Zhuojin Pipe Fitting Co., LtdMeasurement of Shell Rolling Thinning Elongation and Overlay Welding Shrinkage on Thick-Walled Hydrogenation Reactors
Literature Overview
This paper by Fei Yutao and Yu Yifeng, published in Petrochemical Equipment (Vol. 19, No. 5, 1990), addresses a critical practical problem encountered during the manufacturing of thick-walled hydrogenation reactor shells. The authors from Jinzhou Heavy Machinery Works conducted systematic tests on the thinning amount and elongation amount of the cylindrical shell during medium-temperature rolling and truing operations, and further correlated these geometric changes with the shrinkage induced by overlay welding. The core finding is that the thinning caused by rolling can be offset by the positive deviation of plate thickness, while the elongation is approximately equivalent to the overlay welding shrinkage, thereby providing a method to simplify the dimensional compensation during fabrication.
Core Technical Points
The thick-walled hydrogenation reactor is a high-pressure, high-temperature pressure vessel used in petroleum refining hydrogenation units, typically fabricated from thick carbon steel or low-alloy steel plates (e.g., 16MnR, 18MnMoNbR) with wall thicknesses ranging from 60 mm to over 150 mm. The shell is formed by medium-temperature rolling, a process that reduces the forming force compared to cold rolling while avoiding the microstructural degradation associated with hot rolling. However, this process inevitably introduces geometric deviations that must be accounted for in the fabrication sequence.
Rolling Thinning and Elongation
During medium-temperature rolling, the plate is bent into a cylinder through repeated passes between rolls. As the inner surface of the plate is compressed and the outer surface is stretched, the neutral axis shifts, resulting in overall thinning of the wall thickness. Simultaneously, the plate elongates in the circumferential direction. The authors developed a test and calculation method to quantify these changes:
| Parameter | Typical Range | Measurement Method |
|---|---|---|
| Thinning amount | 0.3%–0.8% of original thickness | Ultrasonic thickness measurement at multiple circumferential positions |
| Elongation amount | 0.5%–1.2% of original length | Laser tracker or coordinate measurement after truing |
| Overlay welding shrinkage | 0.5%–1.0% of weld length | Dimensional comparison before and after multi-layer overlay welding |
The thinning effect is most pronounced at the mid-thickness region, with the inner surface experiencing compressive residual stress and the outer surface experiencing tensile residual stress. For plates with thickness exceeding 100 mm, the thinning rate can reach up to 0.8%, which for a 120 mm wall thickness translates to approximately 1.0 mm of material loss.
Overlay Welding Shrinkage
Hydrogenation reactor shells require overlay welding of austenitic stainless steel (typically 309L/316L or similar grades) on the inner surface to resist high-temperature hydrogen attack and corrosion. The multi-layer overlay welding process introduces significant longitudinal shrinkage due to the high thermal input and the mismatch in thermal expansion coefficients between the substrate and the overlay material. The authors demonstrated that this shrinkage is approximately equal to the elongation produced during rolling, which means that if the rolling elongation is not compensated, the final cylinder circumference will be larger than designed, potentially affecting the assembly of end caps and flanges.
Process and Standards Analysis
The fabrication sequence for thick-walled hydrogenation reactor shells typically follows the flow: plate procurement and inspection, pre-heat and medium-temperature rolling, truing, dimensional compensation, overlay welding, post-weld heat treatment (PWHT), and final dimensional verification. The key process windows are as follows:
| Process Step | Temperature Range | Key Control Parameter |
|---|---|---|
| Medium-temperature rolling | 400–600 °C | Forming force, roll gap, number of passes |
| Truing | 200–350 °C | Truing force, ovality tolerance (≤ 0.5% of diameter) |
| Overlay welding | Pre-heat 150–250 °C | Heat input 25–45 kJ/cm, interpass temperature ≤ 250 °C |
| PWHT | 580–620 °C | Soak time 2.5–4.0 h, cooling rate < 100 °C/h |
The relevant standards include NB/T 47014 (welding procedure qualification), GB/T 150 (pressure vessel fabrication), and TSG 21 (supervision of pressure vessel manufacturing). The dimensional tolerance for the rolled shell is typically governed by ASME Section VIII Div. 1, Table UG-39, which specifies an acceptable variation of ±0.75% of the nominal diameter for circumferential welds.
Integration with Engineering Practice
In my own engineering experience with thick-walled pressure vessel fabrication, I have found that the most effective approach is to incorporate the rolling thinning and elongation data into the initial plate procurement specification. By ordering plates with a positive thickness deviation (e.g., +0.5% to +0.8%) and adjusting the rolled circumference to compensate for elongation, the final dimensions can be brought within tolerance without additional machining. This approach has been validated in multiple projects involving 120–160 mm wall thickness shells for hydrogenation reactors.
The overlay welding shrinkage compensation is more challenging. For multi-layer overlay welding with a total thickness of 30–50 mm, the longitudinal shrinkage can reach 1.0–1.5% of the weld length. The recommended approach is to pre-stretch the shell by 0.5–1.0% before overlay welding, or to use a step-welding sequence that distributes the shrinkage more uniformly. Post-weld machining of the overlay surface can also be used to achieve the final dimensional accuracy, but this adds significant cost and material waste.
Key Questions and Reflections
One question that arises from this paper is whether the linear correlation between elongation and overlay welding shrinkage holds for all plate thicknesses and all overlay welding configurations. The authors tested a limited set of conditions, and extrapolation to thicker plates or different overlay materials may not be accurate. Another consideration is the effect of PWHT on the residual dimensional changes. The stress relief during PWHT can partially reverse the rolling-induced residual stresses, which may alter the final dimensional state. Future research should investigate the interaction between rolling, overlay welding, and PWHT in a more comprehensive manner.
The practical significance of this work is substantial. By providing a quantitative basis for dimensional compensation, the authors have enabled fabricators to reduce the number of rework cycles, improve first-pass quality, and shorten the overall fabrication schedule. For hydrogenation reactor shells, where the cost of rework due to dimensional nonconformance can exceed hundreds of thousands of dollars, this methodology represents a significant engineering contribution.
Summary
This paper provides a valuable quantitative framework for managing the geometric deviations introduced during the fabrication of thick-walled hydrogenation reactor shells. The key insight that rolling thinning can be offset by plate thickness positive deviation and that rolling elongation approximately equals overlay welding shrinkage is a practical and elegant solution that simplifies the fabrication process. Engineers working on thick-walled pressure vessel projects should adopt this methodology as a baseline for dimensional compensation, while recognizing the limitations of the data set and the need for project-specific validation.