Metallurgical failures can be extremely costly, resulting in replacement costs, down time, and in the worst case, injury or loss of life. Preventing as many of these failures as possible is of significant interest and importance to industry. In an ideal scenario, failures would never be unexpected, and could then be planned for. Unfortunately, this is not the case and unexpected failures do occur. However, knowing the locations that are more at risk to fail provides useful insight. These areas should have more regular inspection and care, especially in equipment where failures have significant consequences. For more than 11 years, KnightHawk Engineering’s (KHE) Lab has analyzed hundreds of failures from our customers (primarily oil and gas and petrochemical industries) and determined the types of parts that are most likely to fail, based on our experience. The following is a list, in order of the part types that are most likely to fail.
and Pipe Fittings:
By far the most likely components to fail, according to KHE’s Lab experience, are those that are designed to carry liquid or gas from one location to another. Sometimes at high temperatures and/or pressures, and sometimes used to transfer heat into or out of those liquids or gases, i.e. tubes, pipes, and pipe fittings (e.g. flanges, valves, etc.). An example of such a failure is shown in Figure 1. The reasons for these failures are highly varied and too complex to discuss in their entirety here, but likely the most significant of these causes, is simply the fact that there are more of these parts in the oil, gas, and petrochemical industries than there are of the other categories of parts in which failures occur. It is expected that these parts will fail more often than others, because there are more of them to fail.
Figure 1: An Erosion Corrosion Failure of a Tube Bundle from a Heat Exchanger.
Figure 2: High Cycle Fatigue of a Turbine Blade.
Parts within rotating equipment (engine components, bearings, turbine components such as the turbine blades shown in Figure 2, etc.) are the second most likely to fail, according to KHE’s Lab experience. The primary reason for this is somewhat intuitive; parts that make up rotating equipment are generally exposed to higher and more dynamic loads, along with the same environmental factors that affects the rest of the materials in industry. The materials are obviously designed for these higher forces, but where there are moving parts, there are more opportunities for something to go wrong and cause one or more of those parts to fail.
Just over 1 in 10 failures that KHE’s Lab sees, occur in or adjacent to welds. Welding is critical to the construction of almost all metallic equipment at some level. When performed properly, welding serves the critical purpose of permanently joining like, or even unlike, materials, making them critical to the construction of everything from buildings to process equipment to pipelines. However, welds also form natural stress risers in the material and locally alter the materials microstructure (shown in Figure 3), which can cause even a good weld to be the first location to fail in a given system, with high stresses. Additionally, when performed improperly, welds can significantly damage the surrounding material, resulting in failures that would otherwise not have occurred. Despite only making up a tiny fraction of the total volume of material in industry, welds account for a significant percentage of metallurgical failures.
Figure 3: Multi-Pass Weld, HAZ, and Base Metal Regions of a Pipe in a Heat Exchanger.
Figure 4: Cracking in an I-Beam
Static components (i.e. I-Beams, like the one shown in Figure 4, and Pressure Vessels, etc.), account for just under 1 out of 10 failures that KHE’s Lab observes. These components account for a relatively large amount of material in Industry. However, they have a relatively low frequency of failures. This is most likely a result of the fact that, as static components, these parts tend to be designed with a larger safety margin than many other parts, which decreases the number of failures. However, when these parts do fail, the results can often be catastrophic.
Fasteners (e.g. bolts, tie rods, etc.) make up a very small percentage of the volume of materials used in industry, but make up a much larger percentage of the failures that have been seen at KHE’s Lab. Much like parts from moving equipment, fasteners tend to be exposed to more extreme and dynamic loading conditions than the parts that they are installed in. For this reason, fatigue is the dominate cause of fastener failures.
Figure 5: Flange Fasteners
Safety and Sensor Equipment:
Figure 6: Pressure Gauge Sensor
Safety and sensory equipment (rupture disks, thermowells, gauges, etc.) failures only account for approximately 1 out of every 25 failures that KHE’s Lab analyzes. While this is not a large percentage, the critical role that this equipment plays in maintaining a process, means that failures in this category can have potentially catastrophic consequences.
You may have noticed that the percentages presented in this article do not sum to 100; in fact they sum to 105% and there are some failed parts that did not fit into any of the above categories. The reason for this is that many of the weld failures occurred in parts that can also be attributed to one of the other categories (e.g. a pipeline weld failure). The weld category was included due to the frequency of weld specific failures, and the resulting importance of accounting for weld behavior. However, the archived information on the locations at which failures occur still provides useful insight. In particular, the fact that tubes, pipes, and pipe fittings make up nearly half of all failures, is indicative of the importance of process control, in terms of chemistry, temperature and pressure. If failures and the resulting cost, are to be prevented, process control is critical.
Future installments in this series, will discuss the dominant failure modes in the most significant of the above categories, and present sound practices for preventing these failures. Stay tuned for the next installment in our series “Industry Blacksmith Decisions: Reasons and Reliability”