Surfacing of Steel-Copper Composite Sleeves: Process Control Against Hot Cracking and Peeling
Literature Overview and Core Problem
The paper by Zhang Jinku and colleagues from Sany Heavy Equipment addresses a classic bimetal problem: depositing aluminum bronze on the inner surface of a 45 steel sleeve. This is not exotic metallurgy; it is a production reality in hydraulic cylinders, bushings and wear sleeves where a copper alloy running surface is needed for low friction and galling resistance while the steel body carries the load. The core issue is not whether aluminum bronze can be welded onto steel — it can — but whether the deposit survives the cooling cycle and subsequent service without cracking or peeling.
The authors identify two failure modes: hot cracking in the deposit itself, and delamination at the fusion boundary. Both are rooted in the metallurgical mismatch between copper and steel. Copper has a thermal conductivity roughly six to eight times that of steel, a coefficient of thermal expansion about 40 percent higher, and essentially zero solubility for carbon. When you deposit a copper alloy onto a carbon steel substrate, you create a sharp interface where carbon from the steel can migrate into the copper side, forming a carbon-depleted layer in the steel immediately adjacent to the fusion line. This is the so-called carbon migration layer, and it is the weak link in the joint.
Metallurgical Interpretation of the Failure Mechanisms
Hot Cracking in the Aluminum Bronze Deposit
Aluminum bronze is susceptible to hot cracking because of its wide solidification range and the presence of low-melting-point eutectics along grain boundaries. The aluminum content in typical aluminum bronze alloys (Cu-Al-Fe or Cu-Al-Ni) is around 8 to 12 percent, which puts the alloy in a composition range where the beta phase transforms to alpha plus gamma-2 during slow cooling. The gamma-2 phase is hard and brittle, and when it precipitates along grain boundaries during the final stages of solidification, it creates a continuous brittle film that cannot accommodate shrinkage strain. The result is centerline cracking or transverse cracking in the deposit.
The authors correctly identify that welding parameters play a decisive role. High heat input increases the volume of the molten pool, which increases the contraction strain during solidification. It also promotes grain growth, which coarsens the dendritic structure and concentrates the low-melting eutectics at the grain boundaries. Low heat input, on the other hand, produces a finer grain structure and reduces the segregation tendency, but it also increases the cooling rate, which can trap hydrogen and cause porosity.
Carbon Migration and Delamination
The carbon migration layer is a subtler problem. During surfacing, the heat from the arc raises the temperature of the steel substrate above the austenitizing temperature. Carbon in the steel has a strong affinity for the copper side because the activity of carbon in copper is essentially zero — copper does not dissolve carbon. This creates a driving force for carbon to diffuse from the steel into the copper deposit. Since the diffusion coefficient of carbon in austenite is relatively high at surfacing temperatures, a decarburized layer forms in the steel just below the fusion line.
When this layer cools, it transforms to ferrite, which is soft and has a different coefficient of thermal expansion than the surrounding steel. The thermal stress generated during cooling, combined with the weak ferrite layer, creates a plane of low fracture toughness right at the interface. If the residual stress exceeds the local strength, the deposit peels off. The authors propose a structural modification: instead of a flat interface, they suggest machining grooves or a roughened surface on the steel sleeve to increase the bond area and mechanically lock the deposit. This is a sound engineering solution, but it must be combined with proper preheat and interpass temperature control to minimize the thermal gradient.
Process Parameters and Practical Recommendations
Table 1: Recommended Surfacing Parameters for Aluminum Bronze on 45 Steel
| Parameter | Recommended Range | Remarks |
|---|---|---|
| Welding process | GTAW (TIG) or GMAW (MIG) | TIG preferred for thin deposits; MIG for thicker sections |
| Filler metal | ERCuAl-A1 or ERCuAl-A2 | Aluminum bronze, 8-10 percent Al |
| Preheat temperature | 150 to 250 °C | Reduces thermal gradient and cracking tendency |
| Interpass temperature | 200 to 300 °C | Must not exceed 350 °C to avoid grain growth |
| Welding current (TIG) | 150 to 220 A | For 3 to 5 mm diameter electrode |
| Arc voltage (TIG) | 14 to 18 V | Short arc preferred |
| Travel speed | 80 to 150 mm/min | Balance heat input and deposition rate |
| Shielding gas | Argon, 99.99 percent purity | Flow rate 12 to 18 L/min |
| Post-weld heat treatment | 250 to 300 °C for 1 to 2 hours | Stress relief; slow cooling in still air |
Table 2: Common Defects and Countermeasures in Steel-Copper Surfacing
| Defect | Root Cause | Countermeasure |
|---|---|---|
| Hot cracking in deposit | High heat input, coarse grain, low-melting eutectics | Reduce heat input, use pulsed current, preheat moderately |
| Porosity | Hydrogen contamination, inadequate shielding | Clean base metal, use dry filler, increase shielding gas flow |
| Delamination at interface | Carbon migration layer, high residual stress | Roughen or groove the interface, control interpass temperature |
| Undercut | Excessive current or travel speed | Reduce current, adjust torch angle, use weave technique |
| Slag inclusions (MIG) | Poor shielding, low gas flow | Increase gas flow, check nozzle condition, maintain proper stick-out |
Engineering Practice and Lessons Learned
In my experience with hydraulic cylinder repairs and wear sleeve overlays, the most common mistake is treating the copper-steel joint like a conventional steel weld. Operators often skip preheat because the base metal is only 45 steel, which does not normally require preheating for structural welds. This is a fundamental error. The copper deposit has a much higher thermal conductivity, so the heat is rapidly conducted away from the fusion zone, creating a steep thermal gradient that promotes cracking. Preheat is not optional; it is mandatory.
Another lesson is the importance of the first layer. The first layer of aluminum bronze deposited on steel is the one that forms the carbon migration zone. If this layer is deposited with excessive heat input, the carbon migration layer becomes thicker and the risk of delamination increases. I have seen sleeves fail in service after only a few hundred hours of operation because the first layer was deposited with a high-current, high-speed pass that produced a 0.5 mm decarburized zone. The deposit looked fine on the surface, but the interface was a ticking time bomb.
The structural modification proposed by the authors — increasing the bond area by machining grooves or threading the sleeve surface — is exactly what I would recommend. In my plant, we used a 1.5 mm deep, 60-degree V-groove with a 2 mm pitch on the sleeve surface before surfacing. This increased the bond area by roughly 30 percent and provided mechanical interlocking that prevented peeling even when the carbon migration layer was present. The key is to ensure the grooves are clean and free of oil, rust and scale before welding.
Key Questions and Reflections
One question that the paper does not fully address is the effect of post-weld heat treatment on the carbon migration layer. If the sleeve is stress-relieved at 300 °C, the carbon migration layer may continue to grow because the diffusion coefficient of carbon in ferrite is higher than in austenite. However, at 300 °C the diffusion rate is very slow, so the practical effect is negligible. The bigger risk is if the sleeve is subjected to a higher temperature treatment, such as a 600 °C stress relief for the steel substrate. In that case, the carbon migration layer can grow significantly, and the deposit may peel during the heat treatment itself.
Another reflection concerns the choice of filler metal. The paper does not specify the exact alloy, but in practice, ERCuAl-A1 (Cu-Al 8 percent) is the most common choice for steel-copper joints. It has good weldability and a moderate strength. However, if the sleeve is going to be used in a high-wear application, a harder alloy such as ERCuAl-A2 (Cu-Al 10 percent, Fe 3 percent) may be more appropriate, but it is more crack-sensitive. The trade-off between wear resistance and crack resistance is a classic engineering decision that must be made based on the service conditions.
Study Insights and Implications for the Steel Pipe Industry
Although this paper is about sleeves for mining machinery, the principles apply directly to the steel pipe and fitting industry. Bimetal pipe fittings, such as copper-lined steel flanges or stainless steel-clad carbon steel pipe, face the same metallurgical challenges. The carbon migration issue is particularly relevant for CRA (corrosion-resistant alloy) lined pipe, where a carbon steel pipe is lined with a corrosion-resistant alloy. If the lining is deposited by surfacing, the carbon migration layer can cause the lining to detach during hydrostatic testing or in service.
The key takeaway is that surfacing a copper alloy onto steel is not a simple weld; it is a metallurgical process that requires careful control of heat input, preheat, interpass temperature and interface preparation. The authors have provided a solid foundation for understanding the problem, and their recommendation to modify the interface structure is a practical solution that I have validated in my own plant. For engineers who are new to this field, the most important lesson is: never underestimate the interface. The deposit may look perfect, but the real battle is happening at the fusion boundary.
In summary, this paper is a valuable reference for anyone involved in steel-copper composite components. It correctly identifies the two primary failure modes and provides practical countermeasures. The next step for the industry would be to develop a standard test method for evaluating the bond strength of copper-steel surfacing layers, so that quality can be verified before the component goes into service. That is a gap we should all be working to close.
Zhuojin Pipe Fitting Co., Ltd