July 2025
Maintenance, Reliability and Inspection
Case study: Why did the tube sheet of the waste heat boiler deform?
This article analyzes the causes of a 19-mm protrusion of the tube sheet and the tube joint leaks that occurred within a sulfur recovery unit (SRU) waste heat boiler (WHB).
In a previous article in Hydrocarbon Processing,1 improvements aimed at extending the lifespan of tubes operating in a high-temperature acid gas environment in a sulfur recovery unit (SRU) waste heat boiler (WHB) were reviewed and successfully implemented. These best practices were applied to the No. 6 SRP WHB process in 2019. However, after approximately 3 yr of operation, unexpected leaks were detected in many tube joints.
This was particularly surprising, as the WHB had been specifically designed and modified with the goal of extending tube life. A temporary shutdown was initiated, and an internal inspection revealed that the central portion of the tube sheet had deformed and protruded by approximately 19 mm. Additionally, damage to multiple tube joints and several leak points were identified.
This article analyzes the causes of the 19-mm protrusion of the tube sheet and the tube joint leaks that occurred within this relatively short operational period. Two primary causes were considered:
- Insufficient design strength
- Equipment damage due to operational conditions exceeding design limits.
The first potential cause, insufficient design strength, was ruled out because the equipment was verified during its 2019 manufacturing according to the American Society of Mechanical Engineers (ASME) code, with a design safety margin exceeding 10%. Therefore, this study primarily focuses on the second cause: exceeding design conditions during operation. Specifically, the authors investigate the possibility of the tube sheet's temperature exceeding its design limits.
To address this, plausible hypotheses have been established, and computational fluid dynamics (CFD) analyses were conducted to assess temperature increases and the resulting deformation. The results were then compared to the actual measured deformation to validate the hypotheses. Based on these findings, practical recommendations are proposed to prevent similar incidents in the future.
WHB inspection history. The WHB was newly manufactured and installed in 2019. During this time, hexagonal ferrules were also purchased and installed. However, it was discovered that the ferrules could not be inserted due to tolerance issues with the inner diameter of the WHB tubes. Specifically, the external diameter of the ferrules was too large for proper installation (FIG. 1).
FIG. 1. Ferrule installation.
The only feasible solution during the turnaround period was to machine the outer diameter of the ceramic ferrules, remove all sandpaper wrapped around them and reduce the thickness of only the first 100-mm section to forcibly insert them.
Ferrules are originally designed to be smoothly inserted while maintaining their specified thickness. They are installed to mitigate thermal shock on tube welds and the tube itself caused by high-temperature process gases. However, it appears that these functions were significantly compromised due to the alterations made during installation.
In the 2020 turnaround inspection, 1 yr after installation, damage to the necks of two ferrules were discovered and the ferrules were replaced. In the 2021 turnaround inspection, six damaged ferrules were replaced. Due to the high cost of replacing ferrules, comprehensive replacement is not performed unless visible damage is observed, leaving the internal tube sheet condition and tube weld joint corrosion uninspected.
By the 2022 turnaround inspection, 3 yr after the initial ferrule installation, significant damage was observed, including:
- Damage to multiple ferrule necks (FIG. 2)
- Extensive corrosion and damage at many tube weld joints (FIG. 3)
- Protrusion of the tube sheet by approximately 19 mm.
FIG. 2. A damaged ferrule with multiple ferrule necks broken, potentially enabling a high-temperature process gas to infiltrate.
FIG. 3. An image of tube weld joint damage, showing corrosion and fractures at multiple tube joints that caused leaks of high-pressure boiler feed water from the shell side during operation.
During the turnaround period, many damaged tube joints were repaired through welding. However, in March 2025, a significant water leak occurred during operation, necessitating an emergency process shutdown. The equipment was urgently replaced during that year's scheduled maintenance period (FIG. 4).
FIG. 4. WHB inspection results.
Failure scenario. During the October 2019 turnaround, ferrules were arbitrarily machined and installed. This machining reduced the neck thickness of the ferrules, weakening their strength. Additionally, the sandpaper at the ferrule neck area was removed, diminishing its ability to absorb thermal shocks. Furthermore, it is presumed that there was no gap between the ferrule and the inner diameter of the tube, making sliding during operation impossible.
Between 2020 and 2022, multiple ferrule necks fractured during operation. Through these cracks, high-temperature gas is presumed to have infiltrated and expanded to the tube joints and the front surface of the tube sheet. Eventually, the increased temperature of the tube sheet likely caused rapid corrosion and deformation of the weld joints, ultimately leading to tube fractures.
In other words, damage to the ferrule necks likely caused ferrule fragments to obstruct the flow of hot gas on the tube side, while also exposing the tube welds and tube sheet—components that should have been protected by the ferrule system—to elevated temperatures. This likely led to a significant rise in temperature in these areas. When the carbon-steel weld joints are exposed to high-temperature acid gas containing hydrogen sulfide (H2S), corrosion can progress rapidly, resulting in damage to the weld joints. Additionally, it is presumed that the high temperatures could have caused increased bending deformation of the tube sheet (FIG. 5).
FIG. 5. Failure scenario.
The following sections analyze the temperature changes and deformation of the tube sheet caused by the fractures in the ferrule neck area. Additionally, they examine whether the deformation of the tube sheet is further exacerbated when the tube weld joints detach, thereby weakening the stay function that supports the tube sheet.
Analysis results by scenario. It is generally known that damage to the front of the ferrule allows hot gas to penetrate the tube sheet. However, inspection results revealed that there were almost no gaps between the hexagonal ferrules. Instead, damage was observed at the ferrule necks, and this scenario was incorporated to account for the penetration of hot gas through the damaged ferrule necks.
Additionally, a gap of 3 mm–5 mm between the tube sheet and the ceramic board was considered. Finally, scenarios involving the detachment of tube joints were also explored.
Step 1: Analysis of ferrule neck damage cases. Ten cases of ferrule neck damage were considered, and CFD simulations were conducted to calculate the temperature changes of the tube sheet and the resulting deformation for each case. At this stage, only hot gas contact through Route 1 (ferrule neck damage) was considered (FIG. 6). The results showed a maximum temperature of 473.8°C and a minimum temperature of 353.8°C, depending on the gap location and size. Based on the inspection results from the 2022 turnaround, Cases 7–10 were selected for further analysis.
FIG. 6. Ferrule neck damage cases.
The deformation values were analyzed to range from a maximum of 12.24 mm to a minimum of 10.62 mm (TABLE 1 and FIG. 7). Notably, the actual measured deformation was 19 mm, showing a significant discrepancy between the simulated and actual deformation values.
FIG. 7. CFD analysis results for Step 1 (Cases 7–10).
Step 2: Analysis of the gap between the tube sheet and ceramic board. This step involves adding the occurrence of a gap between the tube sheet and the ceramic board to the analysis. This corresponds to Route 2 in FIG. 6. While a 25 mm-thick ceramic board is installed on the tube sheet surface, operational conditions—such as the expansion and contraction of the tube sheet—could result in the formation of gaps. For this specific equipment, it was determined that the significant deformation of the tube sheet made the likelihood of gap formation sufficiently plausible.
Both 3-mm and 5-mm gaps were analyzed, and the 5-mm gap was ultimately selected for further consideration. In Case 7, the presence of a 5-mm gap resulted in an additional temperature rise of approximately 0.46°C–0.84°C (TABLE 2). The CFD analysis results for this scenario are presented in FIG. 8.
FIG. 8. CFD analysis results for Step 2 (Cases 7–10).
Step 3: Analysis of tube joint detachment. The third step incorporated the detachment of tube joints into the analysis. Based on the 2022 turnaround inspection results, damage was observed in 56 tube weld joints. For Case 7, models were created to progressively increase the number of detached tube joints (Models A–D, as shown in FIG. 9). These models were analyzed by combining them with the scenarios from Step 1 (ferrule neck damage) and Step 2 (ferrule neck damage + ceramic board gap).
FIG. 9. The tube damage locations and quantities based on the 2022 turnaround inspection results.
Ferrule neck damage + tube joint detachment. The maximum tube sheet deformation was calculated as 14.66 mm, representing an additional deformation of 2.42 mm compared to cases with only ferrule damage (TABLE 3 and FIG. 10).
FIG. 10. CFD analysis results for Case 7.
Ferrule neck damage + ceramic board gap + tube joint detachment
When ceramic board gaps and tube joint detachment were included, the maximum tube sheet deformation increased to 15.94 mm, which is 2.7 mm greater than cases without tube joint detachment (TABLE 4 and FIG. 11).
FIG. 11. CFD analysis results for Case 7.
As described above, the damage scenario was established based on the inspection results of the damaged equipment, and the deformation of the tube sheet was analyzed step by step using CFD.
The analysis results indicated that the tube sheet ultimately deformed by approximately 16 mm. This is about 3 mm less than the actual measured deformation value of 19 mm. The discrepancy is likely attributable to differences between the analysis modeling and the actual operational environment. While the values do not match perfectly, the analysis results are sufficiently close to the actual measurements, and it is believed that the objective of understanding the causes of tube sheet deformation over 3 yr of operation has been achieved.
Recommendations. This article has identified multiple interrelated factors contributing to the 19-mm deformation of the tube sheet over 3 yr of operation, including improperly machined ferrule installation, ferrule neck fractures during operation, gaps between the tube sheet and the ceramic board, hot gas infiltration, tube corrosion and deformation, and tube joint failure. Based on these findings, the following recommendations are proposed to prevent similar incidents in the future:
- Improved ferrule design and installation: Strengthen the design of ferrule neck thickness and sandpaper to maintain functionality. Conduct precise measurements of tube inner diameters during procurement to ensure a proper fit without the need for machining.
- Gap monitoring and control: Ensure high-quality installation to minimize gaps between the ceramic board and the tube sheet. Consider alternative designs to reduce gap formation caused by thermal expansion and contraction.
- Regular inspection of tube joints: Periodically remove selected ferrules to inspect the condition of tube weld joints for corrosion. Perform proactive repair or reinforcement before significant damage occurs.
LITERATURE CITED
1 Lee, S. and E. Kim, “Case study: Extend tube life in a waste heat boiler,” Hydrocarbon Processing, March 2020.
The Authors
Sang-Mo Lee has more than 26 yr of experience as a Fixed-Equipment Engineer at SK Energy. His main tasks include root cause analysis (RCA), troubleshooting, fitness-for-service (FFS) evaluation, maintenance procedures and technical support for the refining and petrochemical business. He earned a BS degree in mechanical engineering, and is qualified as a professional engineer for welding and metallurgy by the South Korean government.
Comments