The practical answer
A crack can be the visible symptom of overload, fatigue, poor detail, brittle behaviour, distortion, restraint, corrosion, wear or an interaction between several of these factors. Repair quality alone cannot overcome a wrong diagnosis.
The best repair decision usually separates two questions: can the crack be physically repaired, and should the component return to the same service in the same form after repair?
If the repair changes geometry, stiffness, material condition, residual stress or inspection accessibility, that change should be explicitly engineered rather than hidden in workshop language.
When this question usually matters
Welded brackets, gussets, lugs or mounting plates show visible cracking.
A pressure or structural component has local cracking near a weld toe, attachment or cut-out.
The workshop wants to gouge, weld and grind without a broader engineering review.
The asset is fatigue-loaded, safety-critical or difficult to inspect after repair.
Red flags that make the question more significant
Cracks are branching, repeating or appearing in more than one location.
The crack originates from a geometry change, weld toe or attachment detail that will remain after repair.
The parent material grade, thickness, heat treatment or previous repair history is unclear.
The component sees repeated dynamic loads, vibration, impact or pressure cycling.
The repair area will be inaccessible for future inspection after reinstatement.
How to work through a cracked-component repair decision
The sequence matters. Premature grinding and welding can erase the evidence needed to choose the right long-term solution.
Make the area safe and preserve evidence
Quarantine, unload or restrict the plant as needed, then photograph, mark and map the cracking before destructive work begins.
Define the crack and the component
Identify location, orientation, extent, parent material, thickness, weld type, attachment geometry and access for NDT.
Understand how the crack formed
Consider overload, fatigue, vibration, restraint, misalignment, corrosion, wear or previous poor repair practice. The failure mechanism will influence whether repair, redesign or replacement is appropriate.
Develop the repair or replacement concept
Decide whether to restore in kind, redesign locally, remove and replace the part, or temporarily stabilise until a permanent solution is designed.
Specify welding and verification
Define weld preparation, consumables, preheat, sequencing, distortion control, post-repair inspection, load restrictions and acceptance of the finished condition.
Detailed practical examples
These examples show how the question changes a real engineering decision. They are not universal answers; they show what needs to be clarified before the right answer becomes obvious.
Cracking around a conveyor drive support bracket
Cracks are found at the toe of a welded support bracket carrying a conveyor drive-side component. The workshop can access the area easily and believes a full gouge-and-reweld is straightforward.
The easy weld access is not the real issue. The harder question is why the bracket cracked and whether the same detail will crack again under the same cyclic duty.
What made the job difficult
- The bracket is part of a vibrating assembly with repeated start-stop cycles.
- The crack is near a local stiffness change and weld termination.
- There is no clear fatigue assessment in the original file.
How the review should proceed
- Inspect the full bracket region and any mirrored details on the opposite side to understand whether the problem is isolated or systemic.
- Assess the nominal load path and likely stress concentration drivers before choosing a repair geometry.
- Compare a simple weld restore option with a redesigned detail that improves fatigue behaviour.
- Plan post-repair inspection and future monitoring because fatigue-sensitive details may require follow-up even after repair.
The likely answer is not 'yes' or 'no' to weld repair in isolation. It is a repair strategy coupled with a detail-improvement or operating-control decision, documented against the actual fatigue duty.
Cracked support connection adjacent to a vessel nozzle
A support connection near a vessel nozzle shows cracking. The vessel itself is still in service, and the owner wants a rapid weld repair during a short shutdown.
The crack is near a pressure-boundary discontinuity and may be driven by external piping or support loads. A local weld fix could leave the underlying stress source unchanged.
What made the job difficult
- The connection is close to a nozzle reinforcement region.
- Thermal movement and piping reactions may be part of the load case.
- Post-repair access for inspection will be limited once insulation is reinstated.
How the review should proceed
- Recover the design basis for the connection, including any piping loads and support assumptions.
- Inspect for associated distortion or local shell damage, not only the visible crack path.
- Determine whether a temporary repair is justified or whether the component needs redesign, support changes or a broader outage scope.
- Align the repair and examination plan with the pressure-equipment integrity framework.
A weld repair may still be possible, but only as part of a wider integrity decision. The right answer may involve support redesign or piping-load control rather than a stand-alone weld fix.
Information to gather before deciding
Crack map showing orientation, length and location relative to geometry changes
Material, thickness and any heat-treatment or hardness information
Operating duty, load cycles, pressure cycles, vibration or impact history
Previous repair records and earlier crack history
NDT plan and access limitations before and after repair
Repair objective: temporary stabilisation, restoration or detail redesign