Creating a Fatigue Study Based on Random Vibration Results

You can create a fatigue study that is based on stress results (in the form of Power Spectral Density function) from a dynamic - random vibration study.

1. Set up and run a linear dynamic random vibration study.
2. Create a fatigue study. In the Study PropertyManager, under Options select Random vibration - fatigue of random vibration .
3. In the fatigue study tree, right-click Loading (Random Vibration) and select Add Event.
4. In the Add Event (Random Vibration) PropertyManager:
   1. Under Study, select the random vibration study to associate the fatigue event.
   2. Under Duration, select the time duration (in units seconds, minutes, hours, or days) of the random loading input applied to the model.
   3. Click .
5. Right-click the top study icon and click Apply/Edit Fatigue Data to associate an S-N curve to the material.

   You can enter the two S-N Basquin curve constants (B) and (m), or click Estimate Basquin constants from S-N curve and enter multiple data points that capture the stress range (stress variation from the maximum cyclic stress to the minimum cyclic stress) versus the cycles to failure.

6. In the Fatigue - Random Vibration dialog box, select the computational method for estimating the vibration fatigue damage: Narrow Band, Steinberg, or Wirsching.

   The Narrow Band method gives usually the most conservative results. It is recommended to run the fatigue study with all three methods and compare the results for the fatigue damage estimates.

7. After you run the study, you can view plot results for Life (Time to fatigue failure), and Damage (percentage or ratio of the life consumed by the random vibration loading). A damage factor of 1 and above indicates that the part has failed for the given time duration of the random loading.