WeldingTechnology2 English

2005

ISF – Welding and Joining Institute RWTH – Aachen University

Lecture Notes

Welding Technology 2 Welding Metallurgy

Prof. Dr. –Ing. U. Dilthey

Table of Contents Chapter Subject Page

1. Weldability of Metals 3

2. TTT - Diagrams 8

3. Residual Stresses 21

4. Heat Treatment and its

Function During Welding 31

5. Welding Plain and

Low Alloy Steels 44

6. Welding High Alloy Steels 70

7. Welding of Cast Materials 89

8. Welding of Aluminium 96

9. Welding Defects 108

10. Testing of Welded Joints 126

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.

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 1.1

Figure 1.2


Material influence on weldability, i.e. welding

suitability, can be detailed for a better un-

derstanding 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 metallurgi-

cal properties (e.g. type of teeming) of the

steel are obtained.

Another modification of the mate-

rial behaviour takes place during

subsequent treatment, where the

raw material is rolled to processi-

ble semi-finished goods, e.g. like

strips, plates, bars, profiles, etc..

With the rolling process, material-

typical transformation processes,

hardening and precipitation proc-

esses are used to adjust an opti-

mised material characteristics

(see chapter 2).

Figure 1.4

Figure 1.3


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 1.6 depicts the deci-

sive importance of the car-

bon 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 pos-

sible by observing special

precautions (e.g. pre- and

post-weld heat treatment).

Figure 1.5

Figure 1.6


In addition to material behaviour, weldability is also essentially determined through the design

of a component. The influence of the design is designated as welding safety, Figure 1.7.

The influence of the manufac-

turing process to weldability is

called welding possibility,

Figure 1.8. For example, a

pre- and post-weld heat

treatment is not always possi-

ble, or grinding the weld sur-

face before welding the

subsequent pass cannot be

carried out (narrow gap weld-

ing).

Figure 1.7

Figure 1.8

2.

TTT - Diagrams

2. TTT – Diagrams 9

An essential feature of low

alloyed ferrous materials is

the crystallographic trans-

formation of the body-

centred cubic lattice which

is stable at room tempera-

ture (a-iron, ferritic struc-

ture) to the face-centred

cubic lattice (?-iron, aus-

tenitic 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 a, since they consist of a

compound of the components A (80%) and of B (20%). Further, a melting with the composi-

tion c0 is present at the temperature T1. With dropping temperature, the remaining melt is en-

Figure 2.1

Figure 2.2


riched with component B, following the course of line Li (liquidus line, up to point 4). In paral-

lel, always new and B richer a-mixed crystals are forming along the connection line So

(solidus line, points 1, 2, 5). The distribution of the components A and B in the solidified struc-

ture 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 com-

plete insolubility in the solid state shows Figure 2.3 If two components are completely insolu-

ble 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 transfo rmation

(critical point) (Figure 2.2).

Eutectic structures are normally fine-grained and show a characteristic orientation between

the constituents. The alloy L1 will consist of a compound of alloy A and eutectic alloy E in the

solid state.

You can find further in-

formation on transforma-

tion behaviour in relevant

specialist literature.

The definite use of the

principles occurs in the

iron-iron carbide diagram.

Transformation behaviour

of carbon containing iron

in the equilibrium condi-

tion is described by the Figure 2.3


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, 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 and graphite during heat

treatment of carbon rich alloys. However, it is fundamentally valid that the formation of ce-

mentite is encouraged with increasing cooling rate and decreasing carbon content. In a dou-

ble diagram, the stable

system is shown by a

dashed, the metastable by

a solid line, 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 mo-

lar content of 25%. The solid solutions in the phase fields are designated by Greek charac-

ters. According to convention, the transition points of pure iron are marked with the character

A - arrêt (stop point) and distinguished by subjacent indexes. If the transition points are de-

termined 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 °C: solidification temperature (melting point) δ-iron,

- 1392 °C: A4- point γ- iron,

Figure 2.4


- 911 °C: A3- point non-magnetic α- iron,

with carbon containing iron:

- 723 °C: A1- point (perlite point).

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 bi-

nary 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 tem-

perature of 1147°C and a carbon concentration of 4.3 mass%, the eutectic phase called Le-

deburite precipitates from cementite with 6,67% C and saturated γ-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 Lede-

burite are called hypereutectic.

If an alloy solidifies with less than 0,51 mass percent of carbon, a δ-solid solution is formed

below the solidus line A-B (δ-ferrite). In accordance with the peritectic transformation at

1493°C, melt (0,51% C) and δ-ferrite (0,10% C) decompose to a γ-solid solution (austenite).

The transformation of the γ-solid solution takes place at lower temperatures. From γ-iron with

C-contents below 0.8% (hypoeutectoidic alloys), a low-carbon α-iron (pre-eutectoidic ferrite)

and a fine-lamellar solid solution (perlite) precipitate with falling temperature, which consists

of α-solid solution and cementite. With carbon contents above 0,8% (hypereutectoidic alloys)

secondary cementite and perlite are formed out of austenite. Below 723°C, tertiary cementite

precipitates out of the α-iron because of falling carbon solubility.


The most important distinguished feature of the three described phases is their lattice struc-

ture. α- and δ-phases are cubic body-centered (CBC lattice) and γ-phase is cubic face-

centered (CFC lattice), Figure 2.1.

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 inte rstitial solid solution.

Although the cubic face-centred lattice of austenite has a higher packing density than the cu-

bic 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 α-iron) is the result. However, diffusion speed in γ-iron is always at

least 100 times slower than in α-iron because of the tighter packing of the γ-lattice.

Although α- and δ-iron show the same lattice structure and properties, there is also a differ-

ence between these phases. While γ-iron develops of a direct decomposition of the melt (S

→ δ), α-iron forms in the solid phase through an eutectoidic transformation of austenite (γ →

α + Fe3C). For the transformation of non- and low-alloyed steels, is the transformation of δ-

ferrite of lower importance, although this δ-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 alloy-

ing elements as manganese, chromium, nickel and silicon. Principally the equilibrium dia-

gram Fe-C applies also to

such multi-component sys-

tems. 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

Figure 2.5


phase system Fe-M-C characterised by a temperature interval in the three-phase field α + γ +

M3C. The beginning of the transformation of α + M3C to γ is marked by Aclb, the end by Acle.

The indices b and e mean

the beginning and the end

of transformation.

The described equilibrium

diagrams apply only to low

heating and cooling rates.

However, higher heating

and cooling rates are pre-

sent during welding, con-

sequently other structure

types develop in the heat

affected zone (HAZ) and in

the weld metal. The struc-

ture transformations during

heating and cooling are described by transformation diagrams, where a temperature change

is not carried out close to the equilibrium, but

at different heating and/or cooling rates.

A representation of the transformation

processes during isothermal austenitizing

shows Figure 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°C. Inhomogeneous austenite 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 ho-

Figure 2.6

Figure 2.7


mogeneous austenite is passed. A growing of the austenite grain size (to ASTM and/or in

µm) can here simultaneously be observed with longer annealing times.

The influence of heating rate on austeniti zing 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 aus-

tenitizing, two microstructure photographs are shown, both with different grain size classes to

ASTM.

Figure 2.8 shows the rela-

tion between the TTA and

the Fe-C diagram. It's obvi-

ous 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 dif-

ferent time-temperature

passes during austeniti zing

and subsequent cooling

down.

The heating period is com-

posed of a continuous and

an isothermal section.

During cooling down, two

different ways of heat con-

trol can be distinguished:

1. : During continuous

temperature control a

cooling is carried out with a

constant cooling rate out of

Figure 2.8

Figure 2.9


the area of the homogeneous and stable austenite down to room temperature.

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), fol-

lowed by an isothermal holding until all transformation processes are completed. After trans-

formation will be cooled down to room temperature.

Figure 2.10 shows the

time-temperature diagram

of a isothermal transforma-

tion 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 me-

tastable austenite, marked

with an A. The areas

marked with F, P, B, und M

represent areas where fer-

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 marten-

sitic 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

Figure 2.10


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 trans-

formation processes are

strongly postponed in rela-

tion to the mild steel. A

completely martensitic

transformation is carried

out up to a cooling time of

about 1.5 seconds, com-

pared with 0.4 seconds of

Ck 15. In addition, the

completely diffusion con-

trolled transformation proc-

esses of the perlite area

are postponed to clearly

longer times.

The hypereutectoid steel C

100 behaves completely

different, Figure 2.13. With

this carbon content, a pre-

Figure 2.11

Figure 2.12


eutectoid ferrite formation cannot still be car-

ried out (see also Figure 2.3).

The term of the figures 2.9 to 2.11 "austeniti z-

ing temperature“ means the temperature,

where the workpiece transforms to an austen-

itic microstructure in the course of a heat

treatment. Don’t mix up this temperature with

the AC3 temperature, where above it there is

only pure austenite. In addition you can see

that only martensite is formed from the aus-

tenite, provided that the cooling rate is suffi-

ciently 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

called critical cooling rate. Figure 2.13

Figure 2.14 Figure 2.15


Figure 2.14 shows schematically how the TTT diagram is modified by the chemical compo-

sition of the steel.

The influence of an increased austenitizing temperature on transformation behaviour 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 diffu-

sion lengths which must be

passed during the trans-

formation. 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 cool-

ing time. You can see that

with higher austenitizing

temperature the start of

Bainite formation together

with the drop of the mart-

ensite proportion is clearly

shifted to longer times.

As Bainite formation is not

so much impeded by the

coarse austenite grain as

with the completely diffu-

sion controlled processes

of ferrite and perlite forma-

tion, the maximum Bainite

proportion is increased

from about 45 to 75%.

Figure 2.16

Figure 2.17


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 1350°C) and shorter austenitizing 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 equilib-

rium diagram for infinite

long cooling time and that

a TTT diagram applies al-

ways only for one alloy.

Figure 2.18

3.

Residual Stresses


The emergence of residual

stresses can be of very

different nature, see three

examples in Figure 3.1.

Figure 3.2 details the

causes of origin. In a pro-

duced workpiece, material-

, production-, and wear-

caused residual stresses

are overlaying in such a

way that a certain condition

of residual stresses is cre-

ated. Such a workpiece

shows in service more or

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, Fig-

ure 3.4 shows a typical distri-

bution of residual stresses.

Residual stresses, which

build-up around dislocations

and other lattice imperfections

(s III), superimpose within a

grain causing stresses of the

2nd type and if spreading

around several grains, bring

out residual stresses of the 1st

type.

The formation of residual

stresses in a transition-free

Figure 3.1

Figure 3.2


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

homogeneous. The left part of

Figure 3.5 shows the T-t-

curve of three different meas-

urement points in the cylinder.

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

figure). Through the stabilising

effect of the cylinder core,


Figure 3.5


tensile stress builds up at the edge areas while the core is exposed to pressure stress. Re-

sulting volume differences between core and edge are balanced by elastic and plastic defor-

mations. 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.

These changes are principally shown once again in Figure 3.7 with the 3-rod model. A warm-

ing 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, be-

cause 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 resi-

dual 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



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 600°C.

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 mate-

rial , 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 build -

up 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 den-

sity, additional volume increase through la t-

tice deformation). In the case of a homoge-

nous transition, the weld will consequently

unfold pressure stresses. If the transition of

the edge areas happens earlier than the tran-

sition of the slower cooling core, plastic de-

formations of the core area may be present

similar to quenching (see above: quenching

stresses). In this case, the weld surface will

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 remain-

ing residual stress condition difficult.

Figure 3.8


Figure 3.9 shows the building-up of residual

stresses crosswise to a welded seam in anal-

ogy to the 3-rod model of Figure 3.7. This fig-

ure considers only shrinking residual stresses.

Before application of welding heat, the seam

area is stress-free (cut A-A). At the weldpool

the highest temperature of the welding cycle

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

through welding heat but are supported by

areas which are not so close to the seam.

Thus, areas close to the joint show compres-

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

areas close to the seam, the weld metal

shows tensile stresses, the adjacent areas

compression stresses. In cut D-D is the tem-

perature completely balanced, a residual

stress condition is recognised as shown in

the lower right figure.

Figure 3.10 shows how much residual

stresses are influenced by constraining ef-

fects of adjacent material. The resulting

stress in the presented case is calculated

according to Hooke:

s = e ? E

Elongation e is calculated as ? l/a (? l is the

length change due to shrinking). With con-

Figure 3.9

Figure 3.10


stant joint volume will shrinking and ? l always have the same value. Thus the elongation e

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 tempera-

ture- 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 vo lume can be up to 3% in the case of martensite formation.

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 sam-

ple over the time is plotted.



In the temperature range above 600°C the force sensor registers a tensile force which is

caused by the shrinking of the austenite. Between 600 and 400°C a large drop in force can

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 tempera-

ture areas for the individual

zones of the welded joint can

be determined. With these

data and with the course of

temperature it can be clearly

determined in which part of

the curve the force drop is

caused by the transition of the

welding consumable and in

which part by transition in the heat affected

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 af-

fected zone which lead to a decrease of ten-

sile stresses. If a high-strength electrode

which has a martensitic transition, is welded

to a StE 70, then there will be pressure resid-

ual stresses in the weld metal and tensile re-

sidual stresses in the HAZ.

Figure 3.13

Figure 3.14


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 re-

maining residual stresses will be much lower than in case with firm clamping.

Methods to determine resid-

ual stresses can be divided

into destructive, non-

destructive, and condition-

ally destructive methods.

The borehole and ring core

method can be considered

as conditionally destructive,

Figures 3.15 and 3.16.

In both cases, present re-

sidual stresses are released

through partial material re-

moval and the resulting de-

formations are then

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.

With the ring core method,

a crown milling cutter is

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-

ience of the core is meas-

ured, the detection of the

residual stress distribution

Figure 3.15

Figure 3.16


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 meas-

urement methods for re-

sidual stresses and what

causes residual stresses

to be picked-up when us-

ing one of the respective

methods.

Figure 3.18 shows a sur-

vey of the completely de-

structive procedures of

residual stress recognition.

Figure 3.17

Figure 3.18

4.

Heat Treatment and

its Function During Welding

4. Heat Treatment and its Function During Welding 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 ap-

pears 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 de-

termining 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.

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 4.1

Figure 4.2


grains. This zone of the weld is the area,

where the worst toughness values are found.

In Figure 4.4 you can see how much the for-

mation of the individual structure zones and

the zones of unfavourable mechanical prop-

erties 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 ad-

vantage of the material, for example to adjust

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 spe-

cifically optimised to achieve special quality, e.g. corrosion resistance against a certain at-

tacking medium, this post-weld heat treatment is of great importance.

Figure 4.6 shows areas in

the Fe-C diagram of differ-

ent heat treatment meth-

ods. 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 tempera-

tures like e.g. coarse-grain

Figure 4.3

Figure 4.4


heat treatment or normalising.

It can also be seen that the start of martensite formation (MS-line) is shifted to continuously

decreasing temperatures with increasing C-content. This is important e.g. fo r hardening

processes (to be explained 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 indi-

vidual heat treatment meth-

ods will be explained by

their temperature-time-

behaviour in the following.


Figure 4.7


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 A3 with relatively short periods.

Figure 4.8 shows sche-

matically time-temperature

behaviour in a TTT-

diagram. (Note: the curves

explain running structure

mechanisms, they must not

be used as reading off ex-

amples. To determine t8/5,

hardness values, or micro-

structure distribution, are

TTT-diagrams always read

continuously or isother-

mally. Mixed types like

curves 3 to 6 are not a llowed for this purpose!).

The most important heat treatment methods can be divided into sections of annealing, hard-

ening 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 30°C above Ac3

until homogeneous austenite 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).

Figure 4.8


To harden a material, aus-

tenisation and homogeni-

sation is carried out also at

30°C above AC3. Also in

this case one must watch

that the austenite grains

remain as small as possi-

ble. To ensure a complete

transformation to marten-

site, a subsequent quench-

ing follows until the

temperature is far below

the Ms-temperature, Figure

4.10. The cooling rate dur-

ing quenching must be high enough to cool down from the austenite 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 hard-

ening. 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 stop-

ping a water quenching

process at a certain tem-

perature and continuing the

cooling with a milder cooling

medium (oil). With longer

holding on at elevated tem-

perature level, transforma-

tions can also be carried

through in the bainite area

(curves 5 and 6).

Figure 4.9

Figure 4.10


Figure 4.11 shows the quenching and tempering procedure. A hardening is followed by an-

other heat treatment below Ac1. During this tempering process, a break down of martensite

takes place. Ferrite and cementite are formed. As this change causes a very fine micro-

structure, this heat treat-

ment leads to very good

mechanical properties like

e.g. strength and tough-

ness.

Figure 4.12 shows the pro-

cedure of soft-annealing.

Here we aim to adjust a

soft and suitable micro-

structure for machining.

Such a structure is charac-

terised 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 hypoeu-

tectic 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 transfor-

mation is carried out because

the surface area is strongly

reduced ? thermodynami-

cally more favourable condi-

tion). Hypereutectic steels

have in addition to the lamel-

lar structure of the perlite a

cementite network on the

grain boundaries.

Figure 4.11

Figure 4.12


Spheroidizing of cementite is achieved by making use of the transformation processes during

oscillating around A1. When exceeding A1 a transformation of ferrite to austenite 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 20°C below until the trans-

formation is completed, a

re-precipitation of cemen-

tite on existing nuclei takes

place. The repetition of this

process leads to a step-

wise spheroidizing of

cementite and the frequent

transformation avoids a

grain coarsening. A soft-

annealed microstructure

represents frequently the

delivery condition of a ma-

terial.

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 increase is

the commonly used

method to make such de-

formations possible be-

cause the yield strength

limit decreases with in-

creasing temperature. A

stress-relieve heat treat-

ment should not cause any

other change to properties,

so that tempering steels

Figure 4.13

Figure 4.14


are heat treated below tempering temperature.

Figure 4.14 shows a survey of heat treatments which are important to welding as well as their

purposes.

Figure 4.15 shows princi-

pally 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 treat-

ment is applied before

welding to adjust a proper

material condition which for

welding. After welding, al-

most any possible heat treatment can be carried

out. This is only limited by workpiece dimen-

sions/shapes or arising costs. The most important

section of the diagram is the kind of heat treatment

which accom-panies the welding. The most impor-

tant processes are explained in the following.

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 mart-

ensite is formed (curve 1, analogous to Figure 4.8,

hardening). A simple heating before welding with-

out 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 4.15

Figure 4.16


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-T-

course in the HAZ. As an example, welding

with simple preheating is selected.

The plate is preheated in a period tV. After

removal of the heat source, the cooling of the

workpiece starts. When tS 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 tempera-

ture-time-behaviour of the HAZ.

The temperature time course during welding

with simple preheating is shown in Figure

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 covering with

heat insulating materials.

Another variant of welding

with preheating is welding

at constant working

temperature. This is

Figure 4.17

Figure 4.18


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 isother-

mal welding, the workpiece

is heated up to a working

temperature above MS

(start of martensite forma-

tion) and is also held there

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 post-

heat 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 tem-

perature. Such materials are heat treated

immediately after welding at a temperature

between 600 and 700°C, so that a formation

Figure 4.19

Figure 4.20


of martensite is avoided and welding residual stresses are eliminated simultaneously.

Aims of the modified step-

hardening welding should

not be discussed here, Fig-

ure 4.21. Such treatments

are used for transformation-

inert 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 tempera-

ture distribution during multi-

pass welding. The solid line

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 tem-

perature. As a result, working temperature in-

creases with the number of passes. The

second pass is welded under a preheat tem-

perature 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 favourable effect of

multi-pass welding is the warming of the HAZ

of each previous pass above recrystallisation

temperature with the corresponding crystallisa-

Figure 4.21

Figure 4.22


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 inter-

pass temperature). In addition, the workpiece will not heat up to excessively high tempera-

tures.

Figure 4.23 shows a nomogram where working temperature and minimum and maximum

heat input for some steels can be interpreted, depending on carbon equivalent and wall thick-

ness.

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 160°C

- after welding, residual stress relieving between 530 and 600°C.

Steels which are placed in

the hatched area called

soaking area, must be

treated with a hydrogen re-

lieve annealing. Above this

area, a stress relieve anneal-

ing must be carried out. Be-

low this area, a post-weld

heat treatment is not re-

quired.

Figure 4.23

5.

Welding Plain and

Low Alloy Steels

5. Welding Plain and Low Alloy Steels 45

© ISF 200 4b r-er05-01.cdr

Einteilung n ach der chem ischen Zusam mensetzung� unlegie rte Stähle

� nichtro stende Stähle

Einteilung n ach Hauptgüt eklassen� unlegi erte Stähle - unlegier te Qualitätsstä hle - unlegier te Edelstähle � nichtro stende Stähle

Definitio n des Begriff es Stahl

� andere legierte Stäh le

� andere legierte Stäh le - legierte Qualitätsstähl e

- legierte Edelstähle In the European Standard DIN EN

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

chemical composition and the main

quality classes.

In accordance with the chemical compo-

sition the steel grades are classified into

unalloyed, stainless and other alloyed

steels. The mass fractions of the individ-

ual elements in unalloyed steels do not

achieve the limit values which are indi-

cated 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 car-

bon.

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.

Figure 5.1

Definition for theclassification of steels

© ISF 2004br-er05-01.cdr

Classification in accordance with the chemical composition:

l

l

l

unalloyed steels

stainless steels

other, alloyed steels

Classification in accordance with the main quality class:

·

·

·

unalloyed steels - unalloyed quality steels- unalloyed special steels

stainless steels

other, alloyed steels - alloyed quality steels- alloyed special steels

Definition of the term “steel”

Steel is a material with a mass fraction if iron which is higherthan of every other element, ist carbon content is, in general,lower than 2% and steel contains, moreover, also otherelements. A limited number of chromium steels might contain acarbon content which is higher than 2%, but, however, 2% is thecommon boundary between steel and cast iron [DIN EN 10020(07.00)].

Figure 5.2

Boundary between unalloyedand alloyed steels


Determined elementlimit value

Mass fraction in %

a) If just the highest value has been determined for

mangenese, the limit value us 1,80% and the 70%-rule

does not apply.

Al aluminium

B boron

Bi bismuth

Co cobalt

Cr chromium

Cu copper

La lanthanides

(rated individually)

Mn manganese

Mo molybdenum

Nb niobium

Ni nickel

Pb lead

Se selenium

Si silicon

Te tellurium

Ti titanium

V vanadium

W tungsten

Zr zirconium

Others (with the exception

of carbon, phosphorus,

sulphur, nitrogen)

(Each)


As far as the main quality classes are concerned, the steels are classified in accor-

dance 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 manufac-

turing 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 manu-

facturing and control conditions.


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 applica-

tion field, about the mechanical or physical properties or about the composition.

The code designations of the steels are divided into two groups. The code designa-

tions 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.

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).

Figure 5.3

Classification of steels in accordancewith their designated use


l

l

l

l

l

l

l

l

l

l

l

e.g. S235JR, S355J0

P =e.g. P265GH, P355M

L =e.g. L360A, L360QB

E =e.g. E295, E360

B =e.g. B500A, B500B

Y =e.g. Y1770C, Y1230H

R =e.g. R350GHT

H =

e.g. H400LA

D =e.g. DD14, DC04

T =

e.g. TH550, TS550

M =e.g. M400-50A, M660-50D

S = Steels for structural steel engineering

Steels for pressure vessel construction

Steels for pipeline construction

Engineering steels

Reinforcing steels

Prestressing steels

Steels for rails (or formed as rails)

Cold rolled flat-rolled steels with higher-strengthdrawing quality

Flat products made of soft steels for cold reforming

Black plate and tin plate and strips and also speciallychromium-plated plate and strip

Magnetic steel sheet and strip


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 ex-

plained in Figure 5.4.

Figure 5.4


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 spe-

cial demands, symbols for the type of coating and symbols for the treatment con-

dition. These additional symbols are stipulated in the ECISS-note IC 10 and depicted

in Figures 5.5 and 5.6.

© ISF 2004br-er-05-06.cdr

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,for example +TA

Symbol ) )

+ A+ AC+ C

+ Cnnn+ CR+ HC+ LC+ Q+ S+ ST+ U

1 2 treatment condition

softenedannealed for the production of globular carbideswork-hardened (e.g., by rolling and drawing), also a distinguishingmark for cold-rolled narrow strips)cold-rolled to a minimum tensile strength of nnn MPa/mm²cold-rolledthermoformed/cold formedslightly cold-drawn or slightly rerolled (skin passed)quenched or hardenedtreatment for capacity for cold shearingsolution annealeduntreated

Symbols for the treatment condition

Figure 5.6


Symbol ) )

+ A+ AR+ AS+ AZ+ CE+ Cu+ IC+ OC+ S+ SE+ T+ TE+ Z+ ZA+ ZE+ ZF+ ZN

1 2 Coating

hot dippedaluminium, cladded by rollingcoated with Al-Si alloycoated with Al-Tn alloy (>50% Al)electrolytically chromium-platedcopper-coatedinorganically coatedorganically coatedhot-galvanised

upgraded by hot dipping with a lead-tin alloyelectrolytically coated with a lead-tin alloyhot-galvisedcoated with Al-Zn alloy (>50% Zn)electrolytically galvaniseddiffusion-annealed zinc coatings (galvannealed, with diffused Fe)nickel-zinc coating (electrolytically)

electrolytically galvanised

1

2

))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,for example +SA

Symbols for the coating type

Figure 5.5


Figure 5.7 shows an example of the novel designation of a steel for structural steel

engineering which had formerly been labelled St37-2.

Figure 5.8 depicts the chemical composition and the mechanical parameters of dif-

ferent steel grades. The figure explains the influence of the chemical composition on

the mechanical properties.

Figure 5.7


S = steels for structural steel engineeringP = steels for pressure vessel constructionL = steels for pipeline constructionE = engineering steelsB = reinforcing steels

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

R 235 MPa/mmeH

2³

further property(RR = normalised)

test temperature 20°C

impact energy ³ 27 J

Steel designation in accordance with DIN EN 10027-1

Stahl C Si Mn P S Cr Al Cu N Mo Ni Nb VS355J0(St 52-3)S500N(StE500)P295NH(HIV)S355J2G1W(WTSt510-3)S355G3S(EH36)

Stahl

S355J2G3(St 52-3)S500N(StE500)P295NH(HIV)S355J2G1W(WTSt510-3)S355G3S(EH36)

Kerbschlagarbeit AV

[J]

Zugfestigkeit Rm

[N/mm²]BruchdehnungA

[%]

StreckgrenzeReH

[N/mm²]0°C -20°C

27

610-780 500 16 31-47

27355510-680 20-22

285

355

355

>18

22

>22

49(bei +20°C)

76(bei -10°C)

21-39

460-550

510-610

400-490

£0,18

£0,55

£0,35

£0,1-0,35

£0,50

0,1- 0,6

£0,26

£0,15

0,21

£0,20 £1,60 0,040

1- 1,7 0,035

³0,6 £0,05

0,5- 1,3 0,035

0,7- 1,5 £0,05

0,040 /

0,030 0,30

£0,05 /

0,0350,40-0,80

£0,05 /

/ /

0,020 0,20

/ /

/0,25-0,5

/ /

£0,009 /

0,020 0,1

/ /

/ £0,30

/

/ /

1 0,05

/ /

£0,65 /

/

0,22

/

0,02-0,12

// / /

Chemical composition and mechanicalparameters of different steel sorts


impact energy AVelongation after fracture Ayield point ReHTensile strength RmSteel

Steel

Figure 5.8


The steel S355J2G2 represents the basic type of structural steels which are nowa-

days 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°C. This steel shows a relatively low strength but very good toughness values

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 work-

piece 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.


The steel grades are classified into four subgroups according to the chemical com-

position (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 con-

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

elements and numbers, which in the

sequence of the designating alloying

elements give reference about their

content. The individual numbers stand

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.

Codes accordingto the chemical composition


Unalloyed steels (Mo content < 1%)

Unalloyed steels (Mn content > 1%)

Alloyed steels (content of alloying element > 5%)

X10CrNi18-10

Legiert C=10/100=0,1% Cr=18% Ni=10%

10CrMo9-10

C=10/100=0,10% Cr=9/4=2,25% Mo=10/10=1%

C45

Carbon 0,45% Carbon

element factor

Cr, Co, Mn, Ni, Si, W

Al, Be, Cu, Mo, Nb, Pb, Ta, Ti, V, Zr

C, Ce, N, P, S

B

4

10

100

1000

High-speed steels

HS 2-9-1-8

Mo=9% Co=8%W=2% V=1%

Table 5.1

Figure 5.9


The alloyed steels are labelled with the code letter X, a number which again com-

plies 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 de-

creasing 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)


Figure 5.10 specifies the material numbers for the material main group „steel“.

Figure 5.10


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-per-

until length, the

bigger the austen-

ite grains in the

HAZ and the width

of the HAZ in-

creases. Such

coarsened austen-

ite grain decreases

the critical cooling

time, thus increas-

ing the tendency of

the steel to harden.

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 precipita-

tions 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 con-

tinuous grain growth related to temperature. Steel 2 contains AIN precipitations which

are stable up to a temperature of approx. 1100°C, thus preventing a growth of the

austenite grain.

Influence of the energy-per-unitlength on the austenite grain size

13

11

9

7

5

30 0,2 0,4 0,6 0,8 1,0

Au

ste

nite

gra

in s

ize

ind

ex

acc

ord

ing

to D

IN 5

06

01

Distance of the fusion linemm

Energy-per-unit length in kJ/cm

9 12 18 36


Figure 5.11


With higher temperatures, these

precipitations dissolve and cannot

suppress a grain growth any more.

Steel 3 contains mainly titanium car-

bonitrides of a much lower grain-

refining effect than that of AIN. Steel 4

is a combination of the most effective

properties of steels nos. 2 and 3.

The importance of grain refinement

for the mechanical properties of a

steel is shown in Figure 5.13. Pro-

vided the temperature keeps con-

stant, 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:

dKR

iel

1⋅+= σ

According to the

above-mentioned

law, the increase of

the yield point is

inversely propor-

tional to the root of

the medium grain

diameter d. σi

stands for the inter-

nal friction stress of

the material. The

grain boundary

resistance K is a

measure for the

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

Austenite grain size as a functionof the austenitization temperature

Steel % C % Mn % Al % N % Ti

1 0,21 1,16 0,004 0,010 /

2 0,17 1,35 0,047 0,017 /

3 0,18 1,43 0,004 0,024 0,067

4 0,19 1,34 0,060 0,018 0,140

900 1000 1100 1200 1300 1400°C

Austenitization temperature

18

6

4

2

10-1

8

6

4

2

10-2

6 10-3

8

mm

Mediu

m fib

re le

ngth

Gra

in s

ize in

dex

acc

ord

ing to D

IN 5

0601

-4

-2

0

2

4

6

8

10

12

Steel 1Steel 2Steel 3Steel 4


Figure 5.12

Connection betweenyield point and grain size

900

800

700

600

500

400

300

200

N/mm²

10 2 3 4 5 6 7 8 10mm-1/2

Yie

ld p

oin

t or

0,2

boundary

Grain size d-1/2

Temperature in °C:

-193

-185

-180

-155

+20

-40

-100

-170


Figure 5.13


structural steels are concerned, this means the improvement of the mechanical prop-

erties without any further alloying. Modern fine-grained structural steels show im-

proved 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 advan-

tages of microal-

loyed fine-grained

structural steels in

comparison with

conventional struc-

tural steels are

shown in Figure

5.14. Due to the

considerably better

mechanical proper-

ties of the fine-

grained structural

steel in comparison

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 on the

classification of

Figure 5.2, Fig-

ure 5.15 divides the

steels with regard

to their problematic

processes during

welding. When it

comes to unalloyed

steels, only ingot

Figure 5.14

Influence of the steel selection on theproducing costs of welded structures

S235JR S355J2G3 S690Q S890Q S960Q Verhältnis

(St37-2) (St52-3) (StE690) (StE890) (StE960) S235JR - S960Q

Streckgrenze N/mm2215 345 690 890 960 1 : 5

Blechdicke mm 50 31 14,4 11 10 5 : 1

Nahtquerschnitt mm2870 370 100 60 50 17 : 1

Schweißdraht ø 1.2 mm SG2 SG3 NiMoCr X 90 X 96 -

Schweißdrahtkosten Verhältnis 1 1 2,4 3,2 3,3 1 : 3,3

Stahlkosten Verhältnis 1 1,2 1,9 2,3 2,4 1 : 2,4

Schweißgutkosten Verhältnis 5,3 2,3 1,5 1,16 1 5,3 : 1

Spez. Schweißnahtkosten Verhältnis 12 5,1 1,8 1,18 1 12 : 1

Kostenverhältnis inklusiveGrundwerkstoffe

5 : 1

Randbedingungen: Schweißverfahren = MAG

Abschmelzleistung = 3 kg Schweißdraht / h, Nahtform X - 60°

Lohn- und Maschinenkosten = 60 DM / h

Spez. Schweißnahtkosten = Schweißzusatzwerkstoffe + Schweißen

Berechnungsgrundlage =szul = Re / 1.5

Stahlsorte


Yield point

Plate thickness

Weld cross-section

Welding wire Ø 1.2

Welding wire costs

Steel costs

Weld metal costs

Special weld costs

Costs ratio inclusive basematerials

Ratio

Ratio

Ratio

Ratio

Boundary condition: welding process = MAG

Deposition rate = 3 kg welding wire/h, weld shape X -60°

Costs of labour and equipment = 30€/h

Special weld costs = weld filler materials + welding

Calculation base = = Re/1.5szul

Steel type Ratio

Figure 5.15

Classification of steels withrespect to problems during welding

low-alloyed high-alloyed

hardeningspecial properties areachieved, for example:

heat resistance,tempering resistant,

high-pressure hydrogen resistance,toughness at low temperatures,

surface treeatment condition, etc.

corrosionresistant steels

tool steels

Hardening,special properties

are achieved

steels

unalloyed alloyed

mild steel higher-carbon steel

HardeningUnderbead cracking

rimmed steel killed steel duplex killed steel

cutting ofsegregation

zones

cold brittleness(coarse-grained recrystallization

after critical treatment)stress corrosion crackingsafety from brittle fracture

ferritic pearlitic-martensitic austenitic

grain desintegrationstress corrosion

cracking hot cracks(sigma phaseembrittlement)

hardeningembrittlement

formationof chromium

carbide

grain increase inthe weld interfaces

Post-weld treatment forhighest corrosion resistance



casts, rimmed and semi-killed steels are causing problems. “Killing” means the re-

moval 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

segregated zone

where unfavourable

elements like sul-

phur and phospho-

rus are enriched.

During rolling, such

zones are stretched

along the complete

length of the rolling

profile.

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 transforma-

tion-inert than the

outer envelope

and are inclined to

hardening. In

addition, they are

sensitive to hot-

cracking, as, in

these zones, the

elements phospho-

rus and sulphur

are enriched.

Figure 5.16

Ingot cross-sectionsafter different casting methods

Figures: mass content of oxygen in %

fully killed steel semi-killed steel rimmed steel

0,003

0,012

0,025


Figure 5.17

Example of unfavourable (a) andfavourable (b) welds

a b

B CD

E



Therefore, “ touching” such segregation zones during welding must be avoided by all

means.

In the case of low-

alloy steels, the

problem of HAZ

hardening during

welding must be

observed. Fig-

ure 5.18 shows

hardness values of

various microstruc-

tures. The highest

hardness values

can be found with

martensite and

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 C-

content.

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 influ-

ence of the mart-

ensite content in

the HAZ on the

crack formation of

welded joints.

Hardening through

martensite forma-

tion is not to be

expected with pure

carbon steels up to

about 0,22%,

Hardness of Several Microstructures

Microstructures Average Brinell Hardness (Approximately)

Ferrite 80

Austenite 250

Perlite (granular) 200

Perlite (lamellar) 300

Sorbite 350

Troostite 400

Cementite 600 - 650

Martensite 400 - 900

© ISF 2004Br-er-05-18.cdr

Figure 5.18

Influence of Martensite Content

strength,calculated at

max. hardness

with maximummartensite

contentHV HRC N/mm2 %

root crackingpresumable

400 41 1290 70

root crackingpossible

400 - 350 41 - 36 1290 - 1125 70 - 60

no root cracking 350 36 1125 60

sufficient operational safetywithout heat treatment

280 28 900 30

maximum hardness

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.


Figure 5.19


because the critical cooling rate with these low C-contents is so high that it normally

won’t be reached within the welding cycle. In general, such steels can be welded

without special problems (e.g., S. 235).

In addition to car-

bon, all other alloy

elements are im-

portant when it

comes to marten-

site formation in

the welding cycle,

as they have sub-

stantial influence

on the transforma-

tion behaviour of

steels (see

Fig. 2.12 ). It is not

appropriate just

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 ele-

ments to hardening tendency, (Fig. 5.20). As these approximation formulas are em-

pirically determined

and as for the

hardening tendency

the general condi-

tions like plate

thickness, heat

input, etc., are also

of importance, the

carbon equivalent

cannot be a com-

mon limit value for

the weldability.

For the determina-

Figure 5.20

Definition of C - Equivalent

C-Äqu.= carbon equivalent (%) PLS = pipeline steels PCM = (%)cracking parameters

IIW

Stout

Ito and Bessyo

Mannesmann

Hoesch

Thyssen

15

NiCu

5

VMoCr

6

MnCÄqu.C

++

++++=-

40

Cu

20

Ni

10

MnCr

6

MnCÄqu.C ++

+++=-

5B10

V

15

Mo

60

Ni

20

CrCuMn

30

SiCPCM ++++

++++=

40

Ni

20

CuCr

10

MoMnCCET +

++

++=

20

VMoNiCrCuMnSiCÄqu.C

+++++++=-

15

V

40

Mo

60

Ni

20

Cr

16

CuMn

25

SiCÄqu.C PLS ++++

+++=-


Mo

Figure 5.21

Quelle: DIN EN 1011-2br-er05-21.cdr

Calculation of the preheating temperatures

Tp =697 CET + 160 tanh (d/35) + 62 HD + (53 CET - 32) Q - 3280,35

-100

-80

-60

-40

-20

0

20

40

0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 5

Wärmeeinbringen Q [kJ/mm]

delt

aT

p[°

C]

delta Tp = (53 CET - 32) Q - 53 CET + 32

d = 50 mmHD = 8

CET = 0,4 % CET = 0,2 % CET = 0,2 %

delta Tp = (53 CET - 32) Q - 53 CET + 32

CET = 0,4 % CET = 0,2 % CET = 0,2 %

d = 50 mmHD = 8

0

20

40

60

80

100

0 5 10 15 20 25

Wasserstoffgehalt HD des Schweißgutes [%]

de

lta

Tp

[°C

]

delta Tp = 62 HD 0,35 - 100

CET = 0,33 %d = 30 mmQ = 1 kJ/mm

delta Tp = 62 HD - 1000,35

CET = 0,33 %dQ = 1 kJ/mm

= 30 mm

0

50

100

150

200

250

0,2 0,3 0,4 0,5

Kohlenstoffäquivalent CET [%]

Tp

[°C

]

Tp = 750 CET - 150

d = 30 mmHD = 4Q = 1 kJ/mm

Tp = 750 CET - 150

d = 30 mmHD = 4Q = 1 kJ/mm

0

10

20

30

40

50

60

0 20 40 60 80 100

Blechdicke d [mm]

de

lta

Tp

[°C

]

delta Tp = 160 tanh (d/35) - 110

CET = 0,4 %HD 2Q = 1 kJ/mm

delta Tp = 160 tanh (d/35) - 110

CET = 0,4 %HD = 2Q = 1 kJ/mm

© ISF 2005

Heat input

Hydrogen content of the weld metalCarbon aquivalent

Plate thickness

Source:


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 in-

put Q are considered.

The essential factor

to martensite forma-

tion in the welding

cycle is the cooling

time. As a measure

of cooling time, the

time of cooling from

800 to 500°C (t8/5) is

defined (Fig. 5.22).

The temperature

range was selected

in such a way that it

covered the most

important structural transformations and that the time can be easily transferred to the

TTT diagrams.

Figure 5.23 shows

measured time-

temperature distri-

butions in the vicin-

ity of a weld. Peak

values and dwell

times depend obvi-

ously on the loca-

tion of the

measurement and

are clearly strongly

determined by the

heat conduction

conditions.

Figure 5.22

Definition of t8/5

Te

mp

era

ture

T

Time t

Tmax

°C

800

500

t t s800 500

DT

t8/5


Figure 5.23

Temperature-time curvesin the adjacence of a weld

2000

°C

1500

1000

500

00 50 100 150 200 250 s 300

Time t

Tem

pera

ture

T

A

B

C

10mm

A

B

C



With the use of thinner plates with complete heating of the cross-section during weld-

ing, 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

provide a method

of calculating the

cooling time t8/5 of

low-alloyed steels.

In the case of a

three-dimensional

heat dissipation,

t8/5 it independent

of plate thickness.

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 dissi-

pation changes to a three-dimensional heat dissipation.

Figure 5.25 shows

the influence of the

welding method on

the heat dissipa-

tion. With the same

heat input, the

energy which is

transferred to the

base material

depends on the

Figure 5.24

Calculation equation for two- andthree-dimensional heat dissipation

3 - dimensional:

2 - dimensional:


÷÷ø

öççè

æ

--

-×

××

××=

00

5/8800

1

500

1

2 TTv

IUt

lph

( ) 3

00

04

5/8800

1

500

110567,0 N

TTv

IUTt ×¢×÷÷

ø

öççè

æ

--

-×

×××-= - h

úúû

ù

êêë

é÷÷ø

öççè

æ

--÷÷

ø

öççè

æ

-××÷

ø

öçè

æ ××

××××=

2

0

2

02

22

5/8800

1

500

11

4 TTdv

IU

ct

rlph

( ) 22

2

0

2

02

2

05

5/8800

1

500

11103,4043,0 N

TTdv

IUTt ×¢×

úúû

ù

êêë

é÷÷ø

öççè

æ

--÷÷

ø

öççè

æ

-××÷

ø

öçè

æ ×××-= - h

÷÷ø

öççè

æ

-+

-×

××¢×

×-×-

=-

-

0004

05

800

1

500

1

10567,0

103,4043,0

TTv

IU

T

Td

üh

K3

universal formula:

extended formulaFor low-alloyed steel:

universal formula:

extended formulaFor low-alloyed steel:

K2

formula for the transitionthickness of low-alloyed steel:

Figure 5.25

Relative thermal efficiency degreeof different welding methods

0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1

SA welding

Manual arc welding

MAG-(CO )-2 welding

MIG-(Ar)-welding

TIG-(Ar)-welding

TIG-(He)-welding

welding methods

Relative thermal efficiency degree ‘h



welding method. This dependence is described by the relative thermal efficiency ŋ’.

The influence of

the groove ge-

ometry is covered

by seam factors

according to

Fig. 5.26. Empiri-

cally determined,

these factors were

introduced for an

easier calculation.

For other groove

geometries, tests

to measure the

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.

Figure 5.26

Weld factors for differentweld geometries

Type of weldweld factor

2-dimensionalheat dissipation

3-dimensionalheat dissipation

1

0,45 - 0,67

0,9

0,9

1

0,67

0,67

0,9


Figure 5.27

Transition From Two to ThreeDimensional Heat Flow

Heat input E. .N [kJ/cm]h n

0 10 20 30 40 50

5

cm

3

2

1

0

Pla

te thic

kness

cooling time t [s]10 15 20 25

8/5

3040

60100

2-dimensional

3-dimensional

T =20°CA

0 10 20 30 40 50

cooling time t [s]10 20 30 40 50

8/5

2-dimensional

3-dimensional

T =200°CA

60

80

100

150



Fig. 5.28 shows the

possible range of

heat input depend-

ing on the elec-

trode diameter. It is

clear that a rela-

tively large working

range is available

for arc welding

procedures. A

variation of the

energy-per-unit

length can be

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 weld-

ing. The shorter the fused electrode

distance, i.e., the shorter the ex-

tracted length, the higher the energy-

per-unit length.

Figure 5.28

br-er-05-28.cdr

Heat Inputs ofVarious Welding Methods

3,25 4 5 6 0,8 1,0 1,2 1,6 2,5 3,0 4,0 5,0

20

kJ/cm

12

8

4

He

at

inp

ut

Manual metal arc welding MAGC-, MAGM-method

SA-welding

-short arc

-sprayarc

© ISF 2004

Figure 5.29


35

kJ/cm

25

20

15

10

5

0

Energ

y-per-

unit

length

0 50 100 150 200 250 300 350 400 450 500 mm 600

run-out length

Stick electrode(mm)

Current intensity (A)

Current intensity (A)

2,5

90

75

3,25

135

120

4,0

180

140

5,0

235

190

6,0

275

250

Æ6,0mm x 450mm

Æ5,0mm x 450mm

Æ4,0mm x 450mm

Æ3,25mm x 350mm

Æ2,5mm x 350mm

Energy-per-unit length as afunction of the run-out length


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 two-

dimensional heat dissipation, where a dependence between energy-per-unit length,

cooling time and preheating temperature is given, depending on the plate thickness. .

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.

Figure 5.30

br-er05-30.cdr

Dependence of E, t andd During SA - Welding

8/5

Heat input E5 6 7 8 9 10 15 20 30 kJ/cm 50

504030

20

10

7

504030

20

10

7

504030

20

10

7

504030

20

10

7

Coolin

g tim

e t

in s

8/5

d = 7,5 mm

d = 10 mm

d = 15 mm

d = 20 mm

transition to3-dimensional

heat flow

T 200°C150°C100°C

20°C

0

T 200°C150°C100°C

20°C

0

T 200°C150°C100°C

20°C

0

T 200°C150°C100°C

20°C

0

© ISF 2004


With the transition to thicker plates,

the diagrams in Fig. 5.31 apply. The

upper part of the figure determines

whether a two-dimensional or a three-

dimensional heat dissipation is pre-

sent. For the three-dimensional heat

dissipation, the lower diagram applies

where the same information can be

determined, independent of plate

thickness, as with Fig. 5.30.

The relation be-

tween current and

voltage for MAG

welding is shown

in Fig. 5.32 and

the used shielding

gas is one of the

parameters. Weld-

ing voltage and

welding current, or

wire feed speed,

determine the type

of arc.

Figure 5.31

br-er05-31.cdr

Dependence ofE, T , t And d0 8/5 Ü

Heat input E

50s

40

30

20

15

109

87

Co

olin

g t

ime

t 8

/5

5 6 7 8 9 10 15 20 30 kJ/cm 50

T250

°C

200°C

150°C

100°C

20°C

0

Heat input E

50mm

40

30

20

15

109

87

Tra

nsi

tion

th

ickn

ess

dÜ

5 6 7 8 9 10 15 20 30 kJ/cm 50

aera of3-dimensional

heat flow

area of2-dimensional

heat flow

T250 °C 200 °C

150 °C 100 °C

20 °C

0

© ISF 2004

Figure 5.32

br-er-05-32.cdr

Dependence of Current And Voltage DuringMAG-Welding, Solid Wire, 1.2 mmÆ

35V

30

25

20

15

Weld

ing v

olta

ge

Welding current

Wire feed

150 200 250 A 300

3,5 4,5 5,5 7,0 8,0 9,0 10,5 m/min

C1

M21

M23

gas composition:C1 100% COM21 82% Ar + 18% COM23 92% Ar + 8% O

2

2

2

short arc

contact tube distance ~15mm contact tube distance ~19mm

mixed arc spray arc

© ISF 2004


The diagram in Fig. 5.33 demon-

strates the dependence of plate thick-

ness, heat input E and cooling time

t8/5 for fillet welds at a preheating

temperature of T0 = 150°C. 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 three-

dimensional heat dissipation.

Fig. 5.34 shows the same depend-

ence for butt welds with V groove

preparation..

Figure 5.33

br-er05-33.cdr

Permissible E-RangeDuring SA - And MAG - Welding

hh

' = 1' = 0,85

d = 32 mmd = 15 mm

UP

MAG

U max

U min

F = 0,67F = 0,67

3

2

t = 30 st = 6 s8/5 max

8/5 min

E = 66 kJ/cmE = 14 kJ/cm

max

min

60

kJ/cm

50

45

40

35

30

25

20

15

10

5

0

70

kJ/cm

59

53

47

41

35

29

23

18

12

6

0

He

at

inp

ut

ES

A-

we

ldin

g

He

at

inp

ut

EM

AG

- w

eld

ind

Plate thickness

0 5 10 15 20 25 30 mm 40

cracking tendency

toughness affection

fillet weldsT = 150 °C0 30s

25s

20s

15s

10s

6s

© ISF 2004

Figure 5.34

br-er05-34.cdr

Permissible E-RangeDuring SA - And MAG - Welding

hh

' = 1' = 0,85

d = 34 mmd = 15 mm

UP

MAG

U max

U min

F = 0,9F = 0,9

3

2

t = 30 st = 6 s8/5 max

8/5 min

E = 49 kJ/cmE = 10 kJ/cm

max

min

60

kJ/cm

50

45

40

35

30

25

20

15

10

5

0

70

kJ/cm

59

53

47

41

35

29

23

18

12

6

0

Heat

inp

ut

ES

A-

weld

ing

Heat

inp

ut

EM

AG

- w

eld

ing

Plate thickness

0 5 10 15 20 25 30 mm 40

cracking tendency

toughness affection

butt weldsT = 150 °C0

30s

25s

20s

15s

10s

6s

© ISF 2004


The curve family in Fig. 5.35 shows the dependence of the heat input from the weld-

ing speed as well as the acceptable working range. The parameters of the curves 1

to 8 in the table

have been taken

from Figures 5.32

and 5.34 and apply

only for related

conditions like wire

diameter, wire

feed, welding

voltage, etc.

Figure 5.36 shows

a reading example

for such diagrams

(according to DVS-

Reference Sheet

Nr. 0916).

In this example, a

plate thickness of

15 mm and a cool-

ing time t8/5 be-

tween 10 and 20 s

are given. In this

case, the maximum

cooling time for MAG welding is 15 s. A solid wire with a diameter of 1.2 mm at 29V

and 300A is used.

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.

Figure 5.35

br-er-05-35.cdr

E as a Function of Welding Speed,Solid Wire, 1.2mmÆ

MAG/ M21 (82% Ar, 18% CO)

25kJ/cm

20

15

10

5

010 15 20 25 30 35 40 45 50 cm/min 60

Welding speed vS

Heat in

put E

working range

12

34

56

7

8

curve

V

A

v (m/min)Z

29

300

10.5

27

275

9.0

24

250

8.0

22

225

7.0

20

200

5.5

19

175

4.5

18

150

3.5

17

125

3.0

1 2 3 4 5 6 7 8

© ISF 2004

Figure 5.36

br-er-05-36.cdr

Determination of Welding Speedfor MAG - Welding

curve

V

A

v (m/min)Z

29

300

10.5

27

275

9.0

24

250

8.0

22

225

7.0

20

200

5.5

19

175

4.5

18

150

3.5

17

125

3.0

1 2 3 4 5 6 7 860

kJ/cm

50

45

40

35

30

25

20

15

10

5

0

70

kJ/cm

59

53

47

41

35

29

23

18

12

6

0

He

at

inp

ut

E

SA

- w

eld

ing

He

at

inp

ut

E

MA

G -

we

ldin

g

Plate thickness0 5 10 15 20 25 30 mm 40

cracking tendency

toughness affection

butt weldsT = 150 °C0

30s

25s

20s

15s

10s

6s

30s

25s

20s

15s

10s

6s

1613

25kJ/cm

20

15

10

5

010 15 20 25 30 35 40 45 50 cm/min 60

Welding speed vS

he

at

inp

ut

E working range

12

34

56

7

8

16

13

33 41

© ISF 2004


Fig. 5.37 presents a simplification of

the determination of the microstruc-

tural composition and cooling time

subject to peak temperatures which

occur in the welding cycle. In the

lower diagram, the point of the plate

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 deter-

mination of microstructures compared

with the upper TTT diagram is that

a TTT diagram applies only for exactly one peak temperature, other peak tempera-

tures 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.

Figure 5.37

© ISF 2004

Peak temperature/cooling time– diagram for the determination

of t and the structure8/5

bie5-37.cdr

1400

°C

1200

1000

800

600

800

°C

700

600

500

400

300

200

1 10 100 1000Te

mpera

ture

Peak

tem

pera

ture

B

M

M

Arc3

Arc1

B+M F+B

300 200HV30=400

F+P

F

P

s t8/5

40 30 25 20 15 10 9 8 7 6 5 mm 4plate thickness

300 100200three-dimensional

1 2 3 5 10 20 50 100 200 400 s 1000two-dimensional

0 100 °C 200preheating temperature

6 8 10 20 30 40 50 kJ/cm 70energy-per-unit length

t8/5

1000°C1400°C

Peak temperature

6.

Welding High Alloy Steels


Basically stainless steels are characterised by a chromium content of at least 12%. Figure

6.1 shows a classification

of corrosion resistant

steels. They can be sin-

gled out as heat- and

scale-resistant and

stainless steels, depend-

ing on service tempera-

ture. Stainless steels are

used at room temperature

conditions and for water-

based media, whilst heat-

and 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 resis-

tance, 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 tempera-

ture impact resistance.

Based on the simple Fe-C

phase diagram (left figure),

Figure 6.2 shows the ef-

fects of two different

groups of alloying elements

on the equilibrium diagram.

Ferrite developers with

chromium as the most im-

portant element cause a

strong reduction of the aus-

Classification of Corrosion-Resistant Steels

non-stabilized

(austenite withdelta-ferrite)X12CrNi18-8

stabilized

(austenite withoutdelta-ferrite)

X8CrNiNb16-13

ferritic austenitic

stainlesssteels

scale- and heat-resistantsteels

corrosion-resistant steels

semi-ferritic ferritic-austenitic

X40Cr13 X10Cr13 X8Cr13 X20CrNiSi25-4

perliticmartensitic

© ISF 2002br-er-06-01e.cdr

Figure 6.1

Modifications to the Fe-C Diagramby Alloy Elements

ChromiumVanadiumMolybdenumAluminiumSilicon

NickelManganeseCobalt

Alloy elements in %Alloy elements in % Alloy elements in %

T

A4

A3

T

A4

A3

T

A4

A3

gg g

aa

a(d)

d d


Figure 6.2


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 mem-

ber of this group is nickel.

Austenite developers cause an extension of

the austenite area to Figure 6.2 (left figure)

and form a purely austenitic and transforma-

tion-free steel.

The table in Figure 6.3 summarises the ef-

fects of some selected elements on high alloy

steels.

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°C this system contains the intermetallic

σ-phase, which decomposes in the lower

temperature range into a low-chromium α-solid solution and a chromium-rich α’-solid solution.

Both, the development of the σ-phase and of the unary α-α’-decomposition cause a strong

Effects of Some Elementsin Cr-Ni Steel

Element Steel type, no. Effect

Carbon

l

l

l

All types

l

l

l

Increases the strength, supports development

of precipitants which reduce corrosion

resistance, increasing C content reduces

critical cooling rate

Chromium

l

All types

l

Works as ferrite developer, increases

oxidation- and corrosion-resistance

Nickel

l

l

All typesWorks as austenite developer, increases

toughness at low temperature, grain-refining

Oxygen

lSpecial types l

Works as strong austenite developer

(20 to 30 times stronger than Nickel)

Niobium

l

1.4511,1.4550,

1.4580 u.a.

Binds carbon and decreases tendency to

intergranular corrosion

Manganese

l

l

All types

l

l

Increases austenite stabilization, reduces hot

crack tendency by formation of manganese

sulphide

Molybdenum

l

l

1.4401,1.4404,

1.4435 and others.

l

Improves creep- and corrosion-resistance

against reducing media, acts as ferrite

developer

Phosphorus,

selenium, or

sulphur l

1.4005, 1.4104,

1.4305

l

l

Improve machinability, lower weldability,

reduce slightly corrosion resistance

Silicon l

l

All types

l

l

Improves scale resistance, acts as ferrite

developer, all types are alloyed with small

contents for desoxidation

Titanium

l

l

1.4510, 1.4541,

1.4571 and others

l

Binds carbon, decreases tendency to

intergranular corrosion, acts as a grain refiner

and as ferrite developer

Aluminium

l

Type 17-7 PH

l

Works as strong ferrite developer, mainly

used as heat ageing additive

Copper

l

l

l

Type 17-7 PH,

1.4505, 1.4506

l

l

Improves corrosion resistance against certain

media, decreases tendency to stress

corrosion cracking, improves ageing

© ISF 2002br-er06-03e.cdr

Figure 6.3

Binary System Fe - Cr

Te

mp

era

ture

1800

0

200

400

600

800

1000

1200

1400

1600

°C

30Fe 10 20 40 50 60 70 80 90 Cr%

Chromium

S S+a

a

a'a

dd+a d+a'

g+ag


Figure 6.4


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 embrittle-

ment 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

decomposition processes

in the lower temperature

range take place.

Along two cuts through the

ternary system Fe-Cr-Ni,

Figure 6.6 shows the most

important phases which

develop in high alloy steels.

A solidifying alloy with 20%

Cr and 10% Ni (left figure)

forms at first δ-ferrite. δ-

ferrite is, analogous to the

Fe-C diagram, the primary

from the melt solidifying

body-centred cubic solid

solution. However α-ferrite

is developed by transfor-

mation of the austenite, but

is of the same structure

from the crystallographic

point of view, see Figure

6.4.

Binary System Fe - Ni

30Fe 10 20 40 50 60 70 80 90 Ni

0

200

400

600

800

1000

1200

1400

1600

Te

mp

era

ture

°C

%

Nickel

Fe Ni3

Fe N

i 3

a a+g

g

dd+g

S+dS+g


Figure 6.5

Sections of the Ternary System Fe-Cr-Ni

700

800

900

1000

1100

1200

1300

1400

1500

1600

0 5 10 15 20 % Ni

% Cr30 25 20 15 10

70 % Fe

0 5 10 15 20 25700

800

900

1000

1100

1200

1300

1400

1500

1600

40 35 30 25 20 15

% Ni

% Cr

60 % Fe

Tem

pera

ture

°C °C SS

S+d+g

d+gd+g

d+g+s

dd

d+s

d+s

gg

g+sg+s

S+gS+gS+d

S+d

d+g+

s

S+d+g


Figure 6.6


During an ongoing cooling, the binary area ferrite + austenite passes through and a transfor-

mation into austenite takes place. If the cool-

ing is close to the equilibrium, a partial trans-

formation of austenite into the brittle α-phase

takes place in the temperature range below

800°C. Primary ferritic solidifying alloys show

a reduced tendency to hot cracking, because

δ-ferrite can absorb hot-crack promoting ele-

ments like S and P. However primary austen-

itic solidifying alloys show, starting at a certain

alloy content, no transformations during cool-

ing (14% Ni, 16% Cr, left figure). Primary aus-

tenitic solidifying alloys are much more

susceptible to hot cracking than primary fer-

ritic solidifying alloys, a transformation into the

σ-phase normally does not take place with

these alloys.

Figure 6.7 shows some typical compositions

of certain groups of high alloy steels.

The diagram of Strauß 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

6.1 is based on this dia-

gram. If a steel only con-

tains C, Cr and Ni, the

lowest austenite corner will

be at 18% Cr and 6% Ni.

And also other elements

than Ni and Cr work as an

austenite or ferrite devel-

oper. The influence of

these elements is de-

scribed by the so-called

chromium and nickel

Typical Alloy Content ofHigh-Alloy Steels

4.Aus

teni

tic-fe

rritic

ste

els

3.Aus

teni

tic s

teel

s

2. M

arte

nsitic

stee

ls

1. F

errit

ic s

teel

s

C

Si

Mn

Cr

Mo

Ni

Cu

Nb

Ti

Al

V

N

S

£0.1

£

.0 1

£0.1

0.11.2

max.1.0

max.1.0

max.1.0

max.1.0

max.1.0

max.1.5

max.2.0

max.2.0

1518

1218

1726

2428

up to2.0

up to1.2

up to5.0

up to2.0

£1.0

£2.5

726

47.5

up to2.2

+

+

+

+

+

+

+

+

+

+

+ indicates that the alloyelements can be added ina defined content to achievevarious characteristics


Figure 6.7

Maurer - Diagram

0

4

8

12

16

20

24

28

%

0 2 4 6 8 10 12 14 16 18 20 22 24 26%

ferrite / perlite

martensite / troostite / sorbite

austenite / martensite

martensite / ferrite

austenite / martensite / ferrite

austenite / ferrite

austenite

Nic

kel

Chromium


Figure 6.8


equivalents. The Schaeffler diagram reflects additional alloy elements, Figure 6.9. It repre-

sents molten weld metal of high alloy steels and determines the developed microstructures

after cooling down from very high temperatures. The diagram was always prepared consider-

ing 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 de-

fects may occur during welding.

Depending on the composition, purely ferritic chromium steels have a tendency to embrittle-

ment by martensite and therefore to hot cracking (area 2) or to embrittlement due to strong

grain growth (area 1).

A cause for this strong grain

growth during welding is the

greatly increased diffusion

speed in the ferrite com-

pared with austenite. After

reaching a diffusion-start

temperature, Figure 6.10

shows that ferritic steels

have a considerably

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

due to the development of

σ-phase. As explained in

6.6, this risk occurs with in-

creased ferrite contents,

increased chromium con-

tents, and sufficiently slow

cooling speed.

Schaeffler Diagram With Border Lines ofWeld Metal Properties to Bystram

0%Ferri

t

5%

10%

40%

80%

100%

20%

ferrite

martensite

M + FF+M

A+M+F

A + F

A +M

austenite

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

30

28

26

24

22

20

18

16

14

12

10

8

6

4

2

0

hardening crack susceptibility(preheating to 400°C!)

sigma embrittlementbetween 500-900°C

hot cracking susceptibility above 1250°C grain growth above 1150°C

Chromium-equivalent = %Cr + %Mo + 1,5x%Si + 0,5x%Nb

Nic

kel-equiv

ale

nt =

%N

i +

30x%

C +

0,5

x%

Mn


Figure 6.9

Grain Size as a Function of Temperature

0 200 400 600 800 1000 1200

1000

2000

3000

4000

5000

6000

gra

in s

ize

temperature

°C

m²

ferritic steel

austenitic steel


Figure 6.10


Finally, area 4 marks the strongly increased tendency to hot cracking in the austenite. Rea-

son 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 δ-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 exam-

ples for a determination of the weld metal microstructures of so-called 'black and white' joints.

Application Example ofSchaeffler - Diagram

0 4 8 12 16 20 24 28 32 360

4

8

12

16

20

24

28

Chromium-equivalent

Nic

ke

l-e

qu

iva

len

t

F

A

A+M

M

M+F

A+M+F

A+F

² ·: =1:1

²

·

10

20

40

80

100

%

F+

9

·

30%

Weld metal under 30 % dilution (= base metal amount)

² ·

·

9

S235JR (St 37)

Welding consumable

X8Cr17 (W.-Nr. 1.4510)

21% Cr, 14% Ni, 3% Mo

1

2

3


Figure 6.11

·9

Application Example ofSchaeffler - Diagram

0 4 8 12 16 20 24 28 32 360

4

8

12

16

20

24

28

Chromium-equivalent

Nic

kel-equiv

ale

nt

F

A

A+M

M

M+F

A+M+F

A+F² ·: =1:1

²

10

20

40

80

100

%

F

+

20%

Weld metal under 30 % dilution (= base metal amount)

² ·

·

9

S235JR (St 37)

Welding consumable

X10CrNiTi18-9 (W.-No. 1.4541)

21% Cr, 14% Ni, 3% Mo

123

·


Figure 6.12


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 per-

centage, 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 nitro-

gen 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.

The most important feature

of high alloy steels is their

corrosion resistance start-

ing with a Cr content of

12%. In addition to the

problems during welding

described by the Schaeffler

diagram, these steels can

be negatively affected with

view to their corrosion re-

sistance caused by the

welding process. Figure

6.14 shows schematically

the processes of electro-

lytic corrosion under a

drop of water on a piece of

iron. In such a system a

potential difference is a

precondition for the devel-

opment of a local element

consisting of an anode and

a cathode. To develop

De Long Diagram

16 17 18 19 20 21 22 23 24 25 26 27

Chromium-equivalent = %Cr + %Mo + 1,5 x %Si + 0,5 x %Nb

Nic

kel-e

qu

iva

len

t =

%N

i + 3

0 x

%C

+ 3

0 x

%N

+ 0

,5 x

%M

n

21

20

19

18

17

16

15

14

13

12

11

10

austenite

Schaeffler-austenite-martensite-line

austenite + ferrite

form

erly m

agnetically

measu

red

ferri

te c

ontents

in v

ol.-%

ferri

te n

umber

2%

4%

6%

7,6%

9,2%

10,7%

12,3%

13,8%

0%

0

2

4

68

10

12

14

16

18


Figure 6.13

Corrosion Under a Drop of Water

air

water

Fe(OH)3

iron

2Fe +O+H O 2Fe +2OH++ +++ -

2 ®

H O2

O

OH-

cathode

anode

2Fe 2Fe +4e®++ -

4e-

O +2H O+4e 4OH2 2

- -®

O2 OH

Fe+++

2Fe++


Figure 6.14


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 an-

ode, 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 ca-

thodic, 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 sur-

face (note: the processes here described should serve as a principal explanation of electro-

chemical 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 develop-

ment 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.

The examples in Figure 6.15 are more critical, since a complete recovery of the passive layer

is not possible from various reasons.

passive layerpassive layer

passive layerpassive layer

activedissolution

pitting corrosion

tensile stress

active dissolutionof the crack base

active dissolution of the gap

crevice corrosion

activly dissolvedgrain boundary

chromium zones

grain boundarycarbides

depleted

intergranular corrosion

stress corrosion cracking


Figure 6.15

gap

incorrect correct


Figure 6.16


If crevice corrosion is pre-

sent, 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 ac-

cumulate are to be avoided

by introducing a suitable de-

sign, Figure 6.16.

With pitting corrosion, the

chemical composition of the

attacking medium causes a

local break-up of the passive layer. Especially salts, preferably Cl—ions, 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 dis-

solved and the passive

layer redevelops (figures 1-

3). The repeated displace-

ment and repassivation

causes a crack propaga-

tion. Stress corrosion

cracking takes mainly

place in chloride solutions.

The crack propagation is

transglobular, i.e. it does

not follow the grain

boundaries.

Pitting Corrosion of a Storage Container

Steel

br-er-06-17e.cdr

Figure 6.17

Model of Crack PropagationThrough Stress Corrosion Cracking

1 2 3 4 5 6

121110987

offset; passive layer; metal surface; dislocation

br-er-06-18e.cdr

Figure 6.18


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 crack-

ing. 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.

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 tem-

perature, Figure 6.21.

TransgranularStress Corrosion Cracking


Figure 6.20

Influence of Elongation Speed onSensitivity to Stress Corrosion Cracking

SpRK

completecover layer tough fracture

Sensitiv

ity to s

tress c

orr

osio

n c

rackin

g

Elongation speed e

e2 e1

· ·

·

T=RT

© ISF 2002br-er06-19E.cdr

Figure 6.19


The reason is the very high affinity of chro-

mium to carbon, which causes the precipita-

tion of chromium carbides Cr23C6 on grain

boundaries, Figure 6.22. Due to these precipi-

tations, 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 aus-

tenite is considerably lower than that of car-

bon, therefore the chromium reduction cannot

be compensated by late diffusion. In the de-

pleted areas along the grain boundaries (line

2 in Figure 6.22) the steel has become sus-

ceptible to corrosion.

Only after the steel has been subjected to

sufficiently long heat treatment, chromium will

diffuse to the grain boundary and increase the

C concentration along the

grain boundary (line 3 in

Figure 6.22). In this way, the

complete corrosion resis-

tance can be restored (line 4

in Figure 6.22).

Figure 6.23 explains why the

IC is also described as in-

tergranular disintegration.

Due to dissolution of de-

pleted areas along the grain

boundary, complete grains

break-out of the steel.

Carbon Solubility ofAustenitic Cr - Ni Steels

0 0.05 0.1 0.15 0.2 0.25 % 0,3

Carbon content

600

700

800

900

1000

1100

°C

1200

A

He

at

tre

atm

en

t te

mp

era

ture

to Bain and Aborn


Figure 6.21

Sensibility of a Cr - Steel

Chro

miu

m c

onte

nt of auste

nite

resistance limit

1 - homogenuous starting condition2 - start of carbide formation3 - start of concentration balance4 - regeneration of resistance limit

1

2

3

4

Distance from grain boundary© ISF 2002br-er-06-22e.cdr

Figure 6.22


The precipitation and re-

passivation mechanisms

described in Figure 6.22

are covered by intergranu-

lar corrosion diagrams ac-

cording to Figure 6.24.

Above a certain tempera-

ture carbon remains dis-

solved in the austenite

(see also Figure 6.21).

Below this temperature, a

carbon precipitation takes

place. As it is a diffusion

controlled process, the

precipitation occurs after a

certain incubation time

which depends on tem-

perature (line 1, precipita-

tion characteristic curve).

During stoppage at a con-

stant temperature, the

parting limit of the steel is

regained by diffusion of

chromium.

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 inter-

granular corrosion is significantly higher with ferritic steel than with austenite.

Grain Disintegration


Figure 6.23

Area of Intergranular Disintegrationof Unstabilized Cr - Steels

¬R

ecip

roca

l o

f h

ea

t tr

ea

tme

nt

tem

pe

ratu

re 1

/T

oversaturatedaustenite

austenite -chromium carbide (M C )

no intergranular disintegration23 6

unsaturated austenite

Heat treatment time (lgt)

1 incubation time2 regeneration of resistance limit3 saturation limit for chromium carbide

1

2

3

austenite + chromium caride (M C )

to intergranular disintegration23 6 sensitive


Figure 6.24


As carbon is the element that triggers the intergranular corrosion, the intergranular corrosion

diagram is relevantly influenced by the c con-

tent, 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 lower temperatures and longer

times. This fact initiated the development of

so-called ELC-steels (Extra-Low-Carbon)

where the C content is decreased to less

than 0,03%

During welding, the considerable influence of

carbon is also important for the selection of

the shielding gas, Figure 6.27. The higher the

CO2-content of the shielding gas, the

stronger is its carburising effect. The C-

content of the weld metal increases and the

steel becomes more susceptible to inter-

granular corrosion.

An often used method to

avoid intergranular corro-

sion is a stabilisation of the

steel by alloy elements like

niobium and titanium, Fig-

ure 6.28. The affinity of

these elements to carbon is

significantly higher than

that of chromium, therefore

carbon is compounded into

Nb- and Ti-carbides. Now

carbon cannot cause any

chromium depletion. The

Precipitation Curves of VariousAlloyed Cr Steels

Tempering time

Te

mp

erin

g t

em

pe

ratu

re

quenchtemperature

18-8-Cr-Ni steel17% Cr steel

precipitation curves for

cooling curve


Figure 6.25

Figure 6.26

Influence of C-Contenton Intergranular Disintegration

101

102

103

104

105

106

Times

400

500

600

700

800

900

1000

Te

mp

era

ture

°C

0.07%C0.05%C

0.03%C

0.025%C



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 sub-

jected to the same heat treatment (1050°C/W means heating to 1050°C and subsequent wa-

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 inter-

granular corrosion accelerates again. The cause is the dissolution of titanium carbides at suf-

ficiently high temperature. This carbide dissolution causes problems when welding stabilised

steels. During welding, a narrow area of the HAZ is heated above 1300°C, carbides are dis-

solved. During the subsequent cooling and the high cooling rate, the carbon remains dis-

solved.

If a subsequent stress relief treatment around 600°C is carried out, carbide precipitations on

grain boundaries take place again. Due to the large surplus of chromium compared with nio-

bium or titanium, a partial chromium carbide precipitation takes place, causing again inter-

Influence of Shielding Gason Intergranular Disintegration

Shield ing gas Ar [% ] C O2 O2

S 1 99 / 1

M 1 90 5 5

M 2 82 18 /

C omposition

0,2 0,5 1 2,5 5 10 25 50 100 250 h 1000400

450

500

550

600

°C

700

0.058 % C0.53 % NbNb/C = 9

0.030 % C0.51 % NbNb/C = 17 0.018 % C

0.57 % NbNb/C = 32M2

M1

S1

Heat treatment time

Heat tr

eatm

ent te

mpera

ture


Figure 6.27

Influence of Stabilizationon Intergranular Disintegration

800

700

650

600

550

500

450

°C

Heat tr

eatm

ent te

mpera

ture

0,3 1 3 10 30 100 300 1000 h 10000Time

1050°C/W

X5CrNi18-10 unstabilized

800

700

650

600

550

500

450

°C

Heat tr

eatm

ent te

mpera

ture

0,3 1 3 10 30 100 300 1000 h 10000Time

1300°C/W

1050°C/W

X5CrNiTi18-10 stabilized

W.-No.:4301 (0,06%)

W.-No.:4541


Figure 6.28


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 suffi-

ciently 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 in-

creases constantly. They

are ideal materials when it

comes to components

which are exposed to spe-

cial conditions: high tem-

perature, corrosive attack,

low temperature, wear re-

sistance, or combinations

hereof. Figure 6.30 shows

one of the possible group-

ing of nickel-base-alloys.

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

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 an-

nealed or hot-worked con-

dition, and cold-drawn

material is recommended

for best machinability and

smoothest finish.

Knife-Line Corrosion

br-er-06-29e.cdr

Figure 6.29


Alloy Chem. composition Alloy Chem. Composition

Group A Group D1

Nickel 200 Ni 99.6, C 0.08 Duranickel 301 Ni 94.0, Al 4.4, W 0.6

Nickel 212 Ni 97.0, C 0.05, Mn 2.0 Incoloy 925 Ni 42.0, Fe 32.0, Cr 21.0, Mo 3.0, W 2.1, Cu 2.2, Al 0.3

Nickel 222 Ni 99.5, Mg 0.075 Ni-Span-C 902 Y2O3 0.5, Ni 42.5, Fe 49.0, Cr 5.3, W 2.4, Al 0.5

Group B Group D2

Monel 400 Ni 66.5, Cu 31.5 Monel K-500 Ni 65.5, Cu 29.5, Al 2.7, Fe 1.0, W 0.6

Monel 450 Ni 30.0, Cu 68.0, Fe 0.7, Mn 0.7 Inconel 718 Ni 52.0, Cr 22.0, Mo 9.0, Co 12.5, Fe 1.5, Al 1.2

Ferry Ni 45.0, Cu 55.0 Inconel X-750 Ni 61.0, Cr 21.5, Mo 9.0, Nb 3.6, Fe 2.5

Group C Nimonic 90 Ni 77.5, Cr 20.0, Fe 1.0, W 0.5, Al 0.3, Y2O3 0.6

Inconel 600 Ni 76.0, Cr 15.5, Fe 8.0 Nimonic 105 Ni 76.0, Cr 19.5, Fe 112.4, Al 1.4

Nimonic 75 Ni 80.0, Cr 19.5 Incoloy 903 Ni 39.0, Fe 34.0, Cr 18.0, Mo 5.2, W 2.3, Al 0.8

Nimonic 86 Ni 64.0, Cr 25.0, Mo 10.0, Ce 0.03 Incoloy 909 Ni 58.0, Cr 19.5, Co 13.5, Mo 4.25, W 3.0, Al 1.4

Incoloy 800 Ni 32.5, Fe 46.0, Cr 21.0, C 0.05 Inco G-3 Ni 38.4, Fe 42.0, Cu 13.0, Nb 4.7, W 1.5, Al 0.03, Si 0.15

Incoloy 825 Ni 42.0, Fe 30.0, Cr 21.5, Mo 3.0, Cu 2.2, Ti 1.0 Inco C-276 Ni 38.4, Fe 42.0, Cu 13.0, Nb 4.7, W 1.5, Al 0.03, Si 0.4

Inco 330 Ni 35.5, Fe 44.0, Cr 18.5, Si 1.1 Group E

Monel R-405 Ni 66.5, Cu 31.5, Fe 1.2, Mn 1.1, S 0.04

Typical Classification of Ni-Base Alloys

Figure 6.30


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 condi-

tions:

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 (γ') develops which

is the intermetallic compound Ni3(Al, Ti).

During solution heat treatment of X-750 at 1150°C, the number of flaws and dislocations in

the crystal is reduced and soluble carbides dissolve. To achieve best results, the material


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 γ'

phase. However, fracture properties depend more on the microstructure of the grain bounda-

ries. During an 840°C stabilising heat treatment as part of the triple-heat treatment, the fine γ'

phase develops inside the grains and M23C6 precipitates onto the grain boundaries. Adjacent

to the grain boundary, there is a γ' depleted zone. During precipitation hardening (700°C/20

h) γ' phase develops in these depleted zones. γ' particles arrest the movement of disloca-

tions, this leads to improved strength and creep resistance properties.

During the M23C6 transformation, carbon is stabilised to a high degree without leaving chro-

mium 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°C – as required for some

special heat treatments – additional γ' phase is precipitated in smaller particles. This en-

hances 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, elec-

tron 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°C. Figure 6.31 shows typical strength properties of a welded plate at a temperature

range between -423° and 1500°F (-248 – 820°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.


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 in-

sert 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°F, air

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 isola-

tion of the weld metal using gas shielded

processes is hardly possible). If such films

are not removed on a regular basis, they can

become thick enough to cause material sepa-

rations together with a reduced strength.

Brushing with wire brushes only polishes the

surface, the layer surface must be sand-

blasted 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 nor-

malized or solution heat treated condition.


tensile strength

0.2% yield stress

elongation in 1/2”

elongation in 2”

Elo

ngation, %

S

tress, 1000 p

si

Temperature, F°

220

200

180

160

140

120

100

80

60

20

0

10

30

10

20

0-423 0 200 400 600 800 1000 1200 1400 1600

Mechanical Properties ofa Typical Ni-Base Alloy

Figure 6.31

7.

Welding of Cast Materials


Figure 7.1 pro-

vides a summary

of the different

cast iron materi-

als. In this con-

nection it is only

referred to cast

iron, cast steel

and malleable

steel, as special

cast materials,

due to their poor

weldability, are of

no importance in

welding.

Figure 7.2 shows the designation of

the cast material in accordance with

DIN EN 1560. A distinction is made

between the designation “according to

the material code” and the designa-

tion “according to the material num-

ber”. In Figure 7.2, examples of two

materials are specified.

Table of the cast Iron Materials

cast materials

perlitic

alloyed

perlitic

ferritic

cast steel

unalloyed

ferritic austenitic

decarburized not decarburized

malleable iron

decarburizedannealed

malleable cast iron

not decarburizedannealed

malleable cast iron

ferritic perlitic perliticferritic austenitic

ferritic perlitic austenitic

cast iron

lemellar graphitecast iron

nodular graphitecast iron

high alloyedlow alloyed

special cast iron(G...)

hard castiron

clear chillcasting

low Ccontent

high C-content

ledeburitic graphite

Cr-castiron

otherelements

Si-castiron

Al-castiron

plastics, gypsum and s.th.similar

non-iron-metalcast materials

metalliccast materials

non-metalliccast materials

iron-carbon-cast materials


Figure 7.1

Designation according to the material number

e.g.: EN- J L 1271

1 2 3 4,5,6

Position 1: EN - standardised materialPosition 2: J - cast materialPosition 3: L - graphite structure (lamellar graphite)Position 4: 1 - number for the main characteristicPosition 5: 27 - material identification numberPosition 6: 1 - special requirement

Designation of Materials


Designation according to the material code (DIN EN 1560)

e.g.: EN-GJ L F – 150

1 2 3 4 5

Position 1: EN -Position 2: GJ -Position 3: L -Position 4: F -Position 5: 150 - (R = 150 N/mm )

-Position 6: -

m

2

standardised materialcast materialgraphite structure (lamellar graphite)microstructure (ferritic)mechanical properties

chemical composition (high alloyed)optionally

Figure 7.2


Figure 7.3 depicts a survey of the mechanical properties and the chemical composi-

tions of several customary cast materials. As to its analysis and mechanical proper-

ties 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 dif-

ferences between

the cast material

are best explained

this way. Cast iron

with lamellar and

spheroidal graph-

ite has carbon

contents of be-

tween 2,8 and

4,5%. Through the

addition of alloying

elements, above

all Si, these mate-

rials solidify fo llow-

ing the stable sys-

tem, i.e., the car-

bon is precipitated

in the form of

graphite. Malleable

cast iron shows

similar C-contents,

the solidification

from the molten

metal, however,

follows the me-

tastable system.

The C-contents of

cast steel, on the

Figure 7.3

Figure 7.4


other hand, comply with those of

common structural steels, i.e., they

are, as a rule, below 0,8% C.

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 materi-

als are weldable, constructional weld-

ing 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 trans-

formation of the cast structure is ob-

tained, the residual stresses inside the

workpiece are, on the other hand, re-

duced.

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 work-

piece geometry and the welding

method. After welding the cast work-

pieces are subject to stress-relief an-

nealing.

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 7.5

Figure 7.6


improves castability. Besides the poor mechanical properties (elongation after frac-

ture 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. Re-

pair 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 temperatures of approx. 650°C. 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 duc-

tile filler material is applied. The heat input into the base material should be as low as

possible.

Figure 7.7 depicts

the structural con-

stitution of spher-

oidal graphite cast

iron. The graphite

spheroidization is

achieved by the

addition of magne-

sium and cerium.

As, with this type

of graphite, the

notch actions are

considerably

lesser than this is

the case with grey cast iron, this type of cast iron is characterised by substantially

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

Figure 7.7


cast iron. Frequently, nickel-based

alloys are used as filler material. Prob-

lems occur in the HAZ where, besides

the ledeburite eutectic alloy system,

also Ni-Fe-martensite is frequently

formed. Both structures lead to ex-

treme hardening in the HAZ which

can be removed only by time-

consuming heat treatment.

Figures 7.8 and 7.9 show the structures of

Carburized Annealed Malleable Cast Iron

(7.7) and of Decarburized Annealed Malle-

able Cast Iron (7.9). The composition of the

malleable cast iron is thus that during solidifi-

cation, the total of carbon is bound in cemen-

tite and precipitated. During a subsequent

annealing process, the iron carbide disinte-

grates into graphite and iron.

Figure 7.8

Figure 7.9


If annealing is carried out in neutral

atmosphere, the structure of Carbur-

ized Annealed Malleable Cast Iron

develops. Annealing in oxidising at-

mosphere leads to the decarburisa-

tion of the workpiece surfaces and

Decarburized Annealed Malleable

Cast Iron is developed, Figure 7.10.

Carburized Annealed Malleable

Cast Iron is not weldable. Decarbur-

ized Annealed Malleable Cast Iron,

in contrast, is weldable.

You can see in Figure 7.11 that, also

with malleable cast iron, hardening in

the region of the HAZ occurs. For car-

rying out constructional welds made of

malleable cast iron parts, a special

material quality has been developed.

Figure 7.11 shows that this material,

EN-GJMW-400-12, is characterised by

considerably less hardening. This ma-

terial is weldable without any problems

up to a wall thickness of 8 mm.

Figure 7.11

Figure 7.10

8.

Welding of Aluminium


Figure 8.1 compares basic physical properties

of steel and aluminium. Side by side with dif-

ferent mechanical behaviour, the following

differences are important for aluminium weld-

ing:

- 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 ap-

proximately at the same temperature.

Figure 8.2 compares some mechanical prop-

erties of steel with properties of some light

metals. The important advantages of light

metals compared with steel are especially

shown in the right part of the figure. If a comparison should be based on an identical stiff-

ness, 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 quali-

tatively the stress-strain dia-

gram 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 trans-

Deflexions and Weights of Cantilever Beams Under Load

br-er-08-02.cdr

Figure 8.2

Property Al Fe

Atomic weight [g/Mol] 26.9 55.84

Specific weight [g/cm³] 2.7 7.87

Lattice fcc bcc

E-module [N/mm²] 71*10³ 210*10³

R PO,2 [N/mm²] ca. 10 ca. 100

R m [N/mm²] ca. 50 ca. 200

spec. Heat capacity [J/(g*°C)] 0.88 0.53

Melting point [°C] 660 1539

Heat conductivity [W/(cm*K)] 2.3 0.75

Spec. el. Resistance [nWm] 28-29 97

Expansion coeff. [1/°C] 24*10-6

12*10-6

FeO

Oxydes Al2O3 Fe3O4

Fe2O3

1400

Melting point of oxydes [°C] 2050 1600

(1455)

Basic Properties of Al and Fe

pO,2

m


Figure 8.1


formation during cooling, thus there is no structure transformation and consequently no

danger of hardening in the heat affected zone as with steel.

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 weld-

ing is about 900°C below steel, the isothermal curves around the welding point have a clearly

larger extension. This is due to the considerably higher heat conductivity of aluminium com-

pared 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 harde-

nable and non-hardenable (naturally hard) alloys.

Comparison of Stress-ElongationDiagrams of Al and Steel

Elongation

Al-alloy

Steel

Str

ess


Figure 8.3

Isothermal Curves of Steel and Al

4

2

-2

-4

4

2

-2

-4

low carbon steel

aluminium

400

600

200°C

800

10001200

1500

-6

-8

cm

8

6

-18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 cm 6

500

600

400300

200

100°Ccm


Figure 8.4


Figure 8.6 shows typical applications of some

Al alloys together with preferably used weld-

ing consumables.

Aluminium alloys are often welded with con-

sumable 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

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 con-

trast to hardenable alloys which have such a

potential.

The important hardening mechanism for this

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°C, then, after a suffi-

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

temperature in this condi-

tion, no precipitation will

take place. The alloy ele-

ments 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

precipitation of a second

phase takes place in ac-

Classification of Aluminium Alloys

67

86

78

Mg

Si

Mn

Cu

ZnAl

Al Cu Mg

Al Mg Si

Al Zn Mg

Al Zn Mg Cu

Al Si Cu

Al Si

Al Mg

Al Mg Mn

Al Mn

non-h

ard

enable

allo

ys

hard

enable

allo

ys


Figure 8.5

Use and Welding Consumablesof Aluminium Alloys

Al - alloys Typical use W elding consumable

Al99,5electrical engineering

SG-Al 99,5Ti;

SG-Al 99,5

AlCuMg1 mechanical engineering, food

industriesSG-AlMg4,5Mn

AlMgSi0,5 architecture, electrical

engineering, anodizing quality

SG-AlMg5; SG-AlMg4,5Mn;

SG-AlSi5

AlSi5 architecture, anodizing quality SG-AlSi5

AlMg3 architecture, apparatus-, vehicle-,

shipbuilding engineering, furniture

industry

SG-AlMg3;

SG-AlMg4,5Mn

AlMg2Mn0,8 apparatus-, vehicle-, shipbuilding

engineering

SG-AlMg5; SG-AlMg3;

SG-AlMg4,5Mn

AlMn1 apparatus-, vehicle-engineering,

food industrySG-AlMn1;SG-Al99,5T

base material - aluminium

percentage of alloy elements without factor


Figure 8.6


cordance with the binary system, the crystal tries to get back into thermodynamical equilib-

rium.

Depending on the level of

hardening temperature, the

precipitation takes place in

three possible forms: co-

herent particles (i.e. parti-

cles deviating from the

matrix in their chemical

composition but having the

same lattice structure),

partly coherent particles

(i.e. the lattice structure of

the matrix is partly re-

tained), and incoherent

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 tem-

perature (i.e. enabling diffusion).

The precipitations cause a restriction to the

dislocation movement in the matrix lattice, thus

leading to an increase in strength. The finer the

precipitations, the stronger the effect.

At an increased temperature (heat ageing, Fig-

ure 8.7) a maximum of second phase has pre-

cipitated after elapse of a certain time.

Consequently a prolonged stop at this tem-

perature 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).

Ageing Mechanism

solution heat treatment

quenching

ageing at slightly

increased temperature

coherent

precipitations,

cold aged

condition

temperature

rise

temperature

riselonger warm

ageing

longer warm

ageing

partly coherent

precipitations,

warm aged

condition

partly coherent

and incoherent

precipitations,

softening

stable incoherent

equilibrium phase

stable condition

stable condition

solidification of alloy elements

in solid solution

oversaturated solid solution,

metastable condition

coherent and partly coherent

precipitations, transition conditions

cold ageing -- warm ageing

repeated hardening

regeneration

cold ageing (RT ageing)

warm ageing


Figure 8.7

Phase Diagram Al-Cu

700

600

500

400

300

200

100

0 1 2 3 4 5 mass-% 7

Q

P

liquid

liquid and solid

copper containingaluminium solid solution

aluminium solid solutionand copper aluminide(Al Cu)2

copper content ofAlCuMg

Copper

Te

mp

era

ture


Figure 8.8


After a very long heat ageing a stable condi-

tion is reached again with relatively large pre-

cipitations of the second phase in the matrix.

In Figure 8.7 is this stable final condition iden-

tical with the starting condition. A deteriorati-

on of mechanical properties only happens

during hot ageing, if the ageing time is exces-

sively long.

The complete process of hardening at room

temperature is metallographic also called age

hardening, at elevated temperature heat age-

ing. A decrease in strength at too long ageing

time is called over-ageing.

Figure 8.9 shows a schematic representation

of time-temperature curves during hardening

with age hardening and heat ageing.

Figure 8.10 shows the

strength increase of AlZnMg

1 in dependence of time.

The difference between age

hardening and heat ageing

is here very clear. Due to

improved diffusion condi-

tions 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 ear-

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

Temperature - Time DistributionDuring Ageing

solution heat treatment

quenching

heat ageing

age hardening

2 4 6 8 10 12 14h

500

°C

400

300

200

100

0

Q

P

Te

mp

era

ture

Time


Figure 8.9

Increase of Yield Stress DuringAgeing of AlZnMg1

quenched Ageing time in h

0.2

% y

ield

str

ess

in N

/mm

²s

0.2

water quenching (~900°C/min)air cooling (~30°C/min)

10-1

100

101

10² 10³

380

320

260

200

140

80

120°C

RT


Figure 8.10


second phase is precipitated from a single-phase structure. To initiate this process, the struc-

ture 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.

Figure 8.11 shows the effect of the height of

ageing temperature level on both, mechanical

properties of a hardenable Al-alloy and on in-

cubation time. The lower the ageing tempera-

ture, the higher the resulting values of yield

stress and tensile strength. If a low ageing tem-

perature is selected, the ageing time as well as

the incubation time become extremely long.

Figure 8.11 shows that a the maximum yield

stress is reached after a period of about one

year under a temperature of 110°C. An in-

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

leads to a lowering of the

maximum strength. As the

lower part of the figure

shows, the fracture elonga-

tion is counter-proportional

to the strength values, i.e.

the strength increase

caused by ageing is ac-

companied by an embrit-

tlement of the material.

Influence of Ageing Temperatureand -Time on Ageing

260

0 10-2

10-1

100

101

102

103

h 104

190180

150 135

110°C

230260

30min

1day

30

20

10

Fra

ctu

re e

lon

ga

tio

nd

20

.2%

yie

ld s

tre

ss

s0.2

400

300

200

110

135

180

Te

nsile

str

en

gth

sB

110

135

150

180

230

500

400

300

200

Ageing time

%

N/mm²

N/mm²

205

260°C

190

150

190205°C

230

205

1week

1month

1year


Figure 8.11

Age Hardening of Al Alloys

0 30 70

100

200

300

400

N/mm²

% Strain

0

Te

nsile

str

en

gth

Rm

AlMg5

AlMg3

Al99,5


Figure 8.12


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 mecha-

nisms of strength increase of such alloys. On

one hand, tensile strength increases with in-

creasing content of alloy elements (solid solu-

tion 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

process on mechanical properties of a cold-

worked 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.

Figure 8.14 illustrates the

mechanisms in the case of a

hardenable aluminium alloy.

As a consequence of the

welding heat, the precipita-

tions are solution heat treated

and the strength values de-

crease in the weld area. Due

to the age hardening, a re-

strengthening of the alloys

takes place with increasing

time.

Non-Hardenable Al Alloy

Distance from Seam Centre

HV

30

80 60 40 20 0 20 40 60 mm 100

0,7

0,6

0,5

0,4

0,3

0,20

50

100

150

200

250

300

N/mm²

R/R

p0

,2m

Ror

Rm

p0

,2


Figure 8.13

Hardenable Al Alloy

4 mm plates of: AlZnMg1F32start values: R =263N/mm²

R =363 N/mm²

welding method: WIG, both sides,simultaneously

welding consumable: S-AlMg5specimens with machinedweld bead

p0,2

m

Distance from seam centre

Str

ess

90 days RT

21 days RT

1 day RT

80

400

N/mm²

350

300

250

200

150

100

5080 60 40 20 0 20 40 60 100 mm 140

90 days RT

1 day RT

21 days RT

Rp0,2

Rm


Figure 8.14


Figure 8.15 shows another

problematic nature of Al-

welding. Due to the high

thermal expansion of alu-

minium, high tensions de-

velop during solidification

of the weld pool in the

course of the welding cy-

cle. If the welded alloy indi-

cates a high melting

interval, cracks may easily

develop in the weld.

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

content influences signifi-

cantly the hot crack suscep-

tibility. The maximum of this

hot crack susceptibility is

likely with about 1% Mg con-

tent (corresponds with alloy

1). With increasing MG con-

tent, hot crack susceptibility

decreases strongly (see also

alloy 2 and 3, left part).

To avoid hot cracking, partly very different preheat temperatures are recommended for the

alloys. Zschötge proposed a calculation method which compares the heat conductivity condi-

tions of the Al alloy with those of a carbon steel with 0.2% C. The formula is shown in Figure

Hot Cracks in a Al Weld


Figure 8.15

Influence of Preheat Temperatureand Magnesium Content

1: AlMgMn 2: AlMg 2,5 3: AlMg 3,5

Preheat temperature

Weld

cra

ckin

g tendency

Cra

ckin

g s

usceptib

ility

Alloy content

X

X

X

X

100

80

60

40

20

%

0 100 200 300 400 °C 500

1

2

3

0 21 3 % 4

Si

Mg


Figure 8.16


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 major problem

during Al welding is the

strong porosity of 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 solidifica-

tion. Solubility of hydrogen

in aluminium changes

abruptly on the phase

transition melt-crystal, i.e.

the melt dissolves many

times more of the hydro-

gen than the just forming

crystal at the same tem-

perature. This leads to a

surplus of hydrogen in the

melt due to the crystallisa-

tion 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.

Recommendations for Preheating

Welding possible without preheating:AlMg5, AlMg7, AlMg4.5Mn,AlZnMg3, AlZnMg1

T in °C temperature of melt start (solidus temperature)

in J/cm*s*K heat conductivity

S

Al-Leg.

T in °C preheat temperaturevorw.

l

;

745TT

.LegAl

SVorw.

-l

-=

melting point pure aluminium

Increasing better weldability

Recom

mended p

reheat te

mpera

ture

Al 99,9

8R

Al9

9,9

Al9

9,8

Al 99,7

Al 99,5

Al 99

Al R

Mg0,5

Al M

g S

i 0,5

Al M

g S

i 0,8

Al M

g S

i 1

EA

l M

g S

i 1

Al M

g 1

Al S

i 5

Al C

u M

g 1

Al R

Mg 2

Al C

u M

g 0

,5A

l M

nA

l M

g 2

Al C

u M

g 2

Al M

g 3

Al M

g 3

Si

Al M

g M

n

Al Z

n M

g C

u 0

,5A

l Z

n M

g C

u 1

,5

mild

ste

el (0

.2%

C)

without pre

heating

660

600

°C

500

400

300

200

100

0


Figure 8.17

Excessive Porosity in a Al Weld


Figure 8.18


To suppress such pore

formation it is therefore

necessary to minimise the

hydrogen content in the

melt. Figure 8.19 shows

possible sources of hydro-

gen during MIG welding of

Al.

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 af-

fected zones cause in com-

bination with a parallel gap

adjustment a strong distor-

tion of the welded parts. To

minimise this distortion, the

workpieces must be set at a

suitable angle before weld-

ing, Figure 8.21.

Ingress of Hydrogen Into the Weld

too thick and water containing oxyde layerby too long or open storagein non air-conditioned rooms

nozzle deposits and too steep inclinationof the torch cause turbulences

VS

too thick oxyde layer(condensed water)

dirt film(oil, grease)

H2

H2

Grundwerkstoff

Poren

festesSchweißgut

feuchte Luft

poorcurrent transition

irregularwireelectrodefeed

humid air

humid air(nitrogen, oxygen, water)

pores

solid weld metal

base material


Figure 8.19

Weld Gap Adjustment

parallel gap

overlap

weld pool

weld pool

opening gap


Figure 8.20


Examples to Minimise Distortion

wedge flame


Figure 8.21

9.

Welding Defects

WeldingTechnology2 English

Documents

Transcript of WeldingTechnology2 English