Learning Forum

[YJ L]

i want to compare the results with transient analysis and response spectrum. the picture says to use RESP Command, but i don't know exactly which command and how to use them. How do i write the RESP command to convert 20g acceleration- time 5ms data into frequency-acceleration data?

(this picture is my data)

[Mike Rife]

Hi YJ

That RESP command is an Mechanical APDL command.  It is not a WB Mechanical function/feature.  You can look up that command in the Ansys MAPDL Help - there is an example on using it though it assumes that a time transient analysis has been done and uses the displacement time history of a node as the input.

There is a LF post on using this though it looks like the format got messed up somehow; there are "n" instead of new line character so it's all one big paragraph!  Here is the link to that:

https://forum.ansys.com/forums/topic/how-to-create-a-shock-response-spectrum-from-time-history-data-with-apdl-commands/

Mike

[peteroznewman]

Hello YJ L,

You said you want to convert a 20 g amplitude, 5 ms duration signal into a Response input spectrum, but when I look at the plot of your data, the triagular pulse has a magnitude of 74.556 m/s^2 (7.6 g) and a duration of 0.0121 s.  So which is it?

Software tools outside of Ansys are available to convert a time-history into a Shock Response Spectrum. I use the free, Stand-Alone Executable Version of Vibrationdata because it doesn’t require a copy of matlab to be installed.  If you have matlab installed, there is a larger toolkit that is also free. Look for the relevant links on this page: https://vibrationdata.wordpress.com/

Conversion of time-history data requires you to choose the Q factor that defines the damping in your structure.  If you know the critical damping ratio for your structure, Zeta, then Q = 1/(2Zeta). For example, if your structure has a critical damping ratio of 2%, then Q = 25.  You add damping to the excitation because the Modal analysis provides undamped modes. What is the Q factor for your structure?

What is the frequency range you want to study for your structure?  I suspect the sampling frequency used in the representation of this triangular pulse is too low as it is only 909 Hz (sampling delta t = 0.0011 s).

By sampling the triangular pulse at 10,000 Hz, the Shock Response Spectrum can be calculated above 1000 Hz.