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.

When the wreckage which fell upon the Ohio shore was disassembled, Professor W. L. Starkey, under contract with the Federal Highway Administration, was invited to examine freshly opened pin connections in the eyebar chain to look for possible signs of fretting-fatigue. Such a mechanism takes place when metal parts in contact under high loads undergo very small relative movements which tend to break the microscopic asperities of the metal surfaces. These broken pieces undergo corrosion, producing iron oxide, usually $ \mathrm{F e_{2} O_{3}} $ , which is very hard and which occupies more space than the steel particle from which it came. The presence of these hard particles, under great pressure because of their own expansion and some small relative motion, produces a grinding action between the two steel surfaces. If the movements are large, the particles escape; but if small, they are trapped and result in the development of pits and then cracks perpendicular to the mating surfaces. Professor Starkey's inspection of the freshly opened joints, in his opinion, showed some evidence of such fretting action near the "90 degree" positions of the hole, where a plane perpendicular to the longitudinal axis of the bar through the center of the hole intersects the hole surface. Pin surfaces were scored and there was a reddish-brown oxide present (Reference 28). Professor Starkey also believed there was evidence of fretting on the fragment of eyebar No. 330 and on one of the pins.

In order to assess the possible importance of this observation, simulated fretting-fatigue studies were conducted on material taken from the C11 head of eyebar No. 330 and the shank just adjacent to this head. The specimens were prepared in the shops of the FHWA laboratories from blanks cut from the eyebar at the NBS laboratory. Before machining, the ends of the blanks were checked for hardness by the NBS laboratory. The material was found to be typical of other material in eyebar No. 330 in this respect, with hardness between Rockwell C 20 and 25. The 19 specimens themselves were tapered pins 4-1/4 inches long, 0.144 to 0.146 inches in diameter at the small end, 0.322 to 0.324 inches in diameter at the large end, with a Brown and Sharpe taper of 0.502 inches per foot. Collars of the same material were also prepared and each test was conducted on a combination pin-collar set. A contact pressure of 5000 psi was obtained by forcing the pin into the collar until a brittle lacquer on the outside surface showed initial cracking, then forcing the pin an additional fixed distance. The reliability of this technique was checked by dummy specimens in which the collars were strain gaged to measure hoop stress. The specimen combination was placed in a specially constructed rotating beam fatigue testing machine which held the collar so as to support the tapered pin as a loaded cantilever beam, thus causing minute longitudinal relative motions between the pin and collar as the whole assembly rotated. An additional 19 standard rotating beam fatigue specimens were also prepared and tested to get basic fatigue data.