FINITE-ELEMENT ANALYSIS OF THE MODIFIED RING TEST FOR DETERMINING MODE-I FRACTURE-TOUGHNESS

Citation
Mp. Fischer et al., FINITE-ELEMENT ANALYSIS OF THE MODIFIED RING TEST FOR DETERMINING MODE-I FRACTURE-TOUGHNESS, International journal of rock mechanics and mining sciences & geomechanics abstracts, 33(1), 1996, pp. 1-15
Citations number
35
Categorie Soggetti
Geochemitry & Geophysics","Mining & Mineral Processing
ISSN journal
01489062
Volume
33
Issue
1
Year of publication
1996
Pages
1 - 15
Database
ISI
SICI code
Abstract
Plane strain fracture toughness (K-Ic) values are determined for the m odified ring (MR) test through numerical simulation of crack growth to highlight the sensitivity of MR K-Ic values on applied displacement o r force boundary conditions, slip conditions at the specimen-platen in terface, and the Poisson ratio (nu) of the test material. Numerical ca lculation of fracture toughness in the MR test is traditionally conduc ted assuming a uniform force along the specimen loading surfaces and n o slip between the specimen and the loading platens. Under these condi tions K-Ic increases by 30-40% as nu decreases from 0.4 to 0.1. When s lip is allowed at the specimen-platen interface under a uniform force, K-Ic values are independent of nu, and for any given nu, are 5-25% le ss than those determined under a no-slip boundary condition. Under a u niform displacement of the specimen loading surfaces, K-Ic is essentia lly independent of nu, regardless of specimen-platen interaction. More over, although K-Ic values determined under uniform displacement and n o-slip boundary conditions are always higher than those determined und er uniform displacement and slip-allowed boundary conditions, the aver age difference in K-Ic for these two methods is less than 5% for the t wo specimen geometries examined. This suggests that under uniform disp lacement conditions, K-Ic is essentially independent of specimen-plate n interaction. Because K-Ic values determined from MR testing are stro ngly dependent on the modeling procedure, future reports of K-Ic, dete rmined from this test should be accompanied by detailed reports of the modeling procedure. Until further testing reveals the most accurate s imulation technique, we advocate use of a uniform displacement formula tion for K-Ic determination from MR testing because results from this method are insensitive to most modeling parameters. Numerical results from models conducted under uniform force, no-slip boundary conditions should be viewed as an upper bound to K-Ic.