TIMESTEP
Answers
-
Hi,
1. Element time step is used by default. The nodal time step is invoked by nodal mass scaling.
2. Mass scaling is a technique to speed up computation by adding mass on nodes. The time step is related to the speed of sound in the material, c:
c=√E/ρwhere ρ the material density and E is Young’s modulus. Mass scaling manipulates this relation on nodes exceeding specified timestep. Care should be taken the added mass does not change the physics of the problem (kinetic energy), and should preferably be below 2% of the total mass. Advanced mass scaling does not add mass and is even more computationally efficient, but is not recommended for high-velocity impact.For more details refer to the timestep control chapter in the Radioss ebook, where I also made a short video on meshing for explicit:
1 -
Simon Križnik said:
Hi,
1. Element time step is used by default. The nodal time step is invoked by nodal mass scaling.
2. Mass scaling is a technique to speed up computation by adding mass on nodes. The time step is related to the speed of sound in the material, c:
c=√E/ρwhere ρ the material density and E is Young’s modulus. Mass scaling manipulates this relation on nodes exceeding specified timestep. Care should be taken the added mass does not change the physics of the problem (kinetic energy), and should preferably be below 2% of the total mass. Advanced mass scaling does not add mass and is even more computationally efficient, but is not recommended for high-velocity impact.For more details refer to the timestep control chapter in the Radioss ebook, where I also made a short video on meshing for explicit:
1.Is there any particular situation where we use nodal time step
2.also regarding video how did you find ur target nodal timestep?
0 -
Amith Anoop Kumar said:
1.Is there any particular situation where we use nodal time step
2.also regarding video how did you find ur target nodal timestep?
1. Nodal timestep is used whenever we want to take control (enforce) a certain timestep. For more details refer to Radioss ebook or:
2. The target timestep is determined based on computational power and time available. In this video, the target timestep was set to achieve a uniform nodal timestep distribution (all elements having similar timestep). Compare uneven timestep distribution before at 0:50, and after remeshing at 4:50 (improved timestep uniformity = better computational efficiency).
The same technique could be used to target multiple timesteps when meshing for multidomain analysis.
0 -
Simon Križnik said:
1. Nodal timestep is used whenever we want to take control (enforce) a certain timestep. For more details refer to Radioss ebook or:
2. The target timestep is determined based on computational power and time available. In this video, the target timestep was set to achieve a uniform nodal timestep distribution (all elements having similar timestep). Compare uneven timestep distribution before at 0:50, and after remeshing at 4:50 (improved timestep uniformity = better computational efficiency).
The same technique could be used to target multiple timesteps when meshing for multidomain analysis.
First a fall thanks a lot for the detailing sir, just one more thing so the target time step in the video was it a value you assumed and to make the time step uniform
0 -
Amith Anoop Kumar said:
First a fall thanks a lot for the detailing sir, just one more thing so the target time step in the video was it a value you assumed and to make the time step uniform
The target timestep was an approximated average timestep of the rest of the model, estimated from the middle hump of the distribution shown at 0:50.
This technique produces the most computationally efficient models for preliminary analysis. The final model needs a mesh convergence study, which usually results in locally refined mesh, thereby reducing computational efficiency, but improving accuracy.
0
https://www.youtube.com/watch?v=GUiGbCQCc6U