How to visualize the actual size of particles in paraview

Submitted by lenard10 on Fri, 04/01/2011 - 19:40

Hi,

I run a simulation of a sphere packing problem, How should I visualize them by their actual size other than a lot of points without contact in paraview? Thanks!

Lenard

Forums:

cstoltz | Fri, 04/01/2011 - 21:35

Hi Lenard,

Once you've imported the particles, add a glyph (icon looks like a sphere on my machine) to the particle description, and under the Properties tab for the glyph, change the 'Scale Mode' to 'scalar', and the Glyph type to 'Sphere'. There are a lot of additional options to play with too.

Anyone else have a better method?

Chris

ckloss's picture

ckloss | Sat, 04/02/2011 - 18:28

if you have a recent version of paraview and a GPU that is up-to-date, you can render the particles as spheres on the GPU using the point sprite plug-in

Christoph

Silias | Wed, 04/20/2011 - 20:12

Hello all,

I have a similar problem: The Particles don't touch each other, but I already used a Glyph...
I simulated a 2d-problem and my Glyph-properties are like this:

scalars: radius
vectors: v
glyph Type: 2D Glyph

Glyph Type: Circle

Scale Mode: scalar
Set scale factor: 1

the rest is like default....

if it helps, here's my 2d problem:

dimension 2

atom_style granular
atom_modify map array
boundary m m p
newton off
echo log

communicate single vel yes

units si

region reg block -0.03 0.03 -0.03 0.03 -0.0002 0.0002 units box
create_box 1 reg

neighbor 0.002 bin

neigh_modify delay 0

#Material properties required for new pair styles
fix m1 all property/global youngsModulus peratomtype 30e9
fix m2 all property/global poissonsRatio peratomtype 0.25
fix m3 all property/global coefficientRestitution peratomtypepair 1 0.25
fix m4 all property/global coefficientFriction peratomtypepair 1 0.5

#New pair style
pair_style gran/hertz/history 1 0
pair_coeff * *

timestep 0.00001

fix xwalls all wall/gran/hertz/history 1 0 xplane -0.3 0.3 1
fix ywalls all wall/gran/hertz/history 1 0 yplane -0.3 0.3 1
fix gravi all gravity 9.81 vector 0.0 -1.0 0.0

#region and insertion
group nve_group region reg
fix ins nve_group pour/legacy 500 1 1 vol 0.7 1000 diam 0.03 0.03 dens 2500 2500 vel 0. 0. -0.3 region reg

#apply nve integration to all particles that are inserted as single particles
fix integr nve_group nve/sphere

#output settings, include total thermal energy
compute 1 all erotate/sphere

thermo_style custom step atoms ke c_1 vol
thermo 1000
thermo_modify lost ignore norm no
compute_modify thermo_temp dynamic yes

#insert the first particles
run 1
dump dmp all custom 350 post/dump.oW.* id type type x y z ix iy iz vx vy vz fx fy fz omegax omegay omegaz radius

run 50000 upto
unfix ins

lenard10 | Mon, 04/04/2011 - 04:26

Thank you Chris! I've already made it following your instruction. It is so nice for you to reply quickly!

rahulsoni | Wed, 10/30/2013 - 11:00

The glyph method and settings to visualize varying radius in sphere packing simulation is working. But, glyph is very bad when number of particles is large. Can any one tell similar settings to work with Point Sprite?

--
Regards

Rahul Kumar Soni
Scientist, CSIR - IMMT, India

rahulsoni | Wed, 10/30/2013 - 17:33

Paraview opens with the following message when I give command paraview --old-panels.

Regarding the supplemented HowToParaview page; the liggghts_reader and liggghts_binary_reader plugins are not available in paraview that I am running.

--
Regards

Rahul Kumar Soni
Scientist, CSIR - IMMT, India

richti83's picture

richti83 | Wed, 10/30/2013 - 18:56

should be --use-old-panels
about the readers: they are not part of Paraview, because I'm the developer and they are only usefull for liggghts time series data.
see http://techblog.richtisoft.de/?p=27 at the bottom for precompiled reader versions and this thread if you get installation trouble:
http://cfdem.dcs-computing.com/?q=node/620

I'm not an associate of DCS GmbH and not a core developer of LIGGGHTS®
ResearchGate | Contact

rahulsoni | Thu, 10/31/2013 - 06:59

Okay, so the older display is coming for paraview. The variation in radius for the packing example can be visualize with glyph option but not with the point sprite. What would be specific settings for that in point sprite?

--
Regards

Rahul Kumar Soni
Scientist, CSIR - IMMT, India

rahulsoni | Thu, 10/31/2013 - 17:13

No Richter not working in mine case. Following are mine settings:

Representation > Point Sprite
Point Sprite Mode > Sphere (Texture)
max Pix > 64
Scale by > radius
Edit radius transfer function > Free form, use scalar range, proportional, ramp

--
Regards

Rahul Kumar Soni
Scientist, CSIR - IMMT, India

richti83's picture

richti83 | Thu, 10/31/2013 - 21:11

Mhn, I remember there is/was a bug in PV that the particles does not rescale even when the diameter changes.
Could you try hitting the entry in the object inspector tree to retrigger render() in Paraview.
I've seen you asked this problem in the PV mailing list and Utkarsh told you that someone is working to recode the PS Plugin

I'm not an associate of DCS GmbH and not a core developer of LIGGGHTS®
ResearchGate | Contact

cstoltz | Fri, 11/01/2013 - 02:23

Are the particles showing up too small? I have found that it is sometimes necessary to increase the max Pix setting to get it to display properly when my particles remain too small even when changing the radius setting.

Regards,
Chris

rahulsoni | Fri, 11/01/2013 - 06:25

No changing the pixel value only increases particle size uniformly up to certain size, there is no real time visualization. Earlier when the pixel size was 64 then particles were not showing small.

--
Regards

Rahul Kumar Soni
Scientist, CSIR - IMMT, India

cgrohs | Fri, 01/10/2014 - 11:01

Using Point Sprite:
1. In my case it's also necessary to change "Scale by" to something else and then back to "radius" to show the correct particle sizes.
2. When I zoom in at some point the size of the particles does not change any more. This leads to spheres all with the same size -- and gaps in between. (ParaView bug?)

Christian