Answers are grounded in this paper and cite their sources.
Thirty-four strain gage recordings of elastic waves radiated from tensile fractures generated by thermal stresses in glass plates are presented. Fractures associated with the extension of an existing crack as well as initial fractures are represented. There appears to be a prominent event on each record associated with the stopping of the fracture. The identification of this event makes it possible to measure the time interval during which the fracture propagates; taken together with the measured length of the fracture this makes it possible to calculate the average fracture velocity v. For extended (unilateral) fractures v = 1.6 km/sec and for initial (bilateral) fractures v = 2.0 km/sec; the difference appears to be significant. These velocities are close to the value (about 1.8 km/sec) measured by high-speed photography for impact-generated fractures. Given the fracture velocity, the analysis may be reversed to estimate the length of a unilateral fracture or the length of each branch of a bilateral fracture from the time intervals which may be measured on the strain record. Finally, an approximate form of the theory of Ang and Williams is used to estimate the order of magnitude of the separation of the two walls of the fracture (about 3 μ) and also the velocity during fracture of one wall relative to the other (about 10 cm/sec). A few minutes after fracture, optical interferometry indicated that the separation of the fracture walls varied from zero at the ends to about 2 μ near the center; we take this as evidence that the order of magnitude estimate of the dynamic separation given above is correct.
No takes yet. Share an insight, caveat, or question.
Savage et al. (1964) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: