In-service hot-tapping on pressurized pipelines poses significant engineering challenges—one of the most critical being the risk of burn-through during welding. While state-of-the-art commercial software and detailed experimental methods exist, our team chose to develop a cost-effective, simplified Finite Element Analysis (FEA) approach. In this post, I’ll walk you through our methodology, the underlying assumptions, the safety factors we employed, and the limitations we encountered. I’ll also compare our method with established techniques such as those from Battelle, PRCI, and EWI.
Understanding Burn-Through in Hot-Tap Welding
Burn-through occurs when the heat from welding softens the pipe wall so much that it can no longer contain the internal pressure, potentially leading to a catastrophic failure. Traditionally, guidelines like those from API set a critical inner-wall temperature (around 980 °C) beyond which burn-through is likely. However, the actual failure mechanism isn’t only thermal—mechanical stresses play a vital role as well.
Advanced methods often incorporate a full transient thermo-mechanical simulation. These approaches include detailed models of the moving heat source, phase change effects, and residual stresses. Although such methods provide high fidelity, they also require significant computational resources and specialized software licenses. Our aim was to develop a simplified approach that still captures the essential physics without incurring prohibitive costs.
Our Simplified FEA Approach: Key Components
Thermal and Mechanical Coupling
Our model is built around a coupled thermal-stress analysis. We first simulate the transient thermal profile through the pipe wall during the welding process. This thermal history is then used to evaluate the hoop stresses induced by the internal pressure. The critical comparison is made between these stresses and the temperature-dependent yield strength of the material. To maintain a conservative design, we applied a safety factor of 1.5 on the yield—a value that aligns well with pressure vessel codes, even though some standards might call for higher factors (such as tensile divided by 2.4 or 3.5).
Approximating the Moving Weld Heat Source
A significant challenge in welding simulation is accurately capturing the moving heat source. Instead of using a fully continuous model, we approximated the weld by discretizing it into sequential heat-input blocks. Each block represents a segment of the weld bead, and the heat input is applied as the welding arc passes over that segment. While this technique might yield slightly higher temperature predictions compared to a true moving heat source model, it provides a very good approximation for preliminary evaluations.
This block-based approach allowed us to simulate the weld process without the need for expensive software features like element birth and death or advanced moving source algorithms. The resulting thermal profile, though slightly conservative, still delivers the necessary insights into how heat propagates through the pipe wall.
Thermal Expansion Testing
Understanding thermal expansion is critical since temperature gradients induce thermal stresses, which—when combined with internal pressure—can contribute to failure. To validate our thermal model, we conducted a test that focused on the thermal expansion at the final time step of the simulation. We found that thermal stresses are largely self-limiting in nature. In other words, as the material expands due to heating, its capacity to further expand without yielding diminishes.
It is important to note that simply combining thermal expansion effects with internal pressure in an elastic analysis does not capture the full picture. Our approach acknowledges this limitation and treats the thermal expansion verification as a separate check, ensuring that the thermal stresses do not exacerbate the risk of burn-through beyond what our safety criteria predict.
Consideration of Phase Change Effects
Another important aspect is the influence of phase changes during welding. Our model assumes isotropic material behavior with temperature-dependent properties; however, it does not fully capture the complex phenomena associated with phase changes—such as latent heat absorption and microstructural transformations. In real welding conditions, phase changes can absorb significant amounts of heat, temporarily flattening the temperature rise, and subsequently affecting residual stresses after cooling.
Incorporating phase change effects would require a much more sophisticated model and additional material data, which would, in turn, increase the computational complexity and cost. For our purposes, we view this as an acceptable limitation in exchange for a simplified and cost-effective approximation.
How Our Approach Compares with Established Methods
Our simplified method is not developed in isolation. It aligns with several established industry practices and research methodologies, such as those from Battelle, PRCI, and EWI:
- Battelle’s Research: Battelle has been at the forefront of developing detailed empirical and simulation-based methods for predicting burn-through. Their techniques often include comprehensive modeling of moving heat sources and phase transformation effects. While our approach simplifies these aspects, it still fundamentally addresses the interplay between heat input and mechanical strength.
- PRCI and EWI Guidelines: These organizations provide practical guidelines and software tools that incorporate empirical correlations and advanced simulation techniques. They emphasize the importance of balancing the weld heat input with the pipe’s ability to dissipate heat, ensuring that inner wall temperatures remain below critical levels. Our work echoes this philosophy by coupling thermal profiles with stress analysis, albeit in a more simplified manner.
Our approach is intended as an approximation—ideal for preliminary design and feasibility studies where budget and time constraints preclude the use of full-scale simulations. By maintaining a conservative safety factor and carefully validating key aspects of the model, we ensure that our predictions, while not as detailed as those from advanced methods, provide meaningful insights into burn-through risks.
Limitations and Areas for Future Improvement
No model is perfect, and our approach has its limitations:
- Weld Heat Source Approximation: The discretization of the moving weld into blocks is a simplification. It tends to yield slightly higher temperature predictions compared to models that use a continuous moving source. This is acceptable for preliminary evaluations but may require refinement for critical applications.
- Thermal Expansion and Elastic Analysis: Our method of testing thermal expansion at the final time step highlights that thermal stresses are self-limiting. However, combining these effects with internal pressure using a simple elastic analysis does not fully capture the true behavior of the material during welding.
- Neglected Phase Change Effects: The current model does not include phase change phenomena, which can significantly affect the thermal response and residual stresses. Future iterations may incorporate these effects, either by adjusting material properties in the critical temperature range or by adopting more advanced simulation techniques.
Despite these limitations, our approach offers a valuable balance of cost, simplicity, and accuracy—providing engineers with a practical tool for early-stage evaluation of burn-through risk in in-service hot-tapping operations.
Conclusion
Our work represents a pragmatic, simplified FEA method for predicting burn-through during hot-tap welding. By integrating transient thermal analysis with mechanical stress evaluations and applying a conservative safety factor of 1.5 on yield, we achieve a method that aligns well with industry standards while remaining cost-effective.
This method, though an approximation, offers valuable insights that can be further refined with advanced simulations or experimental validations. We see this as a strong starting point and are excited to iterate on it further, particularly in the areas of enhanced heat source modeling and phase change effects.