Vibration and age alone should not cause a bolt to fail because they are engineered to do the job they are designed for.
Fatigue cracks are caused because of "tired" metal. To know why a bolt fails one has to determine what conditions existed that caused the metal to fatigue (get tired).
One reason is sloppy manufacturing (machining) of the bolt another might be cheap metallurgical properties of the bolt itself. Most often however failure due to cracks is caused by improper torque or such where one bolt is doing the work of two or more. (i.e. One bolt over torqued and others under torqued)
In such cases normal vibrations will weaken the metal of the overworked bolt and the origin of a crack will begin at the weakest point of the area under stress (crack origins usually begin at the surface often at a machine scratch or blemish).
This spot is smaller than the point of a needle but as normal vibrations continue (from the serpentine belt),
the microscopic dot begins to grow with each vibration and eventually works its way through the shaft of the bolt. In other words, whether a car is gasoline powered or a diesel, will have no effect in determining if a generator bolt will fail. Such vibrations are inconsequential to AC bolts. It is the combination of the high frequency vibrations of the serpentine belt along with a defect of design, engineering or assembly that causes such bolts to crack and fail.
[Edited by ejsharp on 05-27-2001 at 12:19 PM]
1993 190E 2.3
2000 Toyota 4x4 Tundra