Root Cause Failure Analysis of Steam Condensate Cooler
Introduction
Petromaster Ltd. was engaged to perform root cause failure analysis of two leaked tubes from steam condensate cooler of demineralizer package. The vertical condenser was located downstream of Vacuum Degasifier and played critical role in supporting vacuum created by ejectors by condensing mixture of steam and non-condensable gases in the tube side. The shell side of the condenser contained cooling water.
Technical Investigation
Visual Examination
The visual examination displayed visible circumferential cracks at the tube-sheet end for one of the tube samples, while other tube did not reveal clear signs of cracking at the external surface. The cracking/corrosion damage was concentrated at the transition zone of expanded tube section. The damage was located at ? inch from the tube end which was inside tube-sheet. The external surface of both tubes presented shinny appearance and no signs of localized deposit accumulation were observed. The internal appearance of the as-received samples displayed layer of brown corrosion scale at the transition zone, while the non-expanded section was free from corrosion scale or internal deposition. Under-deposit corrosion in form of pitting and cracking was evident.
Dimensional Examination
Dimensional examination was performed by measuring the tube thickness and out diameters in expanded and non-expanded region and comparing with design specifications. The average measured thickness in expanded section was less than design value of 1.245 mm. Similarly, the OD in expanded section was almost 0.5mm higher than design value of 19.05 mm. TEMA allows 8-10% of thinning because of expansion for non-ferrous and stainless-steel tubes. However, the observed thinning in expanded region was in the range of 20-22% of the non-expanded tube thickness. Moreover, any thickness reduction from uniform corrosion has not been accounted.
领英推荐
Metallographic Examination
The metallographic analysis was performed on the transverse cross-section. The samples were polished and etched with glycergia etchant. The micrographs revealed multiple corrosion pits at the internal side of the tube and the widest corrosion pit was measured by image analysis software to be around 939.3 um wide. The through thickness crack was observed to have initiated from internal pitting damage. Multiple fine cracks were also observed adjacent to the wide crack. All the fine cracks were observed to have initiated from internal pits and followed purely trans-granular path towards OD. The cracks exhibited significant branching along their path displayed typical characteristics of stress corrosion cracking (SCC). Almost all the cracks initiated from the internal side from the corrosion pits.
Scanning Electron Microscope (SEM)
The internal surface of both tube samples was analysed under Scanning Electron Microscopy at the transition zone where the deposit accumulation and corrosion cracking were observed. The analysis displayed local accumulation of thick deposits and highly branched cracks emanating beneath these deposits. The local accumulation of deposits had led to under-deposit corrosion in form of circular and elliptical pits which provided crack initiation sites. The EDS analysis revealed that the internal deposits at the transition area predominantly contained Iron Oxide due to high percentages of Fe (23-43wt%) and O (25-36wt%). Moreover, high percentages of carbon (C) were also observed in these deposits varying from (16-24wt%). The source of high traces of carbon can be attributed to the non-condensable gases in the process stream which would have accumulated at transition zone after mixing with steam at upstream steam ejectors. The other notable element was wt.% of chlorides which varied from (0.24-0.72wt%) equivalent to (2400-7200ppm) which is a significant amount for causing localized pitting and stress corrosion cracking in austenitic stainless steels. Stress corrosion cracking in SS 316L with 0.1-1 wt.% can take place even at 50oC and the rate of this damage will be exacerbated at higher temperatures.
In comparison to this, the unaffected region revealed predominantly iron oxide scale without any significant percentage of Chloride, Sulphur, or any other detrimental elements.
DAMAGE MECHANISM
Based on above findings, it can be concluded that tube leakage resulted due to chloride pitting and chloride induced stress corrosion cracking. The local accumulation of the process deposits at geometrical change at the transition zone led to under-deposit corrosion and caused local concentration of chlorides at that area. Due to higher stress concentration at the transition zone from tube expansion, the area became highly susceptible to Chloride Stress Corrosion Cracking.
Conclusion
Based on the above discussion, the root cause of the tube failure is attributed to chloride induced stress corrosion cracking (ClSCC). The accumulation of the process deposits and chlorides at the transition area led to chloride pitting which assisted in the formation of stress corrosion cracks.