Question 1

Why is my endurance limit so much lower than 0.5 x Sut?

Accepted Answer

Because 0.5·Sᵤₜ is the rotating-beam SPECIMEN value — a polished 7.6 mm bar at room temperature with a 50% chance of survival. The Marin factors correct it to a real component and every one of them is below 1. A machined 25 mm shaft at 600 MPa, designed to 99% reliability, lands at about 59% of the specimen figure.

Question 2

What does ke = 1 mean?

Accepted Answer

Fifty percent reliability — a coin flip on the mean S-N curve, not a safe default. For 99% reliability ke is 0.814, and for 99.9% it is 0.753. It is the most surprising default in the whole Marin method.

Question 3

Does the size factor apply to axial loading?

Accepted Answer

No. There is no size effect in axial loading, so kb = 1 there and the knock-down is carried entirely by kc = 0.85. Applying the diameter formula as well would count the same penalty twice.

Question 4

Which mean-stress criterion should I use?

Accepted Answer

Goodman is the common default and is conservative. Soderberg is more conservative again because it measures against yield rather than ultimate. Gerber fits test data best and is least conservative — useful for understanding, risky for designing.

Fatigue — Mean-Stress Criteria

Calculator

Marin factors — what the part actually is