In the demanding environment of thermal power generation, the integrity of primary equipment is non-negotiable. As an engineer deeply involved in the maintenance and life-extension of such critical assets, I have encountered numerous challenges related to aging infrastructure. One particularly compelling case involved the repair of cracked high-pressure combined main steam valves, a quintessential example of the long-term issues faced by large, heavy-section steel castings in service. This firsthand account details the comprehensive strategy and successful execution of an in-situ repair for these vital components, highlighting the methodologies that can be applied to similar high-value steel casting repairs across the industry.
The units in question were part of a 300 MW supercritical coal-fired power plant. The high-pressure combined valves, which integrate one main stop valve and two control valves into a single casting, are fabricated from ZG15Cr1Mo1V steel, a low-alloy pearlitic heat-resistant steel casting designed for prolonged service at temperatures up to 570 °C. These steel castings are monumental; the inlet chamber diameter is 800 mm with a wall thickness of 260 mm, operating at a design pressure of 23.5 MPa and 540 °C. During a routine inspection outage, non-destructive testing revealed extensive cracking within the steam admission chambers of both valve bodies. Ultrasonic and penetrant testing mapped cracks up to 200 mm in length and 40 mm in depth, located predominantly in the upper transition radii (fillet areas) of the internal cavity. In one valve, a temperature probe penetration point had begun to leak, indicating a crack had propagated through the wall. The severity of these defects threatened unit reliability and necessitated an immediate and robust repair plan.
The genesis of cracks in such massive steel castings is a complex interplay of material science and operational mechanics. The alloy ZG15Cr1Mo1V, while possessing excellent high-temperature strength, is inherently susceptible to casting imperfections due to its complex geometry and substantial wall thickness. Defects like gas porosity, inclusions, and shrinkage cavities can act as potent stress concentrators. Furthermore, if the post-casting heat treatment and stress-relief aging are insufficient, significant residual stresses remain locked within the steel casting. During decades of service, these components endure relentless thermal cycling from plant startups, shutdowns, and load variations. The cyclic thermal stress ($\sigma_{thermal}$) can be approximated by:
$$
\sigma_{thermal} = E \cdot \alpha \cdot \Delta T
$$
where $E$ is Young’s modulus, $\alpha$ is the coefficient of thermal expansion, and $\Delta T$ is the temperature gradient across the section. This, superimposed on residual stresses and localized stress concentration at geometric discontinuities like fillets, drives fatigue crack initiation and propagation from the pre-existing casting flaws. The failure mechanism is thus a classic case of thermal-mechanical fatigue in a defect-laden steel casting.
Conventional repair wisdom for such steel castings dictated a homologous “hot” repair process using matching filler metal (ZG15Cr1Mo1V), requiring pre-heating of the entire component to around 300-400°C and subsequent Post-Weld Heat Treatment (PWHT) to temper the weld zone and relieve stresses. However, the valve’s intricate internal geometry and the location of the cracks made uniform pre-heat and controlled PWHT virtually impossible on-site without risking distortion. We needed a fundamentally different approach.
We opted for a heterogeneous “cold” repair using a nickel-base alloy, specifically ENiCrMo-3 (ERNiCrMo-3 wire). The selection was based on a critical analysis of material properties and interface behavior. The core challenge in welding dissimilar materials, especially for high-temperature service, is carbon migration and the formation of brittle phases at the fusion line. When a low-alloy steel like ZG15Cr1Mo1V is welded with a high-chromium steel filler, carbon tends to diffuse from the steel casting (lower chromium content) towards the weld metal (higher chromium content), creating a soft, carbon-depleted zone in the heat-affected zone (HAZ) and a hard, potentially brittle, carburized zone in the weld interface. This severely compromises creep strength and toughness.
The nickel-base alloy acts as a barrier. Its high nickel content (over 60%) has very low carbon solubility and forms stable carbides with elements like Nb. This drastically reduces the driving force for carbon diffusion from the steel casting parent metal. Furthermore, the coefficient of thermal expansion of ENiCrMo-3 is much closer to that of low-alloy steels than austenitic stainless steels are, reducing interfacial stresses during thermal cycling. The stress ($\sigma_{interface}$) due to differential expansion can be modeled as:
$$
\sigma_{interface} \propto (\alpha_{weld} – \alpha_{steel}) \cdot \Delta T \cdot E
$$
By minimizing $(\alpha_{weld} – \alpha_{steel})$, we minimize this stress. Most importantly, the austenitic nickel-base weld metal possesses exceptional ductility and strain-absorbing capacity, allowing it to yield and accommodate strains without cracking. This enables the repair to be performed with only a modest pre-heat (80-100°C) to remove moisture and no subsequent PWHT, as the stress is managed through mechanical means (peening) and the weld metal’s inherent plasticity.
The material properties underpinning this decision are summarized below. Table 1 and 2 detail the base steel casting material, while Table 3 and 4 show the superior strength and ductility of the chosen nickel-base filler metal.
| Element | C | Mn | Si | Cr | Mo | V | S | P |
|---|---|---|---|---|---|---|---|---|
| Content (wt%) | 0.12-0.20 | 0.40-0.70 | 0.20-0.60 | 1.20-1.70 | 0.90-1.20 | 0.25-0.40 | ≤0.03 | ≤0.03 |
| Property | Tensile Strength (Rm) | Yield Strength (Re) | Elongation (A) | Impact Energy (KV) | Hardness (HB) |
|---|---|---|---|---|---|
| Value | ≥ 490 MPa | ≥ 343 MPa | ≥ 14% | ≥ 29.4 J | 140 – 200 |
| Element | C | Mn | Si | Cr | Mo | Ni | Nb+Ta | Fe |
|---|---|---|---|---|---|---|---|---|
| Content (wt%) | 0.023 | 0.35 | 0.096 | 21.59 | 8.65 | 64.8 | 3.65 | 0.69 |
| Property | Tensile Strength (Rm) | Yield Strength (Re) | Elongation (δ5) | Hardness (HB) |
|---|---|---|---|---|
| Value | ≥ 758 MPa | ≥ 379 MPa | ≥ 30% | 150 – 220 |
The repair process was meticulous and sequenced. After comprehensive NDT mapping, the cracks were removed by careful grinding using rotary files, following the defect’s path until a clean sound metal was revealed, as verified by penetrant testing. The resulting groove was our welding prep. We did not machine a standard V-groove but maintained the natural contour of the defect removal, simply smoothing sharp corners to ensure weldability—a practical adaptation for complex internal geometry in a steel casting.
The welding was performed using the Gas Tungsten Arc Welding (GTAW) process with pulsed current for better arc control and heat input management. We employed a multi-pass, multi-layer technique. The initial root and hot passes were laid using 1.6 mm diameter wire at lower amperage to ensure precise fusion and minimize dilution with the base steel casting. Subsequent fill and cap passes used 2.5 mm wire for deposition efficiency. A critical procedural step was interpass peening. After each weld pass cooled to a temperature below 150°C, the entire bead surface was thoroughly peened using a pneumatic needle scaler or rounded chisel. This mechanically work-hardens the surface layer, inducing compressive stresses that counteract the tensile welding stresses, effectively “stress-relieving” the weld locally without a furnace. The principle can be related to the induced compressive stress ($\sigma_c$) which subtracts from the tensile residual stress ($\sigma_t$):
$$
\sigma_{net} = \sigma_t – \sigma_c
$$
This kept the net stress state favorable. Strict interpass temperature control was maintained below 150°C to prevent excessive heat buildup. The parameters are summarized in Table 5.
| Parameter | Root Pass (Φ1.6 mm) | Fill/Cap Pass (Φ2.5 mm) |
|---|---|---|
| Current | 90 – 130 A (Pulsed) | 120 – 175 A (Pulsed) |
| Voltage | 14 – 18 V | 15 – 20 V |
| Polarity | DCEN (Electrode Negative) | |
| Shielding Gas | 100% Argon | |
| Interpass Temperature | ≤ 150 °C | |
Quality assurance was rigorous and multi-stage. After weld completion and final contour grinding, the entire repaired area and adjacent parent steel casting underwent liquid penetrant inspection to verify the absence of surface-breaking defects like cracks or lack of fusion. Subsequently, ultrasonic testing was performed from the external surface to evaluate the internal soundness of the weld deposit and the adjacent base metal, ensuring no new defects were introduced and that the original flaw was completely removed. Finally, hardness surveys and positive material identification (PMI) via X-ray fluorescence were conducted across the weld, HAZ, and parent metal to confirm no unexpected hardening and correct filler metal deposition.
The success of this repair underscores a vital paradigm for maintaining critical infrastructure. For large, irreplaceable, or difficult-to-handle steel castings exhibiting service-induced damage, a well-engineered heterogeneous weld repair using high-ductility nickel-base alloys presents a powerful alternative to traditional homologous methods. It circumvents the logistical nightmares of in-situ heat treatment, reduces the risk of distortion, and provides a metallurgically compatible joint capable of withstanding the operational thermal-mechanical stresses. The repaired valves were returned to service and have since undergone multiple inspection cycles with no indication of defect recurrence. This case stands as a testament to the fact that with advanced materials science and precise execution, the service life of even the most massive and stressed steel casting components in power plants can be reliably and safely extended, ensuring grid stability and operational economy.