Fit for Service
SwRI engineers use a variety of techniques to ensure the integrity of pressure vessels and other structures
Joseph Crouch (left) is manager of the Marine Structures and Engineering Section in SwRI’s Mechanical Engineering Division. Crouch has extensive experience in the design, fabrication and testing of structures for the oil and gas, offshore, marine, space and aerospace industries. Curtis Sifford is a senior research engineer in the Marine Structures and Engineering Section. He specializes in the design and analysis of pressure-containing structures.
SwRI researchers routinely perform fit for service analyses on a variety of structures such as the hull of the next-generation Alvin manned submersible shown here.
A failure assessment diagram approach is one method SwRI researchers use to determine if a structure is fit for service.
Senior Research Engineer Curtis Sifford went onsite and examined the decompression chamber of the tunnel boring machine.
Data from the analyses plotted on the failure assessment diagram show the values at which weld defects are acceptable.
Finite element analysis was used to determine the stresses in the decompression chamber.
Pressure vessels are common pieces of equipment used worldwide for many applications, such as compressed air cylinders, hyperbaric test chambers, chemical reaction vessels, medical decompression chambers and submarine pressure hulls. They are designed to hold gases or liquids at pressures that are frequently much greater than atmospheric pressure, or in the case of a submarine pressure hull, to withstand the crushing external pressure of the deep ocean.
Compressed air cylinders provide a constant air supply for powering pneumatic tooling. Hyperbaric test and chemical reaction chambers create pressurized environments that simulate deep ocean pressures or geologic formation pressures, or create an environment to allow chemical reactions to occur. Decompression chambers are used for safe decompression for saturation divers who perform work in deep water, or for medical purposes to promote healing.
As with most structures, pressure vessels are designed for a finite useful life based on such parameters as the number of pressurization cycles expected, the magnitude of each pressure cycle and structural changes, such as a reduction in material thickness caused by corrosion. These “design” parameters are usually estimates and are often greater than the actual values experienced during operation. If the values used to predict the design life of the vessel are more excessive than what is actually experienced, the vessel will still possess usable life upon reaching the end of its design life.
Unless the structural integrity of a vessel can be re-assessed to determine that it can still be used safely for a longer period of time, a vessel that has reached the end of its design life should be taken out of service. Replacements can be costly, and the loss of a vessel interrupts productivity. Thus, establishing any remaining usable life is very important.What is “fitness-for-service” analysis?
For more than 35 years, Southwest Research Institute (SwRI) engineers have performed structural integrity assessments on pressure vessels, aircraft and propulsion systems, offshore structures, and other structures in accordance with sound engineering principles and relevant industry practices. Recently, as customer demand for pressure vessels has increased, SwRI re-established a program for fabrication of pressure vessels in accordance with Section VIII, Divisions 1, 2 and 3 of the ASME (American Society of Mechanical Engineers) Boiler and Pressure Vessel Code. ASME rules enable SwRI engineers to perform detailed structural integrity assessments, or fitness-for-service (FFS) analyses, of pressure vessels in accordance with both API (American Petroleum Institute) and ASME standards. The structural integrity of a pressure vessel is generally assessed by comparing the calculated stress with the strength of the material and then considering other potential failure mechanisms like fatigue. This type of assessment usually assumes that the vessel is free of defects such as cracks, voids, weld slag or inclusions.
However, defects that may have occurred during fabrication, or that develop during its service life also are often present. A fracture mechanics assessment is therefore required to determine if these defects will have a negative effect on the future operation of the vessel.
A pressure vessel containing a defect is considered fit-for-service if it is able to withstand the loads (such as pressure, thermal, wind and earthquake) experienced during its desired service life with a suitable safety margin to account for any uncertainty in the assumptions used for the assessment. Key elements for an FFS assessment are the loads applied to the component, the dimensions and shape of any defects, the material’s mechanical properties (such as fracture toughness), and the rate of crack growth for the material. Typically, a fracture mechanics fitness-for-service assessment is performed after a defect or crack has been found following routine inspection, maintenance or safety checks, or when the effect of an undetectable crack needs to be considered. The assessment determines whether the pressure vessel is safe to operate with the defect or to establish inspection intervals for monitoring the defect. If the defect size is unacceptable, then the user must decide whether to repair it, replace the equipment or re-rate the equipment for a safe, lower operating load. A fracture mechanics assessment may also be used as input to a quality control program to determine critical locations for future inspection, the size of a defect that must be detected with a high confidence and the necessary inspection interval.
SwRI engineers use the guidelines presented in Fitness-for-Service, API 579-1/ASME FFS-1, which is “a compendium of consensus methods for reliable assessment of the structural integrity of equipment containing identified flaws,” first issued as a recommended practice document in January 2000. These methods require that the state of the vessel be determined using nondestructive evaluation (NDE) methodologies. This can be done specifically for an FFS effort as part of a safety program, or the FFS assessment can result from the identification of flaws found during routine inspection.
The FFS process begins with an NDE evaluation of the structure. This provides details such as depth, length and location of any defects or cracks that have been detected. Based on the probability of detection, assumptions must be made about other defects that might exist but may not have been detected. Once the defect state of the vessel has been characterized, linear elastic fracture mechanics analysis is used to calculate the stress intensity factor, K, a measure of the “driving force” available in the structure that can cause a crack to propagate unstably. SwRI has a great deal of expertise in NDE methodologies, probabilistic assessment and uncertainty characterization, as well as fracture mechanics and finite element analyses, all of which are important to an FFS assessment.
SwRI engineers use finite element analysis to determine the stresses in the structure that result from the applied loads. When a structure is loaded to a level that is less than the yield strength of its material, it behaves elastically, meaning it returns to its original shape like a spring, when the load is removed. If the structure is loaded to a stress level that is greater than its material yield strength, it may experience a change in shape resulting from permanent plastic deformation, the extent of which could be local (barely noticeable) or global (large deformations) approaching plastic collapse, depending on the magnitude of the applied loads. In the presence of plasticity, the driving force that would continue to propagate a crack may be underestimated using linear elastic fracture mechanics. Thus, to account for the interaction between failure by fracture (crack instability) and failure by plastic collapse under limit load, SwRI uses the failure assessment diagram (FAD) approach. The FAD is a two-parameter graphical representation of the failure envelope of a cracked structure expressed in terms of the ratio of the applied stress intensity factor to the material fracture toughness (the toughness ratio, Kr = Kapp/Kmat) and the ratio of the applied load to the plastic limit load of the structure (the load ratio, Lr = P/PL).
To use the FAD approach, assessment points with coordinates (Lr, Kr) calculated based on the applicable loads, crack type and crack size(s), and material properties are compared with the failure envelope line. Assessment points that lie inside the envelope indicate non-failure, while assessment points outside the envelope indicate failure. For many fatigue crack growth analyses, the assessment points will initially be far inside the failure assessment line envelope and will gradually grow toward the envelope as the crack grows sub-critically. When the load ratio is low, the FAD predicts failure based on fracture instability; however, as the load ratio increases, the interaction of the presence of plasticity decreases the allowable stress intensity factor. If the assessment point is on, or inside, the FAD envelope, which indicates that there is remaining service life, then the pressure vessel is deemed safe, and therefore fit for service. A fatigue crack growth analysis must then be performed to determine how long the pressure vessel will remain fit for service.FFS examples
Recently, SwRI engineers have performed FFS assessments on an in-service section of a gas pipeline, a submarine hull and a decompression chamber for a large tunnel boring machine. During inspection, the gas pipeline was found to have a defect and required FFS assessment to determine if the defect had to be addressed immediately or if it was still usable for a specified period of time. Because the submarine was newly built, researchers had to assume that a defect existed, but was too small to be detectable. The FFS assessment on the submarine was performed to determine if the possible defect would jeopardize the safety of the occupants between scheduled inspection intervals.
The tunnel boring machine was an earth pressure-balanced type, which provides continuous support to the tunnel face by balancing the earth and water pressure against the thrust pressure of the machine. Under normal operating conditions, workers operate the machine in an enclosed environment behind the cutting head, which is maintained at atmospheric pressure. However, if cutting head maintenance is required, workers must travel through the decompression chamber, which exposes them to elevated pressures. Once they complete their activity under pressure, they then have to re-enter the decompression chamber and stay inside while the pressure is slowly brought back down to ambient conditions, much as a diver must decompress in a chamber after spending time deep below the surface. This decompression prevents the maintenance personnel from getting sick from “the bends,” which is caused by gas bubbles forming in their blood.
The problem arose when the tunnel boring machine was required to dig deeper into soil conditions that demanded a higher internal pressure rating. The decompression chamber was re-analyzed for the higher pressure and then subjected to a hydrostatic pressure test in accordance with the ASME Boiler and Pressure Vessel Code to prove the structural integrity of the system. The ASME code also requires an inspection of the welded joints following the pressure test. During this inspection, multiple defects were found in some welds. SwRI engineers were therefore asked to perform an FFS assessment to determine if the weld defects had to be repaired or if the decompression chamber was fit-for-service with the defects in place.
For the decompression chamber, the principal loading was a result of internal pressure. Residual thermal stresses from welding were also considered. Engineers determined a residual stress distribution using a solution from FFS that provided a conservative upper bound for the residual weld stresses based on numerical analysis and a literature survey of published results.
How residual stresses affect the stress intensity factor depends on the level of material plasticity. For elastic conditions, the residual stress can significantly weaken a structure containing cracks. On the other hand, when there is high plasticity, the effect of these stresses can be small. The API/ASME FFS methodology applies a plasticity interaction factor to the stress intensity factor to account for this effect.
The crack dimensions and shape were determined using phased array ultrasonic inspection. With this type of NDE inspection, multiple ultrasonic elements are used and their timing is varied so it is possible to steer, focus and scan the beam, providing a visual image of the defect or crack. Nearly all of the defects found on the decompression chamber were embedded and classified as weld slag or porosity. Although classified as “inclusions,” the defects were considered to be cracks for the FFS assessment. Researchers also determined the defect depth, length and distance from the surface.
The crack-growth rate and fracture toughness are well documented for SA516 Grade 70 steel, the base metal used in the chamber. However, this was not the case for the weld metal. There were additional concerns regarding the welds’ toughness since they were not stress-relieved. For these reasons, SwRI researchers performed fracture toughness testing on the welded plate. They tested the fracture toughness in the weld metal, at the fusion line and in the heat-affected zone. Results were used to establish a fracture toughness value that could be used for the FFS assessment.
Stress intensity factors for the weld defects that were assumed to be cracks were calculated using the SwRI-developed computer code NASGRO®. NASGRO, which earned an R&D 100 Award in 2003, was initially developed and released in the 1980s for fracture control analysis of NASA space hardware and has been continuously improved since 2000 by the NASGRO Consortium under the management of SwRI. It contains a large library of advanced stress intensity factor solutions and material property data combined with extensive analysis capabilities.
SwRI engineers calculated a toughness ratio by dividing the stress intensity factor by the fracture toughness of the material. Given the load and toughness ratios, the assessment points were plotted on the failure assessment diagram to determine if the weld defects were acceptable. Because all of the assessment points were inside the failure assessment diagram envelope, engineers determined that the decompression chamber was fit-for-service for the expected static loads, with the weld defects found during inspection. It is also possible for a crack to grow under cyclic loading. For this reason, a crack-growth analysis was also required. NASGRO was again used. While the tunnel boring machine is in use, the decompression chamber is kept sealed and is only used for maintenance of the cutting head. Therefore, it was expected that relatively few pressurization cycles would occur, so no significant crack growth was expected. The fatigue crack-growth analysis demonstrated that the pressure load could be cycled to the maximum operating pressure more than 9,000 times before the first assessment point on the failure assessment diagram reached the envelope. This remaining life greatly exceeded the service life of the decompression chamber for the tunneling project. Thus, the remaining life was sufficient and engineers determined the decompression chamber was fit-for-service for the remainder of the project, saving the client considerable cost and time.Requisite tools and experience
As illustrated by the decompression chamber project, API- 579-1/ASME FFS-1 FFS provides a preferred means for assessing equipment that no longer meets its original design specification or code of construction. Other assessment techniques are available for dealing with a variety of flaws or damage mechanisms, such as metal loss, pitting, corrosion, lamination, dents, gouges, weld misalignment, shell distortions, creep, fire damage and crack-like flaws. These assessment techniques can sometimes prevent costly repairs or replacement of equipment while still allowing safe operation.
A fitness-for-service assessment depends in large part on how well the defects are understood or how well the uncertainty of the information is characterized. It requires a solid understanding of fracture mechanics and crack growth phenomena. Fitness-for-service uses partial safety factors, which are factors applied to the stress, crack size and material toughness to account for uncertainty in the input parameters used for the assessment and to ensure a minimal probability of failure. Also, probabilistic analysis can be used. For this, SwRI engineers use NESSUS, a 2005 R&D 100 award-winning technology developed for NASA by SwRI.
SwRI develops and uses nondestructive evaluation techniques to characterize flaws in metallic structures. Expertise in these areas, coupled with experience in design, analysis, fabrication and use of various pressure vessels (both manned and unmanned), enables SwRI to work efficiently and effectively to assess fitness-for-service for industry and government, ensuring safe, useful life of a wide range of equipment.