Shaft In Bending

[lenovo1]

Using traditional hand calc techniques, I have calculated the stresses within the shaft below and would like to verify my methodology.

Threaded section
I do not believe this is significantly loaded due to the close tolerance in section 2.
I think it may be worth calculating the load required to deflect the shaft so it is definitely in contact with the block and determine what load is therefore applied to the threaded section (so an appropriate assembly torque can be calculated).

Section 2
I have calculated the bending and shear stress at location 2 and applied an appropriate SCF to account for the section change.
I have calculated the second moment of area based on the diameter of section 2 (should I be working out an average for the shaft?).
The bending moment is based on the force multiplied by x1.
Bending stress is based on My/I where y is half the shaft diameter at location 2.

Section 3
I have based the second moment of area based on the outside diameter of the female connector and applied the material properties of the weaker component.
Bending stress is based on My/I where y is half the outer diameter of shaft 2.
I have reviewed the bearing stress between the two interfacing components.
I have reviewed shear through the shaft taking the diameter of shaft 2 to calculate the area and applying the weaker of the material properties.

Should I also be looking at the bearing stress where shaft 1 meets the block (on the outer face of the block and in the clearance hole)?
Basically shaft 1 is showing excessive bending stresses at location 2 and I'm wanting to check that the methodology is not overly pessimistic.

Thanks for any help

 

[rb1957]

if you're comparing Kt*linear bending stress with Ftu, and calling that "excessive" then i'd say yess, that's overly conservative. I'd consider doing a plastic bending analysis (Cozzone) of location 2.

Quando Omni Flunkus Moritati

 

[corus]

If it's a rotating bending shaft then you should look at fatigue damage. For your stress concentration compare that against the limiting stress for infinite fatigue life, which is approximately 0.5 x UTS, if I recall.

 

[rb1957]

that'd be for a fatigue (typical in-service) loading ... presumably this is a static (ultimate design) loading.

Quando Omni Flunkus Moritati

 

[realtime2]

Critical "hot spot" is probably at the transition between smaller and larger diameters (?)
Calculate the elastic stress at hot spot including any stress concentrations.
Correct for plasticity using e.g. Neuber Plasticity correction which is basically
Elastic calculation Energy = Plastic stress-strain energy (using cyclic stress strain curves for fatigue)
With the calculated local stresses and strains, after plasticity correction, compute life of part.
If its a cyclically loaded part tensile yield is garbage. Use cyclic yield.
1/2 the ultimate is a good stress for infinite life in steel. Not so for aluminum.
Find a fatigue curve for your material.

Repeat for any other suspect hot spot.