Once you have set up the model, learn how to run fast and efficient simulations using
EDEM Simulator. This simulation requires a Nvidia GPU card.
To run the simulation:
Go to Tools > Options > Simulator Engine > CUDA.
Select your Nvidia GPU if displayed, else click Device Start
Test to view the test results and then click
OK.
Click the
icon on the toolbar to open EDEM Simulator.
Set the Time options
The first step in running the simulation is to set the Time Step. The Time Step is
the amount of time between iterations (calculations). The Time Step value is based on the
material properties, and is usually set between 20% and 40% of the automatically calculated
Rayleigh Time Step.
To set the Time Step:
In this simulation, ensure that the Auto Time Step
checkbox is not selected.
Set the Time Step value to 0.0001 s which should be
approximately 12% of the Rayleigh Time Step.
Set the simulation time and data write-out frequency.
The simulation time is the amount of real time your simulation represents.
In the Simulation Time section, set the
Total Time to 15
s.
Set the Target Save Interval to 0.02
s to specify the write-out frequency.
As this is a
mixed shape solver with polyhedral particles, the GPU engine and
settings are disabled. It is recommended to run this simulation on
CUDA GPU only.
Under Advanced Settings, select
Hybrid from the Solver
Precision dialog box.
Double precision is a direct match to the CPU solver which only has the
double precision option. Many graphics cards have slower double precision
calculation speed than single precision, the mixed shape solver allows a mix of
single and double precision, only using double when required for
accuracy.
Run the Simulation
The last step is to run the simulation.
To run the simulation:
Click the Start Progress icon in the Simulation window.
In the Viewer Controls, click the Refresh Viewer
icon or the enable Auto Update
icon at any point to update the Viewer to view the progress of the simulation.
Note: This simulation takes 15-30 minutes to run depending
on the GPU card.