Metal Fatigue Part III:

One and Two Way Bending

The swinging gate, of course has two hinges and two hinge pins.  Interestingly, the other hinge was mounted (welded in place) in such a way that it experienced mostly one way bending.  The fracture surface for the other pin is shown in Photograph A.

Photograph A: Two way bending fatigue in a bolt. Large Arrow at 1 o'clock shows area of fatigue crack initiation.

In this case, the fracture initiated at the 1 o’clock position and propagated toward the 7 o’clock position.  A much smaller fatigue crack propagated (on a lower level) inward on a thread below the main fracture surface, from the 7 o’clock position.

The final, fast fracture region is the small band, denoted with small red arrows, which are oriented from about 9 o’clock to 4:30 o’clock.

Fatigue cannot usually be accurately timed, i.e., this crack has been growing for two years, three months and twenty two days.  The inability to age or time a fatigue crack can be seen by considering a cross-country airplane trip.  During takeoff from NYC, the wing stresses are high (and if small cracks are present, they may grow microscopically), between NYC and the Great Lakes, the stresses are much lower (and there is no microscopic crack growth).  Then passing over the Great Lakes, there are times of extreme air turbulence (notice we have left out complicating factors such as jet stream).  Then over the Great Plains, the weather/winds are calm and wing stress is very low.  When crossing the Rocky Mountains, updrafts (and downdrafts) are experienced.  As can be seen, on a six or seven hour flight, it would be impossible to assign any crack growth to specific stress events in the trip.  The extreme stresses which MIGHT promote crack growth are not continuous/intermittent and thus to use and average stress over the time in the air would be incorrect. However, under certain, very unique circumstances, “timing” can be accomplished.  I have worked on at least two “wheel off” tractor/trailer truck cases where aging or timing of the growing fatigue crack could be and was accurately predicted.  Each case was unique in that new wheel/lug studs were replaced when a flat tire was fixed. Thus when the studs failed on the first trip out, ALL prior stress history of the truck was irrelevant, i.e., the critical component, i.e., the wheel/lug studs which failed were changed at the last stop.

Sometimes, the fracture surface even at a visual level, can give us useful information.  A close-up view of a fractured hexagonal shaft is shown in Photograph B.

Photograph B: Overall view of fatigue failure on hexagonal shaped jackhammer tool.

In Photograph B, the fatigue fracture surface exhibits, in classic “beach marks”, “thumbnail marks”, “stop marks” and are clearly present in the darker portion of the fracture surface (see small white arrows).  However, from Photograph C, it can be seen that newer cyclic fatigue crack evidence (in the areas denoted “new fracture surface”) is an area where fatigue crack growth occurred long after the darker fatigue crack surface shown in Photograph B was created. 

Photograph C: More recent cracks emanating from older, longstanding fatigue.

The upper half of the fracture surface is where "fast fracture" (final separation) occurred.  This final “fast fracture” region is shown in Photograph D. 

Photograph D: Final fast fracture region of fatigue crack was created at the final instant of failure and separation.

Thus, just visually the fracture surface tells us that an “old fatigue fracture” surface is present, newer fatigue fracture is present (created at a distinctly different time and environment) and the fast fracture region created at the time of final failure/separation.

--top--

Part I - Part II - Part III