Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Fatigue analysis from market data

  1. Apr 17, 2005 #1
    I am working on a project for a company to analyse the fatigue life of a component. There is an existing component that the company
    manufactures and the component fails by fatigue on the average 540 days. The component is a tube made of low carbon steel of yield 260N/sqmm and ultimate of 340N/sqmm. But the company gives a guarantee of 700 days to the customers. How can i simulate
    by theory the fatigue life of 540 days for that component.
     
  2. jcsd
  3. Apr 18, 2005 #2

    minger

    User Avatar
    Science Advisor

    What exactly do you mean by "simulate by theory?" I'm a little lost at that phrase.
     
  4. Apr 18, 2005 #3
    Yes. Simulate by basic theory.
     
  5. Apr 18, 2005 #4

    Clausius2

    User Avatar
    Science Advisor
    Gold Member

    I recall there are two ways:

    -Using Paris equation, which is equated with experimental coefficients.

    -Using FEM.

    For a rapid calculation see first case.
     
  6. Apr 18, 2005 #5

    PerennialII

    User Avatar
    Science Advisor
    Gold Member

    Would also consider how you define "fatigue life" of the component, i.e. whether it consists of crack initiation and/or propagation, if only the latter then Paris law is the best way to go, if former is included methods of "classical" fatigue analysis come into play. Depending on the complexity of your component I think you can have decent enough estimates using "desktop" solutions, such as the IIW rules and there are some general lower bounds for Paris law coefficients (or Nasgro if a more general form of FCP law is required) depending on type of material and under what conditions the fatigue occurs.
     
  7. Apr 18, 2005 #6
    can anyone help with classical fatigue analysis
     
  8. Apr 18, 2005 #7

    PerennialII

    User Avatar
    Science Advisor
    Gold Member

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?