Hello,
I'm investigating the effect of shock waves on granular materials. My setup consists of a box of spheres which are impacted on by a high speed CAD granular wall. Some spheres, however, pass directly through the wall upon impact (this is due to the high impact velocity, and is a very undesirable effect, as it prevents me from conducting proper analysis). Any suggestions as to how to deal with this circumstance are highly appreciated (my first instinct is to substantially increase the wall's Young modulus, making it stiffer and harder to be penetrated, but Im not sure whether this would produce any significant difference). Attached is a picture of my setup, in case it is of any help.
Thank you kindly for your time
Carlos
| Attachment | Size |
|---|---|
 sound_propagation_-_setup.png | 30.19 KB |
arnom | Wed, 01/03/2018 - 18:05
You will have to lower the
You will have to lower the time step to avoid such penetration. See the fix check/timestep/gran for more.
c97 | Sun, 01/14/2018 - 04:29
Timestep
Thank you for your help Arnom.
As of now, I have a timestep of .0000000005 (5e-10!!) and a wall impact of 750 m/s. No warnings/errors appear after using the check/timestep/gran fix; however, the spheres continue to penetrate through the wall, and this timestep reduction means calculating on the order of 7 million steps for each simulation. Do you know of any fix for this, or have any alternative ideas as to how to induce a shock wave in the granular matter?
arnom | Mon, 01/22/2018 - 18:01
LIGGGHTS has never been used
LIGGGHTS has never been used in this regime and I'm not sure whether our models actually would produce the right physics for such a case.
AndresMM | Wed, 01/10/2018 - 11:05
Hola Carlos,
Hola Carlos,
I am also studying sound propagation on granular matter. Out of curiosity, what exactly are you trying to simulate?
Based on my experience and as Arnom said, the time-step you are using might be too high. Another possibility is that the granular packing is so compressed that the particles have no other choice than "bounce out" from the container, but that does not seem to be the case.
Saludos,
Andrés
c97 | Sun, 01/14/2018 - 04:34
Hola Andrés,
Hola Andrés,
I am studying how shock waves affect granular media by performing plate impact experiments (basically, I hit a granular packing really hard with a moving wall, and then figure out what is happening). I dont seem to be able to create shock waves just yet, and I have a feeling it may be a software limitation particular to LIGGGHTS. If you have any ideas as to how I might proceed to create a shock, those are greatly appreciated. Suerte con tus investigaciones.
Saludos,
Carlos
arnom | Mon, 01/22/2018 - 18:13
This topic is closed because
This topic is closed because further discussion happens here: https://www.cfdem.com/forums/shock-waves