Cracking of hardfacing alloys - prevent, cure or ignore?

TWI Bulletin, May/June 1991

Neville Gregory's career as a metallurgist spanned 40 years up to his retirement in 1990. After working at The English Electric Company, Rugby, Murex Welding Processes and Enfield Rolling Mills he joined the London office of BWRA as a Welding Technologist. For six years he carried out liaison visits to members and dealt with technical enquiries.

This work continued when he joined the Research Station at Abington and was combined with contract research on various welding processes and materials including zinc coated steels, low alloy steels, reinforcing bars, and armour plate.

From 1975 onwards Neville Gregory headed a Welding Advisory Service in the Arc Welding Department assisted by a team of Welding Engineers.

The consequences of finding cracking in hardfacing alloys need not always be costly. The wear resistance or life of a component may be unaffected. But if cracks are deemed unacceptable precautions can be taken to prevent their recurrence, as Neville Gregory explains.

In the article 'Why do welds crack' in Bulletin 2, 1991 it was noted that cracks in welded fabrications, however minor, are unacceptable and are not allowed by any codes of practice. Furthermore, cracking indicates that the welding procedure was faulty and repair of the cracks is almost invariably required.

Hardfacing is somewhat different because in some cases where alloys having high hardnesses are deposited the brittle nature of the alloy leads to cracking whatever the welding procedure. For example, the higher alloy chromium irons with hardnesses up to 800HV have a fine network of cracks which are caused by shrinkage stresses during cooling. These cracks do not affect the long life of the weld deposit, which gives excellent service under heavy abrasion in many industries including mining, quarrying, and mineral processing. In contrast, some applications require crack free deposits, typical examples being valves and valve seats, steel mill rolls, and hot work dies.

Types of surface that must be free from cracks are:

• Sealing surfaces;
• Surface requiring both wear and corrosion resistance;
• Surfaces subject to fatigue stresses arising from either mechanical or thermal cycling;
• Surfaces that must be clean and avoid pick-up of material being processed which could contaminate subsequent batches.

To prevent cracking in hardfacing deposits correct procedures must be used and these are somewhat different from those used to avoid cracking of arc welded joints.

First, solidification cracking is rare in hardfacing deposits although it has been encountered in somewhat specialised circumstances. There is normally no need to take any particular precautions against this form of cracking.

Contraction cracks caused by the low ductility of the weld metal are the most common type of defect and occasionally hydrogen cracking of the parent metal is a problem. However, hydrogen cracking is far less frequent in hardfacing than in welding because of the much lower level of restraint.

Contraction cracks

The commonest form of cracking in hardfacing deposits is contraction cracking, sometimes referred to as relief cracking because it relieves a high proportion of the residual stresses set up by the shrinkage of the weld ( Fig.1).

Contraction cracking may be defined as a brittle form of cracking caused by low ductility of the weld metal which is unable to accommodate the shrinkage strain during cooling.

Cracking may occur at various temperatures from ambient up to around 500°C or even higher depending on the alloy composition.

Geometrical factors

The shape of the weld bead is far less important in hardfacing than in fabrication welding, in which the width to depth ratio must be maintained between certain limits to avoid solidification cracking. Hardfacing is different because the deposit consists essentially of bead on plate welds in which the direction of solidification is different from that which occurs in a welded joint.

Solidification of a butt weld or a fillet weld starts at the two parent metal surfaces and the solidifying metal grains meet at the centreline where any impurities tend to segregate. This is the site where solidification cracking can occur.

The solidification pattern of hardfacing deposit is different because the metal grains grow parallel to one another and the solidification front is substantially parallel to the plate surface so that there is no central plane of weakness where low melting point impurities might segregate. Hence solidification cracking is rare in hardfacing deposits.

Because of the solidification mode a very high width to depth ratio can be tolerated without any risk of solidification cracking. Apart from this, transverse restraint on a hardfacing deposit is far lower than on a butt weld or a fillet weld. Longitudinal shrinkage assumes far greater importance so that contraction cracks are orientated transverse to the direction of welding ( Fig.1). Occasionally a type of fillet weld is made with a hardfacing alloy, for example when an internal corner is coated ( Fig.2). In this case longitudinal contraction cracking may occur through the throat of the weld.

This can be prevented by welding the corner with a less crack sensitive material and hardfacing over the top. A steel component should have the internal corners welded with a mild steel consumable before hardfacing (Fig.3). Alternatively the corner should be machined or ground to a radius of at least 5mm if this is practicable.

If a recess is machined in a component to receive a hardfacing alloy the corners of the recess should have a minimum angle of 45° to the vertical or the corner should be rounded to at least 5mm radius ( Fig.4).

Metallurgical factors

A tendency to contraction cracking increases with the hardness of a hardfacing alloy because the higher the hardness the lower the ductility and therefore the ability to accommodate shrinkage stresses. Low alloy steel hardfacing alloys do not generally exhibit contraction cracking until the hardness is 500HV or higher.

The metallurgical changes leading to the higher hardness are the formation of carbides of Cr,Mo,V, or W in the martensitic matrix.

The Stellite type alloys based on Co-Cr-W, used for their excellent resistance to heat and corrosion, have hardnesses from 400 to 630HV and even the softest has ductility less than 1% so that contraction cracking may occur if precautions are not taken.

Two layer deposits of Co-Cr-W alloys on steel parent metal are more susceptible to cracking than single layer deposits which have higher ductility because of dilution by iron.

Alloys containing large volumes of Cr,V, or W carbides have high susceptibility to contraction cracking, which generally takes the form of a network of fine cracks, and precautions against this are not usually effective.

The precaution usually taken to prevent contraction cracking is to preheat the component so that the surface expands and when it cools the differential contraction between the hardfacing deposit and the parent metal is reduced. The higher the preheat the lower the contraction stresses and a general preheat applied to the whole component is more effective than a local preheat. This is because a general preheat results in a lower cooling rate than a local.

In any event, the cooling rate of a hardfaced component should generally be reduced if solidification cracking is to be avoided. This can be accomplished by wrapping the component in insulating blankets or by burying it in mica or kieselguhr.

A simple guide for preheating temperatures to prevent contraction cracking is shown in the Table.

Typical preheat temperatures to prevent contraction cracking

To avoid hot spots which could result in uneven heating and cooling the maximum interpass temperature should generally be restricted to 100°C higher than the preheat temperature. This temperature range is a general guideline which may have to be very carefully controlled in critical cases and which may be dispensed with in others.

In complex hardfacing operations, e.g. when coating large areas, it may be necessary after preheating and welding to heat slowly to 600-650°C to relieve residual stresses. This is followed by slow cooling.

The Table suggests that low alloy steels having hardnesses less than 400HV do not require preheating of the parent metal to avoid contraction cracking. This certainly applies to single layer deposits as shown in Fig.5, in which the internal walls of a fluidised bed boiler were hardfaced without any preheat over several square meters with an alloy of 350HV hardness. No cracks were visible at the surface.

Welding procedure tests were carried out with and without preheat before work on the boiler was started.

Hardfacing hardenable parent metals

The medium and higher carbon steels and low alloy steels form hard heat affected zones (HAZs) when hardfaced, and when the HAZ transforms to hard martensite it expands. This expansion can produce additional tensile stresses in the hardfacing deposit and can increase the tendency to cracking in the harder alloys.

To prevent cracking of the hardfacing caused by expansion of the HAZ the preheating temperature must be high enough to suppress transformation partially at least.

Preheat temperatures of 300-350°C restrict martensite formation in the HAZ and, depending on the compositions of the parent metal and the hardfacing alloy, cracking can be prevented by maintaining the temperature for some time after welding so that transformation of the HAZ occurs above the temperature at which martensite starts to form. Alternative more complex heat treatments are possible involving partial transformation to martensite followed by tempering.

Stress level

In some cases, when hardfacing the edges of long components such as shear blades or bulldozer cutting edges, presetting or bending the part in the opposite direction to which it would distort on welding can prevent contraction cracking if the component is allowed to move freely during hardfacing.

When a multilayer deposit is required to build up a component to its original dimensions the initial layers should, if possible, be made with softer hardfacing alloys. For example, a three layer deposit could be made with individual layers having hardnesses of 250, 350, and finally 650HV.

Experience has shown that this gradual increase in hardness, as the deposit is built up, effectively reduces the cracking tendency below that of a multilayer deposit of an alloy having a hardness of 650HV.

An interesting variation of this technique is the hardfacing of sinter breaker hammers which are built up with alternate layers of mild steel and tungsten carbide rich alloys. Multilayer deposits of tungsten carbide alone contain extensive cracking which leads to spalling of the deposit after a short time in service. Mild steel interlayers effectively cushion the impact on the brittle tungsten carbide.

Hydrogen cracks

Weld metal

Some clarification is required on hydrogen cracking of alloy steel hardfacing deposits. It is well known in fabrication welding that low alloy steel weld metals, particularly the Cr-Mo types, may require higher preheating temperatures than parent metal of similar composition to prevent hydrogen cracking. However, as noted above, the restraint on hardfacing deposits is generally lower than that on welded joints and this reduces the susceptibility to cracking to some extent.

This point is borne out by the extensive single layer hardfacing deposit shown in Fig.5, where mild steel was coated without any preheat using a 3%Cr weld metal.

What is surprising is that the electrode used had a rutile coating which would have produced a relatively high hydrogen content in the weld metal.

This example illustrates the significance of restraint in hydrogen cracking.

If, on the other hand, a 3%Cr weld is deposited on a medium carbon steel, dilution of the weld metal by the parent metal produces a higher carbon content in the weld which becomes more susceptible to hydrogen cracking. In this case use of a basic coated electrode dried to give a low hydrogen content would be recommended.

Use of rutile coverings for low alloy steel electrodes may be queried but there is some demand for them because they can be used with small, low open circuit transformers.

In fact, low alloy steel electrodes giving weld metal hardnesses of up to 600HV are available with rutile coatings, but to avoid hydrogen cracking of the deposit preheats of 200-300°C are necessary.

Low hydrogen steel consumables should always be used if crack free deposits are mandatory because it is possible that small hydrogen cracks could initiate contraction cracking which may extend to the weld surface.

Parent metal

The risk of hydrogen cracking of the parent metal must be considered and preheating data are available based on experience and limited experimental work.

Because of the low level of restraint compared with fabrication welding it is often possible to reduce the preheat by a considerable margin.

On the other hand it must be borne in mind that medium carbon and low alloy steels undergo a sudden and rapid expansion when martensite is formed in the HAZ and this occurs over a relatively narrow temperature range.

This can increase the likelihood of contraction cracking of the hardfacing and the appropriate preheat temperature must be chosen to avoid this.

The scarcity of experimental data on the relationship between hardfacing alloys and parent metals and their transformation characteristics dictates that for critical applications procedure testing should be carried out to determine crack free welding conditions.