Trouble with text or images? View this article on your web browser​

Hello Reader

Few welding discontinuities create more concern than cracks.

Unlike porosity, undercut, or excessive reinforcement, cracks are almost never acceptable under structural welding codes because they can significantly reduce the strength and reliability of a welded joint. More importantly, cracks tend to propagate under service loads, meaning a small crack can eventually become a catastrophic failure.

Understanding why welds crack is one of the most valuable skills a welder, welding inspector, supervisor, or welding engineer can develop.

Although there are many different types of weld cracks—including crater cracks, toe cracks, root cracks, and longitudinal cracks—most fall into two broad categories:

• Hot cracking (solidification cracking)
• Cold cracking (hydrogen-assisted cracking)

These two types of cracking often look similar, but they occur for completely different metallurgical reasons. Because their causes are different, the methods used to prevent them are also very different.

Understanding the difference is the first step toward eliminating weld cracking in production.

Why Cracks Are Different from Other Welding Discontinuities

Many welding discontinuities can be tolerated within limits established by welding codes.

For example, structural welding codes may permit limited amounts of porosity, slight undercut, or minor profile variations depending on the application.

Cracks are different.

A crack creates a sharp stress concentration that can continue to grow under cyclic loading or service stresses. Even if the crack is very small, it has the potential to propagate through the weld or base material over time.

For this reason, structural welding codes such as AWS D1.1 generally reject cracks regardless of their length.

Table 8.1 from AWS D1.1 (2025) provides the visual inspection criteria for welds. As you can see, cracks have no allowances for static or cyclically loaded connections.

What Is Hot Cracking?

Hot cracking occurs while the weld is solidifying or immediately afterward, when the weld metal is still at elevated temperatures.

For this reason, it is often referred to as solidification cracking.

The most common form is the familiar centerline crack, which develops near the center of the weld bead as solidification finishes.

Unlike hydrogen cracking, hot cracking originates within the weld metal itself rather than the heat-affected zone.

Top view of a centerline crack. Sometimes these cracks are barely visible to the naked eye and thus non-destructive inspection methods like liquid dye penetrant are necessary to detect them.

Why Hot Cracking Occurs

As molten weld metal cools, it does not solidify all at once.

Instead, solidification begins at the fusion boundaries and progresses toward the center of the weld.

The last portion of the weld to solidify is located near the weld centerline.

This final liquid contains many of the alloying elements and impurities rejected during solidification. Elements such as sulfur and phosphorus have relatively low melting temperatures and tend to concentrate in this final liquid region.

If tensile stresses develop before enough solid metal forms to support the weld, the remaining liquid film cannot carry the load and a crack forms.

Although sulfur and phosphorus contribute to hot cracking susceptibility, they are only part of the story.

Modern research has shown that several factors influence solidification cracking, including:

• Alloy chemistry
• Weld bead geometry
• Solidification temperature range
• Grain structure
• Joint restraint
• Welding parameters

All of these factors influence whether the final stages of solidification can accommodate the stresses created as the weld contracts.

Why Weld Bead Shape Matters

One of the most overlooked contributors to hot cracking is weld bead geometry.

Very narrow, deep welds are generally more susceptible to centerline cracking than wider welds.

Deep penetrating welds are susceptible to solidification cracks. The base material chemistry plays a huge role in this as impurities will increase the probability of a crack.

As the weld freezes from both sides toward the middle, the final liquid becomes trapped along the centerline.

If the weld is excessively deep compared to its width, tensile stresses concentrate in this narrow region during the last stages of solidification.

For many carbon steel applications, a width-to-depth ratio between approximately 1:1 and 1.4:1 produces a favorable weld profile.

When joint geometry or welding parameters create excessively narrow welds, redesigning the joint or adjusting travel speed, amperage, voltage, or electrode angle may reduce cracking susceptibility.

Hot Cracking in Stainless Steel

Many welders associate hot cracking only with carbon steel.

In reality, fully austenitic stainless steel weld metals can be even more susceptible.

For example, filler metals such as ER310, which produce nearly fully austenitic weld deposits, are considerably more prone to solidification cracking than filler metals containing small amounts of ferrite.

That small amount of ferrite found in fillers such as ER308 and ER309 helps reduce hot cracking by interrupting continuous low-melting films that form during solidification.

This is one reason filler metal selection is so important when welding stainless steels.

Preventing Hot Cracking

The most effective methods include:

• Maintain favorable weld bead geometry.
• Select filler metals appropriate for the base material.
• Minimize sulfur and phosphorus levels where possible.
• Reduce excessive restraint through proper joint design.
• Use welding procedures that promote stable solidification.
• Avoid excessively deep, narrow weld profiles.

Unlike hydrogen cracking, increasing preheat alone generally does not solve hot cracking.

What Is Cold Cracking?

Cold cracking is very different.

More accurately cit is called hydrogen-assisted cracking (HAC) or hydrogen-induced cracking (HIC).

Unlike hot cracking, hydrogen cracking occurs after the weld has solidified.

It may appear:

• Several hours after welding
• Overnight
• One or even several days later

This delayed appearance makes hydrogen cracking especially dangerous because welded assemblies may already have passed inspection or even entered service before cracks become visible.

The Three Conditions Required for Hydrogen Cracking

Hydrogen cracking occurs only when three conditions exist simultaneously:

1. A susceptible microstructure
2. Diffusible hydrogen
3. High tensile stress or restraint

In order for hydrogen-induced cracking to occur all three factors need to be present. Eliminate one and it can't happen.

Remove any one of these three factors and hydrogen cracking becomes highly unlikely.

This is perhaps the most important concept in understanding cold cracking.

Understanding the Susceptible Microstructure

The susceptible microstructure is usually hard martensite formed in the heat-affected zone.

Martensite develops when certain steels cool rapidly from welding temperatures.

Contrary to popular belief, carbon content alone does not determine whether martensite forms.

Other alloying elements—including manganese, chromium, molybdenum, nickel, and vanadium—also influence hardenability.

This is why welding engineers often evaluate Carbon Equivalent (CE) rather than carbon content alone.

Higher carbon equivalent generally means:

• Greater hardenability
• Higher HAZ hardness
• Greater hydrogen cracking susceptibility
• Higher preheat requirements

Why Preheat Works

Many welders think preheat simply "warms the steel."

Its real purpose is much more important.

Preheating slows the cooling rate of the weld and heat-affected zone.

Some applications require heating of the parts before, during and after welding to properly control the cooling rates and prevent cracking.

This provides several benefits:

• Reduces martensite formation
• Allows hydrogen to diffuse out of the weld
• Reduces thermal stresses
• Improves ductility during cooling

These effects work together to dramatically reduce hydrogen cracking susceptibility.

This is why AWS D1.1 specifies preheat requirements for many steels based on thickness and material group.

Where Does the Hydrogen Come From?

Hydrogen can enter the weld from many sources, including:

• Moisture in electrode coverings
• Damp fluxes
• Dirty base metal
• Oil or grease
• Rust
• Moisture in shielding gas systems
• Atmospheric humidity under poor welding conditions

Modern low-hydrogen consumables greatly reduce this risk, but only when stored and handled correctly.

The Role of Joint Restraint

Even when hydrogen and martensite are present, cracking is much less likely if the joint is free to contract.

Problems arise when welded components are highly restrained.

Examples include:

• Thick restrained joints
• Heavy structural members
• Rigid fixtures
• Complex assemblies
• Multipass groove welds

As the weld cools, contraction generates high tensile stresses.

Hydrogen naturally diffuses toward these highly stressed regions, increasing the likelihood of cracking.

Why Hydrogen Cracks Often Appear Later

One of the defining characteristics of hydrogen cracking is delayed appearance.

Hydrogen atoms remain mobile after welding.

Over time they diffuse toward areas of high triaxial stress within hardened microstructures.

Once sufficient hydrogen accumulates, cracking begins.

This delayed mechanism explains why AWS D1.1 requires certain weld inspections to be delayed after welding to allow time for hydrogen-assisted cracks to develop before final acceptance.

Table 8.1 of AWS D1.1 (2025) clearly notes that base metals susceptible to hydrogen-induced cracking most be visually inspected no less than 48 hours after welding.

Preventing Hydrogen Cracking

Fortunately, hydrogen cracking can usually be prevented.

Effective methods include:

• Use low-hydrogen filler metals.
• Store electrodes correctly.
• Maintain clean base materials.
• Apply proper preheat.
• Control interpass temperature.
• Avoid excessively low heat input.
• Reduce restraint where possible.
• Perform post-weld hydrogen bake-out or PWHT when required.

Notice that these recommendations differ significantly from those used to prevent hot cracking.

That is because the cracking mechanisms themselves are completely different.

Identifying the Correct Crack Is Critical

One of the most common troubleshooting mistakes is attempting to solve every crack the same way.

For example:

Increasing preheat may reduce hydrogen cracking but have little effect on solidification cracking.

Likewise, changing filler metal chemistry may solve hot cracking while doing very little to prevent hydrogen-assisted cracking.

The first step in solving any weld cracking problem is correctly identifying which type of crack occurred.

Only then can an effective corrective action be selected.

Final Thoughts

Hot cracking and hydrogen-assisted cracking are among the most serious welding discontinuities because they originate from fundamentally different metallurgical mechanisms.

Hot cracking occurs during solidification and is largely influenced by alloy chemistry, weld bead geometry, and solidification behavior.

Hydrogen cracking occurs after welding and requires three conditions: a susceptible microstructure, diffusible hydrogen, and tensile stress.

Understanding these mechanisms allows welders, inspectors, and engineers to prevent cracking before production begins rather than repairing expensive failures later.

References:

​Weld Troubleshooting for Non Welding Engineers​

Welding Metallurgy and Weldability, John C. Lippold