Dissimilar Metal Welding Guide: Why Joints Fail and How to Prevent It

Dissimilar metal welding fails for three reasons: thermal expansion mismatch that fatigues the joint during temperature cycles, brittle intermetallic compounds that crack on first load, and galvanic corrosion that eats the less noble metal in service. Every dissimilar joint you design and weld needs to address all three of these failure modes, not just one. The fix for each is different, and ignoring any one of them produces a joint with a built-in expiration date.

Filler metal selection is the most powerful tool you have. The right filler creates a deposit that’s compatible with both base metals, resists the service environment, and remains ductile enough to accommodate differential expansion without cracking.

The Three Failure Mechanisms

1. Thermal Expansion Mismatch (CTE Mismatch)

Every metal has a coefficient of thermal expansion (CTE) that determines how much it grows or shrinks with temperature changes. When two metals with different CTEs are welded together, each temperature cycle creates stress at the joint. Over time, this cyclic stress causes fatigue cracking.

The stainless-to-carbon-steel joint is the most common dissimilar combination, and it has a 1.5:1 CTE mismatch. This is manageable for most applications. The aluminum-to-steel mismatch at 2:1 creates much more severe thermal stress, which is one of the reasons (in addition to intermetallic formation) that this joint is so problematic.

Design strategies for CTE mismatch:

• Use a filler with intermediate CTE between the two base metals to create a gradual transition
• Design the joint to be flexible (thin sections, bellows, slip joints) rather than rigid
• Locate the dissimilar joint where temperature fluctuations are minimal
• Consider buttering one side with a buffer layer that transitions the CTE gradually

2. Intermetallic Compound Formation

Some metal combinations form hard, brittle intermetallic compounds when they melt and mix. These compounds have ordered crystal structures with extremely high hardness and zero ductility.

The “No” entries in the table are absolute. No filler metal, flux, shielding gas, or technique makes these combinations work through fusion welding. The alternatives are bimetallic transition inserts, mechanical fastening, adhesive bonding, or solid-state joining processes.

3. Galvanic Corrosion

When two metals with different electrochemical potentials contact each other in the presence of an electrolyte (moisture, seawater, condensation), the less noble (more anodic) metal corrodes preferentially. The galvanic series ranks metals from anodic (most active) to cathodic (most noble):

Galvanic corrosion severity depends on the potential difference between the two metals and the area ratio. A large cathode connected to a small anode accelerates corrosion. For example, a small carbon steel bolt in a large stainless steel plate corrodes rapidly because the large cathodic area drives the corrosion reaction on the small anodic bolt.

Mitigation strategies:

• Select a weld filler that’s electrochemically between the two base metals, or slightly noble to both
• Apply barrier coatings (paint, epoxy, rubber lining) to isolate the dissimilar metals
• Use insulating gaskets or washers at mechanical connections
• Design the joint so the less noble metal has the larger surface area

Filler Metal Selection Strategy

The filler must be metallurgically compatible with both base metals. “Compatible” means the diluted weld deposit doesn’t form brittle phases, hard zones, or corrosion-susceptible microstructures.

Decision Framework

*ER309L is sometimes used for carbon steel to chromoly in power plant applications where the joint operates above 800F, because the austenitic deposit resists creep better than a ferritic matching filler. This is a specialized application requiring engineering approval.

The Nickel Filler Default

When you’re unsure which filler to use on a dissimilar joint, a nickel-based filler (ERNiCr-3 or ERNiCrMo-3) is almost always a safe choice for iron-based combinations. Nickel is compatible with iron, chromium, molybdenum, copper, and most other common alloying elements. It doesn’t form brittle phases with any of them. The only combinations where nickel filler fails are those involving aluminum or titanium (intermetallic formation).

Dilution and Its Effects

Dilution is the percentage of base metal that melts into the weld pool. On a dissimilar joint, dilution shifts the deposit chemistry away from the filler composition and toward one or both base metals. Excessive dilution can move the deposit chemistry into a crack-sensitive or corrosion-susceptible range.

Reducing dilution:

• Lower amperage (reduce penetration into base metal)
• Faster travel speed (less time for base metal to melt)
• Aim the arc on the weld bead, not the base metal (especially on fill passes)
• Use buttering to buffer one side before making the joint

Joint Design for CTE Mismatch

Standard joint designs work for most dissimilar metal welds where the CTE mismatch is moderate (under 1.5:1). For higher mismatches or cyclic temperature service:

• Slip joints allow differential expansion. One piece slides inside the other with a seal weld at one end.
• Bellows or expansion loops in piping absorb differential movement between dissimilar sections.
• Transition pieces with intermediate CTE (like Invar for stainless-to-carbon-steel in cryogenic service) create a gradual CTE gradient.
• Butter layer plus stress-relief on one side creates a ductile buffer zone that deforms plastically during thermal cycling rather than concentrating stress at the fusion line.

Common Dissimilar Joint Examples

Carbon Steel to 304/316 Stainless Steel

This is the most frequently welded dissimilar combination. ER309L (or E309L-16 for stick) is the standard filler. The 309L chemistry is “over-alloyed” with chromium (23%) and nickel (13%) compared to 308L. This compensates for the dilution from the carbon steel side, which adds iron and reduces the effective Cr and Ni content. Even with 30% carbon steel dilution, the 309L deposit remains fully austenitic.

Never use ER308L on a carbon-steel-to-stainless joint. The dilution from the carbon steel side drops the chromium and nickel levels below what’s needed to maintain an austenitic structure. The deposit forms martensite, which is hard, crack-prone, and non-corrosion-resistant.

Carbon Steel to Inconel 600/625

Common in power plants (boiler tube-to-header connections) and chemical processing. ERNiCr-3 (Inconel 82) or ERNiCrMo-3 (Inconel 625) filler creates a nickel-rich deposit that tolerates iron dilution from the carbon steel without forming brittle phases. The nickel deposit also acts as a migration barrier that slows carbon diffusion from the carbon steel into the stainless/nickel alloy side during high-temperature service.

Stainless Steel to Copper-Nickel

Found in marine and desalination systems where CuNi piping transitions to stainless steel valves or flanges. ERNiCr-3 is the preferred filler because its nickel-chromium chemistry is compatible with both the copper-nickel base and the stainless base.

For transition joint solutions using explosion-bonded or roll-bonded inserts, see the transition joints guide. For the buttering technique that creates buffer layers on one side of a dissimilar joint, see the buttering technique guide.

Back to the dissimilar metals welding category.