Learning Forum

[mehmetyaylaciinfo]

Is X axis (time) important to fatigue analysis? Although my time inputs have 0.02s period, when I put the just Y axis datas, the X axis periyod is 1 second for every Y axis data.If I put Y axis datas, it is enough? I hope I would tell well

 

Sincerely,

[danielshaw]

Are you using Ansys nCode DesignLife?  If so, the values in the time series file represents scale factors applied to the FE stresses.  The fatigue life is calcuated as repeats of that loading sequence (repeats of the applied time series).  The time associated with the time series is only important if you want to relate the predicted number of loading cycles to time.  The actual time of the individual scale factors within the time series is not relevant.

[mehmetyaylaciinfo]

Yes, I use nCode DesignLife. Actually, I understood what you said, but what I want to know is that when I add my own acceleration-time data, the values in time do not match each other. For example, the acceleration value in 0.02 seconds is 1.2g, but in time-series, 1.2g corresponds to the 1st second. So I just put the g values into the time-series. I want to know exactly this method.

[danielshaw]

If you are using the time series load provider in Ansys nCode DesignLife, the solver multiplies one set of FE results by the scale factors stored in the time series file to create a stress history.  For example, if the time series file contained 1,3,-1, the solver just creates a stress history that varies from 1*FE results to 3*FE results to -1*FE results.  The time series load provider creates a stress history that is independent of time.

If the actual loading time is critical, then you may not be able to use a time series load provider.  For example, if the acceleration loading produces dynamic behavior as a function of time, you may need to run a time-transient analysis to accurately capture the dynamic stresses as a function of time and use a time step load provider with multiple FE result sets to capture the stress history.

[mehmetyaylaciinfo]

The time series load provider creates a stress history that is independent of time, so that even if I put acceleration data without time, it is enough for the fatigue analysis, because my acceleration loading does not produce dynamic behavior. I understood better, thank you so much.