UDC 502 957 7525 Boiler Condition Assessments why and how?

Defining a Strategy for Boiler Condition Assessment "The underlying driver for performing component condition assessment is the need to manage component life to achieve plant operating safety, reliability and economic objectives". Condition assessment is the primary requirement for managing the useful life of plant boiler components, and it relies on the following research:

How long has the unit been in operation and under what conditions such as; Thermal cycling, Fuel types, Atmospheric changes, and operational changes. The degree of damage current in the component The rate of damage accumulation The degree of damage causing failure How many years of service are required Are there upgrades or replacements in the future Will the unit operational conditions remain the same as found in step one

To carry out a boiler condition assessment program successfully, Plant Personnel must have several key elements in place. These elements have been well-established by a successful BTF reduction program and are summarized as follows: Support of Management Cross-functional teaming (including maintenance, operations, and engineering personnel) in performing the program (Boiler Inspection Team) Attention to long-term solutions to root cause problems Training Documentation of results and periodic review

A Final Boiler Condition Assessment Document should include: Date the assessment was performed Summary of assessment activities, such as inspections, material tests, and results Estimate of Component Remaining Life and summary of basis Damage mitigation/prevention actions, if appropriate Follow-up inspections or monitoring actions and their timing, if appropriate Recommendations for next assessment, including operating changes/upsets that would prompt a re-assessment

Determination of assessment priorities The scope depends on factors such as the boiler design type, design temperatures and pressures, materials, fuels, age, unit history and future plans for the boiler or plant. We suggest a multi-level approach in the planning of the survey scope which is similar to the approach developed by the Electric Power Research Institute (EPRI) for utility boilers. Basic to all levels of the fitness survey is a compre hensive inspection by an experienced field Service Engineer. A first level survey depends on minimal if any testing or nondestructive examination (NDE). Level I Evaluate past operating and maintenance history. Provide means for monitoring conditions in the component area such as; Thermal probes, or flue gas analysis. Identify any critical components on basis of history, experience with similar boilers, and objectives for future of unit. Perform complete visual inspections of all accessible areas of the boiler and/ or auxiliaries - photo-document problem areas as needed. Identify the root cause of damage found to the limits of level I survey. Develop a final boiler fitness report with recommendations. Level II includes NDE testing with little material sampling although tube samples may be included.

Level II (in addition to items in Level I) Establish the outage inspection and testing plan based on information gathered during initial research and data collection of the unit. Define support requirements for the inspections including any materials that may be needed. Assess operation - may include hot walk down of boiler and piping, and data gathering to evaluate performance of auxiliary equipment. Implement inspection and testing plan which may include tube samples for basic condition assessment analysis. Perform preliminary life estimates and provide recommendations for immediate action as needed. Prepare for follow up operational testing as required/ planned. Estimate remaining life - analyze data and inspection results. Implement operational testing if required.

Level III surveys incorporate material testing, engineering studies and more extensive analysis in support of the assessment. Proceeding from a level I effort to a level III survey is dictated by the objectives of the project, i.e., the scope of data needed to predict future operation to the extent needed by the owner. Typical multilevel activities are as follows.

Level III (in addition to Levels I and II) Perform engineering analysis such as piping stress analysis, finite element analysis, and boiler performance/upgrade analysis. Remove materials for laboratory analysis, i.e. boat samples, tube materials for accelerated creep rupture tests, fatigue tests, etc. Perform specialized site testing such as strain measurements, support load testing, etc. After the conclusion of the condition assessment program, the plant owner utilizes the results in the planning for the plant - whether long range or short range. The information may simply aid in the planning of re-inspection and regular preventive maintenance to ensure reliable steam production. The assessment may be used to define the scope of a major upgrade or plant overhaul by determining what components need replacement.

Scope of work of Critical Systems and Components Described below are the types of problems found in the various components as well as the recommended NDE methods. In general, visual examination, the most basic of NDE, is done for all components. It is recommended to photo-document the inspection to provide a permanent record in the report. Internal inspections are frequently done by video probe and recorded on tape. 1.) Drums: The steam drum is the single most expensive component in the boiler. Consequently, any assessment program must address the steam drum as well as any other drums in the convection passes of the boiler. In general, problems in the drums are associated with corrosion. Problems in the drums normally lead to indications that are seen on the surfaces - either ID or OD. Assessment: Inspection and testing focuses on detecting surface indications. The suggested NDE method is Wet florescence magnetic particle method. Because WFMT uses fluorescent particles which are examined under ultraviolet light it is more sensitive than dry powder type MT and it is faster

than PT methods. WFMT should include the major welds, selected attachment welds and at least some of the ligaments. If locations of corrosion are found then ultrasonic thickness testing (UTT) may be performed to assess thinning due to metal loss. In rare instances metallographic replication may be performed. Replication is done by polishing the surface of the drum to a mirror finish, etching the polished surface with a nitric acid, and then lifting an image of the metal surface by applying a softened acetate tape (the replica). The procedure, analogous to finger printing, allows the metal grain structure to be examined under a microscope. 1A.) Unit #6 Steam Drum- Routine visual inspections have been conducted inspecting drum internals, and drum externals in both hot and cold conditions. Upon visual inspection numerous attachment cracks, loose bolts, moderate accumulation of magnetite, and slight caustic corrosion have been noted over the last two years. The cracks and surrounding areas were immediately Wet Magnetic Particle tested and all cracks and indications were removed with no reoccurring instances. In addition all loose items were resecured, all debris was removed and the slight corrosion is being monitored. Mud Drum- Routine visual inspections and documentation have also been conducted on the front and rear Mud Drums. The most critical of concerns are indications of circumferential cracks that are located in the downcomer to shell weld. The locations and size of these cracks have been documented and are planned for NDE during the 2001 inspection. Also planned for 2001 outage is the Go no Go testing of the inlet orifices of the front and rear water walls. As with the Steam Drum cracked attachment welds and loose items were corrected without further incidents and slight corrosion is being monitored. 1B.) Unit #7 Steam Drum- Routine visual inspections have been conducted inspecting drum internals, and drum externals in both hot and cold conditions. Upon visual inspection numerous attachment cracks, loose bolts, moderate accumulation of magnetite, and slight caustic corrosion have been noted over the last two years. The cracks and surrounding areas were immediately Wet Magnetic Particle tested and all cracks and indications were removed with no reoccurring instances. In addition all loose items were resecured, all debris was removed and the slight corrosion is being monitored. Mud Drum- Routine visual inspections and documentation have also been conducted on the front and rear Mud Drums. The front water wall orifices were gauge tested with no existing problems found the rear water wall orifices will be tested next scheduled outage. As with the Steam Drum

cracked attachment welds and loose items were corrected without further incidents and slight corrosion is being monitored. 2.) Headers: Boilers designed for temperatures at or above 1000 F can have superheater outlet headers that are subject to Creep, the plastic deformation (strain) of the header from long term exposure to temperature and stress. For high temperature headers, tests should include metallographic replication and ultrasonic shear wave inspections of higher stress weld locations. Lower temperature headers are subject to corrosion or possible erosion. Additionally, cycles of thermal expansion and mechanical loading may lead to fatigue damage. Assessment: Inspection should include testing of the welds by MT or WFMT. In addition, it is advisable to perform internal inspection with a video probe to assess waterside cleanliness, to note any buildup of deposits or maintenance debris that could obstruct flow, and to determine if corrosion is a problem. Inspected headers should include some of the water circuit headers as well as superheater headers. If a location of corrosion is seen then UTT to quantify remaining wall thickness is advisable.

2A.) Unit #6 Final Superheat Outlet- Case history has found that certain seamed headers have failed within Southern Co. system. Since this failure has been recognized we have identified our FSH header as a seamed header and have sandblasted and WMPT the seam to ensure no cracks were present. Radiant Reheat Inlet- Previously documented leaks located in the circumferential terminal tube weld to header combined with thermal fatigue cracks in the Radiant wall tubes led us to re-inspect the header for potential reoccurring problems. In addition to the documented problems repair procedures at the time were felt to be inadequate. Upon repairing the previous cracks WMPT was used to verify that the cracks were entirely removed which led to numerous additional repairs. The entire header was then sandblasted and WMPT leading to additional repairs on all four headers. Research showed that pressure excursions and start up/shut down procedures were the root cause of the problem. 2B.) Unit #7 Final Superheat Outlet- As with Unit #6 the FSH header was identified as a seamed header. The seam was located, sandblasted, and WMPT to ensure no cracks were present. Radiant Reheat Inlet- Unit #7 was a mirror image of Unit#6, documented leaks located in the circumferential terminal tube weld to header combined with thermal fatigue cracks in the Radiant wall tubes led us to reinspect the header for potential reoccurring problems. In addition to the documented problems repair procedures at the time were felt to be inadequate.

Upon repairing the previous cracks WMPT was used to verify that the cracks were entirely removed which led to numerous additional repairs. The entire header was then sandblasted and WMPT leading to additional repairs on all four headers. Research showed that pressure excursions and start up/shut down procedures were the root cause of the problem.
3.) Piping - Main Steam: For lower temperature systems the piping is subject to the same damage as noted above for the boiler headers. In addition the piping supports may experience deterioration and become damaged from excessive or cyclical system loads. Assessment: The NDE method of choice for testing of external weld surfaces is WFMT. MT and PT are sometimes used if lighting or pipe geometry make WFMT impractical. Non drainable sections such as sagging horizontal runs are subject to internal corrosion and pitting. These areas should be examined by internal video probe and or UT measurements. Volumetric inspection, i.e. shear wave, of selected piping welds may be included in the NDE; however, concerns for weld integrity associated with the growth of subsurface cracks is a problem associated with creep of high temperature piping. 4.) Feed water Piping: A piping system often overlooked is feed water piping. Depending upon the operating parameters of the feed water system, the flow rates, and the piping geometry, the pipe may be prone to corrosion or flow assisted corrosion (FAC). This is also referred to as erosion-corrosion. If susceptible, the pipe may experience material loss from internal surfaces near bends, pumps, injection points and flow transitions. Ingress of air into the system can lead to corrosion and pitting. Out-of-service corrosion can occur if the boiler is idle for long periods. Assessment: Internal visual inspection with a video probe is recommended if access allows. NDE can include MT, PT or WFMT at selected welds. UTT should be done in any locations where FAC is suspected to ensure there is not significant piping wall loss. 5.) Deaerators: deaerators have been known to fail catastrophically in utility plants. The damage mechanism is corrosion of shell welds which occurs on the ID surfaces. Assessment: Deaerators' welds should have a thorough visual inspection. All internal welds and selected external attachment welds should be tested by WFMT.

6.) Attemperators: The spray flow attemperator, a device for controlling superheater outlet steam temperature, is normally located in the piping system between the primary (1st stage) superheater outlet and the secondary (2nd stage) superheater inlet. The attemperator is subject to failures associated with thermal fatigue cracking of its components and welds. Since it is in a non viewable area of the boiler, failures may go undetected until pieces of the attemperator lead to other damage, such as superheater tube failures. These steam temperature control systems should also be part of the boiler fitness survey testing. Assessment: For the inspection is recommended by removal of the spray head assembly. The spray head is inspected visually and tested nondestructively by MT/PT methods. Following removal of the spray head from the body of the attemperator, the attemperator thermal liner can be internally inspected with a video probe. 7.) Tubing: Statistically the greatest numbers of forced outages in all types of boilers are related to tube failures. Failure mechanisms vary greatly from the long term to the short term. Tubes are more likely to fail because of abnormal deterioration such as: water/steam-side deposition inhibiting heat transfer, flow obstructions, tube cor rosion (ID and/or OD), fatigue, and tube erosion. Assessment: Tubing is one of the components where visual examination is of great importance because many tube damage mechanisms lead to visual signs such as distortion, discoloration, swelling or surface damage. The primary NDE method for obtaining data used in tube assessment is contact UT for tube thickness measurements. Sample removal for laboratory analysis is by far the best test to conduct for the assessment of life in the component.

7A.) Water wallsUnit #6- Starting in the Coutant throat and extending to the Screen tubes multiple problematic conditions exists in Unit #6; Caustic corrosion, Internal corrosion resulting in corrosion fatigue failures, erosion, overheat, thermal fatigue, reducing atmosphere, and past maintenance repair problems. Caustic corrosion exists on both hot and cool surfaces of the boiler. The external surface of the boiler tube walls appears as heavy oxide deposits that when chipped away left a 1/16 divot. Tube samples have been removed for analysis by Dr. French. Root cause of the boiler external corrosion apparently is caused by the acidic nature of the ash combined with external pre-outage wash downs. It is recommended to target two more samples from the water walls next outage to compare remaining life to the last samples taken. In addition two samples from the bottom side of the Coutant should also be taken for base line analysis due to recent failures in this area.

Internal corrosion has also been documented correlating to failures in the burner corners relating to corrosion fatigue. A combination of internal corrosion, high heat flux zone, and degradation of the support system in this area are believed to have been the root cause for previous leaks in this area. Water wall panels have been replaced in this area to remove the afflicting conditions and this area is now under monitoring status. Erosion was a problematic condition around wall blowers however the plant does not use the wall blowers anymore hence this area is in monitoring status. Soot blower erosion on the other hand has been a concern on the rear side of the Superheat Pendant Platen, front side of the Final Superheat, front side of the vertical outlet legs of the Primary Superheat, Coutant, and Deflection Arch. In the past these areas have been shielded with success except for the rear side of the Superheat Pendant Platen. Due to the intense heat in this area the shields have been unsuccessful in protecting the trailing side of the pendant and multiple repairs have been conducted outage to outage. Flame spray is a potential candidate for this area and budget allowing will be utilized this outage. Other locations of concern are the Coutant area. Previous UT scans across the upper side of the throat and along the side walls revealed four areas of immediate attention during last outage. All areas were cleaned and flame sprayed to extend the life an additional 5 years. During the upcoming outage the upper Coutant transition to front and rear water walls will be accessed for more UT scans to provide a baseline reference for future work. Last but not least the upper side of the Deflection Arch between rear wall Hangers, and Screen tubes, the Arch was found to have lost up to 40% of minimum wall requirements. Due to this area being replaced in the next five years the area was prepared and Flame sprayed. Over the course of the last two years hundreds of swelled tubes have been documented in the water walls from the Coutant to the Radiant Reheat wall elevation. Research discovered that for a length of time one of the boiler circ. Pumps was not being utilized resulting in a starvation of flow leading to the overheated sections of tubing. A few samples were taken two years ago to visually determine the severity however none of the samples were sent for analysis. Two samples are recommended to be sent for analysis of remaining life. All circ. Pumps have been in operation thereof and no failures have been documented since the condition has been recognized. The two planned samples will establish a rate of progressive degradation if any exists. Another problematic area concerning heat related issues is the propagating thermal fatigue cracks located in the membrane of the Deflection Arch. Due to the mass quantity of cracks over the last few outages the most severe areas of membrane have been removed and replaced with new membrane. The new membrane has also been documented as developing cracks in just the few short years it has been installed. Research proved that in the past, due to the unit being positive pressure the membrane was changed from peg to solid; however the gap between the tubes does not allow for proper coolant of the solid welded membrane hence the

propagating crack issue. To solve the problem a long term solution has been made to reengineer the Deflection Arch tube spacing to allow adequate cooling area for the membrane. Reducing atmosphere conditions exist due to the combination of Low Nox Burners and Over Fired Air. Slight corrosion pitting is apparent on the water walls and was ground smooth for UT verification. Remaining wall thickness was .180 leaving the tube well above action criteria. We recommend that a tube be removed in this same general area for analysis to determine a rate of progressive degradation if any exists.

7B.) Radiant Reheat7C.) Pendant Platen Superheat7D.) Pendant Reheat7E.) Final Superheat7F.) Primary Superheat7G.) Economizer-