Stepped Shaft Calculation 1

[Alesek]

I have stepped shaft as per attachement. This shaft is made from stainless steel 416T. I would like to calculate when the shaft will be made from different material (3xx stainless steel) if the shaft is enought strong for radial loads and torque.
Can someone help me to solve it?

 

[corus]

As a simple hand calculation, treat it as a beam of different section properties with simple supports at the bearing positions. You could use a finite element program for this, of which there are some free ones available for this case. At the step change, use Peterson's Stress Concentration factors to evaluate the true stress at fillet radii from the nonminal stresses produced by the FE results.

corus

 

[Alesek]

I'm not familiar with this kind of calculations. Can I ask you to help me to find out some calculation steps to solve it?

 

[corus]

If you're not familiar with doing beam/stress calculations then it's unlikely you'll be able to check or interpret the results properly. I'd look online using pdf search engines to find examples of similar calculations so you can see the methodology.

corus

 

[Alesek]

ok will try to find some methodology.

 

[MartinSr00]

The classical way is straightforward.

Almost by inspection, the highest stress will likely be at the fillet at the end of the smallest step.

Use the classical Mc/I and TR/j to get the the torsional stress in the .5512 diameter section. The bending stress is from your side force. The torsional stress is from the turning.

Then, get a copy of Peterson's Stress Concentration Factors to get the stress concentration factor at the fillet for both torsion and bending.

Construct a Soderberg Diagram for your material. When doing this, safety factors of, say, three or so to the material data.

Then plot your duty point on the soderberg diagram. Your torsional stress times the stress concentration factor is a constant stress. The bending stress times the concentration factor is the alternating stress.

See if your stresses come under the wire.

The problem with finite element models is that the chief action is probably in the fillets, and if the fillet is not meshed fine enough, you'll probably won't get a good idea of concentrated stress.

Look perhaps in Mark's Handbook, or a college stress textbook to help you along.