Failure of a boiler tube: an investigation into the cause and preventions
R-TECH received a section of rear water wall tubes, extracted from a boiler onboard of an LNG tanker. During operation the vessel crew experienced limitations in steam production when attempting to operate the main turbine at high load; a significant amount of feed water consumption was also noted. An on-board inspection subsequently found evidence of considerable water ingress about the manipulated tubes immediate to retractable soot blower lance sleeve, situated at the rear of the superheater arcade.?The leakage was accompanied also by damage to the surrounding monolithic refractory lining.
In the as-received condition (Figure 1) the tube section was noted to comprise of four tubes, joined by way of a welded membrane. At the point of termination for the membrane, and either side of a welded joint, pinholes were observed about the central two tubes. This region was understood to be immediate to the point of entry for the soot blower lance sleeve, allowed for by the manipulation of the waterwall tubes. Three of the pinholes were located on a single tube, with one pinhole located in the tube immediately adjacent and another one located in the inter-tube membrane.
The pinholes were associated with metal loss, the morphology of which was smooth and exhibited an undulating surface topography, almost scalloped in nature (Figure 2 to 4). These scalloped regions also exhibited a corrosion deposit which was orange/brown to dark grey. A blueish deposit was also noted about the outermost scallop of one of the pinholes whilst traces of a white corrosion product were present about the circumference of both pinholes. Further, a pockmarked morphology was noted about the region immediate to the pinholes.
Metallographic examination of one of the pinholes was associated with a shallow gradient of wall loss. See figure 5. Moreover, the external surface was noted to possess a coarse and irregularly undulating topography, with instances of shallow pockmarks as well as a consistent layer of grey corrosion product. Figures 6 and 7. The general microstructure consisted of ferrite and pearlite with limited degradation, see Figure 8.
Analysis of the deposit revealed the presence of magnetite, hematite, and iron sulphate hydrate. Magnetite is a protective grey iron-oxide, formed by the direct reaction between the metal surface and moisture within the atmosphere. In contrast, hematite is a non-protective red iron-oxide formed in oxygen-rich concentrations and can act as binder species; allowing for other phases to accumulate within the deposit. This can be deleterious as it has the potential to allow for the development of a differential corrosion cell. Iron sulphate hydrate is produced when concentrated sulphuric acid forms on steel surfaces in strongly acidic conditions (pH<4) and reacts with the underlying steel. Dissolution is possible whenever metal (and not the flue gas) temperatures are less than the sulphuric acid dewpoint.??
Based on the evidence it is suggested that the waterwall tubes had failed due to the accelerated deterioration of the underlying steel, and ultimately final failure by way of over pressurisation caused by critical wall losses. Moreover, it is postulated that such wall losses are attributable to the dissolution of the steel by sulphuric acid dewpoint corrosion.
领英推荐
During operation, the introduction of sulphur can occur during incomplete fuel combustion, resulting in the generation of sulphur oxides (and trioxides, SO3) and the deposition of volatile sulphur compounds. High concentrations of sulphur trioxide render the tubes susceptible to sulphuric acid corrosion when flue gas or metal temperatures are equal to or less than the sulphuric acid dewpoint. At this temperature sulphur trioxide and water vapour (moisture) react to form vapour-phase sulphuric acid, which can subsequently condense onto the tube surface, producing dilute liquid sulphuric acid. Dilute sulphuric acid (5 to 40%) is more corrosive than higher concentrations and as the water dew point is approached, acid concentrations in this range are likely. This process results in the dissolution of the underlying steel, as well as the formation of hydrogen gas and a protective layer of FeSO4 (iron sulphate). The associated damage is typically characterised by either general material wastage or through thickness defects.
Accordingly, the critical factors governing sulphuric acid dew point corrosion include the presence of sulphur trioxide, the presence of moisture, and the presence of metal whose surface temperature is below the sulphuric acid dew point. This corrosion mechanism is normally termed ‘cold-end corrosion’ since it generally affects the cooler regions of the system. The hydrated nature of the corrosion products suggests that significant levels of moisture were present in the environment. The dew point increases with sulphur trioxide content and moisture content. It is suggested that water ingress may have occurred by way of the retractable soot blower lance sleeve. Moreover, the temperatures at which sulphuric acid first condenses varies between 116 and 166°C or higher, depending on the concentration of both sulphur trioxide and water vapour present. It is postulated that such an environment could exist beneath the refractory lining, where metal temperatures are likely to be lower than that of the flue gas. Sulphuric acid dewpoint corrosion is however generally considered a problem during periods of idleness such as shutdowns, because as the boiler cools the temperature of the external surface may drop below the dewpoint, allowing moisture and therein sulfuric acid to form on tube surfaces.
Some of the sulphur compounds and pyrosulphates that deposit on tube surfaces during service can also be considerably active, owing to low melting points (as low as 371oC, relative to the concentration of SO3). As temperature increases, the amount of SO3 required to form a liquid phase increases significantly. Therefore, sulphur compounds and pyrosulphates tend to only be molten on relatively cool surfaces such as water-wall tubes. The resultant molten sulphates react with and thereby dissolve the protective iron oxide on the tube external. This process enhances the transport of oxygen to the surface to re-form the iron oxide at the expense of the tube wall. Hence the presence of oxygen rich hematite at the external surfaces. This corrosion process can result in the accelerated deterioration of the underlying metal during service and is characterised by metal loss and the development of a dark grey, which forms as the liquid sulphates sinter corrosion debris to the surface. The severity of this corrosion process is also dependent upon the concentration of SO3 within the fuel, as well as exact metal temperature.
The following measures can be implemented to reduce the likelihood of sulphuric acid dew point corrosion:
1) Controlling the corrosive quantities of sulphur trioxide within the flue gas:
2) Controlling the level of moisture in the flue gas:
3) Inspection and monitoring: regular and thorough visual inspection of retractable soot blower system, with particular attention to evidence of water ingress/leakage.
4) Inspection and monitoring: where possible regular ultrasonic thickness surveys of areas susceptible to dewpoint corrosion.
5) Maintain metal surface temperatures above that of sulfuric acid dewpoint corrosion.
6) Weathering steels, such as COR-TEN and S-TEN, have been reported to be highly resistant to sulphuric acid dew point corrosion in comparison to low carbon steels and the majority of stainless steels.
BOAS Trainer and Assessor
1 年Hi Sarah. I may have missed it in the document but was the boiler being fired on LNG boil off or Oil. I suspect the later which has a higher Sulphur content although I am lead to believe that Intermediate fuel oils should only have <1% sulphur these days, however that’s in the UK. Foreign fuels may contain more but this should be identified in the ships documentation from her last bunkering. As you say it is important to keep the exhaust gas temperatures above the sulphuric acid dew point
Asset Integrity Management | PSM | Risk Engineering | Operations Assurance
1 年Thanks for sharing. All too common I am afraid. Did you investigate the chemical injection and water compositions? Normally the basics are not conducted correctly on these units.
Inspection Engineer at Axiom.
1 年Thanks for sharing, I’m assuming D type marine boiler, imo it’s worth considering in leakage of soot blower in operation also and will be seen as wallbox refractory damage and guide tube erosion pitting prior to wall tube failure. Check them as part of gas path inspections.
Trainer at Wilkinson Coutts Engineering Training Ltd
1 年Was quite common on D-type roof-fired steam ship boilers ...keeping the back-end temp above 130 degC was always recommended.Power station boilers with FGD often had to line back-end components with C276 owing to the FGD bringing the gas temp down below dewpoint. Pinholing near sootblowers was another common occurrence