Failures at Welds: is Welding Always to Blame?

Philippa Moore, MA, MSci, PhD, CEng, FWeldI, IWE

Failure at the welding areas are assumed to be caused by the weld quality, however, other factors could've played a role.

Share this post

In engineering failures, we often see examples of pressure vessels bursting along their seams or bridge decks where cracks have started at welds. While at first, it might be tempting to consider that faulty welding could be the cause of the failure, there are also other reasons why failures occur at the location of welds, even with the highest quality of welding.

The surface profile of the weld acts as a geometrical stress concentration at the weld, and excessive structural loading or over-pressurization could initiate rupture at the weld location. Vibrations can also initiate fatigue cracking at welds due to this stress concentration effect. In addition, residual stresses are intrinsic to welded joints, as a small region of hot weld metal cools within the surrounding ambient temperature material, forming thermal strains that are trapped as internal stresses as it cools. This residual stress can make the region close to welds susceptible to stress corrosion cracking in components if exposed to a sufficiently corrosive environment. These failures might happen years later in the service life of the component even with welds fabricated by the most skilled welder.

Safety Standards of Welding

So how can an investigator identify whether the weld was faulty or not? First off, the welding quality should be investigated as part of a forensic investigation into the engineering failure. Fortunately, welding of safety critical equipment in industries such as oil & gas, power generation and manufacturing is covered by a robust set of quality requirements, encapsulated in ISO 3834 ‘quality requirements for fusion welding of metallic materials’, and the other standards cited within it. No weld is perfect after it’s created, but fabrication following welding best practice is intended to ensure that any weld flaws are not detrimental to the performance of that weld. Not all fabricators need to have sought official certification to ISO 3834, but nonetheless, the pack of documentation that could be requested to support an investigation into a failed weld should, among other records, include:

  • The welding procedure specification (WPS), listing the essential parameters for making that weld.
  • The welding procedure qualification records (WPQR), comprising the test and inspection certificates that demonstrates the suitability of the WPS to produce a quality weld.
  • Welder qualification certificates for the welders that did the fabrication, demonstrating their ability to perform the welding with sufficient quality.
  • Inspection and non-destructive testing records, which would demonstrate the presence or absence of flaws associated with the welds after fabrication is completed.

The types of weld flaws that could have been present since fabrication include: misalignment of the weld joint fit-up; “lack-of-fusion” in areas that have not fully melted and fused during welding; and cracks associated with the solidification of the liquid weld metal. But by first demonstrating whether the fabrication has followed the established quality procedures for welding, it allows the investigation to subsequently look at the conditions experienced by the weld once in service to understand the failure, both in terms of the design, and the actual situation during operation. Depending on how long the weld has been in operation, materials science and analysis will focus on the maintenance and inspection records and an examination of the component itself. This is to seek and understand evidence of any degradation mechanisms, such as corrosion, embrittlement, stress-corrosion cracking or fatigue cracking, supported by microscopy, analysis, metallography and testing where necessary.

When components fail at welds, the welding shouldn’t always be the first to blame and it’s important that a thorough forensic investigation has taken place to objectively establish the origin and cause of component failures.

Stay up to date. Sign Up for the Blog
Comments

More blog posts from Jensen Hughes


When Alarm Bells Ring: How Forensic Fire Investigators Uncover Arson Fraud

8 Jul 2021

Recognition of warning signs are key to identifying arson fraud and obtaining evidence needed fraudulent insurance claims

Read more
Lessons Learned from the Marine Insurance Asia Conference

28 Apr 2021

The Marine Insurance Asia conference discussed how lessons learned could improve on-board fire protection and safety.

Read more
Failures at Welds: is Welding Always to Blame?

12 Apr 2021

Failure at the welding areas are assumed to be caused by the weld quality, however, other factors could've played a role.

Read more