Collapse of U.S. 35 Highway Bridge, Point Pleasant, West Virginia, December 15, 1967 (HAR-71/01)

Read from the scan by GLM-OCR; expect the odd misread word.

must be concluded that brittle fracture was possible at a stress about equal to the yield strength of the steel.

The validity of these data was confirmed later by a test of a full scale eyebar head under loading conditions which produced a strain distribution across one limb of the eye similar to that existing in the prototype structure under load. The purpose of this latter test was to determine whether the type of fracture found in the lower limb of the C13 eye of eyebar No. 330 could have been produced by static load alone. A static load test at room temperature of a piece of the south eyebar from C9-C11 of the north chain, which contained a crack similar to those found in eyebar No. 330, had been conducted at the U. S. Steel laboratories. The fracture had shown an intermediate zone of ductile fracture between the pre-existent crack and the zone of brittle fracture. The Battelle laboratories were, therefore, asked to break a full scale eye at a temperature of about $ 32^{\circ} $ Fahrenheit.

The specimen for this test was prepared from one of the eyebar heads in which the non-destructive test work had indicated the existence of a pre-existent crack on the pinhole surface in approximately the same location as that which produced the fracture of eyebar No. 330. The general arrangement is shown in Figure 19. The left side of the eye was severed by a saw cut. Two 2.74" diameter holes were drilled on a line parallel to the longitudinal axis of the eyebar at a location such that a tensile load applied through pins at these holes generated a stress gradient across the right limb of the eye similar to that existing in the case of normal loading of the eye in the chain. Strains along the expected line of fracture were monitored by bonded electrical strain gages.

On the first attempt at loading, in which a stress of 90,000 psi was produced near the crack, no failure occurred. The specimen was partially unloaded and again checked by dye penetrant. It was found that the "crack" was actually a shallow surface defect. An artificial defect was therefore introduced by means of a hardened steel wedge and then enlarged and sharpened by fatigue loading. At the second attempt at loading to failure, fracture took place during a partial loading imposed to check instrumentation. The stress in the vicinity of the crack was about 57,000 psi. Fortunately, all instrumentation worked properly, so it was possible to measure the flaw size (0.345" x 0.780") and to compute the effective critical stress intensity factor, $ K_{Ic} $ . A value of 46.2 ksi in. was obtained, which compares closely with the average value of 46.4 ksi in. at $ 32^{\circ} $ Fahrenheit measured on the surface flaw specimens. No intermediate ductile zone was present in the fracture.