A statistical approach on the support type effect on mode I fracture toughness determined using semi-circular bend (SCB) specimen. (1st March 2020)

• Record Type: Journal Article
• Authors: Sedighi, Iman; Ayatollahi, Majid R.; Bahrami, Bahador
• Is Part Of: Engineering fracture mechanics. Volume 226(2020)
• Publication Date: 2020-03-01
• Subjects: Mode I fracture toughness -- Support type effect -- Statistical approach -- Semi-circular bend; Fracture mechanics -- Periodicals; Rupture, Mécanique de la -- Périodiques; Fracture mechanics; Periodicals; 620.112605
• DOI: 10.1016/j.engfracmech.2020.106891
• Languages: English
• ISSNs: 0013-7944

Abstract

Highlights: Effect of support friction on mode I fracture of SCB specimen was statistically assessed. 65 fracture tests with four different types of supports were conducted. Friction in supports caused errors up to 90% for calculated fracture toughness values. Results showed the importance of using accurate support type in the fracture tests. Weibull's parameters were successfully predicted by employing a shift factor method. Abstract: The semi-circular bend (SCB) specimen has been utilized to investigate the effect of support type on the mode I fracture toughness of polymethyl-methacrylate (PMMA). A total of 68 samples have been tested using four types of supports: roller, roller-in-groove, fixed roller-in-groove and fixed. The experimental results show that, when the roller support type is used, the fracture toughness values obtained from the SCB tests are in good agreement with those of the compact tension (CT) standard tests. Meanwhile, using the other types of supports results in different values for fracture toughness. Then, in order to analyze the statistical nature of the fracture toughness results for each type of support, two and three-parameter Weibull distribution models are used. The applicability of Weibull model on predicting the support type effect is proved with its good estimations. Furthermore, it is shown that the distribution parameters obtained for roller type support can be successfully predicted by using the distribution parameters of other support types if the friction coefficient between the supports and the SCB specimen is taken into account.