Refineries and chemical plants own and operate numerous process heaters (e.g. gas reformers, CCRs, etc.) as part of the standard assets throughout the facilities. Many heater coil configuration designs are flanged at both ends; however, there are also coil designs which contain common headers, linking the individual coil passes together at the inlet, outlet or even at both ends in some cases (see Figures 1 and 2). Plant inspectors are presented with a significant challenge to perform the necessary condition assessment inspections on serpentine coils, and common headers complicate this further by adding additional challenges for decoking and inspection processes. For more than a decade, the use of intelligent or smart pigs has become a standard practice when inspecting regular flanged coil configurations, which do not contain common headers. However, with new common header snorkel delivery systems, these particular heater designs are also able to be inspected using the same approach.
As described in a previous article in the
November/December 2012 issue of Inspectioneering
Journal ("New Technology Allows Access to Coils with Common Headers in the
Process Industry”), advanced engineering firms and mechanical decoking
companies have developed unique common header snorkel delivery systems enabling
both cleaning and inspection of connected serpentine coils. Once access to the headers’ interior has been
provided, by removing the blind flange or cutting off the bonnet, these systems
are then temporarily inserted into the common header while the fired heater is off-line
(see Figures 3 and 4), and guide mechanical decoking pigs and intelligent pigs
into the heater’s individual coil passes.
The delivery systems are modular by design, eliminating the challenges
associated with accessing individual coils, and they can be attached to short
or long common headers.
By designing the snorkel system in small, compact individual sections, they can be assembled as the cleaning and inspection activities move into individual coils further down the length of the common header (see Figure 5). The modular design also offers the user smaller, compact hardware which is easier to maneuver in tight spaces, in addition to addressing safety concerns.
Once the mechanical cleaning (decoking) effort has concluded, inspection of the individual coils utilizing intelligent pigging technology can be performed immediately afterwards. The intelligent pig is launched into each individual heater coil via the common header delivery system. Generally the instrument is placed into the temporary launcher (see Figure 5) and hydraulically propelled forward via the common header delivery system, which purposely directs the tool into the individual serpentine heater coil. After reaching the intended heater coil, the remote intelligent pig travels through the coil at an average speed of 2-feet per second, collecting millions of inspection data points. If the heater coil configuration permits removal at the coils’ far end, the intelligent pig is removed at this point. In the event that no access exists at the coils’ exit location, the intelligent pig's travel direction can be reversed allowing removal from the coil to take place at the original launch location.
The evolution of intelligent pigging technology has been continuous since its introduction in the mid 1990s. When the technology first surfaced within the refining and petrochemical markets, inspection coverage, resolution and accuracy were limited at best. Today’s advanced models contain dozens (50-400) of fixed mounted ultrasonic (UT) transducers on each tool, ensuring 100% overlapping inspection coverage is achieved. Advancements in ultrasonic transducer technology, signal processing algorithms and high speed electronics have had a significant positive impact on resolution, senility and data accuracy. The combinations of all noted advancements have evolved the technology beyond what anyone originally thought obtainable (see Figure 6).
The initial intelligent pig tool design in the 1990’s contained limited ultrasonic (UT) sensors, sometimes as few as 8 to 16 per tool. The limited number of UT sensors mounted within the early prototype systems provided limited inspection coverage (20 to 30%) of the heater coil’s surface, which resulted in the inability to detect small flaws such as fretting, pitting, etc. In reality, the earlier versions of the tools were nothing more than a glorified thickness measurement device, which were very poor at detecting flaws unless the physical size of the flaw encompassed a very large surface area. Today’s advanced tool designs contain significantly more sensors (e.g. 48 to 366, depending upon tool size), ensuring 100% overlapping inspection coverage. Ultrasonic-based intelligent pigs with this number of UT sensors provide a much higher level of assurance that small flaws such as fretting and pitting are detected.
When providing 100% overlapping inspection coverage, there are naturally millions of single ultrasonic data points captured over the full length of a heater’s serpentine coil. Proper management of this volume of inspection data is essential to ensure accurate and repeatable test results. In addition to the ability to extract and examine single point remaining pipe wall thickness readings, another important capability is the application of all the data when producing high resolution two-dimensional (2D) and three-dimensional (3D) graphics. High resolution graphical models produced utilizing the inspection data collected has been referred to as an "ultrasonic photograph" that clearly illustrates areas of concern and highlights general degradation patterns, regardless of severity (see Figures 7 and 8).
Remaining life assessment and fitness-for-service approaches have evolved in parallel with inspection technology. With the latest release of the API-579/ASME FFS-1 Standard, a comprehensive evaluation can be performed while considering the interactive effect of corrosion, erosion, creep and other damage mechanisms. Software has been developed to automatically apply the API-579/ASME FFS standards, utilizing 100% of the ultrasonic inspection data within the complex calculations.
Inspection data can lose a significant portion of its value if not utilized properly. Fitness-for-service (FFS) and remaining life assessment (RLA) calculations can be conducted in accordance with API-579/ASME FFS-1 (see Figures 9 and 10, remaining life assessment 2D Views) utilizing 100% of the inspection data. Plant reliability engineers then have the necessary confidence to make critical repairs enabling extended run times.
Plant operators who own and operate heaters which contain common headers now have the same cleaning and inspection options that were previously available only for those heaters which did not contain common headers. Considering the limited access to the external surface of heater coils, this is a significant advancement in the industry because these systems reduce the risk of heater failure due to individual coil failure caused by damage mechanisms such as corrosion, erosion and creep strain.