August 10, 2021 at 3:33 am

cwk_ansys

Subscriber

Once again, thanks for your great explanation!

I always though that Explicit Dynamics time step=sampling time step that I set.

It turns out that there is a Explicit Dynamics time step, the simulation only record the data of sampling time step to our hard drive.

However, I can not find and set the ls-dyna time step in Ansys mechanical.

Even if I could, I don't have the storage size to store all the raw data.

5000 points already occupied 80GB of hard drive, If I were to record all the data of every time increment of the simulation, it would take more than 10TB.

This is the acceleration graph that plotted with IACCOP flag equal to 2, it is still very noisy

These are the graph that applied low pass filter, the peak value of IACCOP=2 is lower than IACCOP=0.

Even with IACCOP=2 and low pass filter applied, the graph is not smooth and fluctuating.

Finally, can you help to confirm the following statements?

1) The acceleration graph after applying low pass filter is very noisy and incorrect because data has been aliased due to the time step that I set.

Every single increment of time step from the simulation need to be taken and applied to low pass filter in order to get the correct acceleration curve.

2) The velocity graph is smooth because it is integrated from acceleration in every Explicit Dynamic time step, and then decimated to my time step that I set.

I always though that Explicit Dynamics time step=sampling time step that I set.

It turns out that there is a Explicit Dynamics time step, the simulation only record the data of sampling time step to our hard drive.

However, I can not find and set the ls-dyna time step in Ansys mechanical.

Even if I could, I don't have the storage size to store all the raw data.

5000 points already occupied 80GB of hard drive, If I were to record all the data of every time increment of the simulation, it would take more than 10TB.

This is the acceleration graph that plotted with IACCOP flag equal to 2, it is still very noisy

These are the graph that applied low pass filter, the peak value of IACCOP=2 is lower than IACCOP=0.

Even with IACCOP=2 and low pass filter applied, the graph is not smooth and fluctuating.

Finally, can you help to confirm the following statements?

1) The acceleration graph after applying low pass filter is very noisy and incorrect because data has been aliased due to the time step that I set.

Every single increment of time step from the simulation need to be taken and applied to low pass filter in order to get the correct acceleration curve.

2) The velocity graph is smooth because it is integrated from acceleration in every Explicit Dynamic time step, and then decimated to my time step that I set.