Dissimilar metal welds concentrate almost everything that makes welding engineering difficult: carbon migration across the fusion line, mismatched thermal expansion driving cyclic stress, oxide notch development in service, and residual-stress states that depend heavily on the original fabrication. When one fails — a ferritic-to-austenitic joint on a header stub, a nickel-transition weld in high-energy piping — the organization suddenly needs a repair philosophy it should have selected in advance.
Diagnose the driver before selecting the repair
A DMW repair chosen without a cause is a repeat failure with a newer date stamp. The essential first question is which driver dominated: creep-driven cracking along the carbon-depleted zone, thermal-fatigue from expansion mismatch and cycling, an original fabrication defect finally propagating, or an operating change that moved the joint into harsher duty. Each points to a different combination of filler, geometry, and post-weld treatment.
The main approaches, and what each one buys
Like-for-like replacement restores the original condition — appropriate when the joint achieved acceptable life and the failure was end-of-life rather than premature. Nickel-based filler repairs reduce carbon migration and better bridge expansion mismatch, at the cost of tighter welding control. Redesigning the transition — relocating the joint to a lower-temperature or lower-stress position, or introducing an engineered transition piece — is the most expensive option and the only one that removes the underlying problem rather than managing it. Weld overlay and local repair techniques can defer a full replacement when the remaining section is sound, provided the residual-stress consequences are understood.
- Confirm the damage driver with metallurgical evidence before committing to an approach
- Select filler and procedure against service temperature, cycling, and the expansion mismatch of the specific material pair
- Decide post-weld heat treatment deliberately: it is not automatically beneficial in dissimilar joints
- Document the repair basis so the next outage team inherits a position, not a mystery
The best time to select a DMW repair philosophy is before the first failure. The second-best time is immediately after it, with the metallurgy in hand.