again
tension s
s III
s II
+
Residual stresses of the I. type are almost homogenuous across larger material areas (several grains). Internal forces related to residual stresses of I. type are in an equilibrium with view to any cross-sectional plane throughout the complete body. In addition, the internal torques related to the residual stresses with reference to each axis disappear. When interfering with force and torque equilibrium of bodies under residual stresses of the I. type, macroscopic dimension changes always develop.
sI
x
0
-
Residual stresses of the II. type are almost homogenuous across small material areas (one grain or grain area). Internal forces and torques related to residual stresses of the II. type are in an equilibrium across a sufficient number of grains. When interfering with this equilibrium, macroscopic dimension changes may develop.
x
0
Residual stresses of the III. type are inhomogenuous across smallest material areas (some atomic distances). Internal forces and torques related to residual stresses of the III. type are in an equilibrium across small areas (sufficiently large part of a grain). When interfering with this equilibrium, macroscopic dimension changes do not develop.
grain boundaries
sIII
= residual stresses between several grains = residual stresses in a single grain = residual stresses in a point
ISF 2002
br-er03-03e.cdr
ISF 2002
Figure 3.3 homogeneous. The left part of Figure 3.5 shows the T-tcurve of three different measurement points in the cylinder.
Temperature
Figure 3.4
1000 C 900
1000 C 750
0s
10 s
5s
1s
800
2 3
Temperature
15 s
20 s
1
35 mm diameter water cooling
25 s
Figure 3.6 shows the results of quenching on the stress condition in the cylinder. At the beginning of cooling, the cylinder edge starts shrinking faster than the core (upper
500
250
MS
35 s
400 300 200
45 s
53 s
0 -2 10
68 s
10-1 10-0 101 102 Cooling time 103
s 104
100
280 s
0 17,5
br-eI-03-05e.cdr
14 10,5
7 3,5
0 3,5
Radius
7 10,5
ISF 2002
mm 17,5
figure). Through the stabilising effect of the cylinder core, Figure 3.5
3. Residual Stresses
24
tensile stress builds up at the edge areas while the core is exposed to pressure stress. Resulting volume differences between core and edge are balanced by elastic and plastic deformations. When cooling is completed, edge and core are on the same temperature level, the plastically stretched edge now supports the unstressed core, so that pressurestresses are present in the edge areas and tensile residual stresses in the core.
tension pressure
300
N/mm
E D
tension
200
100
tension
pressure
tension
-100
pressure tension
-200 B
B'
pressure
-300
0
ISF 2002
br-er03-06e.cdr
200
400
600
Figure 3.6
Figure 3.7
These changes are principally shown once again in Figure 3.7 with the 3-rod model. A warming of the middle rod causes at first an elastic expansion of the outer rods, the inner rod is exposed to pressure stress (line A-B). Along the line B-C the rod is plastically deformed, because pressure stresses have exceeded the yielding point. At point C, the cooling of the rod starts, it is exposed to tensile stress due to shrinking. Along the line D-E the rod is plastically deformed due to the influence of the counter members beeing in tension. At the point E the system has cooled down to its initial temperature. This point represents the remaining residual stress condition of this construction. If heating is stopped before point C is reached and cooled down to the initial temperature, then stress increase in the centre rod will be in parallel
3. Residual Stresses
25
with the elastic areas. Starting with point B, the same residual stress condition is present as in a case of heating up to a temperature above 600C. Figure 3.8 divides the development of residual stresses in welded seams in three different mechanisms. Shrinking stresses: these are stresses formed through uniform cooling of the seam. Caused by expansion restriction of the colder areas at the edge of the weld and base material , tensile stresses develop along and crosswise to the seam. Quenching stresses: If cooling is not homogenous, the surface of the weld cools down faster than the core areas. If the high-temperature limit of elasticity is exceeded due to buildup stress differences, pressure stresses will be present at the weld surface after cooling. In contrast, the core shows tensile stresses in cold condition (see also Figure 3.6). Transition stresses: Transitions in the ferrite and perlite stage cause normally only residual stresses, because within this temperature range the yield strength of the steel is so low that generated stresses can be undone by plastic deformations. This is not the case with transitions in the Bainite and martensite stage. A transition of the austenite causes an increase in volume (transition cfc in cbc, the cfc lattice has a higher density, additional volume increase through lattice deformation). In the case of a homogenous transition, the weld will consequently unfold pressure stresses. If the transition of
+x +y -x
the edge areas happens earlier than the transition of the slower cooling core, plastic deformations of the core area may be present similar to quenching (see above: quenching stresses). In this case, the weld surface will
3. Transformation stresses
-x
-y
1. Shrinking stresses
+s +y
2. Quenching stresses
-x
+x
+x
-s -y
show tensile stresses after cooling. Generally these mechanisms cannot be separated accurately from each other, thus the residual stress condition of a weld will represent an overlap of the cases as shown in the 3rd figure. This overlap of the different mechanisms makes a forecast of the remaining residual stress condition difficult.
homogenuous transformation
+s +y
inhomogenuous transformation
-x
+x
-x
+x
-s -y
br-er03-08.cdr
-s -y
ISF 2002
Figure 3.8
3. Residual Stresses
26
Stress distribution sX
stresses crosswise to a welded seam in analogy to the 3-rod model of Figure 3.7. This figure considers only shrinking residual stresses. Before application of welding heat, the seam area is stress-free (cut A-A). At the weldpool
2. cutt B-B
tension
weldpool B
area of plastic deformations
can be found (cut B-B), metal is liquid. At this point, there are no residual stresses, because molten metal cannot transmit forces at the weldpool. Areas close to the joint expand
3. cut C-C
M'
4. cut D-D
DT = 0
areas which are not so close to the seam. Thus, areas close to the joint show compres ISF 2002
br-er03-09e.cdr
sion stress, areas away from the joint tensile stress. In cut C-C the already solidified weld metal starts to shrink and is supported by
Figure 3.9 areas close to the seam, the weld metal shows tensile stresses, the adjacent areas compression stresses. In cut D-D is the temperature completely balanced, a residual stress condition is recognised as shown in the lower right figure.
1 3 15 mm
15 mm
material S235JR (St 37)
103 a a
Figure 3.10 shows how much residual stresses are influenced by constraining effects of adjacent material. The resulting stress in the presented case is calculated according to Hooke: = E Elongation is calculated as l/a ( l is the length change due to shrinking). With conFigure 3.10
1. 2. 3. 4. 5. a = 100 mm a = 150 mm a = 200 mm a = 250 mm a = 300 mm s = 800 N/mm s = 530 N/mm s = 400 N/mm s = 300 N/mm s = 270 N/mm
br-er03-10e.cdr
ISF 2002
3. Residual Stresses
27
stant joint volume will shrinking and l always have the same value. Thus the elongation depends only on the value a. The smaller the a is chosen, the higher are the resulting stresses. Effects of transition on cooling can be estimated from Figure 3.11. Here curves of temperature- and length-changes of ferritic and austenitic steels are drawn. It is clear that a ferritic lattice has a higher volume than an austenitic lattice at the same temperature. A steel which transforms from austenite to one of the ferrite types increases its volume at the critical point. This sudden rise in volume can be up to 3% in the case of martensite formation.
Longitudinal expansion Dl
firm clamping
fe
ste
nit ic
ste
rri
tic
ste
el
el
force sensor
thermo couples
links
to calculator
au
800
1000
C
800
Temperature
600
14
elektrode
Force
400
600
200
400
force
0 200
Temperature [C]
temperature
-200 0 -1 10 100 101 102 Time
br-er03-12e.cdr
103
104
105
br-er03-11e.cdr
ISF 2002
ISF 2002
Figure 3.11
Figure 3.12
To record the effects of this behaviour on the stress condition of the weld, sample welds are carried out in the test device outlined in Figure 3.12. Thermo couples measure the T-t curve at the weld seam, a force sensor records the force which tries to bend the samples. The lower picture shows the results of such a test. The temperature behaviour at the fusionline as well as the force necessary to hold the sample over the time is plotted.
3. Residual Stresses
28
In the temperature range above 600 the force sensor registers a tensile force which is C caused by the shrinking of the austenite. Between 600 and 400 a large drop in force can C be seen, which is caused by the transition of the austenite. The repeated increase of the force is based on further shrinking of the ferrite. With the help of TTT diagrams of base material and welding consumable, the transition temperatures and/or temperature areas for the individual zones of the welded joint can be determined. With these temperature it can be clearly determined in which part of
br-eI-03-13e.cdr
steel consumable electrode sample shape (V-groove, 60) type of welding position of the HAZ
austenitic austenitic
surface weld
surface weld
surface weld
pressure
tension
ISF 2002
the curve the force drop is caused by the transition of the welding consumable and in Figure 3.13
zone (HAZ). These results can be used to determine the longitudinal residual stresses transversal to the joint, as shown in Figure 3.13. During welding of austenitic transition-free materials only tensile residual stresses are caused in the welded area according to Figure 3.8. If an austenitic electrode is welded to a StE 70, transitions occur in the area of the heat affected zone which lead to a decrease of tensile stresses. If a high-strength electrode which has a martensitic transition, is welded
28'
151'
140 %
Angle change
100 80 60 40 20
f = 1
f = 3
f = 7
f = 13
a=5
a=7
a=9
a = 12,5
to a StE 70, then there will be pressure residual stresses in the weld metal and tensile residual stresses in the HAZ.
br-er03-14e.cdr
ISF 2002
Figure 3.14
3. Residual Stresses
29
If parts to be welded are not fixed, the shrinking of the weld will cause an angular distortion of the workpieces, Figure 3.14 . If the workpieces can shrink unrestricted in this way, the remaining residual stresses will be much lower than in case with firm clamping. Methods to determine residual stresses can be divided into destructive, nona
WSG
section
plan
destructive, and conditionally destructive methods. The borehole and ring core method can be considered as conditionally destructive, Figures 3.15 and 3.16. In both cases, present rebr-eI-03-15e.cdr
b
workpiece
ISF 2002
sidual stresses are released through partial material removal and the resulting deformations are then Figure 3.15
measured by wire strain gauges. An essential advantage of the borehole method is the very small material removal, the diameter of the borehole is only 1 to 5 mm, the bore depth is 1- to 2-times the borehole diameter. The disadvantage here is that only surface elongations can be measured, thus the results are limited residual stresses in the surface area of the workpiece.
wire expansion gauge
b(sb) 45 c(ec)
z
t0.21
used to mill a ring groove around a three-axes wire strain gauge. The core is released from the force effects and stress-relieved. At the time when the resil-
D1.58
a(sa)
s1 (z)
s2 (z)
measurement point
br-eI-03-16e.cdr
ISF 2002
ience of the core is measured, the detection of the residual stress distribution
Figure 3.16
3. Residual Stresses
30
across the depth is also possible. Both methods are limited in their suitability for measuring welding residual stresses, because steep strain gradients in the HAZ may cause wrong measurements. The table in Figure 3.17 shows a survey of measurement methods for residual stresses and what causes residual stresses to be picked-up when using one of the respective methods.
causes
thermal processes mechanical deformations surfacetreatment
measurement
non-destructive
others
ring groove
optical procedures
bore hole
ultra sonic
ring core
A E A E
A E A E
A E A E
A E
x - ray
A E A E A E
magnetic
Figure 3.17
measured variable
bending deflection f curves reduced curves
residual stresses
sy sz tzy
biaxial any uniaxial locally different linear, tensile residual stresses on top, down pressure stresses tripleaxial independent of smple length sL, sT, sR
cutting-in
f
tear f
drilling
e45 eT eL
sL sT sR
slitting
0.46f
tear f
br-eI-03-18e.cdr
ISF 2002
Figure 3.18
optical procedures
A E A E
cam web
32
When welding a workpiece, not only the weld itself, but also the surrounding base material (HAZ) is influenced by the supplied heat quantity. The temperature-field, which appears around the weld when different welding procedures are used, is shown in Figure 4.1.
Figure 4.2 shows the influence of the material properties on the welding process. The determining factors on the process presented in this Figure, like melting temperature and interval, heat capacity, heat extension etc, depend greatly on the chemical composition of the material. Metallurgical properties are here characterized by e.g. homogeneity, structure and texture, physical properties like heat extension, shear strength, ductility. Figure 4.1 Structural changes, caused by the heat input
(process 1, 2, 7, and 8), influence directly the mechanical properties of the weld. In addition, the chemical composition of the weld metal and adjacent base material are also influenced by the processes 3 to 6.
Based on the binary system, the formation of the different structure zones is shown in Figure 4.3. So the coarse grain zone occurs in areas of intensely elevated
austenitising temperature for example. At the same time, hardness peaks appear in these areas because of
greatly reduced critical cooling rate and the coarse austenite Figure 4.2
4. Heat Treatment and its Function During Welding grains. This zone of the weld is the area, where the worst toughness values are found.
33
In Figure 4.4 you can see how much the formation of the individual structure zones and the zones of unfavourable mechanical properties can be influenced. Applying an electroslag one pass weld of a 200 mm thick plate, a HAZ of approximately 30 mm width is achieved. Using a three pass technique, the HAZ is reduced to only 8 mm.
With the use of different procedures, the differences in the formation of heat affected zones become even clearer as shown in Figure 4.5. These effects can actively be used to the advantage of the material, for example to adjust Figure 4.3
calculated mechanical properties to one's choice or to remove negative effects of a welding. Particularly with high-strength fine grained steels and high-alloyed materials, which are specifically optimised to achieve special quality, e.g. corrosion resistance against a certain attacking medium, this post-weld heat treatment is of great importance.
Figure 4.6 shows areas in the Fe-C diagram of different heat treatment methods. It is clearly visible that the carbon content (and also the content of other alloying elements) has a distinct influence on the level of annealing temperatures like e.g. coarse-grain
Figure 4.4
4. Heat Treatment and its Function During Welding heat treatment or normalising.
34
It can also be seen that the start of martensite formation (MS-line) is shifted to continuously
Figure 4.5
Figure 4.6
decreasing temperatures with increasing C-content. This is important e.g. fo r hardening processes (to be e xplained later).
As this diagram does not cover the time influence, only constant stop-
temperatures can be read, predictions about heating-up and cooling-down rates are not possible. Thus the individual heat treatment methods will be explained by their temperature-time-
35
Figure 4.7 shows in the detail to the right a T-t course of coarse grain heat treatment of an alloy containing 0,4 % C. A coarse grain heat treatment is applied to create a grain size as large as possible to improve machining properties. In the case of welding, a coarse grain is unwelcome, although unavoidable as a consequence of the welding cycle. You can learn from Figure 4.7 that there are two methods of coarse grain heat treatment. The first way is to austenite at a temperature close above A3 for a couple of hours followed by a slow cooling process. The second method is very important to the welding process. Here a coarse grain is formed at a temperature far above A 3 with relatively short periods. Figure 4.8 shows schematically time-temperature behaviour in a TTT-
diagram. (Note: the curves explain running structure mechanisms, they must not be used as reading off examples. To determine t8/5, hardness values, or microstructure distribution, are TTT-diagrams always read continuously Figure 4.8 curves 3 to 6 are not a llowed for this purpose!). mally. Mixed or isotherlike
types
The most important heat treatment methods can be divided into sections of annealing, hardening and tempering, and these single processes can be used individually or combined. The normalising process is shown in Figure 4.9. It is used to achieve a homogeneous ferrite perlite structure. For this purpose, the steel is heat treated approximately 30C above Ac3 until homogeneous auste nite evolves. This condition is the starting point for the following hardening and/or quenching and tempering treatment. In the case of hypereutectoid steels, austenisation takes place above the A1 temperature. Heating-up should be fast to keep the austenite grain as fine as possible (see TTA-diagram, chapter 2). Then air cooling follows, leading normally to a transformation in the ferrite condition (see Figure 4.8, line 1; formation of ferrite and perlite, normalised micro-structure).
36 To harden a material, austenisation and homogenisation is carried out also at 30C above AC3. Also in this case one must watch that the austenite grains remain as small as possible. To ensure a complete transformation to marte nsite, a subsequent quenc hing follows until the
temperature is far below Figure 4.9 the Ms-temperature, Figure 4.10. The cooling rate during quenching must be high enough to cool down from the auste nite zone directly into the martensite zone without any further phase transitions (curve 2 in Figure 4.8). Such quenching processes build-up very high thermal stresses which may destroy the workpiece during hardening. Thus there are variations of this process, where perlite formation is suppressed, but due to a smaller temperature gradient thermal stresses remain on an uncritical level (curves 3 and 4 in Figure 4.8). This can be achieved in practice for example- through stopping a water quenc hing
process at a certain temperature and continuing the cooling with a milder cooling medium (oil). With longer holding on at elevated temperature level, transformations can also be carried through in the bainite area (curves 5 and 6). Figure 4.10
37
Figure 4.11 shows the quenching and tempering procedure. A hardening is followed by another heat treatment below Ac1. During this tempering process, a break down of
C 900 austenite + ferrite
Temperature
austenite
martensite
takes
place.
Ferrite and cementite are formed. As this change causes a very fine microstructure, this heat treatment leads to very good mechanical properties like e.g. strength and toughness.
A1
quenching
slow cooling
300 0,4
br-eI-04-11.cdr
0,8 C-Content
Time
Figure 4.11
Figure 4.12 shows the procedure of soft-annealing. Here we aim to adjust a soft and suitable micro-structure for machining. Such a structure is characterised by mostly globular formed cementite particles, while the lamellar structure of the perlite is resolved (in Figure 4.12 marked by the circles, to the left: before, to the right: after soft-annealing). For hypoeutectic steels, this spheroidizing of cementite is achieved by a heat treatment close below A1. With these steels, a part of the cementite bonded carbon dissolves during heat treating close below A1, the remaining cementite lamellas transform with time into balls, and the bigger ones grow at the expense of the smaller ones (a transformation is carried out beTemperature
C 900
austenite
cause the surface area is strongly reduced thermodynamically more favourable condition). Hypereutectic steels have in addition to the lamellar structure of the perlite a
A3 A1
or
br-eI-04-12.cdr
Soft Annealing
Figure 4.12
38
Spheroidizing of cementite is achieved by making use of the transformation processes during oscillating around A1. When exceeding A1 a transformation of ferrite to auste nite takes place with a simultaneous solution of a certain amount of carbon according to the binary system Fe C. When the temperature drops below A1 again and is kept about 20C below until the transformation is completed, a re-precipitation of cementite on existing nuclei takes place. The repetition of this process leads to a stepwise spheroidizing of
Figure 4.13 shows the principle of a stress-relieve heat treatment. This heat treatment is used to eliminate dislocations which were caused by welding, deforming, transformation etc. to improve the toughness of a workpiece. Stress-relieving works only if present dislocations are able to move, i.e. plastic structure deformations must be executable in the micro-range. A temperature the increase is
commonly
used
cause the yield strength limit decreases with increasing temperature. A stress-relieve heat treatment should not cause any other change to properties, so that tempering steels
Figure 4.14
4. Heat Treatment and its Function During Welding are heat treated below tempering temperature.
39
Figure 4.14 shows a survey of heat treatments which are important to welding as well as their purposes.
Figure 4.15 shows principally the heat treatments in connection with welding.
Heat treatment processes are divided into: before, during, and after welding. Normally a stress-relieving or normalizing heat treatment is applied before
welding to adjust a proper material condition which for welding. After welding, alFigure 4.15 most any possible heat treatment can be carried out. This is only limited by workpiece dimensions/shapes or arising costs. The most important section of the diagram is the kind of heat treatment which accom-panies the welding. The most important processes are e xplained in the follo wing.
Figure 4.16 represents the influence of different accompanying heat treatments during welding, given within a TTT-diagram. The fastest cooling is achieved with welding without preheating, with addition of a small share of bainite, mainly martensite is formed (curve 1, analogous to Figure 4.8, hardening). A simple heating before welding without additional stopping time lowers the cooling rate according to curve 2. The proportion of martensite is reduced in the forming structure, as well as the Figure 4.16
40
level of hardening. If the material is hold at a temperature above MS during welding (curve 3), then the martensite formation will be completely suppressed (see Figure 4.8, curve 4 and 5).
To explain the temperature-time-behaviours used in the following, Figure 4.17 shows a superposition of all individual influences on the materials as well as the resulting T-Tcourse in the HAZ. As an example, welding with simple preheating is selected. The plate is preheated in a period t . After V removal of the heat source, the cooling of the workpiece starts. When t S is reached, welding starts, and its temperature peak overlays the cooling curve of the base material. When the welding is completed, cooling period tA starts. The full line represents the resulting temperature-time-behaviour of the HAZ.
The temperature time course during welding with simple preheating is shown in Figure Figure 4.17 4.18. During a welding time tS a drop of the working temperature TA occurs. A further air cooling is usually carried out, however, the cooling rate can also be reduced by cove ring with heat insulating materials.
Another variant of welding with preheating is welding at Figure 4.18 constant working This is
temperature.
41 achieved through further warming during welding to avoid a drop of the working temperature. In Figure 4.19 is this case (dashed line, TA needs not to be above MS) as well as the special case of isothermal welding illustrated. During isothermal welding, the workpiece is heated up to a working temperature above MS
Figure 4.19
after welding until a transformation of the austenitised areas has been completed. The aim of isothermal welding is to cool down in accordance with curve 3 in Figure 4.16 and in this way, to suppress martensite formation.
Figure 4.20 shows the T-T course during welding with post-warming (subsequent heat treatment, see Figure 4.15). Such a treatment can be carried out very easy, a gas welding torch is normally used for a local preheating. In this way, the toughness properties of some steels can be greatly improved. The lower sketch shows a combination of pre- and postheat treatment. Such a treatment is applied to steels which have such a strong tendency to hardening that a cracking in spite of a simple preheating before welding cannot be avoided, if they cool down directly from working temperature. Such materials are heat treated immediately after welding at a temperature between 600 and 700C, so that a formation Figure 4.20
4. Heat Treatment and its Function During Welding of martensite is avoided and welding residual stresses are eliminated simultaneously.
42
not be discussed here, Figure 4.21. Such treatments are used for transformationinert materials. The aim of the figure is to show how complicated a heat treatment can become for a material in combination with welding.
Figure 4.22 shows temperaFigure 4.21 represents the T-T course of a point in the HAZ in the first pass. The root pass was welded without preheating. Subsequent passes were welded without cooling down to a certain temperature. As a result, working temperature increases with the number of passes. The second pass is welded under a preheat temperature which is already above martensite start temperature. The heat which remains in the workpiece preheats the upper layers of the weld, the root pass is post-heat treated through the same effect. During welding of the last pass, the preheat temperature has reached such a high level that the critical cooling rate will not be surpassed. A f vourable effect of a multi-pass welding is the warming of the HAZ of each previous pass above recrystallisation temperature with the corresponding crystallisaFigure 4.22 ture distribution during multipass welding. The solid line
43
tion effects in the HAZ. The coarse grain zone with its unfavourable mechanical properties is only present in the HAZ of the last layer. To achieve optimum mechanical values, welding is not carried out to Figure 4.22. As a rule, the same welding conditions should be applied for all passes and prescribed t8/5 times must be kept, welding of the next pass will not be carried out before the previous pass has cooled down to a certain temperature (keeping the interpass temperature). In addition, the workpiece will not heat up to excessively high temperatures.
Figure 4.23 shows a nomogram where working te mperature and minimum and maximum heat input for some steels can be interpreted, depending on carbon equivalent and wall thickness. If e.g. the water quenched and tempered fine grain structural steel S690QL of 40 mm wall thickness is welded, the following data can be found:
- minimum heat input between 5.5 and 6 kJ/cm - maximum heat input about 22 kJ/cm - preheating to about 160C - after welding, residual stress relieving between 530 and 600C.
soaking
must
treated with a hydrogen relieve annealing. Above this area, a stress relieve annealing must be carried out. Below this area, a post-weld heat treatment is not required.
Figure 4.23
45
o t i e D n f
e d f s n i B r g
t a l e h S s
g n u e l t i E
h r a m c e d t l S
h Z s a n i c m e u
s n m e g u t z
a n d e r
c n r h i o t e t S n s l d h
t i l h e S r g
l e
a n d e r
t i l h e S r g
l e r g t i
h l u t a Q s i
c r h o g n u l t e i E
e S n s l d r t g p u h a H c
u g n ir ls e k a
s t u e a Q l i d E h
e h l
e lt r g i
l h s d E e t
e r 0 d 5 c 1 .
2 0 S F I
Definition of the term steel Steel is a material with a mass fraction if iron which is higher than of every other element, ist carbon content is, in general, lower than 2% and steel contains, moreover, also other elements. A limited number of chromium steels might contain a carbon content which is higher than 2%, but, however, 2% is the common boundary between steel and cast iron [DIN EN 10020 (07.00)].
10020 (July 2000), the designations (main symbols) for the classification of steels are standardised. Figure 5.1 shows the definition of the term steel and the classification of the steel grades in accordance with their
Classification in accordance with the main quality class: unalloyed steels stainless steels other, alloyed steels - unalloyed quality steels - unalloyed special steels
br-er05-01.cdr
Figure 5.1 In accordance with the chemical composition the steel grades are classified into
Al Determined element aluminium boron bismuth chromium limit value Mass fraction in %
unalloyed, stainless and other alloyed steels. The mass fractions of the individual elements in unalloyed steels do not achieve the limit values which are indicated in Figure 5.2. Stainless steels are grades of steel with a mass fraction of chromium of at least 10,5 % and a maximum of 1,2 % of carbon. Other alloyed steels are steel grades which do not comply with the definition of stainless steels and where one alloying element exceeds the limit value indicated in Figure 5.2.
B Bi Cr La
Co cobalt Cu copper lanthanides (rated individually) Mn manganese Mo molybdenum Nb niobium Ni nickel Pb lead Se selenium Si Te Ti V W silicon tellurium titanium vanadium tungsten
Zr zirconium Others (with the exception of carbon, phosphorus, sulphur, nitrogen) (Each) a) If just the highest value has been determined for mangenese, the limit value us 1,80% and the 70%-rule does not apply.
br-er05-02.cdr
ISF 2004
Figure 5.2
46
As far as the main quality classes are concerned, the steels are classified in accordance with their main characteristics and main application properties into unalloyed, stainless and other alloyed steels. As regards unalloyed steels a distinction is made between unalloyed quality steels and unalloyed high-grade steels. Regarding unalloyed quality steels, prevailing demands apply, for example, to the toughness, the grain size and/or the forming properties. Unalloyed high-grade steels are characterised by a higher degree of purity than unalloyed quality steels, particularly with regard to non-metal inclusions. A more precise setting of the chemical composition and special diligence during the manufacturing and monitoring process guarantee better properties. In most cases these steels are intended for tempering and surface hardening. Stainless steels have a chromium mass fraction of at least 10,5 % and maximally 1,2 % of carbon. They are further classified in accordance with the nickel content and the main characteristics: corrosion resistance, heat resistance and creep resistance. Other alloyed steels are classified into alloyed quality steels and alloyed high-grade steels. Special demands are put on the alloyed quality steels, as, for example, to toughness, grain size and/or forming properties. Those steels are generally not intended for tempering or surface hardening. The alloyed high-grade steels comprise steel grades which have improved properties through precise setting of their chemical composition and also through special manufacturing and control conditions.
47
The European Standard DIN EN 10027-1 (September 1992) stipulates the rules for the designation of the steels by means of code letters and identification numbers. The code letters and identification numbers give information about the main application field, about the mechanical or physical properties or about the composition. The code designations of the steels are divided into two groups. The code designations of the first group refer to the application and to the mechanical or physical properties of the steels. The code designations of the second group refer to the chemical composition of the steels.
l l l l l l l l
S = Steels for structural steel engineering e.g. S235JR, S355J0
According to the utilization of the steel and also to the mechanical or physical properties, the steel grades of the first group are designated with different main symbols (Fig. 5.3).
P = Steels for pressure vessel construction e.g. P265GH, P355M L = Steels for pipeline construction e.g. L360A, L360QB E = Engineering steels e.g. E295, E360 B = Reinforcing steels e.g. B500A, B500B Y = Prestressing steels e.g. Y1770C, Y1230H R = Steels for rails (or formed as rails) e.g. R350GHT H = Cold rolled flat-rolled steels with higher-strength drawing quality e.g. H400LA D = Flat products made of soft steels for cold reforming e.g. DD14, DC04 T = Black plate and tin plate and strips and also specially chromium-plated plate and strip e.g. TH550, TS550 M = Magnetic steel sheet and strip e.g. M400-50A, M660-50D
ISF 2004
l l
br-er05-03.cdr
Figure 5.3
48
An example of the code designation structure with reference to the usage and the mechanical or physical properties for steels in structural steel engineering is explained in Figure 5.4.
Figure 5.4
49
For designating special features of the steel or the steel product, additional symbols are added to the code designation. A distinction is made between symbols for special demands, symbols for the type of coating and symbols for the treatment condition. These additional symbols are stipulated in the ECISS-note IC 10 and depicted in Figures 5.5 and 5.6.
Symbol1)2) +A + AR + AS + AZ + CE + Cu + IC + OC +S + SE +T + TE +Z + ZA + ZE + ZF + ZN
1 2
Coating hot dipped aluminium, cladded by rolling coated with Al-Si alloy coated with Al-Tn alloy (>50% Al) electrolytically chromium-plated copper-coated inorganically coated organically coated hot-galvanised electrolytically galvanised upgraded by hot dipping with a lead-tin alloy electrolytically coated with a lead-tin alloy hot-galvised coated with Al-Zn alloy (>50% Zn) electrolytically galvanised diffusion-annealed zinc coatings (galvannealed, with diffused Fe) nickel-zinc coating (electrolytically)
) The symbols are separated from the preceding symbols by plus-signs (+) ) In order to avoid mix-ups with other symbols, the figure S may precede,
Figure 5.5
Symbol ) ) +A + AC +C + Cnnn + CR + HC + LC +Q +S + ST +U
1 2
treatment condition softened annealed for the production of globular carbides work-hardened (e.g., by rolling and drawing), also a distinguishing mark for cold-rolled narrow strips) cold-rolled to a minimum tensile strength of nnn MPa/mm cold-rolled thermoformed/cold formed slightly cold-drawn or slightly rerolled (skin passed) quenched or hardened treatment for capacity for cold shearing solution annealed untreated
1 2
) The symbols are separated from the preceding symbols by plus-signs (+) ) In order to avoid mix-ups with other symbols, the figure T may precede,
Figure 5.6
50
Figure 5.7 shows an example of the novel designation of a steel for structural steel engineering which had formerly been labelled St37-2.
The steel St37-2 (DIN 17100) is, according to the new standard (DIN EN 10027-1), designated as follows:
S235 J 2 G3
Steel for structural steel engineering further property (RR = normalised) test temperature 20C impact energy 27 J
S = steels for structural steel engineering P = steels for pressure vessel construction L = steels for pipeline construction E = engineering steels B = reinforcing steels
br-er-05-07.cdr ISF 2002
Figure 5.7
Steel Stahl S355J0 (St 52-3) S500N (StE500) P295NH (HIV) S355J2G1W (WTSt510-3) S355G3S (EH36)
Steel Stahl
Si 0,55
Mn 1,60
Al / 0,020 / / /
N 0,009 0,020 / / /
Mo / 0,1 / 0,30 /
Ni / 1 / 0,65 /
Nb / 0,05 / / /
0C S355J2G3 (St 52-3) S500N (StE500) P295NH (HIV) S355J2G1W (WTSt510-3) S355G3S (EH36)
br-er-05-08.cdr
27 31-47
27 21-39
49 (bei +20C)
76 (bei -10C)
ISF 2004
Figure 5.8 Figure 5.8 depicts the chemical composition and the mechanical parameters of different steel grades. The figure explains the influence of the chemical composition on the mechanical properties.
51
The steel S355J2G2 represents the basic type of structural steels which are nowadays commonly used. Apart from a slightly increased Si content for desoxidisation it this an unalloyed steel. S500N is a typical fine-grained structural steel. A very fine-grained microstructure with improved tensile strength values is provided by the addition of carbide forming elements like Cr and Mo as well as by grain-refining elements like Nb and V. The boiler steel P295NH is a heat-resistant steel which is applied up to a temperature of 400 This steel shows a relatively low strength but very good toughness values C. which are caused by the increased Mn content of 0,6%. S355J2G1W is a weather-resistant structural steel with mechanical properties similar to S355J2G2. By adding Cr, Cu and Ni, formed oxide layers stick firmly to the workpiece surface. This oxide layer prevents further corrosion of the steel. S355G3S belongs to the group of shipbuilding steels with properties similar to those of usual structural steels. Due to special quality requirements of the classification companies (in this case: impact energy) these steels are summarised under a special group.
52
The steel grades are classified into four subgroups according to the chemical composition (Fig 5.9): Unalloyed steels (except free-cutting steels) with a Mn content of < 1 % Unalloyed steels with a medium Mn content > 1 %, unalloyed free-cutting steels and alloyed steels (except high-speed steels) with individual alloying element contents of less than 5 percent in weight Alloyed steels (except high-speed steels), if, at least for one alloying element the content is 5 percent in weight High-speed steels
The unalloyed steels with Mn conUnalloyed steels (Mo content < 1%)
tents of < 1% are labelled with the code letter C and a number which complies with the hundredfold of the mean value which is stipulated for the carbon content. Unalloyed steels with a medium Mn content > 1 % are labelled with a number which also complies with a hundredfold of the mean value which is stipulated for the carbon content, the chemical symbols for the alloying elements, ordered according to the decreasing contents of the alloying
br-er05-09.cdr
C45
Carbon
0,45% Carbon
10CrMo9-10
C=10/100=0,10% Cr=9/4=2,25%
element Cr, Co, Mn, Ni, Si, W Al, Be, Cu, Mo, Nb, Pb, Ta, Ti, V, Zr C, Ce, N, P, S B
Mo=10/10=1%
factor 4 10 100 1000
Table 5.1
X10CrNi18-10
Legiert C=10/100=0,1% Cr=18% Ni=10%
High-speed steels
HS 2-9-1-8
W=2% Mo=9% V=1% Co=8%
ISF 2004
elements and numbers, which in the sequence of the designating alloying elements give reference about their content. The individual numbers stand Figure 5.9
for the medium content of the respective alloying element, the content had been multiplied by the factor as indicated in Fig. 5.9/Table 5.1 and rounded up to the next whole number.
53
The alloyed steels are labelled with the code letter X, a number which again complies with the hundredfold of the mean value of the range stipulated for the carbon content, the chemical symbols of the alloying elements, ordered according to decreasing contents of the elements and numbers which in sequence of the designating alloying elements refer to their content. High-speed steels are designated with the code letter HS and numbers which, in the following sequence, indicate the contents of elements:: tungsten (W), molybdenum (Mo), vanadium (V) and cobalt (Co).
The European Standard DIN EN 10027-2 (September 1992) specifies a numbering system for the designation of steel grades, which is also called material number system.. The structure of the material number is as follows: 1. XX XX (XX) Sequential number The digits inside the brackets are intended for possible future demands. Steel group number (see Fig. 5.10) Material main group number (1=steel)
54
Figure 5.10 specifies the material numbers for the material main group steel.
Figure 5.10
55
The influence of the austenite grain size on the transformation behaviour has been explained in Chapter 2. Figure 5.11 shows the dependence between grain size of the austenite which develops during the welding cycle, the distance from the fusion line and the energy-per-unit length from the welding method. The higher the energy-peruntil length, the bigger the austen13
Energy-per-unit length in kJ/cm
11 9 9 12
of
the
HAZ
in-
creases.
Such
3 0
br-er-05-11.cdr
0,2
0,8
mm
1,0
ISF 2004
Figure 5.11
With fine-grained structural steels it is tried to suppress the grain growth with alloying elements. Favourable are nitride and carbide forming alloys. They develop precipitations which suppress undesired grain growth. There is, however, a limitation due to the solubility of these precipitations, starting with a certain temperature, as shown in Figure 5.12. Steel 1 does not contain any precipitations and shows therefore a continuous grain growth related to temperature. Steel 2 contains AIN precipitations which are stable up to a temperature of approx. 1100 thus preventing a growth of the C, austenite grain.
56
With
mm 1 8 6 -4
higher
temperatures,
these
precipitations dissolve and cannot suppress a grain growth any more. Steel 3 contains mainly titanium carbonitrides of a much lower grainrefining effect than that of AIN. Steel 4 is a combination of the most effective properties of steels nos. 2 and 3.
Steel 1 Steel 2 Steel 3 Steel 4
1000 1100 1200 Austenitization temperature 1300 C 1400
-2
10 8 6 4
-1
10 8
-2
10
The importance of grain refinement for the mechanical properties of a steel is shown in Figure 5.13. Provided the temperature keeps constant, the yield strength of a steel increases with decreasing grain size. This influence on the yield point Rel is specified in the Hall-Petch-law:
ISF 2004
6 10-3 12 900
Steel 1 2 3 4
br-er05-12.cdr
% Ti / / 0,067 0,140
Rel = i + K
1 d
above-mentioned law, the increase of the yield point is inversely proportional to the root of the medium grain diameter d. i
-180
stands for the internal friction stress of the grain measure material. The is a the
10
br-er-05-13.cdr
ISF 2004
resistance K
influence of the grain size on the forming mechanisms. Apart from this increase of the yield point, grain refinement also results in improved toughness values. As far as
57
structural steels are concerned, this means the improvement of the mechanical properties without any further alloying. Modern fine-grained structural steels show improved mechanical properties with, at the same time, decreased content of alloying elements. As a consequence of this chemical composition the carbon equivalent decreases, the weldability is improved and processing of the steel is easier. The major advanSteel type Stahlsorte
Yield point Streckgrenze Plate thickness Blechdicke Weld cross-section Nahtquerschnitt Welding wire 1.2 Schweidraht 1.2 Welding wire costs Schweidrahtkosten Steel costs Stahlkosten Weld metal costs Schweigutkosten Special weld costs Spez. Schweinahtkosten Costs ratio inclusive base Kostenverhltnis inklusive materials Grundwerkstoffe
S690Q (StE690) 690 14,4 100 NiMoCr 2,4 1,9 1,5 1,8
tages of microalloyed fine-grained with are structural steels in comparison tural shown steels in conventional strucFigure
Ratio Verhltnis
Ratio Verhltnis
Ratio Verhltnis
welding process = MAG Schweiverfahren = MAG Deposition rate = 3 kg welding wire/h, weld /shape X -60 X - 60 Abschmelzleistung = 3 kg Schweidraht h, Nahtform Costs ofund Maschinenkosten = 60 DM / h Lohn- labour and equipment = 30/h Special weld costs = weld filler Schweizusatzwerkstoffe + Schweien Spez. Schweinahtkosten = materials + welding
Figure 5.14
with unalloyed structural steel, substantial savings of material are possible. This leads also to reduced joint cross-sections and, in total, to lower costs when making welded steel constructions. Based Figure on 5.2, the
unalloyed steels
classification
ure 5.15 divides the steels with regard to their problematic processes during welding. When it comes to unalloyed
br-er-05-15.cdr
hardening corrosion tool steels special properties are resistant steels achieved, for example: Hardening, special properties heat resistance, are achieved tempering resistant, high-pressure hydrogen resistance, toughness at low temperatures, surface treeatment condition, etc. ferritic grain increase in the weld interfaces pearlitic-martensitic hardening embrittlement formation of chromium carbide austenitic grain desintegration stress corrosion cracking hot cracks (sigma phase embrittlement)
cold brittleness (coarse-grained recrystallization after critical treatment) stress corrosion cracking safety from brittle fracture
Figure 5.15
58
casts, rimmed and semi-killed steels are causing problems. Killing means the removal of oxygen from the steel bath. Figure 5.16 shows cross-sections of ingot blocks with different oxygen contents. Rimming steels with increased oxygen content show, from the outside to the inside, three different zones after solidification: 1.: a pronounced, very pure outer envelope, 2.: a typical blowhole formation (not critical, blowholes are forged together during rolling), 3.: in the centre a clearly zone segregated
semi-killed steel
rimmed steel
Figure 5.16
Figure 5.17 shows important points to be observed during welding such steels. Due to their enrichment with alloy elements, the segregation zones are more transformation-inert than the outer envelope In to as, hotin
B
ISF 2004
59
Therefore, touching such segregation zones during welding must be avoided by all means. In the case of lowalloy
Microstructures Ferrite Austenite Perlite (granular) Perlite (lamellar) Sorbite Troostite Cementite Martensite Average Brinell Hardness (Approximately) 80 250 200 300 350 400 600 - 650 400 - 900
steels, of must
the HAZ
during be Figshows
Br-er-05-18.cdr
tures. The highest hardness martensite values and can be found with
Figure 5.18
cementite. Hardness values of cementite are of minor importance for unalloyed and low-alloy steels because its proportion in these steels remains low due to the low Ccontent. However, hardening because of martensite formation is of greatest importance as the martensite proportion in the microstructure depends mainly on the cooling time. Figure 5.19 shows the essential influence of the martensite content in the HAZ on the crack formation of welded martensite joints. formaHardening through tion is not to be
Br-er-05-19.cdr ISF 2004
maximum hardness HV root cracking presumable root cracking possible no root cracking sufficient operational safety without heat treatment 400 HRC 41
41 - 36 36 28
70 - 60 60 30
If too much martensite develops in the heat affected zone during welding (below or next to the weld), a very hard zone will be formed which shows often cracks.
60
because the critical cooling rate with these low C-contents is so high that it normally wont be reached within the welding cycle. In general, such steels can be welded without special problems (e.g., S. 235). In addition to carIIW
C - qu. = C + Mn Cr + Mo + V Cu + Ni + + 6 5 15
Mo Mn Cr + Mn Ni Cu + + + 6 10 20 40
bon, all other alloy elements are important site when it in comes to martenformation the welding cycle, as they have substantial influence on the transformation behaviour of steels (see
Stout
C - qu. = C +
PCM = C +
Si Mn + Cu + Cr Ni Mo V + + + + + 5B 30 20 60 15 10
Si Mn + Cu Cr Ni Mo V + + + + + 25 16 20 60 40 15
Mannesmann
C - qu.PLS = C +
Hoesch
C - qu. = C +
Si + Mn + Cu + Cr + Ni + Mo + V 20
Thyssen
C ET
Mn + Mo Cr + Cu Ni = C+ + + 10 20 40
PLS = pipeline steels PCM = cracking parameters (%)
ISF 2002
Definition of C - Equivalent
Figure 5.20
to take the carbon content as a measure for the hardening tendency of such steels. To estimate the weldability, several authors developed formulas for calculating the so-called carbon equivalent, which include the contribution of the other alloy elements to hardening tendency, (Fig. 5.20). As these approximation formulas are empirically determined and as for the
Tp [ C]
250 100
0,35
delta Tp [C]
150
60
100
40
50
20
0 0 5 10 15 20
60
40
delta Tp = CET - 32) 32) CET + CET + 32 deltaTp = (53(53 CET -Q - 53Q - 5332
20
CET = 0,4 %
0
CET = 0,4 %
CET = 0,2 %
CET = 0,2 %
CET = 0,2 %
CET = 0,2 %
40
delta Tp [C]
delta Tp [C]
input, etc., are also of importance, the carbon equivalent cannot be a common limit value for the weldability.
-20
30
-40
20
-60
10
0 0 20 40 60 80
-80
d =d50 mm = 50 mm HDHD = 8 =8
4,5 5
0,35
61
tion of the preheating temperature Tp, the formula as shown in Fig. 5.21 is used. The effects of the chemical composition which is marked by the carbon equivalent CET, the plate thickness d, the hydrogen content of the weld metal HD and the heat input Q are considered. The essential factor to martensite formaTmax
tion in the welding cycle is the cooling time. As a measure of cooling time, the
DT
Temperature T
800
t800
Time t
br-er-05-22.cdr
The
temperature
Definition of t8/5
important structural transformations and that the time can be easily transferred to the
timeC
distriTemperature T
1500
A
butions in the vicinity of a weld. Peak values and dwell times depend obviously on the location of the and
10mm
1000
B
500
C
200
250
300
measurement
are clearly strongly determined by the heat conduction Figure 5.23 conditions.
62
With the use of thinner plates with complete heating of the cross-section during welding, the heat conductivity is only carried out in parallel to the plate surface, this is the two-dimensional heat dissipation. With thicker plates, e.g. during welding of a blind bead, heat dissipation can also be carried out in direction of plate thickness, heat dissipation is three-dimensional. These two cases are covered by the formulas given in Figure 5.24, which
2 - dimensional:
universal formula:
3 - dimensional:
universal formula:
K3
t8 / 5 = h U I 1 1 2 p l v 500 - T0 800 - T0
t8 / 5 = 0,67 - 5 10 - 4 T0
) Uv I 5001- T
1 h N 3 800 - T0
K2
t8 / 5 =
2 2 2 h2 1 1 U I 1 - 4 p l r c v d 2 500 - T0 800 - T0
provide a method of calculating the cooling time t8/5 of low-alloyed steels. In the case of a
ISF 2004
d =
1 1 500 - T + 800 - T 0 0
br-er-05-24.cdr
Figure 5.24
In the case of two-dimensional heat dissipation it is clear that t8/5 becomes the shorter the thicker the plate thickness d is. Provided, the cooling times are equal, the plate thickness can be calculated from these relations where a two-dimensional heat dissipation changes to a three-dimensional heat dissipation. Figure 5.25 shows
welding methods
the influence of the welding method on the heat dissipation. With the same heat energy base depends input, which the is
TIG-(He)-welding TIG-(Ar)-welding MIG-(Ar)-welding MAG-(CO2)- welding Manual arc welding SA welding 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1
on
63
welding method. This dependence is described by the relative thermal efficiency . The influence of the groove geType of weld 2-dimensional heat dissipation 1
0,45 - 0,67
0,67
Fig. 5.26. Empirically determined, these factors were introduced for an easier calculation. For other groove geometries, tests to measure the Figure 5.26
br-er-05-26.cdr
0,9
0,67
0,9
0,9
ISF 2004
cooling time are recommended. Fig. 5.27 shows the transition of the two-dimensional to the three-dimensional heat dissipation for two different preheating temperatures in form of a curve according to the equation of Fig. 5.24. Above the curve, t8/5 depends only on the energy input, but not on the plate thickness, heat dissipation is carried out three-dimensionally.
cm
25
40
50
TA=20C
Plate thickness
3
TA=200C
30 40 60 100
3-dimensional
2
3-dimensional
60 80 100 150
2-dimensional
0 0 10 20 30 40 50
2-dimensional
0 10 20 30 40 50
Figure 5.27
64 Fig. 5.28 shows the possible range of heat input depending on the electrode diameter. It is clear that a relatively large working range is available for arc welding A the
procedures. variation of
carried out by alteration of the welding current, the welding voltage and the welding speed.
Fig. 5.29 depicts variations of the heat input during manual metal arc welding. The shorter the fused electrode distance, i.e., the shorter the extracted length, the higher the energyper-unit length.
Figure 5.29
65
In order to minimize calculation efforts in practice, the specified relations were transferred into nomograms from which permissible welding parameters can be read out, provided some additional data are available. Fig. 5.30 shows diagrams for twodimensional heat dissipation, where a dependence between energy-per-unit length, cooling time and preheating temperature is given, depending on the plate thickness. .
50 40 30 20 10 7 50 40 30 20 10 7 50 40 30 20 10 7 50 40 30 20 10 7 5
br-er05-30.cdr
20C
d = 20 mm
7 8 9 10
15 20
30
kJ/cm 50
ISF 2004
Heat input E
Figure 5.30
If a fine-grained structural steel is to be welded, the steel manufacturer presets a certain interval of cooling times, where the steel characteristics are not too negatively affected. The user lays down the plate thickness and, through the selection of a welding method, a specified range of heat input E. Based on the data E and t8/5 the diagram provides the required preheating temperature for welding the respective plate thickness.
66
50 mm 40
Transition thickness d
30
upper part of the figure determines whether a two-dimensional or a threedimensional heat dissipation is present. For the three-dimensional heat dissipation, the lower diagram applies
kJ/cm 50
20 15
10 9 8 7
7 8 9 10
15 20
30
Heat input E
50 s 40 30
C 0 C 0 C 10 0 C 20 C 15
where the same information can be determined, independent of plate thickness, as with Fig. 5.30.
20 15
25 T
0
20
10 9 8 7
br-er05-31.cdr
7 8 9 10
15 20
30
kJ/cm 50
ISF 2004
Heat input E
Figure 5.31
The
relation
be35 V 30
gas composition: C1 100% CO2 M21 82% Ar + 18% CO2 M23 92% Ar + 8% O2
C1 M21 M23
welding is shown in Fig. 5.32 and the used shielding gas is one of the parameters. Welding voltage and welding current, or
25
20
15
short arc
mixed arc
spray arc
150 3,5
br-er-05-32.cdr
A 9,0
67
The diagram in Fig. 5.33 demonh'UP = 1 h'MAG = 0,85 dU max = 32 mm dU min = 15 mm F3 = 0,67 F2 = 0,67 t8/5 max = 30 s t8/5 min = 6 s Emax = 66 kJ/cm Emin = 14 kJ/cm
strates the dependence of plate thickness, heat input E and cooling time t8/5 for fillet welds at a preheating
60 kJ/cm 50
toughness affection
25s
70
fillet welds T0= 150 C
30s
kJ/cm 59 53 47
C. temperature of T0 = 150 If d and t8/5 are given, the acceptable range of heat input can be determined with the help of this diagram. The kinks of the curves mark the transition between two-dimensional and threedimensional heat dissipation.
45
40
20s
35 30 25 20 15
6s 10s 15s
41 35 29 23 18 12
cracking tendency
10 5 0 0 5 10 15
6 mm 0 40
ISF 2004
20 25 30 Plate thickness
br-er05-33.cdr
Figure 5.33 Fig. 5.34 shows the same dependence for butt welds with V groove preparation..
h'UP = 1 h'MAG = 0,85 dU max = 34 mm dU min = 15 mm F3 = 0,9 F2 = 0,9 t8/5 max = 30 s t8/5 min = 6 s Emax = 49 kJ/cm Emin = 10 kJ/cm
60 kJ/cm 50
toughness affection
70
butt welds T0= 150 C
kJ/cm 59 53
30s
45
40 35 30 25 20 15 10 5 0 0 5 10 15
cracking tendency
15s 25s
47 41 35 29 23 18 12 6 0 40
20s
10s
6s
20 25 30 Plate thickness
mm
br-er05-34.cdr
ISF 2004
Figure 5.34
68
The curve family in Fig. 5.35 shows the dependence of the heat input from the welding speed as well as the acceptable working range. The parameters of the curves 1 to 8 in the table
curve V
1
have been taken from Figures 5.32 and 5.34 and apply only for related wire welding conditions like wire diameter, feed,
29
27
24
22
20
19
18
17
25 kJ/cm 20
2
1
300 275 250 225 200 175 150 125 5.5 4.5 3.5 3.0
Heat input E
15
5
wor king ra
nge
10 5 0 10
15
20
25
30 35 40 45 Welding speed vS
50 cm/min 60
voltage, etc.
ISF 2004
Figure 5.35
Figure 5.36
shows
60 kJ/cm 50
SA - welding
toughness affection
a reading example for such diagrams (according to DVSReference Nr. 0916). In this example, a plate thickness of 15 mm and a cooling time t8/5 between 10 and 20 s are given. In this case, the maximum and 300A is used. Sheet
70
butt welds T0= 150 C
curve V A
MAG - welding
kJ/cm 59 53
30s
29
27
24
22
20
19
18
17
300 275 250 225 200 175 150 125 5.5 4.5 3.5 3.0
45 40 35
25s
47 41 35 29
15s
Heat input E
Heat input E
30 25 20 15 10 5 0 0 5 10 15
cracking tendency
20s
25 kJ/cm 20
2
1
23 18
heat input E
10s
16 15 13
6
work
ing
rang
6s
16 12 13
6 0 40
10 5 0 10
33
15 20 25
41
50 cm/min 60
20 25 30 Plate thickness
mm
30 35 40 45 Welding speed vS
br-er-05-36.cdr
ISF 2004
Figure 5.36
cooling time for MAG welding is 15 s. A solid wire with a diameter of 1.2 mm at 29V The left diagram provides heat input values between 13 and 16 kJ/cm, based on the given data. Using these values, the acceptable range of welding speeds can be taken from the diagram on the right.
69
Fig. 5.37 presents a simplification of the determination of the microstructural composition and cooling time subject to peak temperatures which occur in the welding cycle. In the lower diagram, the point of the plate
Peak temperature Temperature
800 C 700
F
600
B
HV30=400
1400 C 1200
M
300
200
thickness at the top line is linked with the point of heat input at the lower line. The point of intersection of the linking line with the middle scale represents the cooling time t8/5 . If the peak temperature of the welding cycle is known, one can read from the middle diagram in which transition field the final microstructures are formed. The advantage of the determination of microstructures compared
B+M
F+B
F+P
1000
Arc3
800
Arc1
600
10
100
s
7 6
1000
t8/5
plate thickness 40
30
25
20
15
10 9 8
5 mm 4
three-dimensional
two-dimensional
2 3
10
20
t8/5
preheating temperature
energy-per-unit length 6
bie5-37.cdr
10
20
30
40
50 kJ/cm 70
ISF 2004
Peak temperature/cooling time diagram for the determination of t8/5 and the structure
with the upper TTT diagram is that Figure 5.37 a TTT diagram applies only for exactly one peak temperature, other peak temperatures are disregarded. The disadvantage of the PTCT diagram (peak temperature cooling time diagram) is the very expensive determination, therefore, due to the measurement efforts a systematic application of this concept to all common steel types is subject to failure.
71
Basically stainless steels are characterised by a chromium content of at least 12%. Figure 6.1 shows a classification of corrosion resistant
stainless steels
corrosion-resistant steels
steels. They can be singled out as heat- and scale-resistant and stainless steels, depending on service temperature. Stainless steels are used at room temperature conditions and for water-
perlitic martensitic
semi-ferritic
ferritic
ferritic-austenitic
austenitic
X40Cr13
X10Cr13
X8Cr13
X20CrNiSi25-4
based media, whilst heatand scale-resistant steels are applied in elevated temperatures and gaseous media. Depending on their microstructure, the alloys can be divided into perlitic-martensitic, ferritic, and austenitic steels. Perlitic-martensitic steels have a high strength and a high wear resistance, they are used e.g. as knife steels. Ferritic and corrosion resistant steels are mainly used as plates for household appliances and other decorative purposes. The most important group are austenitic steels, which can be used for very many applications and which are corrosion resistant against most media. They have a very high low temperature impact resistance. Based on the simple Fe-C
T d A4 g A3 a a A4 g a(d) A3 g T T A4 d
Figure 6.1
phase diagram (left figure), Figure 6.2 shows the effects of two different groups of alloying elements on the equilibrium diagram.
Alloy elements in %
Nickel Manganese Cobalt
A3
Alloy elements in %
Alloy elements in %
Chromium Vanadium Molybdenum Aluminium Silicon
Ferrite
developers
with
br-er-06-02e.cdr
Figure 6.2
72
tenite area, partly with downward equilibrium line according to Figure 6.2 (central figure). With a certain content of the related element, there is a transformation-free, purely ferritic steel. An opposite effect provide austenite developers. In addition to carbon, the most typical member of this group is nickel.
Element Carbon l l l Chromium l Nickel l l Oxygen l Steel type, no. All types l l l All types l All types Special types l Effect Increases the strength, supports development of precipitants which reduce corrosion resistance, increasing C content reduces critical cooling rate Works as ferrite developer, increases oxidation- and corrosion-resistance Works as austenite developer, increases toughness at low temperature, grain-refining
Austenite developers cause an extension of the austenite area to Figure 6.2 (left figure) and form a purely austenitic and transformation-free steel. The table in Figure 6.3 summarises the effects of some selected elements on high alloy steels.
Works as strong austenite developer (20 to 30 times stronger than Nickel) 1.4511,1.4550, Binds carbon and decreases tendency to Niobium 1.4580 u.a. intergranular corrosion l All types Increases austenite stabilization, reduces hot Manganese l l crack tendency by formation of manganese l l sulphide Improves creep- and corrosion-resistance Molybdenum 1.4401,1.4404, 1.4435 and others. against reducing media, acts as ferrite l developer l l 1.4005, 1.4104, Phosphorus, Improve machinability, lower weldability, 1.4305 selenium, or l reduce slightly corrosion resistance l sulphur l All types Improves scale resistance, acts as ferrite l Silicon developer, all types are alloyed with small l l contents for desoxidation l 1.4510, 1.4541, Binds carbon, decreases tendency to Titanium l 1.4571 and others intergranular corrosion, acts as a grain refiner l l and as ferrite developer Type 17-7 PH Works as strong ferrite developer, mainly Aluminium l l used as heat ageing additive Type 17-7 PH, Copper Improves corrosion resistance against certain l 1.4505, 1.4506 media, decreases tendency to stress l l corrosion cracking, improves ageing l l
br-er06-03e.cdr
S+a
ISF 2002
Temperature
1200
g g+a
1000 800
d+a d d+a' a'
Figure 6.3
600
a
The binary system Fe-Cr in Figure 6.4 shows the influence of chromium on the iron lattice. Starting with about 12% Cr, there is no more transformation into the cubic face-centred lattice, the steel solidifies purely as ferritic. In the temperature range between 800 and 500 this system contains the intermetallic C -phase, which decomposes in the lower
400 200 0 Fe
10
20
30
40
50
Chromium
60
70
80
90 % Cr
br-er06-04e.cdr
ISF 2002
Binary System Fe - Cr
Figure 6.4
temperature range into a low-chromium -solid solution and a chromium-rich -solid solution. Both, the development of the -phase and of the unary --decomposition cause a strong
73
embrittlement. With higher alloy steels, the diffusion speed is greatly reduced, therefore both processes require a relatively long dwell time. In case of technical cooling, such embrittlement processes are suppressed by an increased cooling speed. Nickel is a strong austenite developer, see Figure 6.5 Nickel and iron develop in this system under elevated temperature a complete series of face-centred cubic solid solutions. Also in the binary system Fe-Ni
1600 C 1400
S+d
S+g
Fe Ni3
processes
d
1200
d+g
Temperature
1000 800
Along two cuts through the ternary system Fe-Cr-Ni, Figure 6.6 shows the most
Fe Ni3
600
a
400 200 0 Fe
br-er-06-05e.cdr
a+g
important
90 % Ni
ISF 2002
phases
which
Binary System Fe - Ni
A solidifying alloy with 20% Cr and 10% Ni (left figure) forms at first -ferrite. ferrite is, analogous to the
Figure 6.5
70 % Fe
60 % Fe
1600
C
S S+d+g
1600 C 1500
S+g
S
S+d+g
S+d
from the melt solidifying body-centred cubic solid solution. However -ferrite
1200
d
d+g
d+g
700
700
0 30
br-er-06-06e.cdr
5 25
10 20
15 15
20 % Ni 10 % Cr
0 40
5 35
ISF 2002
Figure 6.6
74
During an ongoing cooling, the binary area ferrite + austenite passes through and a transformation into austenite takes place. If the cooling is close to the equilibrium, a partial transls
formation of austenite into the brittle -phase takes place in the temperature range below 800 Primary ferritic solidifying alloys show C. a reduced tendency to hot cracking, because -ferrite can absorb hot-crack promoting elements like S and P. However primary austenitic solidifying alloys show, starting at a certain alloy content, no transformations during cooling (14% Ni, 16% Cr, left figure). Primary austenitic solidifying alloys are much more susceptible to hot cracking than primary ferritic solidifying alloys, a transformation into the -phase normally does not take place with these alloys. Figure 6.7 shows some typical compositions of certain groups of high alloy steels.
st ee l
si tic
st ee
te ni ti
Fe rri tic
te n
Au s
C Si Mn Cr Mo Ni Cu Nb Ti Al V N S
br-er06-07e.cdr
1.
2.
3.
+ + + +
+ +
4.
Au s
ar
+ + +
te n
+ indicates that the alloy elements can be added in a defined content to achieve various characteristics
iti c-
fe rri tic
st ee l
ls
st ee
ISF 2002
Figure 6.7
The diagram of Strau and Maurer in Figure 6.8 shows the influence on the microstructure formation of steels with a C-content of 0,2%. The classification of high-alloy steels in Figure
28 % 24 20 16 12 8 4
6.1 is based on this diagram. If a steel only contains C, Cr and Ni, the
austenite
lowest austenite corner will be at 18% Cr and 6% Ni. And also other elements
austen it e / mart ensite
Nickel
austenite / ferrite
austenite / martens
20
than Ni and Cr work as an austenite or ferrite developer. these The influence is of denickel elements and
ite / ferrite
24 % 26
martensite / ferrite
6 8 10 12 14 Chromium 16 18 22
ISF 2002
Maurer - Diagram
Figure 6.8
75
equivalents. The Schaeffler diagram reflects additional alloy elements, Figure 6.9. It represents molten weld metal of high alloy steels and determines the developed microstructures after cooling down from very high temperatures. The diagram was always prepared considering identical cooling conditions, the influence of different cooling speeds is here disregarded. The areas 1 to 4 in this diagram limit the chemical compositions of steels, where specific defects may occur during welding. Depending on the composition, purely ferritic chromium steels have a tendency to embrittlement by martensite and therefore to hot cracking (area 2) or to embrittlement due to strong grain growth (area 1).
Nickel-equivalent = %Ni + 30x%C + 0,5x%Mn
30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 00
100%
A cause for this strong grain growth during welding is the greatly increased diffusion speed in the ferrite compared with austenite. After reaching a diffusion-start Figure 6.10 temperature, have a
0%
r Fe
rit
austenite
5%
10
%
%
0 A+F 2
40%
A +M
80%
martensite F + M
2 4 6 8 10
A+M+F
M+F
ferrite
12 14 16 18 20 22 24 26 28 30 32 34 36 38 40
hardening crack susceptibility (preheating to 400C!) hot cracking susceptibility above 1250C
stronger grain growth than austenites. Therefore high alloyed ferritic steels are to be considered as of limited weldability. The area 3 marks a possible embrittlement of the material
4000 6000
m
Figure 6.9
5000
grain size
due to the development of -phase. As explained in 6.6, this risk occurs with increased increased ferrite contents, conchromium
3000
2000
austenitic steel
1000
ferritic steel
0
br-er-06-10e.cdr
200
400
600 temperature
800
1000
1200
ISF 2002
Figure 6.10
76
Finally, area 4 marks the strongly increased tendency to hot cracking in the austenite. Reason is, that critical elements responsible for hot cracking like e.g. sulphur and phosphorous have only very limited solubility in the austenite. During welding, they enrich the melt residue, promoting hot crack formation (see also chapter 9 - Welding Defects). There is a Z-shaped area in the centre of the diagram which does not belong to any other endangered area. This area of chemical composition represents the minimum risk of welding defects, therefore such a composition should be adjusted in the weld metal. Especially when welding austenitic steels one tries to aim at a low content of -ferrite, because it has a much greater solubility of S and P, thus minimising the risk of hot cracking. The Schaeffler diagram is not only used for determining the microstructure with known chemical composition. It is also possible to estimate the developing microstructures when welding different materials with or without filler metal. Figures 6.11 and 6.12 show two examples for a determination of the weld metal microstructures of so-called 'black and white' joints.
28
28
24
24
A
A
10
10
20
Nickel-equivalent
20
Nickel-equivalent
A+M
20
A+M
16 12 8
20
16 12 8 4 0 0 4 8 12 16 20 24 Chromium-equivalent 28 32
40
40
3 2 1 30%
9
A+F
80
20%
: =1:1
80
+
A+M+F
M+F
br-er06-12e.cdr
Figure 6.12
123
A+F
100 % F
: =1:1
4 0 0
36
A+M+F
M+F
F
4
100 % F
8 12 16 20 24 28 32 36
Chromium-equivalent
ISF 2002
br-er06-11e.cdr
ISF 2002
Figure 6.11
77
The ferrite content can only be measured with a relatively large dispersal, therefore DeLong proposed to base a measurement procedure on standardized specimens. Such a system makes it possible to measure comparable values which don't have to match the real ferrite content. Based on these measurement values, the ferrite content is no longer given in percentage, but steels are grouped by ferrite numbers. In addition to ferrite numbers, DeLong proposed a reworked Schaeffler diagram where the ferrite number can be determined by the chemical composition, Figure 6.13. Moreover, DeLong has considered the influence of nitrogen as a strong austenite developer (effects are comparable with influence of carbon). Later on, nitrogen was included into the nickel-equivalent of the Schaeffler diagram.
21
20 19 18 17 16 15 14 13 12 11 10 16
r fe
austenite
e rit
nu
r be m 0 2
d re su ea m lly -% ica l. et in vo n ag s m nt 0% ly te er con 2% rm fo rrite 4% Sc fe ha effl 6%% er6 au 7, 2% ste nite 9, 7% , -m art 10 ,3% en site 12 ,8% -lin 13 e
8 10 12 14 16 18
corrosion resistance starting with a Cr content of 12%. In addition to the problems during welding described by the Schaeffler diagram, these steels can
26
ISF 2002
austenite + ferrite
17
br-er-06-13e.cdr
18
27
De Long Diagram
Figure 6.13
welding process. Figure
air O 2Fe+++O+H2O 2Fe++++2OHOH
-
water
drop of water on a piece of iron. In such a system a potential difference is a precondition for the development of a local element consisting of an anode and a cathode. To develop
H2O
cathode 4e-
Fe(OH)3
2Fe++
iron
O2+2H2O+4e 4OH
br-er-06-14e.cdr
ISF 2002
Figure 6.14
78
such a local element, a different orientation of grains in the steel is sufficient. If a potential difference under a drop of water is present, the chemically less noble part reacts as an anode, i.e. iron is oxidised here and is dissolved as Fe2+-ion together with an electron emission. Caused by oxygen access through the air, a further oxidation to Fe3+ takes place. The cathodic, chemically nobler area develops OH- ions, absorbing oxygen and the electrons. Fe3+and OH--ions compose into the water-insoluble Fe(OH)3 which deposits as rust on the surface (note: the processes here described should serve as a principal explanation of electrochemical corrosion mechanisms, they are, at best, a fraction of all possible reactions). If the steel is passivated by chromium, the corrosion protection is provided by the development of a very thin chromium oxide layer which separates the material from the corrosive medium. Mechanical surface damages of this layer are completely cured in a very short time.
passive layer
passive layer
active dissolution
gap
tensile stress
crevice corrosion
intergranular corrosion
incorrect
correct
br-er06-15e.cdr
ISF 2002
br-er06-16e.cdr
ISF 2002
Figure 6.15
Figure 6.16
The examples in Figure 6.15 are more critical, since a complete recovery of the passive layer is not possible from various reasons.
79
If crevice corrosion is present, corrosion products built up in the root of the gap and oxygen has no access to restore the passive layer. Thus narrow gaps where the corrosive medium can accumulate are to be avoided by introducing a suitable debr-er-06-17e.cdr
With pitting corrosion, the chemical composition of the attacking medium causes a
Figure 6.17
local break-up of the passive layer. Especially salts, preferably Clions, show this behaviour. This local attack causes a dissolution of the material on the damaged points, a depression develops. Corrosion products accumulate in this depression, and the access of oxygen to the bottom of the hole is obstructed. However, oxygen is required to develop the passive layer, therefore this layer cannot be completely cured and pitting occurs, Figure 6.17. Stress-corrosion cracking occurs when the material displaces under stress and the passive layer tears, Figure 6.18. Now the unprotected area is subjected to corrosion, metal is dissolved and the passive layer redevelops (figures 13). The repeated displace1
ment tion.
and
10
11
12
offset;
passive layer;
metal surface;
dislocation
br-er-06-18e.cdr
not
follow
the
grain
boundaries.
Figure 6.18
80
Figure 6.19 shows the expansion-rate dependence of stress corrosion cracking. With very low expansion-rates, a curing of the passive layer is fast enough to arrest the crack. With very high expansion-rates, the failure of the specimen originates from a ductile fracture. In the intermediate range, the material damage is due to stress corrosion cracking. Figure 6.20 shows an example of crack propagation at transglobular stress corrosion cracking. A crack propagation speed is between 0,05 to 1 mm/h for steels with 18 - 20% Cr and 8 20% Ni. With view to welding it is important to know that already residual welding stresses may release stress corrosion cracking.
tough fracture
T=RT
SpRK
e2
e1
Elongation speed e
br-er06-19E.cdr
ISF 2002
br-er06-20e.cdr
ISF 2002
Figure 6.19
Figure 6.20
The most important problem in the field of welding is intergranular corrosion (IC). It is caused by precipitation of chromium carbides on grain boundaries. Although a high solubility of carbon in the austenite can be expected, see Fe-C diagram, the carbon content in high alloyed Cr-Ni steels is limited to approximately 0,02% at room temperature, Figure 6.21.
81
The reason is the very high affinity of chroto Bain and Aborn
1200 C 1100 1000 900 800 700 600 0 0.05 0.1 0.15 0.2 Carbon content 0.25 % 0,3
mium to carbon, which causes the precipitation of chromium carbides Cr23C6 on grain boundaries, Figure 6.22. Due to these precipitations, the austenite grid is depleted of chromium content along the grain boundaries
and the Cr content drops below the parting limit. The diffusion speed of chromium in austenite is considerably lower than that of carbon, therefore the chromium reduction cannot be compensated by late diffusion. In the depleted areas along the grain boundaries (line 2 in Figure 6.22) the steel has become susceptible to corrosion.
br-er06-21e.cdr
ISF 2002
Only after the steel has been subjected to sufficiently long heat treatment, chromium will diffuse to the grain boundary and increase the
Figure 6.21 C concentration along the grain boundary (line 3 in Figure 6.22). In this way, the complete corrosion resisChromium content of austenite
1 - homogenuous starting condition 2 - start of carbide formation 3 - start of concentration balance 4 - regeneration of resistance limit 1 2 4 resistance limit 3
tance can be restored (line 4 in Figure 6.22). Figure 6.23 explains why the IC is also described as intergranular disintegration. Due to dissolution of depleted areas along the grain boundary, complete grains break-out of the steel.
br-er-06-22e.cdr
ISF 2002
Sensibility of a Cr - Steel
Figure 6.22
82
The precipitation and repassivation mechanisms described in Figure 6.22 are covered by intergranular corrosion diagrams according to Figure 6.24. Above a certain temperature carbon remains dissolved in the austenite (see also Figure 6.21). Below this temperature, a carbon precipitation takes place. As it is a diffusion controlled certain process, the time
3
Reciprocal of heat treatment temperature 1/T
br-er-06-23e.cdr ISF 2002
Grain Disintegration
Figure 6.23
which depends on temperature (line 1, precipitation characteristic curve). During stoppage at a constant temperature, the parting limit of the steel is
2
austenite + chromium caride (M23C6) sensitive to intergranular disintegration
oversaturated austenite
1 incubation time 2 regeneration of resistance limit 3 saturation limit for chromium carbide
ISF 2002
Figure 6.24 Figure 6.25 depicts characteristic precipitation curves of a ferritic and of an austenitic steel. Due to the highly increased diffusion speed of carbon in ferrite, shifts the curve of carbon precipitation of this steel markedly towards shorter time. Consequently the danger of intergranular corrosion is significantly higher with ferritic steel than with austenite.
83
As carbon is the element that triggers the intergranular corrosion, the intergranular corrosion diagram is relevantly influenced by the c content, Figure 6.26. By decreasing the carbon content of steel, the start of carbide precipitation and/or the start of intergranular corrosion are shifted towards so-called than 0,03% During welding, the considerable influence of
cooling curve
and
longer
times. This fact initiated the development of where the C content is decreased to less
Tempering temperature
(Extra-Low-Carbon)
carbon is also important for the selection of the shielding gas, Figure 6.27. The higher the CO2-content of the shielding gas, the
br-er06-25e.cdr
Tempering time
ISF 2002
stronger is its carburising effect. The Ccontent of the weld metal increases and the steel becomes more susceptible to intergranular corrosion.
Figure 6.25
700
Temperature
600
ure 6.28. The affinity of these elements to carbon is significantly higher than that of chromium, therefore
500
400 1 10
br-er-06-26e.cdr
10
103 Time
104
105
10
ISF 2002
carbon is compounded into Nb- and Ti-carbides. Now carbon cannot cause any chromium depletion. The
Figure 6.26
84
proportion of these alloy elements depend on the carbon content and is at least 5 times higher with titanium and 10 times higher with niobium than that of carbon. Figure 6.28 shows the effects of a stabilisation in the intergranular corrosion diagram. If both steels are subjected to the same heat treatment (1050 C/W means heating to 1050 and subsequent waC ter quenching), then the area of intergranular corrosion will shift due to stabilisation to significantly longer times. Only with a much higher heat treatment temperature the intergranular corrosion accelerates again. The cause is the dissolution of titanium carbides at sufficiently high temperature. This carbide dissolution causes problems when welding stabilised steels. During welding, a narrow area of the HAZ is heated above 1300 carbides are disC, solved. During the subsequent cooling and the high cooling rate, the carbon remains dissolved.
700
Heat treatment temperature
600 550
M2 M1
1050C
/W
S1
1000
300
1000
10000
X5CrNi18-10
Heat treatment temperature
800 C 700 650 600 550 500 450 0,3
unstabilized
1300C
/W
1050C
/W
1 3 W.-No.:4541
10
30
100 Time
300
1000
10000
X5CrNiTi18-10
stabilized
br-er06-27e.cdr
ISF 2002
br-er06-28e.cdr
ISF 2002
Figure 6.27
Figure 6.28
If a subsequent stress relief treatment around 600C is carried out, carbide precipitations on grain boundaries take place again. Due to the large surplus of chromium compared with niobium or titanium, a partial chromium carbide precipitation takes place, causing again inter-
85
granular susceptibility. As this susceptibility is limited to very narrow areas along the welded joint, it was called knife-line attack because of its appearance. Figure 6.29. In stabilised steels, the chromium carbide represents an unstable phase, and with a sufficiently long heat treatment to transform to NbC, the steel becomes stable again. The stronger the steel is over-stabilised, the lower is the tendency to knife-line corrosion. Nowadays the importance of Nickel-Base-Alloys increases constantly. They are ideal materials when it comes to components which are exposed to special conditions: high temperature, corrosive attack, low temperature, wear rebr-er-06-29e.cdr
sistance, or combinations hereof. Figure 6.30 shows one of the possible grouping of nickel-base-alloys.
Knife-Line Corrosion
Figure 6.29
Materials listed there are selected examples, the total number of available materials is many times higher. Group A consists of nickel alloys. These alloys are
Alloy Group A Nickel 200 Nickel 212 Nickel 222 Group B Monel 400 Monel 450 Ferry Group C Inconel 600 Nimonic 75 Nimonic 86 Incoloy 800 Incoloy 825 Inco 330 Ni 66.5, Cu 31.5 Ni 30.0, Cu 68.0, Fe 0.7, Mn 0.7 Ni 45.0, Cu 55.0 Ni 76.0, Cr 15.5, Fe 8.0 Ni 80.0, Cr 19.5 Ni 64.0, Cr 25.0, Mo 10.0, Ce 0.03 Ni 32.5, Fe 46.0, Cr 21.0, C 0.05 Ni 42.0, Fe 30.0, Cr 21.5, Mo 3.0, Cu 2.2, Ti 1.0 Ni 35.5, Fe 44.0, Cr 18.5, Si 1.1 Ni 99.6, C 0.08 Ni 97.0, C 0.05, Mn 2.0 Ni 99.5, Mg 0.075 Chem. composition Alloy Group D1 Duranickel 301 Ni 94.0, Al 4.4, W 0.6 Incoloy 925 Ni 42.0, Fe 32.0, Cr 21.0, Mo 3.0, W 2.1, Cu 2.2, Al 0.3 Ni-Span-C 902 Y2O3 0.5, Ni 42.5, Fe 49.0, Cr 5.3, W 2.4, Al 0.5 Group D2 Monel K-500 Inconel 718 Ni 65.5, Cu 29.5, Al 2.7, Fe 1.0, W 0.6 Ni 52.0, Cr 22.0, Mo 9.0, Co 12.5, Fe 1.5, Al 1.2 Chem. Composition
characterized by moderate mechanical strength and high degree of toughness. They can be hardened only by cold working. The alloys are quite gummy in the annealed or hot-worked condition, and cold-drawn material is recommended
Inconel X-750 Ni 61.0, Cr 21.5, Mo 9.0, Nb 3.6, Fe 2.5 Nimonic 90 Ni 77.5, Cr 20.0, Fe 1.0, W 0.5, Al 0.3, Y2O3 0.6 Nimonic 105 Incoloy 903 Incoloy 909 Inco G-3 Inco C-276 Group E Monel R-405 Ni 66.5, Cu 31.5, Fe 1.2, Mn 1.1, S 0.04 Ni 76.0, Cr 19.5, Fe 112.4, Al 1.4 Ni 39.0, Fe 34.0, Cr 18.0, Mo 5.2, W 2.3, Al 0.8 Ni 58.0, Cr 19.5, Co 13.5, Mo 4.25, W 3.0, Al 1.4 Ni 38.4, Fe 42.0, Cu 13.0, Nb 4.7, W 1.5, Al 0.03, Si 0.15 Ni 38.4, Fe 42.0, Cu 13.0, Nb 4.7, W 1.5, Al 0.03, Si 0.4
br-er-06-30e.cdr
ISF 2002
Figure 6.30
86
Group B consists mainly of those nickel-copper alloys that can be hardened only by cold working. The alloys in this group have higher strength and slightly lower toughness than those in Group A. Cold-drawn or cold-drawn and stress-relieved material is recommended for best machinability and smoothest finish. Group C consists largely of nickel-chromium and nickel-iron-chromium alloys. These alloys are quite similar to the austenitic stainless steels. They can be hardened only by cold working and are machined most readily in the cold-drawn or cold-drawn and stress-relieved condition. Group D consists primary of age-hardening alloys. It is divided into two subgroups: D 1 Alloys in the non-aged condition. D 2 Aged Group D-1 alloys plus several other alloys in all conditions. The alloys in Group D are characterized by high strength and hardness, particularly when aged. Material which has been solution annealed and quenched or rapidly air cooled is in the softest condition and does machine easily. Because of softness, the non-aged condition is necessary for trouble free drilling, tapping and all threading operations. Heavy machining of the age-hardening alloys is best accomplished when they are in one of the following conditions: 1. Solution annealed 2. Hot worked and quenched or rapidly air cooled Group E contains only one material: MONEL R-405. It was designed for mass production of automatically machined screws. Due to the high number of possible alloys with different properties, only one typical material of group D2 is discussed here: Material No. 2.4669, also known as e.g. Inconel X-750. The aluminium and titanium containing 2.4669 is age-hardening through the combination of these elements with nickel during heat treatment: gamma-primary-phase (') develops which is the intermetallic compound Ni3(Al, Ti). During solution heat treatment of X-750 at 1150 the number of flaws and dislocations in C, the crystal is reduced and soluble carbides dissolve. To achieve best results, the material
87
should be in intensely worked condition before heat treatment to permit a fast and complete recrystallisation. After solution heat treatment, the material should not be cold worked, since this would generate new dislocations and affect negatively the fracture properties. The creep rupture resistance of X-750 is due to an even distribution of the intercrystalline ' phase. However, fracture properties depend more on the microstructure of the grain boundaries. During an 840 stabilising heat treatment as part of the triple-heat treatment, the fine ' C phase develops inside the grains and M23C6 precipitates onto the grain boundaries. Adjacent to the grain boundary, there is a ' depleted zone. During precipitation hardening (700 C/20 h) ' phase develops in these depleted zones. ' particles arrest the movement of dislocations, this leads to improved strength and creep resistance properties. During the M23C6 transformation, carbon is stabilised to a high degree without leaving chromium depleted areas along the grain boundaries. This stabilisation improves the resistance of this alloy against the attack of several corrosive media. With a reduction of the precipitation temperature from 730 to 620 as required for some C special heat treatments additional ' phase is precipitated in smaller particles. This enhances the hardening effect and improves strength characteristics. Further metallurgical discussions about X-750, can be taken from literature, especially with view to the influence of heat treatment on fracture properties and corrosion behaviour.
The recommended processes for welding of X-750 are tungsten inert gas, plasma arc, electron beam, resistance, and pressure oxy arc welding. During TIG welding of INCONEL X-750, INCONEL 718 is used as welding consumable. Joint properties are almost 100% of base material at room temperature and about 80% at 700 820 Figure 6.31 shows typical strength properties of a welded plate at a temperature C. range between -423 and 1500 (-248 820 F C). Before welding, X-750 should be in normalised or solution heat treated condition. However, it is possible to weld it in a precipitation hardened condition, but after that neither the seam nor the heat affected zone should be precipitation hardened or used in the temperature range of precipitation hardening, because the base material may crack. If X-750 was precipitation hardened and then welded, and if it is likely that the workpiece is used in the temperature range of precipitation hardening, the weld should be normalised or once again precipitation hardened. In any case it must be noted that heat stresses are minimised during assembly or welding.
88
X-750 welds should be solution heat treated before a precipitation hardening. Heating-up speed during welding must be from the start fast and even touching the temperature range of precipitation hardening only as briefly as possible. The best way for fast heating-up is to insert the welded workpiece into a preheated furnace. Sometimes a preheating before welding is advantageous if the component to be welded has a poor accessibility, or the welding is complex, and especially if the assembly proves to be too complicated for a post heat treatment. Two effective welding preparations are: 1. 1550 F/16 h, air cooling 2. 1950 F/1 h, furnace cooling with 25 -100 F/h up to 1200 air F, A repair welding of already fitted parts should be followed by a solution heat treatment (with a fast heating-up through the temperature range of precipitation hardening) and a repeated precipitation hardening. A cleaning of intermediate layers must be carried out to remove the oxide layers which are formed during welding. (A complete isolaprocesses is hardly possible). If such films are not removed on a regular basis, they can become thick enough to cause material separations together with a reduced strength. Brushing with wire brushes only polishes the
Elongation, % Stress, 1000 psi
220 200 180 160 tensile strength 140 120 100 0.2% yield stress 80 60
surface, the layer surface must be sandblasted or ground with abrasive material. The frequency of cleaning depends on the mass of the developed oxides. Any sand must be removed before the next layer is welded. X-750 can be joined also by spot-, projection, seam-, and flash butt welding. The welding equipment must be of adequate performance. X-750 is generally resistance welded in normalized or solution heat treated condition.
30 20 10 0 20 10 0 -423 0 elongation in 2 200 400 600 800 1000 1200 1400 1600 elongation in 1/2
Temperature, F
br-er06-31e.cdr
ISF 2002
Figure 6.31
90
cast materials
Figure 7.1 provides a summary of the different cast iron materials. In this connection it is only referred iron, and cast to cast steel cast
ferritic cast steel
iron-carboncast materials
malleable iron
cast iron
unalloyed
perlitic
alloyed
low alloyed
nodular graphite cast iron
lemellar graphite cast iron
high alloyed
hard cast clear chill low C iron casting content ledeburitic Cr-cast iron other elements
ferritic
perlitic
austenitic
decarburized
perlitic
austenitic
malleable materials,
ferritic
br-er-07-01e.cdr
perlitic
ferritic
perlitic
austenitic
ISF 2002
Figure 7.1
the cast material in accordance with DIN EN 1560. A distinction is made between the designation according to the material code and the designation according to the material number. In Figure 7.2, examples of two materials are specified.
2 34
standardised material cast material graphite structure (lamellar graphite) microstructure (ferritic) 2 mechanical properties (Rm= 150 N/mm ) chemical composition (high alloyed) optionally
4,5,6
standardised material cast material graphite structure (lamellar graphite) number for the main characteristic material identification number special requirement
br-er07-02e.cdr
ISF 2004
Designation of Materials
Figure 7.2
91
Figure 7.3 depicts a survey of the mechanical properties and the chemical compositions of several customary cast materials. As to its analysis and mechanical properties which are very different from other cast materials, cast steel constitutes an exception to the rule.
In Figure 7.4 the stable and the metastable iron-carbon diagram are shown. The differences between the cast material are best explained this way. Cast iron with lamellar and spheroidal ite has graphcarbon of 2,8 beand
contents tween
all Si, these materials solidify fo llowing the stable system, i.e., the carbon is precipitated in the form of
similar C-contents, the from metal, follows tastable solidification the molten
system.
7. Welding of Cast Materials other hand, comply with those of common structural steels, i.e., they are, as a rule, below 0,8% C.
92
The structure of a normalised cast iron which is composed of ferrite (bright) and pearlite (dark) is shown in Figure 7.5. Since the properties are similar to those of structural steels these materials are weldable, constructional welding is also possible. It is recommended to normalise the cast steel parts before welding. Through this type of heat treatment, on the one hand the transformation of the cast structure is obtained, the residual stresses inside the workpiece are, on the other hand, reduced. Figure 7.5
From a C-content in the steel cast of 0,15% up, it is recommended to carry out preheating during welding, the preheating temperature should follow the analysis of the material, the workpiece geometry and the welding method. After welding the cast workpieces are subject to stress-relief annealing.
Figure 7.6 shows the structure of cast iron with lamellar graphite (grey cast iron). Apart from their carbon content, these materials are characterised by increased contents of S and P which
Figure 7.6
93
improves castability. Besides the poor mechanical properties (elongation after fracture of approx. 1%), these chemical properties also impede welding with ordinary means. It is not possible to carry out constructional welding with grey cast iron. Repair welds of grey cast iron are, in contrast, carried out more frequently as damaged cast parts are not easily replaceable. For those repair welds, the cast parts must be preheated (entirely or partly) to te mperatures of approx. 650C. Heating and cooling must be done very slowly as the cast piece may be destroyed already by the thermal stresses. The highly liquid weld metal also constitutes a problem, and thus the molten pool must be supported by a carbon pile. Welding may be carried out with similar filler material (materials of the same composition as the base). If grey cast iron is to be welded without any preheating, the filler material must, as a rule, be dissimilar (of different composition to the base metal). During this type of welding, there are always strong structural changes in the region of the weld which lead to high hardening and high residual stresses. For the minimisation of these structural changes, a highly ductile filler material is applied. The heat input into the base material should be as low as possible. Figure 7.7 depicts the structural constitution of sphe roidal graphite cast iron. The graphite spheroidization achieved by is the
addition of magnesium and cerium. As, with this type of graphite, the
notch actions are Figure 7.7 considerably lesser than this is the case with grey cast iron, this type of cast iron is characterised by substa ntially better mechanical parameters with a considerably higher elongation after fracture and improved ductility. For this reason, the risk of material failure caused by weld residual stresses or thermal stresses is considerably reduced for spheroidal graphite
94 cast iron. Frequently, nickel-based alloys are used as filler material. Problems occur in the HAZ where, besides the ledeburite eutectic alloy system, also Ni-Fe-martensite is frequently formed. Both structures lead to extreme hardening in the HAZ which can be removed only by time-
Figure 7.8
Figures 7.8 and 7.9 show the structures of Carburized Annealed Malleable Cast Iron (7.7) and of Decarburized Annealed Malleable Cast Iron (7.9). The composition of the malleable cast iron is thus that during solidification, the total of carbon is bound in cementite and precipitated. During a subsequent annealing process, the iron carbide disintegrates into graphite and iron. Figure 7.9
95 If annealing is carried out in neutral atmosphere, the structure of Carburized Annealed Malleable Cast Iron develops. Annealing in oxidising atmosphere leads to the decarburisation of the workpiece surfaces and Decarburized Annealed Malleable Cast Iron is developed, Figure 7.10. Carburized Annealed Malleable
Cast Iron is not weldable. Decarburized Annealed Malleable Cast Iron, in contrast, is weldable.
Figure 7.10
You can see in Figure 7.11 that, also with malleable cast iron, hardening in the region of the HAZ occurs. For carrying out constructional welds made of malleable cast iron parts, a special material quality has been developed. Figure 7.11 shows that this m aterial, EN-GJMW-400-12, is characterised by considerably less hardening. This material is weldable without any problems up to a wall thickness of 8 mm.
Figure 7.11
8. Welding of Aluminium
8. Welding of Aluminium
97
Figure 8.1 compares basic physical properties
Property Atomic weight Specific weight Lattice E-module R pO,2 PO,2 R mm spec. Heat capacity Melting point Heat conductivity Spec. el. Resistance Expansion coeff. [N/mm] [N/mm] [N/mm] [J/(g*C)] [C] [W/(cm*K)] [nWm] [1/C] [g/Mol] [g/cm]
Al 26.9 2.7 fcc 71*10 ca. 10 ca. 50 0.88 660 2.3 28-29 24*10 -6
Fe 55.84 7.87 bcc 210*10 ca. 100 ca. 200 0.53 1539 0.75 97 12*10 -6 FeO
of steel and aluminium. Side by side with different mechanical behaviour, the following differences are important for aluminium welding: - considerably lower melting point compared with steel - three times higher heat conductivity - considerably lower electrical resistance - double expansion coefficient - melting point of Al203 considerably higher than that of Al; metal and iron oxide melt approximately at the same temperature. Figure 8.2 compares some mechanical properties of steel with properties of some light metals. The important advantages of light metals compared with steel are especially
Oxydes
Al2O 3
Fe 3O 4 Fe 2O 3 1400
[C]
2050
1600 (1455)
br-er08-01.cdr
ISF 2002
Figure 8.1
shown in the right part of the figure. If a comparison should be based on an identical stiffness, then the aluminium supporting beam has a 1.44 times larger cross-section than the steel beam, however only about 50% of its weight. Figure 8.3 compares qualitatively the stress-strain diagram of Aluminium and steel. In contrast to steel, aluminium has a fcc (face centred cubic)-lattice at room temperature. This is why there is no distinct yield point as being the case in a bcc (body centred cubic)lattice. Aluminium is not subject to a lattice transbr-er-08-02.cdr
Figure 8.2
8. Welding of Aluminium
98
formation during cooling, thus there is no structure transformation and consequently no danger of hardening in the heat affected zone as with steel.
4 cm 2
400
200C
1000 1200
800
-2
1500
Steel
-4
8 cm aluminium 6
100C
200
Stress
Al-alloy
2
300 400 500
600
-2 -4
-6
Elongation
br-er08-03.cdr
-8 -18
-16
-14
-12
-10
-8
-6
-4
-2
cm
ISF 2002
br-er08-04.cdr
ISF 2002
Figure 8.3
Figure 8.4
Figure 8.4 illustrates the effect of the considerably higher heat conductivity on the welding process compared with steel. With aluminium, the temperature gradient around the welding point is considerably smaller than with steel. Although the peak temperature during Al welding is about 900 below steel, the isothermal curves around the welding point have a clearly C larger extension. This is due to the considerably higher heat conductivity of aluminium compared with steel. This special characteristic of Al requires a input heat volume during welding equivalent to steel. Figure 8.5 lists the most important alloy elements and their combinations for industrial use. Due to their behaviour during heat treatment can Al-alloys be divided into the groups hardenable and non-hardenable (naturally hard) alloys.
8. Welding of Aluminium
99
Figure 8.6 shows typical applications of some Al alloys together with preferably used welding consumables. Aluminium alloys are often welded with consumable of the same type, however, quite often over-alloyed consumables are used to compensate burn-off losses (especially with Mg and Zn because of their low boiling point) and to improve the mechanical properties of
Al Si
678
Al Cu Mg
Cu
Al Mg Si
Al Zn Mg
Mg
Al Zn Mg Cu
Al
Zn
Al Si Cu
the seam. The classification of Al alloys into two groups is based on the characteristic that the group of the non-hardenable alloys cannot increase the strength through heat treatment, in contrast to hardenable alloys which have such a potential. The important hardening mechanism for this Figure 8.5
br-er08-05.cdr
Si
Al Mg
Mn
Al Mg Mn
Al Mn
ISF 2002
second group is explained by the figures 8.7 und 8.8. Example: If an alloy containing about 4.2% Cu, which is stable at room temperature, is heat treated at 500 then, after a suffiC, ciently long time, there will be only a single phase structure present. All alloy elements were dissolved, Figure 8.8 between point P and Q. When quenched to room
Al - alloys Al99,5 AlCuMg1 AlMgSi0,5 AlSi5 AlMg3 Typical use electrical engineering mechanical engineering, food industries architecture, electrical engineering, anodizing quality architecture, anodizing quality architecture, apparatus-, vehicle-, shipbuilding engineering, furniture industry apparatus-, vehicle-, shipbuilding engineering apparatus-, vehicle-engineering, food industry W elding consumable SG-Al 99,5Ti; SG-Al 99,5 SG-AlMg4,5Mn SG-AlMg5; SG-AlMg4,5Mn; SG-AlSi5 SG-AlSi5 SG-AlMg3; SG-AlMg4,5Mn SG-AlMg5; SG-AlMg3; SG-AlMg4,5Mn SG-AlMn1;SG-Al99,5T
temperature in this condition, no precipitation will take place. The alloy elements are forced to remain dissolved, the crystal is out of equilibrium. If such a structure is subjected to an age hardening at room or elevated temperature, a
AlMg2Mn0,8 AlMn1
Figure 8.6
non-hardenable alloys
hardenable alloys
678
8. Welding of Aluminium
100
cordance with the binary system, the crystal tries to get back into thermodynamical equilibrium. Depending on the level of hardening temperature, the
warm ageing
stable condition
repeated hardening
quenching
oversaturated solid solution, metastable condition
precipitation takes place in three possible forms: copartly coherent precipitations, warm aged condition
herent particles (i.e. particles deviating from the matrix in their chemical composition but having the same partly lattice coherent structure), particles
Ageing Mechanism
(i.e. the lattice structure of the matrix is partly retained), and incoherent
Figure 8.7
particles (lattice structure completely different from the matrix), Figure 8.7. Coherent particles formed at room temperature can be transformed into incoherent particles by increase of temperature (i.e. enabling diffusion). The precipitations cause a restriction to the dislocation movement in the matrix lattice, thus
600
700 liquid liquid and solid Q copper containing aluminium solid solution 500 P
leading to an increase in strength. The finer the precipitations, the stronger the effect. At an increased temperature (heat ageing, Figure 8.7) a maximum of second phase has precipitated after elapse of a certain time. Consequently a prolonged stop at this temperature does not lead to an increased strength, but to coarsening of particles due to diffusion processes and to a decrease in strength (less bigger particles in an extended space).
br-er08-08.cdr
Temperature
400
300
aluminium solid solution and copper aluminide (Al2Cu)
200
Copper
ISF 2002
Figure 8.8
8. Welding of Aluminium
101
After a very long heat ageing a stable condition is reached again with relatively large precipitations of the second phase in the matrix. In Figure 8.7 is this stable final condition iden-
500 P C 400
tical with the starting condition. A deterioration of mechanical properties only happens during hot ageing, if the ageing time is excessively long.
quenching
Temperature
300
200
heat ageing
100
The complete process of hardening at room temperature is metallographic also called age hardening, at elevated temperature heat age-
age hardening
0 2 4 6 8 10 12 h 14
Time
br-er08-09.cdr
Figure 8.9 shows a schematic representation of time-temperature curves during hardening with age hardening and heat ageing.
Figure 8.9
Figure
8.10
shows
the
380
water quenching (~900C/min) air cooling (~30C/min)
strength increase of AlZnMg The difference between age hardening and heat ageing is here very clear. Due to improved diffusion conditions is the strength increase in the case of heat ageing much faster than in the case of age hardening. The strength maximum is also reached considerably earbr-er-08-10.cdr
1 in dependence of time.
320
260
120C
200
RT
140
80 10-1
quenched
100
Ageing time in h
101
10
10
ISF 2002
Figure 8.10
lier. The curve of hot ageing shows clearly the begin of strength loss when held at a too long stoppage time. This figure shows another specialty of the process of ageing. During ageing, a
8. Welding of Aluminium
102
second phase is precipitated from a single-phase structure. To initiate this process, the structure must contain nuclei of the second phase. However, a certain time is required to develop such nuclei. Only after formation of nuclei can the increase in strength start. The period up to this point is called incubation time.
500
Tensile strength sB
Figure 8.11 shows the effect of the height of ageing temperature level on both, mechanical properties of a hardenable Al-alloy and on incubation time. The lower the ageing temperastress and tensile strength. If a low ageing temperature is selected, the ageing time as well as the incubation time become extremely long. ture, the higher the resulting values of yield
N/mm
400
110
135
150
180
300
230
190 205
260C
200
N/mm 400
0.2% yield stress s0.2
110
300
135
150
200
230 260
Fracture elongation d2
30 %
190
205
180
150
135
Figure 8.11 shows that a the maximum yield stress is reached after a period of about one year under a temperature of 110 An inC. crease of the ageing temperature shortens the duration of the complete precipitation process by a certain value raised by 1 to a power. On the other hand, such an acceleration of ageing
400 N/mm 300
Tensile strength Rm
20
110C
10
260
230
30 min
1 day
0
1 week
1 1 month year
10
-2
10
-1
10 10 Ageing time
10
10
h 10
br-er08-11.cdr
ISF 2002
Figure 8.11 leads to a lowering of the maximum strength. As the lower part of the figure shows, the fracture elongation is counter-proportional to the strength values, i.e. the strength increase caused by ageing is accompanied by an embrittlement of the material.
AlMg5 AlMg3
200
100
Al99,5
0 0
br-er-08-12.cdr
30
70
Strain
ISF 2002
Figure 8.12
8. Welding of Aluminium
103
Figure 8.12 shows a method of how to increase the strength of non-hardenable alloys. As no precipitations are present to reduce the movement of dislocations, such alloys can only be strengthened by cold working. Figure 8.12 illustrates two essential mechanisms of strength increase of such alloys. On
300 N/mm 250
one hand, tensile strength increases with increasing content of alloy elements (solid solution strengthening), on the other hand, this increase is caused by a stronger deformation of the lattice. Figure 8.13 shows the effect of the welding
0,7 0,6
Rp0,2/Rm
200
Rm or Rp0,2
150
process on mechanical properties of a coldworked alloy. Due to the heat input during welding, the blocked dislocations are released (recovery), in addition, a grain coarsening will start in the HAZ. This is followed by a strong drop in yield point and tensile strength. This strength loss cannot be overcome in the case of a welding process.
100
0,5 50
HV30
br-er08-13.cdr
ISF 2002
Non-Hardenable Al Alloy
Figure 8.13
400
Figure
8.14
illustrates
the
N/mm 350
90 days RT
Rm
21 days RT
300
1 day RT
As a consequence of the welding heat, the precipitations are solution heat treated and the strength values decrease in the weld area. Due to the age hardening, a restrengthening of the alloys takes place with increasing time.
250
Rp0,2
90 days RT 21 days RT
200
150
1 day RT
100
4 mm plates of: AlZnMg1F32 start values: Rp0,2=263N/mm Rm=363 N/mm welding method: WIG, both sides, simultaneously welding consumable: S-AlMg5 specimens with machined weld bead
20 20 60 0 40 Distance from seam centre
50 80
br-er-08-14.cdr
60
40
80
100
mm
140
ISF 2002
Hardenable Al Alloy
Figure 8.14
8. Welding of Aluminium
104
Figure 8.15 shows another problematic nature of Alwelding. Due to the high thermal expansion of aluminium, high tensions develop during solidification of the weld pool in the course of the welding cycle. If the welded alloy indicates a high melting
br-er-08-15.cdr
ISF 2002
Figure 8.15 A relief can be afforded by preheating of the material, Figure 8.16. With an increasing preheat temperature, the amount of fractured welds decreases. The different behaviour of the three displayed alloys can be explained using the right part of the figure. One can see that the manganese
Weld cracking tendency
100 % 80
Cracking susceptibility
content
influences
signifi-
2 60 1 40
X X
Mg
cantly the hot crack susceptibility. The maximum of this hot crack susceptibility is likely with about 1% Mg content (corresponds with alloy 1). With increasing MG content, hot crack susceptibility decreases strongly (see also alloy 2 and 3, left part).
Si
3
X X
20
100
200
300
400
500
Preheat temperature
2
Alloy content
br-er-08-16.cdr
ISF 2002
Figure 8.16
To avoid hot cracking, partly very different preheat temperatures are recommended for the alloys. Zschtge proposed a calculation method which compares the heat conductivity conditions of the Al alloy with those of a carbon steel with 0.2% C. The formula is shown in Figure
8. Welding of Aluminium
105
8.17, together with the related calculation result. These results are only to be regarded as approximate, the individual application is subject to the information of the manufacturer. Another strong major porosity problem of the
TS Tvorw. lAl-Leg.
660
TVorw. = TS in C in C in J/cm*s*K
during Al welding is the welded joint. It is based on the interplay of several characteristics and hard to suppress. Pores in Al are mostly formed by hydrogen, which is driven out of the weld pool during solidification. Solubility of hydrogen in aluminium on the changes phase
745 l Al-Leg.;
temperature of melt start (solidus temperature) preheat temperature heat conductivity
ISF 2002
Figure 8.17
abruptly
transition melt-crystal, i.e. the melt dissolves many times more of the hydrogen than the just forming crystal at the same temperature. This leads to a surplus of hydrogen in the melt due to the crystallisation during solidification. This surplus precipitates in form of a gas bubble at the solidifying front. As the melting point of Al is very low and Al has a very high heat conductivity, the solidification speed of Al is relatively high. As a result, in the melt ousted gas bubbles have often no chance to rise all the way to the surface. Instead, they are passed by the solidifying front and remain in the weld metal as pores, Figure 8.18. Figure 8.18
br-er-08-18.cdr ISF 2002
500
Al Zn Mg Cu 0,5 Al Zn Mg Cu 1,5
Al Si 5 Al Cu Mg 1 Al R Mg 2 Al Cu Mg 0,5 Al Mn Al Mg 2 Al Cu Mg 2 Al Mg 3 Al Mg 3 Si Al Mg Mn
600
8. Welding of Aluminium
106
To suppress such pore
too thick and water containing oxyde layer by too long or open storage in non air-conditioned rooms
formation it is therefore necessary to minimise the hydrogen content in the melt. Figure 8.19 shows possible sources of hydrogen during MIG welding of Al.
humid air
VS
too thick oxyde layer (condensed water) dirt film (oil, grease)
base material
br-er-08-19.cdr
Grundwerkstoff
ISF 2002
Figure 8.19
Figure 8.20 and 8.21 show the effect of pure thermal expansion during Al welding. The large thermal expansion of the aluminium along with the relatively large heat affected zones cause in combination with a parallel gap adjustment a strong distortion of the welded parts. To minimise this distortion, the workpieces must be set at a suitable angle before welding, Figure 8.21. Figure 8.20
br-er-08-20.cdr
parallel gap
weld pool
overlap
opening gap
weld pool
ISF 2002
8. Welding of Aluminium
107
wedge
flame
br-er08-21.cdr
ISF 2002
Figure 8.21
9. Welding Defects
9. Welding Defects
109
Figures 9.1 to 9.4 give a rough survey about the classification of welding defects to DIN 8524. This standard does not classify existing welding defects according to their origin but only to their appearance.
Figure 9.1
Figure 9.2
Figure 9.3
9. Welding Defects
110
A distinction of arising defects by their origin is shown in Figure 9.5. The development of the most important welding defects is explained in the following paragraphs.
Lack of fusion is defined as unfused area between weld metal and base material or previously welded layer. This happens when the base metal or the previous layer are not completely or insufficiently
molten. Figure 9.6 explains the influence of welding parameters on the development of lack of fusion. In Figure 9.4 the upper part, arc characteristic lines of MAG welding are shown using CO2 and mixed gas. The welding voltage depends on welding current and is selected according to the joint type. With present tension, the welding current is fixed by the wire feed speed (thus also
melting rate) as shown in the middle part of the figFigure 9.5 ure.
Melting rate (resulting from selected welding parameters) and welding speed define the heat input. As it can be changed within certain limits, melting rate and welding speed do not limit each other, but a working range is created (lower part of the figure). If the heat input is too low, i.e. too high welding speed, a definite melting of flanks cannot be ensured. Due to the
9. Welding Defects
111
poor power, lack of fusion is the result. With too high heat input, i.e. too low welding speed, the weld pool gets too large and starts to flow away in the area in front of the arc. This effect prevents a melting of the base metal. The arc is not directed into the base metal, but onto the weld pool, and flanks are not entirely molten. Thus lack of fusion may occur in such areas.
Figure 9.6
Figure 9.7
Figure 9.7 shows the influence of torch position on the development of weak fusion. The upper part of the figure explains the terms neutral, positive and negative torch angle. Compared with a neutral position, the seam gets wider with a positive inclination together with a slight reduction of penetration depth. A negative inclination leads to narrower beads. The second part of the figure shows the torch orientation transverse to welding direction with multi-pass welding. To avoid weak fusion between layers, the torch orientation is of great importance, as it provides a reliable melting and a proper fusion of the layers. The third figure illustrates the influence of torch orientation during welding of a fillet weld. With a false torch orientation, the perpendicular flank is insufficiently molten, a lack of fusion occurs. When welding an I-groove in two layers, it must be ensured that the plate is com-
9. Welding Defects
112
pletely fused. A false torch orientation may lead to lack of fusion between the layers, as shown in the lower figure. Figure 9.8 shows the influence of the torch orientation during MSG welding of a rotating workpiece. As an example, the upper figure shows the desired torch orientation for usual welding speeds. This orientation depends on parameters diameter groove Figure 9.8 like and shape, workpiece thickness, melting
variations of torch orientation on seam formation. A torch orientation should be chosen in such a way that a solidification of the melt pool takes place in 12 o'clock position, i.e. the weld pool does not flow in front or behind of the arc. Both may cause lack of fusion.
In contrast to faulty fusion, pores in the weld metal due to their globular shape are less critical, provided that their size does not exceed a certain value. Secondly, they must occur isolated and keep a minimum distance from each other. There are two possible mechanisms to develop cavities in the weld Figure 9.9
metal: the mechanical and the metallurgical pore formation. Figure 9.9 lists causes of a mechanical pore formation as well as possibilities to avoid them. To over-weld a cavity (lack of
9. Welding Defects
113
fusion, gaps, overlaps etc.) of a previous layer can be regarded as a typical case of a mechanical pore formation. The welding heat during welding causes a strong expansion of the gasses contained in the cavity and consequently a development of a gas bubble in the liquid weld metal. If the solidification is carried out so fast that this gas bubble cannot raise to the surface of the weld pool, the pore will be caught in the weld metal.
a) X-ray photograph
b) Surface cross-section
c) Transverse section
br-er09-10.cdr ISF 2002
Figure 9.10
which developed in this way, as well as a surface and a transverse section. This pore formation shows its typical pore position at the edge of the joint and at the fusion li ne of the top layer.
Figure 9.12
9. Welding Defects
114
Figure 9.11 summarises causes of and measures to avoid a metallurgical pore formation. Reason of this pore formation is the considerably increased solubility of the molten metal compared with the solid state. During solidification, the transition of liquid to solid condition causes a leapwise reduction of gas solubility of the steel. As a result, solved gasses are driven out of the crystal and are ena) X-rax photograph
riched as a gas bubble ahead of the solidification front. With a slow growth of the
crystallisation front, the bubbles have enough time to raise to the surface of the weld pool, Figure 9.12 upper part. Pores will not be develb) surface section
oped. However, a higher solidification speed may lead to a case where gas bubbles are passed by the crystallisation front and are trapped as pores in the weld metal, lower part of the figure.
c) transverse section
br-er09-13.cdr ISF 2002
Figure 9.13 shows a X-ray photograph, a surface and a transverse section of a seam with
Figure 9.13 metallurgical pores. The evenly distributed pores across the seam and the accumulation of pores in the upper part of the seam (transverse typical. section) are
Figure 9.14 shows the ways of ingress of gasses into the weld pool as an example during MAG
9. Welding Defects
115
is mainly caused by hydrogen and nitrogen. Oxygen is bonded in a harmless way when using universal electrodes which are alloyed with Si and Mn.
br-er09-15.cdr
ISF 2002
9. Welding Defects
116
Figure 9.15 classifies cracks to DIN 8524, part 3. In contrast to part 1 and 2 of this sta ndard, are cracks not only classified by their appearance, but also by their development.
Figure
9.16
allocates
welding heat cycle. Principally there is a distinction between the group 0010 (hot cracks) and 0020
(cold cracks).
Figure 9.16 A model of remelting development and solidification cracks is shown in Figure 9.17. The upper part illustrates solidification conditions in a simple case of a binary system, under the provision that a complete concentration balance takes place in the melt ahead of the solidification front, but no diffusion takes place in the crystalline solid. When a melt of a composition C0 cools down, a crystalline solid is formed when the liquidus line is reached. Its concentration can be taken from the solidus line. In the course of the ongoing solidification, the rest of molten metal is enriched with alloy elements in accordance with the liquidus line. As defined in the beginning, no diffusion of alloy elements in the already solidified crystal takes place, thus the crystals are enriched with alloy elements much slower Figure 9.17
117
As a result, the concentration of the melt exceeds the maximum equilibrium concentration (C 5), forming at the end of solidification a very much enriched crystalline solid, whose melting point is considerably lower when compared with the firstly developed crystalline solid. Such concentration differences between first and last solidified crystals are called segregations. This model of segregation development is very much simplified, but it is sufficient to understand the mechanism of hot crack formation. The middle part of the figure shows the formation of solidification cracks. Due to the segregation effects described above, the melt between the crystalline solids at the end of solidification has a considerably decreased solidus temperature. As indicated by the black areas, rests of liquid may be trapped by dendrites. If tensile stresses exist (shrinking stress of the welded joint), the liquid areas are not yet able to transfer forces and open up. The lower part of the figure shows the development of remelting cracks. If the base material to be welded contains already some segregations whose melting point is lower than that of the rest of the base metal, then these zones will melt during welding, and the rest of the material remains solid (black areas). If the joint is exposed to te nsile stress during solidification, then these areas open up (see above) and cracks occur. A hot cracking tendency of a steel is above all promoted by sulphur and phosphorus, because these elements form with iron very low melting phases (eutectic point Fe-S at 988C) and these elements segregate i ntensely. In addition, hot crack te ndency increases with increasing melt interval.
As shown in Figure 9.18, also the geometry of the groove is important for hot crack tendency. With narrow, deep grooves a crystallisation takes place of all sides of the bead, entrapping the remaining melt in the bead centre. With the occurrence of shrinking
stresses, hot cracks may Figure 9.18 develop. In the case of flat beads as shown in the
118
solidifies at the surface of the bead. The melt cannot be trapped, hot cracking is not possible. The case in figure c shows no adva ntage, because a remelting crack may occur in the centre (segregation zone) of the first layer during welding the second layer. Figure 9.19
The example of a hot crack in the middle of a SA weld is shown in Figure 9.19. This crack developed due to the unsuitable
groove geometry.
Figure 9.20 shows an example of a remelting crack which started to develop in a segregation zone of the base metal and spread up to the bead centre. The section shown in Figure 9.21 is similar to case c in Figure 9.18. One can Figure 9.20
clearly see that an existing crack develops through the follo wing layers during over-welding. Figure 9.22 classifies cold cracks depending on their position in the weld metal area. Such a classification does not provide an explanation for the origin of the cracks.
9. Welding Defects
119
Figure 9.21
Figure 9.22
Figure 9.23 shows a summary of the three main causes of cold crack formation and their main influences. As explained in previous chapters, the resulting welding microstructure depends on both, the composition of base and filler materials and of the cooling speed of the joint. An unsatisfactory structure composition promotes very much the formation of cold cracks (hardening by martensite).
Another cause for increased cold crack susceptibility is a higher hydrogen content. The hydrogen content is very much influenced by the condition of the welding filler material (humidity of electrodes or flux, lubricating grease on welding wire etc.) and by humidity on the groove edges. The cooling speed is also important because it determines the remaining time for hydrogen effusion out of the bead, respectively how much hydrogen remains in the weld. A measure is t8/1 because only below 100C a hydrogen e ffusion stops.
9. Welding Defects
120
Figure 9.23
Figure 9.24
A crack initiation is effected by stresses. Depending on material condition and the two already mentioned influencing factors, even residual stresses in the workpiece may actuate a crack. Or a crack occurs only when superimpose of residual stresses on outer stress.
Figure 9.24 shows typical cold cracks in a workpiece. An increased hydrogen content in the weld metal leads to an increased cold crack tendency. Mechanisms of hydrogen cracking were not completely understood until today. However, a spontaneous occurrence is typical of hydrogen cracking. Such cracks do not appear directly after welding but hours or even days after cooling. The weld metal hydrogen content depends on humidity of the electrode coating (manual metal arc welding) and of flux (submerged arc welding).
9. Welding Defects
121
Figure 9.25 shows that the moisture pick-up of an electrode coating greatly depends on ambient conditions and on the type of electrode. The upper picture shows that during storage of an electrode type the water content of the coating depends on air humidity. The water content of the coating of this electrode type advances to a maximum value with time. The lower picture shows that this behaviour does not apply to all electrode types. The
characteristics of 25 welding electrodes stored under identical conditions are plotted here. It can clearly be seen that a behaviour as shown in the upper picture applies only to some electrode types, but basically a very different behaviour in connection with storage Figure 9.25 can be noticed.
In practice, such constant storage conditions are not to be found, this is the reason why electrodes are backed before welding to limit the water content of the coating. Figure 9.26 shows the effects of this measure. The upper curve shows the water content of the coating of electrodes which were stored at constant air humidity before rebaking. Humidity values after rebaking are plotted in the lower curve. It can be seen that even electrodes stored under Figure 9.26
9. Welding Defects
122
very damp conditions can be rebaked to reach acceptable values of water content in the coating. Figure 9.27 shows the influence of cooling speed and also the preheat temperature on hydrogen content of the weld metal. The values of a high hygroscopic cellulose-coated
electrode are considerably worse than of a basiccoated one, however both show the same tendency: increased cooling speed Figure 9.27 leads to a raise of diffusible hydrogen content in weld metal. Reason is that hydrogen can still effuse all the way down to room temperature, but diffusion speed increases sharply with temperature. The longer the steel takes to cool, the more time is available for hydrogen to effuse out of the weld metal even in higher quantities. The table in Figure 9.28 shows an assessment of the quantity of diffusible hydrogen in weld metal according to DIN 8529.
Based on this assessment, a classification of weld metal to DIN 32522 into groups depending on hydrogen is carried out, Figure 9.29. Figure 9.28
9. Welding Defects
means of sound emission measurement. Figure 9.30 represents the result of such a measurement of a welded component. A
solid-borne sound microphone is fixed to a component which measures the sound pulses generated by Figure 9.29 crack development. The intensity of the pulses pro-
vides a qualitative assessment of the crack size. The observation is carried out without applying an external tension, i.e. cracks develop only caused by the internal residual stress condition. Figure 9.32 shows that most cracks occur relatively short after welding. At first this is due to the cooling process. Ho wever, after completed cooling a multitude of deve loping sounds can be registered. It is remarkable that the intensity of late occurring pulses is especially high. This behaviour is typical for hydrogen induced crack fo rmation.
Figure 9.31 shows a characteristic occurrence of lamellar cracks (also called lamellar tearing). This crack type occurs typically during stressing a plate across its thickness (perpendicular to rolling direction). The upper picture shows joint types which are very much at risk to formation of such cracks. The two lower pictures show the
cause of that crack fo rmation. During steel production, a formation of segregation cannot be avoided due to Figure 9.30
9. Welding Defects
124
the casting process. With follo wing production steps, such segregations are stretched in the rolling direction. Zones enriched and depleted of alloy elements are now close together. These concentration diffe rences influence the transformation behaviour of the individual zones. During cooling, zones with enriched alloy elements develop a different microstructure than depleted zones. This effect which can be well recognised in Figure 9.31, is called structure banding. In practice, this formation can be hardly avoided. Banding in plates is the reason for worst mechanical properties perpendicular to rolling direction. This is caused by a different mechanical behaviour of different microstructures. When stressing lengthwise and transverse to rolling direction, the individual structure
br-er09-31.cdr ISF 2002
bands may support each other and a mean strength is provided. Such support cannot be obtained perpendicular to rolling direction, thus the strength of the
Figure 9.31 workpiece is that of the weaker microstructure areas. Consequently, a lamellar crack propagates through weaker micro-
structure areas, and partly a jump into the next band takes place.
Figure 9.32 illustrates why such t-joints are particularly vulnerable. DependFigure 9.32 ing on joint shape, these welds show to some extent
9. Welding Defects
125
a considerable shrinking. A welded construction which greatly impedes shrinking of this joint, may generate stresses perpendicular to the plane of magnitude above the tensile strength. This can cause lamellar tearing.
Precipitation cracks occur mainly during stress relief heat treatment of welded components. They occur in the coarse grain zone close to fusion line. As this type of cracks occurs often during post weld heat treatment of cladded materials, is it also called undercladding crack, Figure 9.33.
Especially susceptible are steels which contain alloy elements with a precipitation hardening effect (carbide developer like Ti, Nb, V). During welding such steels, carbides are dissolved in an area close to the fusion line. During the following cooling, the carbide developers are not completely re-precipitated.
If a component in such a condition is stress relief heat treated, a re-precipitation of carbides takes place (see hot ageing, chapter 8). With this re-precipitation, precipitation-free zones may develop along grain boundaries, which have a considerably lower deformation stress limit compared with strengthened areas. Plastic deformations during stress relieving are carried out almost only in these areas, causing the cracks shown in Figure 9.33.
Figure 9.33
Ls
in test area
a
S S
behaviour is the tensile test. Generally, it is carried out using a round specimen. When determining the strength of a welded joint, also standardised flat specimens are used. Figure 10.1 shows both standard specimen shapes for that test. A specimen is ruptured by a test machine while the actual force and the elongation of the specimen is measured. With these measurement values, tension and strain are calculated. If is plotted over , the drawn diagram is typical for this test, Figure 10.2. Normally, if a steel with a bcc lattice structure is tested, a curve with a clear yield point is obtained (upper picture). Steels with a fcc lattice structure show a curve without yield point. The most important characteristic values which are determined by this test are: yield
in test area
L0 Lc Lt
Lt b1 b b
Lc parallel length ) ) radius of throat r ) for pressure welding and beam welding, L S = 0. 2 ) for some other metallic materials (e.g.aluminium, copper and their alloys) __ L c L S +100 may be required
1
depends on test unit b + 12 12 with a 2 25 with a > 2 6 with D 50 12 with 50 < D 168,3 L S + 60 25
LO LC Lt d = specimen diameter d1 = head diameter depending on clamping device LC = test length = L0 + d/2 r = 2 mm
br-er10-01.cdr
L0 = measurement length (L0 = k S0 with k = 5,65) Lt = total length S0 = initial cross-section within test length
ISF 2002
Flat and Round Tensile Test Specimen to EN 895, EN 876, and EN 10 002
Figure 10.1
s
d1
b1
Rm ReH Rel sf
stress ReL, tensile strength Rm, and elongation A. To determine the deformability of a weld, a
ALud Ag A e
bending test to DIN EN 910 is used, Figure 10.3. In this test, the specimen is put onto two supporting rollers and a former is pressed through between the rollers. The distance of the supporting rollers is Lf = d + 3a (former diameter + three times specimen thickness). The backside of the specimen (tension side) is observed. If a surface crack develops, the
ISF 2002
s Rm RP0,2 RP0,01 sf
0,2 % 0,01 % Ag A
br-er10-02.cdr
test will be stopped and the angle to which the specimen could be bent is measured. The
Figure 10.2
128
test result is the bending angle and the diameter of the used former. A bending angle of 180 is reached, if the specimen is pressed through the supporting rollers without development of a crack. In Figure 10.3 specimen shapes of this test are shown. Depending on the direction the weld is bent, one distinguishes (from top to bottom) transverse, side, and longitudinal bending specimen. The tension side of all three specimen types is machined to eliminate any influences on the test through notch effects. Specimen thickness of transverse and longitudinal specimens thickness. is Side the plate bending
br-er10-03.cdr
r
a
a tension side
B Lt
distance of supporting rollers specimen length former diameter supporting roller diameter: 50 mm specimen thickness radius of specimen edge specimen width
specimens are normally only used with very thick plates, here the specimen thickness is fixed at 10 mm.
Lt
Figure 10.3
A determination of the toughness of a material or welded joint is carried out with the notched bar impact test. A cuboid specimen with a V-notch is placed on a support and then hit by a pendulum ram of the im55 10 r = 0,25 10 8
pact testing machine (with very tough materials, the specimen will be bent and
40
45 J 40 Charpy impact energy 35 30 25 20 15 10 -80 -60
br-er10-04.cdr
45
D im e nsio ns leng th width hight notc h angle thic knes s in notch g roove notc h rad ius notc h d is tanc e from end of s p ecim en angle b etwee n no tch axis and long itudinal axis
drawn through the supports). The used energy is measured. notch Figure 10.4 (Iso-Vrepresents sample shape, shape specimen), and a schematic presentation of test results.
20 40 C
Figure 10.4
B Lt
r r
Designation
Weld centre
b
VWS a/b
Dicke
RL
the range of scatter are entered on the impact energy-temperature diagram (AV-T curve).
b
VWT 0/b
Dicke
VHT 0/b
a
This graph is divided into an area of high impact energy values, a transition range, and an
b
VWT a/b
VHT a/b
a a
VWT 0/b
RL
assigned to the transition range, i.e. the rapid drop of toughness values. When the temperature falls below this transition temperature, a transition of tough to brittle fracture behaviour takes place. As this steep drop mostly extends across a certain area, a precise assignment of transition temperature cannot be carried out. Following DIN 50 115, three definitions of the transition temperature are useful, i.e. to fix T to:
VHT a/b
a RL
VWT a/b
a RL
VHT a/b
a RL
V = Charpy-V notch W = notch in weld metal; reference line is centre line of weld H = notch in heat affected zone; reference line is fusion line or bonding zone (notch should be in heat affected zone) S = notched area parallel to surface T = notch through thickness a = distance of notch centre from reference line (if a is on centre line of weld, a = 0 and should be marked) b = distance between top side of welded joint and nearest surface of the specimen (if b is on the weld surface, then b = 0 and should be marked)
br-er10-05.cdr ISF 2002
Figure 10.5
1.) a temperature where the level of impact values is half of the level of the high range, 2.) a temperature, where the fracture area of the specimen shows still 50% of tough fracture behaviour 3.) a temperature with an impact energy value of 27 J. Figure 10.5 illustrates a specimen position and notch position related to the weld according to DIN EN 875. By modifying the notch position, the impact energy of the individual areas like HAZ, fusion line, weld metal, and base metal can be determined in a relatively accurate way. Figure 10.6 presents the influence of various alloy elements on the AV-T - curve. Three basically different influences can be seen. Increasing manganese contents increase the impact values in the area of the high level and move the transition temperature to lower values. The values of the low levels remain unchanged, thus the steepness of the drop becomes clearer with increasing Mn-content. Carbon acts exactly in the opposite way. An increasing carbon content increases the transition temperature and lowers the values of the high level, the steel becomes more brittle. Nickel decreases slightly the values of the high level, but increases the
J 300
tent. Starting with a certain Nickel content (depends also from other alloy elements), a steep drop does not happen, even at lowest temperature the steel shows a tough fracture behaviour.
0% Mn
200
100
commonly used steels are collected. These curves are marked with points for impact en0,1% C
27 200 J 100
ergy values of AV = 27 J as well as with points where the level of impact energy has fallen to half of the high level. It can clearly be seen that mild steels have the lowest impact energy values together with the highest transition temperature. The development of finegrain structural steels resulted in a clear improvement of impact energy values and in
0,4% C
27 -150
br-er10-06.cdr
0,8% C
-100
-50 0 Temperature
50
C 100
ISF 2002
Figure 10.6
addition, the application of such steels could be extended to a considerably lower temperature range. With the example of the steels St E 355 and St E 690 it is clearly visible that an increase of strength goes mostly hand in hand with a decrease of the impact energy level. Another improvement showed the application of a thermomechanical treatment (controlled rolling during heat treatment). The application of this treatment resulted in an increase of strength and impact energy values together with a parallel saving of alloy elements. To make a comparison, the AV-T - curve of the cryogenic and high alloyed steel X8Ni9 was plotted onto the diabr-er10-07.cdr
specimen position: weld centre, notch parallel to surface specimen shape: standard specimen with V-notch J 300 Charpy impact energy AV X8Ni9 S460M S355N
100
S355J2G3
Figure 10.7
10. Testing of Welded Joints under very high test speed in the impact enP
0,55h 0,25 1,2h 0,25
131
ergy test, thus there are no reliable findings about crack growth and fracture mechanisms.
a
C P
Figure 10.8 shows two commonly used specimen shapes for a fracture mechanics test to determine crack initiation and crack growth. The lower figure to the right shows a possibility how to observe a crack propagation in a compact tensile specimen. During the test, a current I flows through the specimen, and the tension drop above the notch is measured. As soon as a crack propagates through the material, the current conveying cross section decreases, resulting in an increased voltage drop. Below to the left a measurement graph
b L
CT - specimen
h 1,25h 0,13 specimen height h = 2b 0,25 specimen width b total crack length a = (0,50 0,05)h test load P
2,1h
br-er10-08.cdr
ISF 2002
Figure 10.8
of such a test is shown. If the force F is plotted across the widening V, the drawn curve does not indicate precisely the crack initiation. Analogous to the stress-strain diagram, a decrease of force is caused by a reduction of the stressed cross-section. If the voltage drop is plotted over the force, then the start of crack initiation can be determined with suitable accuracy, and the crack propagation can
F
F
be observed. Another typical characteristic of material behaviour is the hardness of the work-
(standardised
to
DIN 50 351) and Vickers (DIN 50 133). When testing to Brinell, a steel ball is
Hardness Testing to Brinell and Vickers
br-er-10-09.cdr
Figure 10.9
132
to the surface of the tested workpiece. The diameter of the resulting impression is measured and is a magnitude of hardness. The hardness value is calculated from test load, ball diameter, and diameter of rim of the impression (you find the formulas in the standards). The hardness information contains in addition to the hardness magnitude the ball diameter in mm, applied load in kp and time of influence of the test load in s. This information is not required for a ball diameter of 10 mm, a test load of 3000 kp (29420 N), and a time of influence of 10 to 15 s. This hardness test method may be used only on soft materials up to 450 BHN
1
0,200 mm
3 6 7
2
100 0
0,200 mm
4 5 3 8
3 10
130 30 0
1 6
3 7
4 5 3 8
3 10
6
hardness scale
This method is standardised to DIN 50133. Instead of a ball, a diamond pyramid is pressed into the workpiece. The lengths of the two diagonals of the impression are measured and the hardness value is calculated from their average and the test load. The impressions of the test body are always geometrically similar, so that the hardness value is normally independent from the size of the test load. In practice, there is a hardness increase under a lower test load because of an increase of the elastic part of the deformation.
0,200 mm
0,200 mm
8,9 10
6 8,9 7 10
30 0
Abbreviation 1 2 3 4 5 6 7 8 9 F0 F1 F t0 t1 tb e HRC HRA cone angle = 120 test preload test load total test load = F0 + F1
penetration depth in mm under test preload F0. This defines the reference level for measurement of tb. total penetrationn depth in mm under test load F1 resulting penetration depth in mm, measured after release of F1 to F0 resulting penetration depth, expressed in units of 0,002 mm: tb / 0,002 Rockwell hardness = 100 - e HRB HRF e =
10
br-er10-10.cdr
ISF 2002
Figure 10.10
Hardness testing to Vickers is almost universally applicable. It covers the entire range of materials (from 3 VHN for lead up to 1500 VHN for hard metal). In addition, a hardness test can be carried out in the micro-range or with thin layers. Figure 10.10 illustrates a hardness test to Rockwell. In DIN 50103 are various methods standardised which are based on the same principle. With this method, the penetration depth of a penetrator is measured.
133
At first, the penetrator is put on the workpiece by application of a pre-test load. The purpose is to get a firm contact between workpiece and penetrator and to compensate for possible play of the device. Then the test load is applied in a shock-free way (at least four times the pre-force) and held for a certain time. Afterwards it is released to reach minor load. The remaining penetration depth is characteristic for the hardness. If the display instrument is suitably scaled, the hardness value can be read-out directly. All hardness test methods to Rockwell use a ball (diameter 1.5875 mm, equiv. to 1/16 Inch) or a diamond sphero-conical penetrator (cone angle 120 as the penetrating body. There are ) differences in size of pre- and test load, so different test methods are scaled for different hardness ranges. The most commonly used scale methods are Rockwell B and C. The most considerable advantage of these test methods compared with Vickers and Brinell are the low time duration and a possible fully-automatic measurement value recognition. The disadvantage is the reduced accuracy in contrast to the other methods. Measured hardness numbers are only comparable under identical conditions and with the same test method. A comparison of hardness values which were determined with different methods can only be carried out for similar materials. A conversion of hardness values of different methods can be carried out
piston
for steel and cast steel according to a table in DIN 50150. A relation of hardness and tensile strength is also given in that table. All the hardness test methods described above require a coupon which must be taken from the workpiece and whose hardness is then determined in a test machine. If a workpiece on-site is to be tested, a dynamical hardness test method will be applied. The advantage of these methods is that measurements can be taken on completed constructions with handheld units in any position. Figure 10.11 illustrates a Figure 10.11
br-er10-11.cdr ISF 2002
reference bar
specimen
Poldi - Hammer
134
hardness test using a Poldi-Hammer. With this (out of date) method, the measurement is carried out by a comparison of the workpiece hardness with a calibration piece. For this purpose a calibration bar of exactly determined hardness is inserted into the unit, which is held by a spring force play-free between a piston and a penetrator (steel ball, 10 mm diameter). The unit is put on the workpiece to be tested. By a hammerblow to the piston, the penetrator penetrates the workpiece and the calibration pin simultaneously. The size of both impressions is measured and with the known hardness of the calibration bar the hardness of the workpiece can be determined. However, there are many sources of errors with this method which may influence the test result, e.g. an inclined resting of the unit on the surface or a hammerblow which is not in line with the device axis. The major source of errors is the measurement of the ball impression on the workpiece. On one hand, the edge of the impression is often unsharp because of the great ball diameter, on the other hand the measurement of the impression using magnifying glasses is subjected to serious errors. Figure 10.12 shows a modern measurement method which works with ultrasound and combines a high flexibility with easy handling and high accuracy. Here a test tip is pressed manually against a workpiece. If a defined test load is passed, a spring mechanism inside the test tip is triggered and the measurement starts.
Test force
5.0 kp 4.0
the steel. The measurement tip is excited to emit ultrasonic oscillations by a piezoelectric crystal. The test tip (diamond pyramid) penetrates the workpiece under the test pressure
3.0
2.0 Federweg
caused by the spring force. With increasing penetration depth the damping of the ultrasonic oscillation changes and consequently the frequency. This change is measured by the device. The damping of the ultrasonic os- little work on surface preparation of specimens (test force 5 kp) - Data Logger for storage of several thousands of measurement points - interfaces for connection of computers or printers - for hardness testing on site in confined locations
cillation depends directly on penetration depth thus being a measure for material hardness. The display can be calibrated for all commonly used measurement methods, a measurement is carried out quickly and easily.
br-er10-12.cdr
ISF 2002
Figure 10.12
135
Measurements can also be carried out in confined spaces. This measurement method is not yet standardised.
+ tension
m > a
m < a
m = a
Description
m = 0
Dye penetrant method crack is free, surface is clean
Application
all materials with surface cracks
compression -
m < a
m = a
m > a
time
crack and surface with penetrant liquid cleaned surface, dye penetrant liquid in crack surface with developer shows the crack by coloring Magnetic particle testing
alternating range
Whler line
A workpiece is placed between the poles of a magnet or solenoid. Defective parts disturb the power flux. Iron particles are collected.
III 0 1 10 102 103 104 105 106 Fatigue strength (endurance) number lg N
I area of overload with material damage II area of overload without material damage III area of load below fatigue strength limit
br-er10-14.cdr
107
Surface cracks and cracks up to 4 mm below surface. However: Only magnetizable materials and only for cracks perpendicular to power lines
br-er10-13.cdr
ISF 2002
ISF 2002
Figure 10.13
Figure 10.14
To test a workpiece under oscillating stress, the fatigue test is standardised in DIN 50100. Mostly a fatigue strength is determined by the Whler procedure. Here some specimens (normally 6 to 10) are exposed to an oscillating stress and the number of endured oscillations until rupture is determined (endurance number, number of cycles to failure). Depending on where the specimen is to be stressed in the range of pulsating tensile stresses, alternating stresses, or pulsating compressive stresses, the mean stress (or sub stress) of a specimen group is kept constant and the stress amplitude (or upper stress) is varied from specimen to specimen, Figure 10.13. In this way, the stress amplitude can be determined with a given medium stress (prestress) which can persist for infinite time without damage (in the test: 107 times). Test results are presented in fatigue strength diagrams (see also DIN 50 100). As an example the extended Whler diagram is shown in Figure 10.13. The upper line, the Whler line, indicates after how many cycles the specimen ruptures under tension amplitude a. The
136
Description X-ray or isotope radiation penetrate a workpiece. The thicker the workpiece, the weaker the radiation reaching the underside.
W ire diameter
Tolerated deviation
W ire number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
mm 3,2 2,5 2 1,6 1,25 1 0,8 0,63 0,5 0,4 0,32 0,25 0,2 0,16 0,125 0,1
mm 0,03
0,02
0,01
0,005
Abbreviation FE 1/7 FE 6/12 FE 10/16 CU 1/7 CU 6/12 W ire number to Table 1 1 to 7 6 to 12 10 to 16 1 to 7 6 to 12 10 to 16 1 to 7 6 to 12 10 to 16 W ire length mm 50 50 or 25 50 or 25 50 50 50 or 25 50 50 50 or 25 aluminium copper W ire material Material groups to be tested
mild steel
iron materials
radiation source
workpiece
copper, zink, tin and its alloys aluminium and its alloys
film (displayed in distance from workpiece) defect in radiation direction; difficult to identify (flank lack of fusion) defect in radiation direction; easy to identify
br-er10-15.cdr ISF 2002
br-er10-16.cdr
ISF 2002
Figure 10.15
Description US-head generates high-frequency sound waves, which are transferred via oil coupling to the workpiece. Sound waves are reflected on interfaces (echo).
Application Mainly for defects with an orientation transverse to sound input direction.
damage to the material starts in form of cracks. Below this line, a material damage does not occur.
Test
methods
described
above
require
specimens taken out of the workpiece and a partly very accurate sample preparation. A testing of completed welded constructions is impossible, because this would require a destruction of the workpiece. This is the reason why various non-destructive test methods were developed, which are not used to determine technological properties but test the workpiece for defects. Figure 10.14 shows
sound head oil coupling workpiece defect ultrasonic test device radiation pulse defect echo backwall echo
br-er10-17.cdr
ISF 2002
Figure 10.17
10. Testing of Welded Joints two methods to test a workpiece for surface defects.
137
Figure 10.15 illustrates the principle of radiographic testing which allows to identify also defects in the middle of a weld. The size of the minimum detectable defects depends greatly on the intensity of radiation, which must be adapted to the thickness of the workpiece to be radiated. As the film with documented defects does not permit an estimation of the plate thickness, a scale bar must be shown for estimation of the defect size. For that purpose, a plastic template is put on the workpiece before radiation which contains metal wires with different thickness and incorporated metallic marks, Figure 10.16. The size of the thinnest recognisable wire indicates the Figure 10.18 size of the smallest visible defect. Radiation testing provides information about the defect position in the plate plane, but not about the position within the thickness depth. A clear advantage is the good documentation ability of defects. An information about the depth of the defect is provided by testing the workpiece with ultrasound. The principle is shown in Figures 10.17 and 10.18 (principle of a sonar). The
br-er10-19.cdr
display
of
original
Figure 10.19
138
This method provides not only a perpendicular sound test, but also inaccessible regions can be tested with the use of so called angle testing heads, Figure 10.19.
30
Pores between 10 and 20 mm depth provide an unbroken echo sequence across the entire display starting from 10mm. The backwall echo sequence of 30 mm is not yet visible. Echo sequence of 20 mm depth. The backwall is completely screened.
Wall thickness is below 40 mm. The roughness provides smaller and wider echos.
40 The oblique and rough defect from 20 to 30 mm provides a wide echo of 20 to 30 mm. Starting with SKW 4, an unbroken echo sequence follows. The inclination of the reflector is recornised by a change of the 1st echo when shifting the test head. Echo sequence of 10 mm depth. The reflector in 30 mm depth is completely screened.
The perpendicular crack penetrating the material does not provide a display because the reflecting surface (tip of crack) is too small.
The oblique backwall reflects the soundwaves against the crack. this is the reason why an impossible depth of 65 mm is displayed.
br-er10-20.cdr
ISF 2002
br-er10-21.cdr
ISF 2002
Figure .10.20
schematically
the
base material
ferrite + perlite
bainite
rect interpretation of all the signals requires great experience, because the
2,5 mm
ferrite + perlite
fusion line Steel: S355N (T StE 355) weld metal cast structure
bainite
shape of the displayed signals is often not so clear. Figure 10.22 illustrates the potential of metallographic examination. Grinding and
br-er10-22.cdr
Figure 10.22
139 etching with an acid makes the microstructure visible. The reason is that depending on structure and orientation, the individual grains react very differently to the acid attack thus
25
100
Fe
% Cr
% Fe 80
reflecting the light in a different way. The macrosection, i.e. without magnification, gives a complete survey about the weld and fusion line, size of the HAZ, and sequence of solidification. Under adequate magnification, these areas can still not be distinguished precisely, however, an assessment of the developed
Cr
20 60 40 15
20 0
10 % Ni 8 10
Ni
6 4 2 0 0 200 mm 100 0 100 5
microstructure is possible. An assessment of the distribution of alloy elements across the welded joint can be carried out by the electron beam micro-analysis. An example of such an analysis is shown in Figure 10.23. If a solid body is exposed to a
Figure 10.23
focused electron beam of high energy, its atoms are excited to radiate X-rays. There is a simple relation between the wave length of this radiation and the atomic number of the chemical elements. As the intensity of the radiation depends on the concentration of the elements, the chemical composition of the solid body can be concluded from a survey of the emitted X-ray and spectrum quantita50 20 20 50
50
qualitatively
tively. A detection limit is about 0.01 mass % with this method. Microstructure areas of a minimum diameter of about 5 m can be analysed. If the electron beam is moved across the specimen (or the specimen under the
1. weld 2. weld
0 10
50
20 20
weld
weld
Agents: - electrolytic copper in the form of chips (min. 50 g/l test solution) - 100 ml H2SO4 diluted with 1 l water and then . 110 g CuSO 5 H2O are added Test: The specimens remain for 15 h in the boiling test solution. Then the specimens are bent across a former up to an angle of 90 and finally examined for grain failure under a 6 to 10 times magnification.
br-er-10-24.cdr
beam), the element distribution along a line across the Figure 10.24
Strau - Test
140
solid body can be determined. Figure 10.23 presents the distribution of Ni, Cr, and Fe in the transition zone of an austenitic plating in a ferritic base metal. The upper part shows the related microsection which belongs to the analysed part. This microanalysis was carried out along a straight line between two impressions of a Vickers hardness test. The impressions are also used as a mark to identify precisely the area to be analysed. The so called Strau test is
12
web
80
sion. Figure 10.24 shows the specimen shape which is normally used for that test. In addition, some details of the test method are explained.
12
a a a
80 120
br-er-10-25.cdr
Figure 10.25 Figure 10.25 presents a specimen shape for testing the crack susceptibility of welding consumables. For this test, weld number 1 is welded first. The 2. weld is welded not later than 20 s in reversed direction after completion of the first weld. Throat thickness of weld 2 must be 20% below of weld 1. After cooling down, the beads are examined for cracks. If
tensioning bolt hexagon nut min. M12 DIN 934 a tensioning plate specimen base body guidance plates
br-er-10-26.cdr
Figure 10.26
a a
a a
cracks are found in weld 1, the test is void. If weld 1 is free from cracks, weld 2 is examined for crack with magnifying glasses. Then weld 1 is machined off and weld 2 is cracked by bending the weld from the root. Test results record any
141
surface and root cracks together with information about position, orientation, number, and length. The welding consumable is regarded as 'non-crack-susceptible' if the welds of this test are free from cracks. Figure 10.26 presents two proposals for self-stressing specimens for plate tests regarding their hot crack tendency. Such tests are not yet standardised to DIN.
thermo couple groove shape 60 cross-section 60 weld metal support plate Wd./2 Wd. Wd./2 H Hc implant electrode welding direction
specimen shape
start
end crater
crack coefficient
c
150
C=
x 100 (in %)
800 500
2 3
br-er10-27.cdr
ISF 2002
br-er10-28.cdr
ISF 2002
Tekken Test
Implant Test
Figure 10.27
Figure 10.28
There are various tests to examine a cold crack tendency of welded joints. The most important ones are the self-stressing Tekken test and the Implant test where the stress comes from an external source. In the Tekken test which is standardised in Japan, two plates are coupled with anchor joints at the ends as a step in joint preparation see Figure 10.27. Then a test bead is welded along the centre line. After storing the specimen for 48 hours, it is examined for surface cracks. For a more precise examination, various transverse sections are planned. The value to be determined is the minimum working temperature at which cracks no longer occur. The specimen shape simulates the conditions during welding of a root pass.
142
The most commonly used cold crack test is the Implant test, Figure 10.28. A cylindrical body (Implant) is inserted into the bore hole of a support plate and fixed by a surface bead. After the bead has cooled down to 150 the implant is exposed to a constant load. The time is C measured until a rupture or a crack occurs (depending on test criterion 'rupture' or 'crack'). Varying the load provides the possibility to determine the stress which can be born for 16 hours without appearance of a crack or rupture. If a stress is specified to be of the size of the yield point as a requirement, a preheat temperature can be determined by varying the working temperature to the point at which cracks no longer appear. As explained in chapter 'cold cracks' the hydrogen content plays an important role for cold crack development. Figure 10.29 shows results of trials where the cold crack behaviour was examined using the Tekken and Implant test. Variables of these tests were hydrogen content of the weld metal and preheat temperature. The variation of the hydrogen content of the weld metal was carried out by different exposure to humidity (or rebaking) of the used stick electrodes. Based on the hydrogen content, the preheat temperature was increased test by test. Consequently, the curves of Figure 10.29 represent the limit curves for the related test. Specimens above these
heat input: 12 kJ/cm basic coated stick electrode plate and support plate thickness: 38 mm
curves remain free from cracks, below these curves cracks are present. Evident for both graphs is that with increased preheat
C Implant-Test 150 Preheat temperature Rcr = Rp0,2 = 358 N/mm Preheat temperature
C Tekken-Test 150
100
100
temperature
50 starting cracks crack-free ml/ 40 100 g
considerably
50
higher hydrogen contents are tolerated without any crack development because of the much better hydrogen effusion. If both graphs are com-
20
20 0 10 20 30
Figure 10.29 pared it becomes obvious that the tests produce slightly different findings, i.e. with identical hydrogen content, the determined preheat temperatures required for the avoidance of cracking, differ by about 20 C.
143
Figure 10.30 illustrates a method to measure the diffusible hydrogen content in welds which is standardised in DIN 8572. Figure a) shows the burette filled with mercury before a specimen is inserted. The coupons are inserted into the opened burette and drawn with a magnet through the mercury to the capillary side (density of steel is lower than that of mercury, coupons surface). Then the burette is closed and evacuated. The hydrogen, which effuses of the coupons but does not diffuse through the mercury, collects in the capillary. The samples remain in the evacuated burette 72 hours for degassing. To determine the hydrogen volume the burette is ventilated and the coupons are removed from the capillary side. The volume of the effused hydrogen can be read out from the capillary; the height difference of the two mercury menisci, the air pressure, and the temperature provide the data to calculate the norm volume conditions. under This
capillary side meniskus1 M meniskus2 mercury coupons a) starting condition
br-er-10-30.cdr
air pressure B
standard
volume and the coupons weight are used to calculate, as measured value, the hydrogen volume in ml/100 g weld metal. This is the most commonly used method to determine the hydrogen Figure 10.30 content in welded joints.
b) during degassing