Application ofWeibull Probability to Predict the Size of Inclusion in Metallic Material

• Main Author: Megat Shahair, Megat Atif
• Format: Final Year Project
• Language: English
• Published: Universiti Teknologi PETRONAS 2008
• Subjects: TJ Mechanical engineering and machinery
• Online Access: http://utpedia.utp.edu.my/9738/1/2008%20-%20Action%20of%20Weibull%20Probability%20to%20Pridict%20the%20Size%20of%20Inclusion%20in%20Metallic%20Material.pdf; http://utpedia.utp.edu.my/9738/

Large inclusions can be the initiation site for fatigue failure in metal components. As melting processes are becoming more refined, the size of the inclusions falls below the level of detectability of the non-destructive testing methods. This final year project is divided into three parts. In the first part of the project, Weibull probability was applied to predict largest oxide inclusion size and compare to the actual observation under scanning electron microscope. The results showed that Weibull probability prediction is accurate with margin of ± 3 microns. In the second part of project, the Weibull probability was tested using nodular cast iron. The nodules were measured for their true and apparent sizes, respectively. Based on the data, the effect on Weibull probability was found to be negligible. In the third part of the project, rotating fatigue test was performed under cantilevered loading by using two sets of medium carbon steel specimens. The specimens were annealed at 840 °C, held for one hour and furnace cooled before being polished and tested. Step-size method was selected where each specimen was subjected to 2.52 x 10 cycles at initial load of 5 N. The load was increased progressively until the specimen eventually fails. Only those specimens failed due too oxide inclusion at fatigue initiation site were regarded. Based on observational results, the two sets had different probability of survivalwhich corresponded to their respectively largestoxide inclusion size.