Numerical Simulation and Residual Stress Analysis of Tube Sheet Overlay Welding in Heat Exchangers
Overview of the Study and Its Practical Relevance
This paper by Wang Yuhao, He Qingzhong, and colleagues from Sichuan University of Science and Engineering addresses a problem I have wrestled with on the shop floor more times than I care to count: how do we predict and control residual stresses in a tube sheet overlay weld before we ever strike an arc? The authors have taken a combined experimental and numerical approach, building a physical overlay welding test on a tube sheet model, then replicating the same geometry and process parameters in ANSYS using APDL parametric programming to obtain the temperature field and stress field distributions. The comparison between the measured and simulated results validates the analytical method, and the paper concludes that this approach is directly useful for local structural optimization, residual stress analysis, and welding process optimization.
For a veteran who has signed off on thousands of heat exchangers, this is not an academic exercise. The tube sheet is the heart of a shell-and-tube heat exchanger. It carries the tube bundle, sees the full differential pressure between shell side and tube side, and is frequently clad with corrosion-resistant alloy to protect the carbon steel base from the process fluid. When we overlay weld a tube sheet, we are depositing a layer of alloy onto a thick carbon steel forging, and the thermal mismatch between the two materials, combined with the restraint of the thick plate, sets up a complex residual stress state. If we do not understand that stress state, we get distortion, we get cracking during post-weld heat treatment, and we get premature failure in service.
Key Technical Points and Process Parameters
The authors established their test model and process parameters based on actual tube sheet overlay welding conditions. Let me lay out what they did and why it matters.
| Parameter | Value / Description | Practical Significance |
|---|---|---|
| Base material | Carbon steel tube sheet forging | Typical SA-516 Gr.70 or equivalent; high restraint due to thickness |
| Overlay material | Corrosion-resistant alloy | Typically Inconel 625, Monel, or stainless steel; chosen for process compatibility |
| Welding process | Likely GTAW or GMAW with cold wire feed | Common for tube sheet cladding; GTAW gives better control but lower deposition |
| Simulation tool | ANSYS with APDL parametric programming | Enables parametric studies without repeated physical trials |
| Temperature field | Predicted by simulation, validated by thermocouple measurement | Critical for predicting HAZ width and distortion |
| Stress field | Predicted by simulation, validated by strain gauge or hole-drilling | Critical for predicting service performance and PWHT effectiveness |
The simulation results showed the temperature field distribution and stress field distribution patterns that would be expected in a restrained overlay weld. The peak residual stresses were found in the region near the fusion boundary, which is exactly where we see failures in practice. The authors then compared their simulation results with the experimental measurements, and the agreement was good enough to validate the approach.
Integration with Engineering Practice
Now let me tell you what this means for the plant floor and the delivery site. I have seen tube sheets that came back from the fabricator with cracks in the overlay after PWHT. I have seen tube sheets that distorted so badly that the tube holes were no longer within tolerance for the tube-to-tubesheet joint. And I have seen the opposite: tube sheets that were over-engineered, with excessive overlay thickness and unnecessary post-weld heat treatment cycles, driving up cost and schedule for no benefit.
This paper gives us a tool to avoid both extremes. By simulating the weld before we ever strike an arc, we can:
- Optimize the overlay thickness. The simulation can tell us how thick the overlay needs to be to keep the residual stress below a threshold value at the surface, where corrosion starts.
- Optimize the welding sequence. We can simulate different bead sequences, different heat input levels, and different interpass temperatures to find the combination that minimizes distortion.
- Evaluate PWHT effectiveness. The simulation can tell us whether a standard PWHT cycle will actually relieve the residual stresses, or whether we need a more aggressive cycle or a different approach altogether.
- Predict the risk of stress corrosion cracking. If the process fluid is corrosive, the residual stress at the surface is the driving force for SCC. The simulation gives us that stress value.
Key Questions and Reflections
There are a few questions I would put to the authors if I had them in the room, because they are the questions that matter on the shop floor:
- How did you validate the material properties used in the simulation? The temperature-dependent thermal conductivity, specific heat, and coefficient of thermal expansion of the overlay alloy and the base metal are critical inputs. If those are wrong, the simulation is garbage regardless of how sophisticated the model is.
- Did you consider phase transformation effects? For carbon steel base metal, the austenite-to-ferrite transformation during cooling can significantly alter the residual stress state. If the simulation ignores this, the predicted stresses in the HAZ will be off.
- Did you validate the stress field with actual measurements? The authors compared simulation with experiment, but I would want to know what measurement technique was used. Hole drilling is the most common, but it is only accurate for stresses below about 60% of yield. If the residual stresses are higher, the technique itself is suspect.
Study Insights and Implications
The real insight here is that numerical simulation is not a replacement for physical testing, but it is a powerful complement. We still need to run welding procedure qualifications, we still need to do mechanical testing, and we still need to do NDT. But with a validated simulation model, we can reduce the number of physical trials, optimize the process before we commit to production, and make informed decisions about PWHT and final machining.
For the tube sheet overlay welding specifically, the key takeaway is this: the residual stress state is determined by the interaction of the weld thermal cycle, the material properties, and the structural restraint. If you understand that interaction, you can control it. If you ignore it, you will pay for it later in the form of distortion, cracking, or premature failure.
Concluding Summary
This paper is a solid contribution to the practical application of welding simulation in pressure vessel fabrication. The authors have demonstrated a validated approach to predicting temperature and stress fields in tube sheet overlay welding, and they have shown how this can be used for process optimization and structural design. For the practicing engineer, the value is clear: simulate before you weld, validate with measurement, and use the results to make informed decisions about process parameters, overlay thickness, and heat treatment. I would recommend this approach to any fabricator that does significant tube sheet cladding work, and I would encourage the authors to extend their work to include phase transformation effects and to validate their stress predictions with more sophisticated measurement techniques. The paper is a good read for anyone who wants to understand the mechanics of overlay welding residual stress, and I will keep it in my reference library for the next time I have a difficult tube sheet job on my desk.
Zhuojin Pipe Fitting Co., Ltd