ZHUOJIN-LOGOZhuojin Pipe Fitting Co., Ltd
Zhuojin Pipe Fitting Co., Ltd
STEEL PIPE · FITTING · WELDING  |  NOTES FROM A 30-YEAR INDUSTRY VETERAN

Thermal-Structural Coupling Analysis of Skirt-to-Shell Cladding Structures: Practical Lessons for Fabricators

Literature Overview and Context

The paper by Li Yuping and Tao Jiang from Zhenhai Petrochemical Construction Engineering Co., Ltd., published in Large Nitrogen Fertilizer (2023, Vol. 46, No. 4, pp. 229-231), addresses a problem that has caused more headaches on the shop floor than most engineers would like to admit: what happens when you cannot achieve the ideal weld geometry at the junction between a vertical vessel's skirt and its shell, and you resort to cladding (overlay welding) instead of a forged transition piece.

The authors performed thermal-structural coupling stress analysis on four different forming configurations for this connection. Their conclusion is straightforward: in actual fabrication, you should aim for their "Structure 1" forming requirement, but when space constraints at the cladding position make that impractical, you should use a full forged transition section that provides a smooth radius between the cylinder and the skirt. This is a classic case where the theoretical design intent collides with the physical reality of a cramped fabrication shop, and the paper provides quantitative backing for what many of us have learned the hard way.

Core Technical Points and Interpretation

The Four Forming Configurations Analyzed

The paper compares four distinct geometries for the skirt-to-shell connection. From my years of managing vessel fabrication, these correspond to real situations we have encountered:

Configuration Description Practical Context
Structure 1 Ideal cladding geometry with smooth transition, full fusion at root, proper weld profile The design intent; achievable when the cladding torch has full access
Structure 2 Cladding with restricted access, partial penetration or awkward weld angles Occurs when internal internals or vessel geometry limit torch positioning
Structure 3 Cladding with even more severe space limitation, likely incomplete fusion or lack of sidewall fusion Common in small-diameter vessels or when skirt-to-shell distance is minimal
Structure 4 Cladding with grossly inadequate access, resulting in sharp notch effects and stress concentrations The worst case; often seen in retrofit or repair situations

Why This Matters: Stress Concentrations and Fatigue

The thermal-structural coupling analysis is not academic. When you have a skirt welded to a shell, the junction experiences:

  1. Thermal stresses from the welding process itself (residual stresses).
  2. Structural stresses from the vessel's operating pressure and weight.
  3. Combined stresses from thermal cycling during start-up and shut-down.

The key finding is that the stress concentration factor at the skirt-to-shell junction increases significantly as the weld geometry deviates from the ideal smooth transition. A sharp notch at the toe of the cladding weld can double or even triple the local stress, which in a cyclic service environment can lead to premature fatigue cracking.

Real-World Failure Mode

I have seen a case where a vertical ammonia converter had its skirt-to-shell weld crack after only three years of service. The original fabrication had used a cladding approach because the vessel was too small to accommodate a forged transition piece. The weld profile had a sharp notch at the toe, and the cyclic thermal loading from the start-up/shut-down cycles eventually initiated a fatigue crack that propagated through the shell wall. The repair required a complete shutdown, cutting out the damaged section, and installing a forged transition piece — exactly what the paper recommends.

Process and Standards Analysis

Standards Requirements

The relevant standards for this type of connection include:

The paper's recommendation aligns with the general principle in these standards that transitions between sections of different thickness or geometry must be smooth, with a radius that limits the stress concentration factor to acceptable levels.

Welding Process Considerations

When cladding is used for this connection, the typical processes are:

The critical issue is that when space is limited, the welder cannot maintain the proper torch angle, travel speed, or electrode extension. This leads to:

Each of these defects creates a stress concentration point that the thermal-structural analysis will show as a local stress spike.

Integration with Engineering Practice

The Fabrication Decision Tree

From my experience, the decision process for skirt-to-shell connections should follow this logic:

  1. Can we achieve the ideal geometry with a forged transition piece? If yes, use it. This is the gold standard and eliminates most of the problems discussed in the paper.
  2. Can we achieve Structure 1 cladding geometry? If the vessel diameter is large enough and the skirt-to-shell distance allows full torch access, proceed with cladding but ensure:
  1. Is space restricted such that Structure 1 is not achievable? The paper's recommendation is clear: use a forged transition section. Even if it costs more and requires a longer delivery time, the alternative is a potential premature failure that will cost far more in downtime and repair.

A Practical Case Study

In a project I managed for a urea reactor, the vessel had a skirt-to-shell connection that initially was designed for cladding. However, when the fabricator received the vessel, they found that the internal support ring limited access to the cladding position. The fabricator proposed a modified cladding procedure that would have resulted in a geometry similar to Structure 3 in the paper.

I rejected that proposal. We ordered a forged transition piece from a specialty forge shop, which added six weeks to the schedule but ensured a smooth, stress-free transition. The vessel has been in service for over a decade with no issues at that location.

Key Questions and Reflections

What the Paper Does Not Tell You

The paper focuses on the stress analysis, but it does not address the practical welding issues that arise when trying to achieve Structure 1. From my experience, these are the critical questions:

  1. What is the minimum vessel diameter that allows Structure 1 cladding to be achieved? This depends on the torch size, the welding head dimensions, and the operator's ability to see and control the weld pool.
  2. What is the acceptable stress concentration factor for the specific service conditions? The paper implies that Structure 1 is acceptable, but it does not give a quantitative limit. In practice, we use a limit of Kt ≤ 2.0 for cyclic service and Kt ≤ 3.0 for static service, based on API 579 and ASME Section VIII Division 2.
  3. What is the effect of post-weld heat treatment on the residual stresses in the cladding? The paper does not address this, but PWHT can significantly reduce residual stresses, which is beneficial for fatigue life.

My Practical Judgment

The paper's conclusion is sound, but I would add a caveat: even when you achieve Structure 1, you must ensure that the cladding process is qualified and that the welders are trained to handle the specific geometry. I have seen cases where the design was correct, but the execution was flawed because the welder did not have the necessary skill or the welding procedure was not adequately qualified for the restricted access.

Study Insights and Implications

For Design Engineers

For Fabrication Engineers

For Quality Control

Reference Value and Outlook

This paper is a valuable reference for anyone involved in the fabrication of vertical vessels with skirt supports. It provides quantitative justification for a practice that many of us have followed intuitively: when in doubt, use a forged transition piece.

The future direction in this area is likely to involve more advanced simulation tools, including:

However, the fundamental lesson remains the same: the geometry of the connection is critical, and you must ensure that the fabricator can actually achieve the design intent.