pic: Snap Rings Or Shaft Collars?

[Rob Stehlik]

Strictly speaking, there is a good reason not to use aluminum in shafts. As a material, aluminum doesn’t handle fatigue well.
From wikipedia: Fatigue occurs when a material is subjected to repeated loading and unloading. If the loads are above a certain threshold, microscopic cracks will begin to form at the surface. Eventually a crack will reach a critical size, and the structure will suddenly fracture.
Since shafts are rotating, they are almost always subject to fatigue. Actually, the lab test for fatigue involves a bending load applied to the end of a rotating shaft. As one would expect, the lower the stress on the material, the longer it lasts. And here is where materials differ: With steel, there is what is referred to as the “Endurance Limit” at which point the degradation of the material is so minimal, it will last indefinitely. With aluminum, there isn’t an endurance limit. Here’s a graph showing a typical fatigue response of steel and aluminum:
http://en.wikipedia.org/wiki/File:S-N_curves.PNG
Notice that at a stress of 30 ksi, the curve for steel levels out, while aluminum continues down.
This being said, fatigue is not an exact science, but is based on lab tests and observation. The results offer guidelines for good design. Here is a great primer on fatigue:
http://www.epi-eng.com/mechanical_en..._in_metals.htm
Another important point about fatigue is that it begins at the surface. So grooves for snap rings act as stress risers, and are often the point of failure if they are in an area of high stress. I suspect that the original shaft in question failed due to bending. A cantilevered load on the end of the shaft would cause bending stress between the bearing supports, which is right where the snap ring groove is. Switching to a higher strength aluminum would make it last longer, but getting rid of the snap ring groove altogether would be an even better solution.
Obviously a lot of teams use aluminum shafts in FRC with great success, and our team has as well. If the stress is low, they will last long enough. Keep in mind that the average lifetime of an FRC robot is measured in hours. It’s a calculated risk. But in the real world, you would almost never see an aluminum shaft because of fatigue.