Learning Forum

[MTLCY]

Could someone tell me why they build a SPH layer at the bottom?

[patjoy]

The presence of downstream particles influences the wave profile, for dam-break waves there are typically two types of simulations, a dry-bed and a wet-bed simulation. Furthermore, this example is based on a SPH validation method termed a bore-in-a-box simulation, you can see the model domain and details in the paper they outline on the example page.n

[MTLCY]

Thank you very much for yor response. Another detail is that the bottom of the pillar is lower than the bottom rigid wall. Is it necessary?nnI modified this model by changing the shape of obstacle and size of the whole model, however, some unreasonable results occurs. Particles penetrated the obstacle and no interaction occurs(CONTACT_AUTOMATIC_NODES_TO_SURFACE is used to define the interaction between particles and FEM obstacle). Furthermore, particles moves upwards along the left wall. Could you tell me how to handle the penetration?nn

[Ian Do]

MTLCY,nIt does not matter is the column sticks a little lower than the bottom surface. It is conservative to have it lower rather than higher.nnAs for contact, maybe you may first try with something like:n$---+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8n*CONTROL_TIMESTEPn$# dtinit tssfac isdo tslimt dt2ms lctm erode ms1stn 0.0 0.45 0 0.0 0.0 0 0 0n*CONTACT_AUTOMATIC_NODES_TO_SURFACE_MPP_IDn$# cid titlen 1SPH to Columnn$# ignore bucket lcbucket ns2track inititer parmax unused cparm8n 0 1 0 3 2 1.0005 0n$# ssid msid sstyp mstyp sboxid mboxid spr mprn 2 1 3 3 0 0 0 0n$# fs fd dc vc vdc penchk bt dtn 0.0 0.0 0.0 0.0 0.0 0 0.01.00000E20n$# sfs sfm sst mst sfst sfmt fsf vsfn 0.0 0.5 0.0 0.0 1.0 1.0 1.0 1.0n$# soft sofscl lcidab maxpar sbopt depth bsort frcfrqn 1 0.1 0 1.025 2.0 2 50 1n$---+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8nnElse, try these additionally, one at a time or combination:n- PENMAX=1000 in the option card B of *CONTACT_ keyword-n - XPENE=1000 in *CONTROL_CONTACT keywordn - try SOFT=1 and set SFS = SFM = 1.0e-12 in *CONTACT_AUTOMATIC_NODES_TO_SURFACE n - SOFSCL=1.0 (increase from 0.1 default if contact leaks, maybe for very high velocity impact or explosion) n n

[MTLCY]

Thank you so much for your kind suggestions. My issues have been addressed.n

[Miner]

Hi all, I have a new issue in this simulation. After the water droped down, there are some particles move upwards along the left rigidwall. It is like those particles adhere to the wall. I have tried to reduce the gap between wall and water, but doesn't work. Could you help me to resolve this?nn

[Miner]

Hi all, I have a new issue in this simulation. After the water droped down, there are some particles move upwards along the left rigidwall. It is like those particles adhere to the wall. I have tried to reduce the gap between wall and water, but doesn't work. Could you help me to resolve this?https://us.v-cdn.net/6032193/uploads/AQFXOQNSY04K/image.png/forum/discussion/comment/96759#Comment_96759

The rigidwall must align the part surface. When change the particle spacing, small gap may generated between the rigidwall and part surface, thus the contact between rigidwall and part produced weird results. If particle spacing is changed, the rigidwall must be updated.n