Fatigue evaluation acc. to section 5.5.3, ASME S8 Div2 2019

[dc_1]

Hi

It would be great if somebody could answer this two (hopefully) simple questions:

1. Are the component stress ranges (equation 5.28) based on linearized stress tensor components or directly on the actual stress tensor?

2. Is it possible to show that the von Mises stress computed from the actual tensor components is always larger/equal than the von Mises stress computed with the linearized components?

Thank you!

 

[TGS4]

1. Actual stress tensor, not linearized.

2. Not sure what you’re driving at with this question. Please provide additional context.

 

[dc_1]

Thank you! That was fast!

The background is parametric optimization, with fatigue being one of the optimization goals. Question 2 was a kind of a back-up to the first one. I would have used the actual von Mises stress range as an upper bound for the one based on the linearized stresses (in order to avoid the ambiguity of SCL placement, which would be quite difficult to automate if the geometrical changes are singificant). But now everything is clear, and I can simply use the acutal stresses to compute my ranges.

Thank you again!

 

[DriveMeNuts]

Div 2 specifies Actual Stress sensor.
However, table 5.12 specifies seeking more information from WRC 432 for intricate rules for establishing the fatigue reduction factor. This WRC bulletin has exceptions to use the lineraised stress value for certain sharp corners and stress risers. Judgement is required.