Hi everyone,
I was wondering if someone has any clue about how different factors are taken into account in a fatigue criteria. For example, if the contributions of temperature and strain in one criteria will be additive or multiplicative.
I'm not sure to be clear but if you have any paper that talk about that, I'm really curious to check it.
Best regards,
Benjamin Martin
Fatigue design will have a factor of 2 for scatter, 4 for surface finish and 3 for volume . The factor 2x4x3, which is 24. We have to divide the uniaxial fatigue life by this number to get the design fatigue life. As a conservative approach, specimen fatigue life is divided by 30 to get the design fatigue life.
As far as the surface finish is concerned, a mirror finish surface will have a factor of 1 while the roughest surface will have a factor of 4. Other factors (scatter and volume) are not alterable.
Most machine design textbooks have a section in the chapter on fatigue about various influence factors. Given the fatigue strength for the lab test results, SN, the fatigue strength of the part will be Sn = k1xk2xk3xk4xk5xk6x . . . xSN where each k is due to some effect. Scatter (assuming 30 is NOT always conservative) should really be considered in the regression of the data. The decrease due to temperature is often assumed to be the same decrease as the drop in ultimate or yield strength (whichever is worst) but it would be better to actually test the fatigue strength at temperature, but the amount of time at the temperature may also be important and no one wants to do a fatigue test that lasts 10,000 hours. Other influence factors include the surface finish (look for charts in textbooks), load (axial, bending, torsion) compared to the test load (usually axial or bending), any coatings applied to the material, the method by which the surface was machined (more charts), atmosphere (dry air, salt spray, humid air, something nasty), and there are a few dozen other things that can affect the fatigue strength of the material in service.
Modifying factors are applied at the Endurance Limit. Your probably best reference is the book "Mechanical Engineering Design", by Shigley et al. I'd be cautious in using these factors or your results can end up being extremely conservative.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs