Make your own free website on

Welding Metallurgy

Phase Diagrams
Welding Metallurgy
Stainless Steel Production
Welding of Stainless Steel
Mechanical Properties of Stainless Steel
Favorite Links
My Resume
What Is Stainless?
About Istanbul


The Metallurgy Of Carbon Steel

The best way to understand the metallurgy of carbon steel is to study the ‘Iron Carbon Diagram’.  The diagram shown below is based on the transformation that occurs as a result of slow heating.  Slow cooling will reduce the transformation temperatures; for example: the A1 point would be reduced from 723C to 690 C.  However the fast heating and cooling rates encountered in welding will have a significant influence on these temperatures, making the accurate prediction of weld metallurgy using this diagram difficult.
  • Austenite   This phase is only possible in carbon steel at high temperature.  It has a Face Centre Cubic (F.C.C) atomic structure which can contain up to 2% carbon in solution.
  • Ferrite  This phase has a Body Centre Cubic structure (B.C.C) which can hold very little carbon; typically 0.0001% at room temperature.  It can exist as either: alpha or delta ferrite. 
  • Carbon  A very small interstitial atom that tends to fit into clusters of iron atoms.  It strengthens steel and gives it the ability to harden by heat treatment.  It also causes major problems for welding , particularly if it exceeds 0.25% as it creates a hard microstructure that is susceptible to hydrogen cracking.  Carbon forms compounds with other elements called carbides.  Iron Carbide, Chrome Carbide etc.
  • Cementite  Unlike ferrite and austenite, cementite is a very hard intermetallic compound consisting of 6.7% carbon and the remainder iron, its chemical symbol is Fe3C.  Cementite is very hard, but when mixed with soft ferrite layers its average hardness is reduced considerably. Slow cooling gives course perlite; soft easy to machine but poor toughness.  Faster cooling gives very fine layers of ferrite and cementite; harder and tougher
  • Pearlite  A mixture of alternate strips of ferrite and cementite in a single grain.  The distance between the plates and their thickness is dependant on the cooling rate of the material;  fast cooling creates thin plates that are close together and slow cooling creates a much coarser structure possessing less toughness.  The name for this structure is derived from its mother of pearl appearance under a microscope.  A fully pearlitic structure occurs at 0.8% Carbon.  Further increases in carbon will create cementite at the grain boundaries, which will start to weaken the steel.
  • Cooling of a steel below 0.8% carbon   When a steel solidifies it forms austenite.  When the temperature falls below the A3 point, grains of ferrite start to form.  As more grains of ferrite start to form the remaining austenite becomes richer in carbon.  At about 723C the remaining austenite, which now contains 0.8% carbon, changes to pearlite.  The resulting structure is a mixture consisting of white grains of ferrite mixed with darker grains of pearlite.  Heating is basically the same thing in reverse.

  • Martensite  If steel is cooled rapidly from austenite, the F.C.C structure rapidly changes to B.C.C leaving insufficient time for the carbon to form pearlite.  This results in a distorted structure that has the appearance of fine needles. There is no partial transformation associated with martensite, it either forms or it doesn’t.  However, only the parts of a section that cool fast enough will form martensite; in a thick section it will only form to a certain depth, and if the shape is complex  it may only form in small pockets.  The hardness of martensite is solely dependant on carbon content, it is normally very high, unless the carbon content is exceptionally low.
  • Tempering  The carbon trapped in the martensite transformation can be released by heating the steel below the A1 transformation temperature.  This release of carbon from nucleated areas allows the structure to deform plastically and relive some of its internal stresses. This reduces hardness and increases toughness, but it also tends to reduce tensile strength.  The degree of tempering is dependant on temperature and time; temperature having the greatest influence. 
  • Annealing  This term is often used to define a heat treatment process that produces some softening of the structure.  True annealing involves heating the steel to austenite and holding for some time to create a stable structure.  The steel is then cooled very slowly to room temperature.  This produces a very soft structure, but also creates very large grains, which are seldom desirable because of poor toughness.
  • Normalising  Returns the structure back to normal.  The steel is heated until it just starts to form austenite; it is then cooled in air. This moderately rapid transformation creates relatively fine grains with uniform pearlite.
  • Welding  If the temperature profile for a typical weld is plotted against the carbon equilibrium diagram, a wide variety of transformation and heat treatments will be observed. 
Note, the carbon equilibrium diagram shown above is only for illustration, in reality it will be heavily distorted because of the rapid heating and cooling rates involved in the welding process. 




Mixture of ferrite and pearlite grains; temperature below A1, therefore microstructure not significantly affected.

Pearlite transformed to Austenite, but not sufficient temperature available to exceed the A3 line, therefore not all ferrite grains transform to Austenite.  On cooling, only the transformed grains will be normalised. 

Temperature just exceeds A3 line, full Austenite transformation.  On cooling all grains will be normalised

Temperature significantly exceeds A3 line permitting grains to grow.  On cooling, ferrite will form at the grain boundaries, and a course pearlite will form inside the grains.  A course grain structure is more readily hardened than a finer one, therefore if the cooling rate between 800C to 500C is rapid, a hard microstructure will be formed.  This is why a brittle fracture is most likely to propagate in this region.

  • Welds  The metallurgy of a weld is very different from the parent material.  Welding filler metals are designed to create strong and tough welds, they contain fine oxide particles that permit the nucleation of fine grains.  When a weld solidifies, its grains grow from the course HAZ grain structure, further refinement takes place within these course grains creating the typical acicular ferrite formation shown opposite. 

Recommended Reading
  • Metals and How To Weld Them :-  Lincoln Arc Foundation
    A very cheap hard backed book covering all the basic essentials of welding metallurgy.
    Welding Metallurgy Training Modules:- (Devised by The Welding Institute of Canada) Published in the UK by Abington Publishing.  Not cheap but the information is clearly represented in a very readable format.

Residual Stress




Magnitude Of Stresses- A Simple Analogy


Strain Age Embrittlement

This phenomenon applies to carbon and low alloy steel.  It involves ferrite forming a compound with nitrogen; iron-nitride (Fe4N).  Temperatures around 250C, will cause a fine precipitation of this compound to occur.  It will tend to pin any dislocations in the structure that have been created by cold work or plastic deformation.

Strain ageing increases tensile strength but significantly reduces ductility and toughness.

Modern steels tend to have low nitrogen content, but this is not necessarily true for welds.  Sufficient Nitrogen, approximately 1 to 2 ppm, can be easily picked up from the atmosphere during welding.

Weld root runs are particularly at risk because of high contraction stresses causing plastic deformation.  This is why impact test specimens taken from the root or first pass of a weld can give poor results.

Additions of Aluminium can tie up the Nitrogen as Aluminium Nitride, but weld-cooling rates are too fast for this compound to form successfully.  Stress relief at around 650 degrees C will resolve the problem


The above picture is of a new pressure vessel that failed during its hydraulic test.  The vessel had been stress relieved, but some parts of it did not reach the required temperature and consequently did not experience adequate tempering.  This coupled with a small hydrogen crack, was sufficient to cause catastrophic failure under test conditions.  It is therefore important when considering PWHT or its avoidance, to ensure that all possible failure modes and their consequences are carefully considered before any action is taken.

The post weld heat treatment of welded steel fabrications is normally carried out to reduce the risk of brittle fracture by: -

  • Reducing residual Stresses.  These stresses are created when a weld cools and its contraction is restricted by the bulk of the material surrounding it.  Weld distortion occurs when these stresses exceed the yield point.  Finite element modelling of residual stresses is now possible, so that the complete welding sequence of a joint or repair can be modelled to predict and minimise these stresses.

  • Tempering the weld and HAZ microstructure.  The microstructure, particularly in the HAZ, can be hardened by rapid cooling of the weld.  This is a major problem for low and medium alloy steels containing chrome and any other constituent that slow the austenite/ferrite transformation down, as this will result in hardening of the micro structure, even at slow cooling rates. 

The risk of brittle fracture can be assessed by fracture mechanics.  Assuming worst-case scenarios for all the relevant variables.  It is then possible to predict if PWHT is required to make the fabrication safe.  However, the analysis requires accurate measurement of HAZ toughness, which is not easy because of the HAZ’s small size and varying properties.  Some approximation is possible from impact tests, providing the notch is taken from the point of lowest toughness. 

If PWHT is to be avoided, stress concentration effects such as: - backing bars, partial penetration welds, and internal defects in the weld and poor surface profile, should be avoided.  Good surface and volumetric NDT is essential.  Preheat may still be required to avoid hydrogen cracking and a post weld hydrogen release may also be beneficial in this respect (holding the fabrication at a temperature of around 250C for at least 2 hours, immediately after welding).

Nickel based consumables can often reduce or remove the need for preheat, but their effect on the parent metal HAZ will be no different from that created by any other consumable, except that the HAZ may be slightly narrower.  However, nickel based welds, like most austenitic steels, can make ultrasonic inspection very difficult.

Further reduction in the risk of brittle fracture can be achieved by refining the HAZ microstructure using special temper bead welding techniques.

Alloying Elements

Increases strength and hardness; forms a carbide; increases hardenability; lowers the transformation temperature range.  When in sufficient quantity produces an austenitic steel; always present in a steel to some extent because it is used as a deoxidiser

Strengthens ferrite and raises the transformation temperature temperatures; has a strong graphitising tendency.  Always present to some extent, because it is used with manganese as a deoxidiser

Increases strength and hardness; forms hard and stable carbides.  It raises the transformation temperature significantly when its content exceeds 12%. Increases hardenability; amounts in excess of 12%, render steel stainless.  Good creep strength at high temperature.

Strengthens steel; lowers its transformation temperature range; increases hardenability, and improves resistance to fatigue. Strong graphite forming tendency; stabilizes austenite when in sufficient quantity.  Creates fine grains and gives good toughness.

Nickel And Chromium
Used together for austenitic stainless steels; each element counteracts disadvantages of the other.

Forms hard and stable carbides; raises the transformation temperature range, and tempering temperatures.  Hardened tungsten steels resist tempering up to 6000C

Strong carbide forming element, and also improves high temperature creep resistance; reduces temper-brittleness in Ni-Cr steels.  Improves corrosion resistance and temper brittleness.

Strong carbide forming element; has a scavenging action and produces clean, inclusion free steels. Can cause re-heat cracking when added to chrome molly steels.

Strong carbide forming element. Not used on its own, but added as a carbide stabiliser to some austenitic stainless steels.

Increases strength and hardnability, reduces ductility and toughness.  Increases machineability and corrosion resistance

Reduces toughness and strength and also weldabilty.  
Sulphur inclusions, which are normally present, are taken into solution near the fusion temperature of the weld.  On cooling sulphides and remaining sulphur precipitate out and tend to segregate to the grain boundaries as liquid films, thus weakening them considerably.  Such steel is referred to as burned.  Manganese breaks up these films into globules of maganese sulphide; maganese to sulphur ratio > 20:1,  higher carbon and/or high heat input during welding > 30:1, to reduce extent of burning. 


Austenitic stainless steels

Austenitic stainless steels have high ductility, low yield stress and relatively high ultimate tensile strength, when compare to a typical carbon steel.

A carbon steel on cooling transforms from Austenite to a mixture of ferrite and cementite. With austenitic stainless steel, the high chrome and nickel content suppress this transformation keeping the material fully austenite on cooling (The Nickel maintains the austenite phase on cooling and the Chrome slows the transformation down so that a fully austenitic structure can be achieved with only 8% Nickel). 

Heat treatment and the thermal cycle caused by welding, have little influence on mechanical properties.  However strength and hardness can be increased by cold working, which will also reduce ductility.  A full solution anneal (heating to around 1045C followed by quenching or rapid cooling) will restore the material to its original condition, removing alloy segregation, sensitisation, sigma phase and restoring ductility after cold working.  Unfortunately the rapid cooling will re-introduce residual stresses, which could be as high as the yield point.  Distortion can also occur if the object is not properly supported during the annealing process.

Austenitic steels are not susceptible to hydrogen cracking, therefore pre-heating is seldom required, except to reduce the risk of shrinkage stresses in thick sections.  Post weld heat treatment is seldom required as this material as a high resistance to brittle fracture; occasionally stress relief is carried out to reduce the risk of stress corrosion cracking, however this is likely to cause sensitisation unless a stabilised grade is used  (limited stress relief can be achieved with a low temperature of around 450C ). 

Austenitic steels have a F.C.C atomic structure which provides more planes for the flow of dislocations, combined with the low level of interstitial elements (elements that lock the dislocation chain), gives this material its good ductility. This also explains why this material has no clearly defined yield point, which is why its yield stress is always expressed as a proof stress. Austenitic steels have excellent toughness down to true absolute (-273C), with no steep ductile to brittle transition. 

This material has good corrosion resistance, but quite severe corrosion can occur in certain environments. The right choice of welding consumable and welding technique can be crucial as the weld metal can corrode more than the parent material.

Probably the biggest cause of failure in pressure plant made of stainless steel is stress corrosion cracking (S.C.C).  This type of corrosion forms deep cracks in the material and is caused by the presence of chlorides in the process fluid or heating water/steam (Good water treatment is essential ), at a temperature above 50C, when the material is subjected to a tensile stress (this stress includes residual stress, which could be up to yield point in magnitude). Significant increases in Nickel and also Molybdenum will reduce the risk. 

Stainless steel has a very thin and stable oxide film rich in chrome. This film reforms rapidly by reaction with the atmosphere if damaged.  If stainless steel is not adequately protected from the atmosphere during welding or is subject to very heavy grinding operations, a very thick oxide layer will form. This thick oxide layer, distinguished by its blue tint, will have a chrome depleted layer under it, which will impair corrosion resistance.  Both the oxide film and depleted layer must be removed, either mechanically (grinding with a fine grit is recommended, wire brushing and shot blasting will have less effect), or chemically (acid pickle with a mixture of nitric and hydrofluoric acid).  Once cleaned, the surface can be chemically passivated to enhance corrosion resistance, (passivation reduces the anodic reaction involved in the corrosion process).

Carbon steel tools, also supports or even sparks from grinding carbon steel, can embed fragments into the surface of the stainless steel.  These fragments can then rust if moistened. Therefore it is recommended that stainless steel fabrication be carried out in a separate designated area and special stainless steel tools used where possible.

 If any part of stainless-steel is heated in the range 500 degrees to 800 degrees for any reasonable time there is a risk that the chrome will form chrome carbides (a compound formed with carbon) with any carbon present in the steel.  This reduces the chrome available to provide the passive film and leads to preferential corrosion, which can be severe. This is often referred to as sensitisation.  Therefore it is advisable when welding stainless steel to use low heat input and restrict the maximum interpass temperature to around 175, although sensitisation of modern low carbon grades is unlikely unless heated for prolonged periods.  Small quantities of either titanium (321) or niobium (347) added to stabilise the material will inhibit the formation of chrome carbides.

To resist oxidation and creep high carbon grades such as 304H or 316H are often used.  Their improved creep resistance relates to the presence of carbides and the slightly coarser grain size associated with higher annealing  temperatures.  Because the higher carbon content inevitably leads to sensitisation, there may be a risk of corrosion during plant shut downs, for this reason stabilised grades may be preferred such as 347H.

The solidification strength of austenitic stainless steel can be seriously impaired by small additions of impurities such as sulphur and phosphorous, this coupled with the materials high coefficient of expansion can cause serious solidification cracking problems.  Most 304 type alloys are designed to solidify initially as delta ferrite, which has a high solubility for sulphur, transforming to austenite upon further cooling. This creates an austenitic material containing tiny patches of residual delta ferrite, therefore not a true austenitic in the strict sense of the word.  Filler metal often contains further additions of delta ferrite to ensure crack free welds.

The delta ferrite can transform to a very brittle phase called sigma, if heated above 550C for very prolonged periods  (Could take several thousand hours, depending on chrome level.  A duplex stainless steel can form sigma phase after only a few minutes at this temperature)

The very high coefficient of expansion associated with this material means that welding distortion can be quite savage.  I have seen thick ring flanges on pressure vessel twist after welding to such an extent that a fluid seal is impossible.  Thermal stress is another major problem associated with stainless steel; premature failure can occur on pressure plant heated by a jacket or coils attached to a cold veesel.  This material has poor thermal conductivity, therefore lower welding current is required (typically 25% less than carbon steel) and narrower joint preparations can be tolerated.  All common welding processes can be used successfully, however high deposition rates associated with SAW could cause solidification cracking and possibly sensitisation, unless adequate precautions are taken. 

To ensure good corrosion resistance of the weld root it must be protected from the atmosphere by an inert gas shield during welding and subsequent cooling.  The gas shield should be contained around the root of the weld by a suitable dam, which must permit a continuous gas flow through the area.  Welding should not commence until sufficient time has elapsed to allow the volume of purging gas flowing through the dam to equal at least the 6 times the volume contained in the dam (EN1011 Part 3 Recommends 10).   Once purging is complete the purge flow rate should be reduced so that it only exerts a small positive pressure, sufficient to exclude air.  If good corrosion resistance of the root is required the oxygen level in the dam should not exceed 0.1%(1000 ppm); for extreme corrosion resistance this should be reduced to 0.015% (150 ppm).  Backing gasses are typically argon or helium; Nitrogen Is often used as an economic alternative where corrosion resistance is not critical, Nitrogrn + 10% Helium is better.  A wide variety of proprietary pastes and backing materials are available than can be use to protect the root instead of a gas shield.  In some applications where corrosion and oxide coking of the weld root is not important, such as large stainless steel ducting, no gas backing is used.
A pdf guide to weld purging
Huntingdon Fusion Techniques Limited

Carbon content:
304 L grade  Low Carbon, typically    0.03% Max
304   grade   Medium Carbon, typically   0.08% Max
304H grade  High Carbon, typically  Up to 0.1% 

The higher the carbon content the greater the yield strength.  (Hence the stength advantage in using stabilised grades)

Typical Alloy Content
316 Ti
(18-20Cr, 8-12Ni)
(16-18Cr, 10-14Ni + 2-3Mo)
(316 with Titanium Added)
(Same as 316Ti)
(17-19Cr, 9-12Ni + Titanium)
(17-19Cr, 9-13Ni + Niobium)
(19-22Cr, 9-11Ni)
(22-24Cr, 12-15Ni)

304 + Molybdenum
304 + Moly + Titanium
304 + Titanium
304 + Niobium
304 + Extra 2%Cr
304 + Extra 4%Cr + 4% Ni

All the above stainless steel grades are basic variations of a 304. All are readily weldable and all have matching consumables, except for a 304 which is welded with a 308 or 316, 321 is welded with a 347 (Titanium is not easily transferred across the arc) and a 316Ti is normally welded with a 318. 

Molybdenum has the same effect on the microstructure as chrome, except that it gives better resistance to pitting corrosion.  Therefore a 316 needs less chrome than a 304.
310 (24-26Cr,19-22Ni) True Austenitic.  This material does not transform to ferrite on cooling and therefore does not contain delta ferrite.  It will not suffer sigma phase embrittlement but can be tricky to weld.
904L (20Cr,25Ni,4.5Mo) Super Austenitic Or Nickel alloy.  Superior corrosion resistance providing they are welded carefully with low heat input (less than 1 kJ/mm recommended) and fast travel speeds with no weaving.  Each run of weld should not be started until the metal temperature falls below 100C.  It is unlikely that a uniform distribution of alloy will be achieved throughout the weld (segregation), therefore this material should either be welded with an over-alloyed consumable such as a 625 or solution annealed after welding, if maximum corrosion resistance is required

Carbon Steel To Austenitic Steel

When a weld is made using a filler wire or consumable, there is a mixture in the weld consisting of approximately 20% parent metal and  80% filler metal alloy ( percentage depends on welding process, type of joint and welding parameters).

Any reduction in alloy content of 304 / 316 type austenitics is likely to cause the formation of matensite on cooling.  This could lead to cracking problems and poor ductility.  To avoid this problem an overalloyed filler metal is used, such as a 309, which should still form austenite on cooling providing dilution is not excessive.

The Shaeffler diagram can be used to determine the type of microstructure that can be expected when a filler metal and parent metal of differing compositions are mixed together in a weld.

The Shaeffler Diagram
The Nickel and other elements that form Austenite, are plotted against Chrome and other elements that form ferrite,  using the following formula:-

Nickel Equivalent =  %Ni + 30%C + 0.5%Mn

Chrome Equivalent = %Cr + Mo + 1.5%Si + 0.5%Nb

Example, a typical 304L = 18.2%Cr, 10.1%Ni, 1.2%Mn, 0.4%Si, 0.02%C

Ni Equiv = 10.1 + 30 x 0.02 + 0.5 x 1.2 =  11.3
Cr Equiv = 18.2 + 0 + 1.5 x 0.4 + 0 = 18.8

A typical 309L welding consumable Ni Equiv = 14.35, Cr Equiv = 24.9

The main disadvantage with this diagram is that it does not represent Nitrogen, which is a very strong Austenite former. 

Ferrite Number
The ferrite number uses magnetic attraction as a means of measuring the proportion of delta ferrite present.  The ferrite number is plotted on a modified Shaeffler diagram, the Delong Diagram. The Chrome and Nickel equivalent is the same as that used for the Shaeffler diagram, except that the Nickel equivalent includes the addition of 30 times the Nitrogen content.  

Delong diagram


The Shaeffler diagram above illustrates a carbon steel C.S , welded with 304L filler. Point A represents the anticipated composition of the weld metal, if it consists of a mixture of filler metal and 25% parent metal. This diluted weld, according to the diagram, will contain martensite.  This problem can be overcome if a higher alloyed filler is used, such as a 309L, which has a higher nickel and chrome equivalent that will tend to pull point A into the austenite region. 

If the welds molten pool spans two different metals the process becomes more complicated.  First plot both parent metals on the shaeffler diagram and connect them with a line.  If both parent metals are diluted by the same amount, plot a false point B on the diagram midway between them.  (Point B represents the microstructure of the weld if no filler metal was applied.)

Next, plot the consumable on the diagram, which for this example is a 309L. Draw a line from this point to false point B and mark a point A along its length equivalent to the total weld dilution.  This point will give the approximate microstructure of the weld metal. The diagram below illustrates 25% total weld dilution at point A, which predicts a good microstructure of Austenite with a little ferrite.

The presence of martensite can be detected by subjecting a macro section to a hardness survey, high hardness levels indicate martensite. Alternatively the weld can be subjected to a bend test ( a side bend is required by the ASME code for corrosion resistant overlays), any martensite present will tend to cause the test piece to break rather than bend.

However the presence of martensite is unlikely to cause hydrogen cracking, as any hydrogen evolved during the welding process will be absorbed by the austenitic filler metal. 

Evaluating Dilution
Causes Of High Dilution
  • High Travel Speed. Too much heat applied to parent metal instead of on filler metal.
  • High welding Current.  High current welding processes, such as Submerged Arc Welding can cause high dilution.
  • Thin Material.  Thin sheet TIG welded can give rise to high dilution levels.
  • Joint Preparation.  Square preps generate very high dilution. This can be reduced by carefully buttering the joint face with high alloy filler metal.
  Large Schaeffler/Delong Diagram (

Duplex stainless steels

Typically twice the yield of austenitic stainless steels.  Minimum Specified UTS typically 680 to 750N/mm2 (98.6 to 108ksi).  Elongation typically > 25%.

Superior corrosion resistance than a 316.  Good Resistance to stress corrosion cracking in a chloride environment. 

Duplex materials have improved over the last decade; further additions of Nitrogen have been made improving weldability. 

Because of the complex nature of this material it is important that it is sourced from good quality steel mills and is properly solution annealed.  Castings and possibly thick sections may not cool fast when annealed causing sigma and other deleterious phases to form. 

The material work hardens if cold formed; even the strain produced from welding can work harden the material particularly in multi pass welding.  Therefore a full solution anneal is advantageous, particularly if low service temperatures are foreseen.

The high strength of this material can make joint fit up difficult.

Usable temperature range restricted to, -50 to 280C

Used in Oil & Natural Gas production, chemical plants etc.

Standard Duplex
S31803 22Cr 5Ni 2.8Mo 0.15N PREn = 32-33

Super Duplex:   Stronger and more corrosion resistant than standard duplex.
S32760(Zeron 100) 25Cr 7.5Ni 3.5Mo 0.23N  PREn = 40
Micro Of Standard Duplex

Dark Areas:- Ferrite

Light Areas:- Austenite

Duplex solidifies initially as ferrite, then transforms on further cooling to a matrix of ferrite and austenite.  In modern raw material the balance should be 50/50 for optimum corrosion resistance, particularly resistance to stress corrosion cracking.  However the materials strength is not significantly effected by the ferrite / austenite phase balance. 

The main problem with Duplex is that it very easily forms brittle intermetalic phases, such as Sigma, Chi and Alpha Prime.  These phases can form rapidly, typically 100 seconds at 900C.  However shorter exposure has been known to cause a drop in toughness, this has been attribute to the formation of sigma on a microscopic scale. 
Prolonged heating in the range 350 to 550C can cause 475C temper embrittlement. 
For this reason the maximum recommended service temperature for duplex is about 280C.

Sigma (55Fe 45Cr) can be a major problem when welding thin walled small bore pipe made of super duplex, although it can occur in thicker sections.  It tends to be found in the bulk of the material rather than at the surface, therefore it probably has more effect on toughness than corrosion resistance.  Sigma can also occur in thick sections, such as castings that have not been properly solution annealed (Not cooled fast enough).

However most standards accept that deleterious phases, such as sigma, chi and laves, may be tolerated if the strength and corrosion resistance are satisfactory.

Nitrogen is a strong austenite former and largely responsible for the balance between ferrite and austenite phases and the materials superior corrosion resistance.  Nitrogen can’t be added to filler metal, as it does not transfer across the arc. It can also be lost from molten parent metal during welding.  Its loss can lead to high ferrite and reduced corrosion resistance.  Nitrogen can be added to the shielding gas and backing gas, Up to about 10%; however this makes welding difficult as it can cause porosity and contamination of the Tungsten electrode unless the correct welding technique is used.  Too much Nitrogen will form a layer of Austenite on the weld surface.  In my experience most duplex and super duplex are TIG welded using pure argon.

Backing / purge gas should contain less than 25ppm Oxygen for optimum corrosion resistance.

Fast cooling from molten will promote the formation of ferrite, slow cooling will promote austenite. During welding fast cooling is most likely, therefore welding consumables usually contain up to 2 - 4% extra Nickel to promote austenite formation in the weld.  Duplex should never be welded without filler metal, as this will promote excessive ferrite, unless the welded component is solution annealed. Acceptable phase balance is usually 30 – 70% Ferrite

Duplex welding consumables are suitable for joining duplex to austenitic stainless steel or carbon steel; they can also be used for corrosion resistant overlays.  Nickel based welding consumables can be used but the weld strength will not be as good as the parent metal, particularly on super duplex. 

  • Low levels of austenite: - Poor toughness and general corrosion resistance. 
  • High levels of austenite: - Some Reduction in strength and reduced resistance to stress corrosion cracking. 
Good impact test results are a good indication that the material has been successfully welded.  The parent metal usually exceeds  200J.  The ductile to brittle transition temperature is about –50C. The transition is not as steep as that of carbon steel and depends on the welding process used. Flux protected processes, such as MMA; tend to have a steeper transition curve and lower toughness.  Multi run welds tend to promote austenite and thus exhibit higher toughness

Tight controls and the use of arc monitors are recommended during welding and automatic or mechanised welding is preferred.  Repair welding can seriously affect corrosion resistance and toughness; therefore any repairs should follow specially developed procedures.  See BS4515 Part 2 for details. 

Production control test plates are recommended for all critical poduction welds.

Welding procedures should be supplemented by additional tests, depending on the application and the requirements of any application code:-

  • A ferrite count using a Ferro scope is probably the most popular.  For best accuracy the ferrite count should be performed manually and include a check for deleterious phases.
  • Good impact test results are also a good indication of a successful welding procedure and are mandatory in BS4515 Part 2.
  • A corrosion test, such as the G48 test, is highly recommended. The test may not model the exact service corrosion environment, but gives a good qualative assessment of the welds general corrosion resistance; this gives a good indication that the welding method is satisfactory. G48 test temperature for standard duplex is typically 22C, for super duplex 35C


Typical Welding Procedure For Zeron 100 (Super Duplex)

Pipe 60mm Od x 4mm Thick     Position 6G

Maximum Interpass 100C     Temperature at the end of welding < 250C

1.6mm Filler Wire          85 amps  2 weld runs (Root and Cap)

Arc energy  1 to 1,5 KJ/mm         Travel speed  0.75 to 1 mm/sec

Recommended Testing

  1. Ferric Chloride Pitting Test To ASTM G48 : Method A
  2. Chemical analysis of root
  3. Ferrite count