The thought that a equipment breaks down due to a single mechanical error in the installation or a single design error, may seem very convincing, but if we study further we will see that this does not have much grip in the real world.
The real causes usually go beyond a metallurgical problem or an error by the mechanic maintenance; even beyond the fact that the machine was operating at twice the capacity for which it was designed.
The truth is that when all human errors and deficiencies of the management system are combined, the typical machine failure is the result of five and ten significant contributors.
The depth of the analysis at the cause of the fault is the key to accurately finding all its sources.
In an effort to better understand the general sources of failures of a plant, a fault analysis company decided to analyze the projects of the last three years to determine the main aspects that contributed to the presence of failures.
There were a total of 131 analyzes and below we list the main failure mechanisms:
23 Corrosion 18%
57 Fatigue 44%
15 Use 11%
17 Fatigue corrosion 13%
19 Overload 15%
Total = 131
Testing Details
It is important to highlight that when defining these five categories there is a possibility of confusion between corrosion fatigue and fatigue.
The practice that the inspectors applied was to assign fatigue as a mechanism in those cases in which the component would eventually have failed and no corrosion was needed to effect the failure.
In those situations, in which the component would not have failed without the action of corrosion, let say, there was a cyclic load but it was not severe enough to cause cracks without corrosion, the cause was mentioned as corrosion fatigue.
In addition, it should be borne in mind that only the main fault mechanism has been listed here, since many of these failures actually had multiple physical causes.
After analyzing the data of the mechanisms, the information was examined to determine the main causes of failure.
These were defined as those causes that were evident from the inspection of components, field analysis and discussions with the operational and maintenance personnel of the plant.
Based on the discussions with the personnel and the experience of the inspectors, six error categories were selected: operational, design, maintenance practice, supervision, manufacturing and installation of original equipment. Data were sorted in these groups.
According to a previous study reported by Congleton, the attribution of responsibility for corrosion failures investigated by a large chemical company based in the United States was broken down into the following:
Lack of evidence: new design, material or process: 36
Lack of specifications or incorrect: 16
Bad inspection: 10
Human Error: 12
Poor planning and coordination 14
Other 4
Unpredictable 8
Final results
This data set indicates that only 8% of corrosion failures are unpredictable. In other words: 92% of corrosion failures could be prevented.
If this is true, the small percentage of corrosion damage attributed to the lack of human performance by the Hoar report, Battelle’s studies and many others, would be approximately 30 to 40% of total corrosion damage.
With information from “Understanding why it failed” by Neville W. Sachs, P.E. Sachs, Salvaterra & Associates, Syracuse, NY.
