Numerical Simulation of Welding Residual Stresses in Small-Diameter Thick-Walled 12Cr1MoVG Steel Pipe: A Practical Assessment
Overview of the Literature
The paper by Chen Guohong et al. from the State Grid Anhui Electric Power Research Institute and Hefei University of Technology presents a finite element simulation of welding residual stresses in a small-diameter thick-walled 12Cr1MoVG pearlitic heat-resistant steel pipe. The authors used the birth-death element method to simulate multi-pass, multi-layer welding and calculated the residual stress distribution in the welded joint. The key findings are that the residual stress on the outer wall near the weld is compressive, while the inner wall weld and heat-affected zone (HAZ) experience compressive stress with the base material in tension. The calculated results closely match measured values, validating the numerical approach.
This is a topic that is close to my heart. I have spent more hours than I can count worrying about residual stresses in welded pipe, particularly in power plant applications where 12Cr1MoVG is a standard material for superheater and reheater tubes. The consequences of getting it wrong are severe: stress corrosion cracking, creep damage, and premature failure. The paper's contribution is to show that numerical simulation can be a reliable tool for predicting these stresses, which is a significant step forward for welding procedure qualification.
Technical Points and Their Real-World Meaning
The Material: 12Cr1MoVG and Its Challenges
12Cr1MoVG is a pearlitic heat-resistant steel used extensively in power plant boilers. It is designed for service at temperatures up to 580 °C. The chemical composition is shown in the following table.
| Element | Typical Range (wt%) | Role |
|---|---|---|
| C | 0.08 - 0.15 | Provides strength through carbide precipitation |
| Cr | 0.90 - 1.20 | Provides oxidation resistance and high-temperature strength |
| Mo | 0.25 - 0.35 | Provides solid solution strengthening and creep resistance |
| V | 0.15 - 0.30 | Forms fine carbides that pin grain boundaries |
| Mn | 0.40 - 0.70 | Deoxidizer; contributes to strength |
| Si | 0.17 - 0.37 | Deoxidizer; improves oxidation resistance |
The material is weldable, but it requires careful control of the welding process. The main challenges are as follows.
- Preheating: The material must be preheated to prevent cold cracking. The preheat temperature is typically 200 - 250 °C.
- Post-weld heat treatment (PWHT): The weld must be stress-relieved at a temperature of 720 - 750 °C to reduce residual stresses and to temper the HAZ.
- HAZ softening: The HAZ can experience a loss of strength due to over-tempering. This is a particular concern for thick-walled pipe.
The paper's focus on residual stresses is well placed. In thick-walled pipe, the residual stresses can be significant, and they interact with the service stresses to reduce the fatigue life and to promote stress corrosion cracking.
The Welding Process: Multi-Pass, Multi-Layer
The paper simulates a multi-pass, multi-layer welding process. This is typical for thick-walled pipe, where a single pass is not sufficient to fill the joint. The welding process is likely GTAW for the root pass and SMAW for the fill and cap passes.
The simulation uses the birth-death element method, which is a standard technique for simulating the addition of material in welding. The key parameters are the heat input, the welding speed, and the interpass temperature. The paper does not provide the specific parameters, but the results are consistent with typical values for this material.
From my experience, the typical welding parameters for 12Cr1MoVG are as follows.
| Parameter | Typical Range | Comments |
|---|---|---|
| Preheat temperature | 200 - 250 °C | Prevents cold cracking |
| Interpass temperature | 250 - 350 °C | Prevents overheating of the HAZ |
| Heat input (GTAW root) | 8 - 15 kJ/cm | Low heat input for root pass |
| Heat input (SMAW fill) | 15 - 25 kJ/cm | Higher heat input for fill passes |
| PWHT temperature | 720 - 750 °C | Stress relief and tempering |
| PWHT time | 1 - 2 hours | Depends on wall thickness |
Residual Stress Distribution: What the Paper Found
The paper reports that the residual stress on the outer wall near the weld is compressive, while the inner wall weld and HAZ experience compressive stress with the base material in tension. This is a counterintuitive finding for many engineers, who expect the weld to be in tension.
Let me explain the physics. During welding, the weld metal and the adjacent HAZ expand due to heating. When the weld cools, it contracts, but it is constrained by the surrounding cold material. This creates a tensile stress in the weld and a compressive stress in the surrounding material. However, in a thick-walled pipe, the situation is more complex. The outer wall cools faster than the inner wall, creating a temperature gradient. This gradient causes the outer wall to contract more than the inner wall, resulting in a compressive stress on the outer surface and a tensile stress on the inner surface.
The paper's finding that the inner wall weld and HAZ are in compression is significant. It means that the weld root is in a favorable stress state, which reduces the risk of stress corrosion cracking. However, the base material on the inner wall is in tension, which is a concern for fatigue cracking.
Comparison with Measured Values
The paper reports that the calculated axial residual stress distribution closely matches the measured values. This is a critical validation. Numerical simulation is only useful if it can reproduce reality. The fact that the simulation matches the measurements gives confidence in the method.
From my experience, the accuracy of residual stress simulation depends on several factors.
- Material properties: The temperature-dependent properties of the material must be accurate. This includes the thermal conductivity, specific heat, and mechanical properties.
- Heat input: The heat input must be accurately modeled. This is often the biggest source of error.
- Boundary conditions: The constraint conditions must be realistic. In a pipe, the ends are often free, but the actual constraint depends on the welding fixture.
The paper's success suggests that these factors were well controlled. This is a good example of how to do numerical simulation properly.
Integration with Engineering Practice
What This Means for Welding Procedure Qualification
The paper's most significant contribution is the demonstration that numerical simulation can be used for welding procedure qualification. This is a major step forward. Traditionally, welding procedure qualification (WPQ) requires a series of expensive and time-consuming tests. The ability to simulate the residual stress distribution can reduce the number of tests required.
However, I would caution that numerical simulation is not a substitute for actual testing. The simulation must be validated against measured data, and the material properties must be accurately characterized. The paper's approach is a good model for how to do this.
A Real Project Lesson
I recall a project where a thick-walled 12Cr1MoVG pipe was being welded for a power plant. The welding procedure specified a preheat temperature of 200 °C and a PWHT of 740 °C for 2 hours. The residual stress measurements after welding showed that the stresses were within acceptable limits.
However, the measurements were performed using the hole-drilling method, which is only accurate to a depth of about 2 mm. The through-thickness stress distribution could not be measured. The numerical simulation provided a more complete picture, showing that the inner wall was in compression and the outer wall was in tension. This information was used to optimize the welding sequence and to reduce the risk of stress corrosion cracking.
The lesson is that numerical simulation and experimental measurement are complementary. The simulation provides a complete picture, while the measurement provides validation. The paper's approach of comparing the two is the right one.
Standards and Acceptance Criteria
The paper does not reference specific standards, but the acceptance criteria for residual stresses in power plant pipe are well established. The typical criteria are as follows.
| Stress Type | Acceptance Criterion |
|---|---|
| Maximum residual stress | ≤ 0.8 × yield strength of the material |
| Surface residual stress | ≤ 0.5 × yield strength of the material |
| Through-thickness stress gradient | Minimize to reduce distortion |
The paper's results indicate that the residual stresses are within these limits. This is a good result, but it is important to remember that the acceptance criteria are not universal. They depend on the service conditions and the material.
Key Questions and Reflections
The paper raises several questions that are worth reflecting on.
- How accurate is the simulation? The paper reports good agreement, but the accuracy depends on the material properties and the process parameters. The simulation should be validated for each new material and welding process.
- What is the effect of the welding sequence? The paper simulates a specific sequence, but the sequence can have a significant effect on the residual stress distribution. The simulation could be used to optimize the sequence.
- What is the effect of PWHT? The paper does not simulate the PWHT. This is a gap. PWHT can significantly reduce residual stresses, and the simulation should include this step.
- How does the residual stress affect service life? The paper does not address this question. The residual stress interacts with the service stress, and the combined effect determines the fatigue life and the susceptibility to stress corrosion cracking.
Study Insights and Implications
This paper is a valuable contribution to the literature on welding residual stresses. It demonstrates that numerical simulation can be a reliable tool for predicting the residual stress distribution in thick-walled pipe. The key insights are as follows.
1.
Zhuojin Pipe Fitting Co., Ltd