Re: [Yade-users] [Question #680662]: Small density sphere in micron size makes segment fault (core dumped)
Question #680662 on Yade changed: https://answers.launchpad.net/yade/+question/680662 gaoxuesong posted a new comment: Sorry to add the ubuntu and yade version used. The ubuntu is Ubuntu 18.04.1 LTS, and the yade is 2018.02b. Much thanks. -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680662]: Small density sphere in micron size makes segment fault (core dumped)
Question #680662 on Yade changed:
https://answers.launchpad.net/yade/+question/680662
gaoxuesong posted a new comment:
Hi, Jan. Yes, you are right. I have tried the above code and no segment fault
occurred. Then I add some lines which i thought were unimportant in the past
and the problems happened at about 341300 iteration. As you suggested, i leave
the snap and VTK code out. Also, to make the crash easy to happen, i amplify
the density by a small factor as 1e3.
The following is the code,
from yade import geom,pack
import math
from yade import qt
import os
#os.system('rm -rf plt;mkdir ./plt/')
#os.system('rm -rf out;mkdir ./out/')
# PhysicalParameters
muS = 0.57735 # Friction coefficient (30 degree)
muLB = 0.0001 # Friction coefficient
muF = 0.17632 # Friction coefficient (10 degree)
FricAngleS = math.atan(muS)
FricAngleLB = math.atan(muLB)
FricAngleF = math.atan(muF)
### mass amplification factor ###
masscof = 1e3
# PhysicalParameters
matSph = CohFrictMat(
density = 7990*masscof,
young = 193e9,
poisson = 0.3,
frictionAngle = FricAngleS,
momentRotationLaw = True)
SMat = O.materials.append(matSph)
# create empty sphere packing
sp = pack.SpherePack()
# generate spheres1 with a certain diameter distribution
sp.makeCloud((0,0,0.35e-3), (1.0e-3,1.0e-3,1.4e-3),
psdSizes=[0.012e-3,0.0186e-3,0.0312e-3,0.0484e-3,0.053e-3],
psdCumm=[0,0.1,0.5,0.9,1.0])
# add the sphere pack to the simulation
sp.toSimulation(material=SMat)
# create rectangular box1(Left) from facets
O.bodies.append(geom.facetBox((0.5e-3,0.5e-3,0.5e-3),(0.5e-3,0.5e-3,0.5e-3),
wallMask=31, material=SMat))
Cylinder1IDs=O.bodies.append(geom.facetBox((0.5e-3,0.5e-3,0.15e-3),(0.5e-3,0.5e-3,0.2e-3),
wallMask=63, color=(0,1,0), wire=False, material=SMat))
## add these lines, problem happens #
## add these lines, problem happens #
def changeValues():
# Standstill for stabilization
if O.time < 0.2:
O.engines = O.engines + [
TranslationEngine(dead=False, translationAxis= [0,0,1],
velocity=0.0, ids=Cylinder1IDs)
]
# Vibrate Piston1
elif 0.2 <= O.time < 0.4:
O.engines = O.engines + [
HarmonicMotionEngine(A=(0.0,0.0,1.5e-5), f=(0.0,0.0,3600/60.0),
ids=Cylinder1IDs)
]
# Vibrate Stop
elif 0.4 <= O.time < 0.5129:
O.engines = O.engines + [
HarmonicMotionEngine(A=(0.0,0.0,0.0), f=(0.0,0.0,0.0),
ids=Cylinder1IDs)
]
O.dt = 0.85*utils.PWaveTimeStep()
## add these lines, problem happens #
## add these lines, problem happens #
def print_ite():
print("ite is %d" % O.iter)
O.engines=[
ForceResetter(),
InsertionSortCollider([Bo1_Sphere_Aabb(),Bo1_Facet_Aabb()]),
InteractionLoop(
# handle sphere+sphere and facet+sphere collisions
[Ig2_Sphere_Sphere_ScGeom6D(),Ig2_Facet_Sphere_ScGeom6D()],
[Ip2_CohFrictMat_CohFrictMat_CohFrictPhys()],
[Law2_ScGeom6D_CohFrictPhys_CohesionMoment()]
),
NewtonIntegrator(damping=0.75, exactAsphericalRot=True,
gravity=(0,0,-9.81)),
PyRunner(iterPeriod=100,command='changeValues()'),
PyRunner(iterPeriod=100,command='print_ite()'),
#qt.SnapshotEngine(fileBase="./plt/",iterPeriod=200,label='snapshot'),
# VTKRecorder(iterPeriod=200, recorders=['spheres','colors'],
fileName='./out/'),
]
#show geometry:
qtr = qt.Renderer()
qtr.bgColor = [1,1,1]
qt.Controller()
v = qt.View()
#v.center()
v.axes=False
v.viewDir=Vector3(0,1,0)
v.eyePosition=Vector3(1.8e-3,-5.2e-3,1.0e-3)
O.run()
--
You received this question notification because your team yade-users is
an answer contact for Yade.
___
Mailing list: https://launchpad.net/~yade-users
Post to : yade-users@lists.launchpad.net
Unsubscribe : https://launchpad.net/~yade-users
More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680662]: Small density sphere in micron size makes segment fault (core dumped)
Question #680662 on Yade changed: https://answers.launchpad.net/yade/+question/680662 Status: Open => Needs information Jan Stránský requested more information: Hello, thanks for the code. I have tried the script for a while without any problem. Please provide more info: - At which iteration the problem occur? - is the SnapshotEngine and VTKRecorder needed to reproduce the problem? If no, please remove them to make the script a true MWE [1] - please use a 'seed' parameter to sp.makeCloud thanks Jan [1] https://yade-dem.org/wiki/Howtoask -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680334]: cohesion between polyhedral particles
Question #680334 on Yade changed: https://answers.launchpad.net/yade/+question/680334 Jan Stránský proposed the following answer: Hi, sorry for late answer. A MWE: ### O.materials.append(JCFpmMat(young=10e9,cohesion=1e20,tensileStrength=1e20)) O.engines = [ ForceResetter(), InsertionSortCollider([Bo1_Polyhedra_Aabb()]), InteractionLoop( [Ig2_Polyhedra_Polyhedra_ScGeom()], [Ip2_JCFpmMat_JCFpmMat_JCFpmPhys(cohesiveTresholdIteration=1)], [Law2_ScGeom_JCFpmPhys_JointedCohesiveFrictionalPM()], ), NewtonIntegrator(), ] def poly(x0,**kw): pts = [(x0+x,y,z) for y in (0,1) for z in (0,1) for x in (0,1)] return polyhedron(pts,**kw) ps = p1,p2,p3 = [poly(x,color=color,wire=False) for x,color in zip((0,1-1e-9,3),((1,0,0),(1,1,0),(0,0,1)))] O.bodies.append(ps) p3.state.vel = (-500,0,0) O.dt = 1e-8 ### red and yellow particles are conented with a cohesive link, the blue one is just an "impactor" ScGeom6D is not implemented, but Ig2_Plyhedra_Polyhedra_ScGeom should be a good inspiration if needed HTH Jan -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680636]: how to choose the unbalanced force ratio
Question #680636 on Yade changed: https://answers.launchpad.net/yade/+question/680636 Status: Answered => Solved procrastinator confirmed that the question is solved: Thanks Bruno Chareyre, that solved my question. -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680636]: how to choose the unbalanced force ratio
Question #680636 on Yade changed: https://answers.launchpad.net/yade/+question/680636 Bruno Chareyre proposed the following answer: Inertial number is just like unbF, it's just an indication (except that Iv is defined from input, unbF is defined from output and in that sense it's closer to what you need). It doesn't tell you in advance, quantitatively, how much non-static perturbation you get in an arbitrary problem. Also, Iv is defined only in homogeneous situations. Bruno -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680641]: Segmentation fault(core dumped)
Question #680641 on Yade changed: https://answers.launchpad.net/yade/+question/680641 Janek Kozicki posted a new comment: https://answers.launchpad.net/yade/+question/680642 -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680636]: how to choose the unbalanced force ratio
Question #680636 on Yade changed: https://answers.launchpad.net/yade/+question/680636 procrastinator posted a new comment: Hi, So we don't have something like the inertial number to determine the "unbF" thanks -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680636]: how to choose the unbalanced force ratio
Question #680636 on Yade changed: https://answers.launchpad.net/yade/+question/680636 Bruno Chareyre proposed the following answer: Hi, "unbF" is just an indication. Everything is problem dependent, and the first questions are more "which value are you interested in?" and "which tolerance do you have on the determination of that value". On this basis you may find which unbF is ok. Bruno -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
[Yade-users] [Question #680662]: Small density sphere in micron size makes segment fault (core dumped)
New question #680662 on Yade:
https://answers.launchpad.net/yade/+question/680662
In my case, the sphere used is in micron size. When i use the real material
property, especially the density, the calculation crashes after some
time(Segment fault (core dumped)). Then i try to magnify the density and find
the run goes well when the density amplified by one million times.
Following is the testing code where the parameter masscof is the density
amplification factor,
from yade import geom,pack
import math
from yade import qt
import os
os.system('rm -rf plt;mkdir ./plt/')
os.system('rm -rf out;mkdir ./out/')
# PhysicalParameters
muS = 0.57735 # Friction coefficient (30 degree)
FricAngleS = math.atan(muS)
### mass amplification factor ###
masscof = 1e6
# PhysicalParameters
matSph = CohFrictMat(
density = 7990*masscof,
young = 193e9,
poisson = 0.3,
frictionAngle = FricAngleS,
momentRotationLaw = True)
SMat = O.materials.append(matSph)
# create empty sphere packing
sp = pack.SpherePack()
# generate spheres1 with a certain diameter distribution
sp.makeCloud((0,0,0.35e-3), (1.0e-3,1.0e-3,1.4e-3),
psdSizes=[0.012e-3,0.0186e-3,0.0312e-3,0.0484e-3,0.053e-3],
psdCumm=[0,0.1,0.5,0.9,1.0])
# add the sphere pack to the simulation
sp.toSimulation(material=SMat)
# create rectangular box1(Left) from facets
O.bodies.append(geom.facetBox((0.5e-3,0.5e-3,0.5e-3),(0.5e-3,0.5e-3,0.5e-3),
wallMask=31, material=SMat))
O.bodies.append(geom.facetBox((0.5e-3,0.5e-3,0.15e-3),(0.5e-3,0.5e-3,0.2e-3),
wallMask=63, color=(0,1,0), wire=False, material=SMat))
O.dt = 0.85*utils.PWaveTimeStep()
O.engines=[
ForceResetter(),
InsertionSortCollider([Bo1_Sphere_Aabb(),Bo1_Facet_Aabb()]),
InteractionLoop(
# handle sphere+sphere and facet+sphere collisions
[Ig2_Sphere_Sphere_ScGeom6D(),Ig2_Facet_Sphere_ScGeom6D()],
[Ip2_CohFrictMat_CohFrictMat_CohFrictPhys()],
[Law2_ScGeom6D_CohFrictPhys_CohesionMoment()]
),
NewtonIntegrator(damping=0.75, exactAsphericalRot=True,
gravity=(0,0,-9.81)),
qt.SnapshotEngine(fileBase="./plt/",iterPeriod=200,label='snapshot'),
VTKRecorder(iterPeriod=200, recorders=['spheres','colors'],
fileName='./out/'),
]
#show geometry:
qtr = qt.Renderer()
qtr.bgColor = [1,1,1]
qt.Controller()
v = qt.View()
#v.center()
v.axes=False
v.viewDir=Vector3(0,1,0)
v.eyePosition=Vector3(1.8e-3,-5.2e-3,1.0e-3)
#O.run()
--
You received this question notification because your team yade-users is
an answer contact for Yade.
___
Mailing list: https://launchpad.net/~yade-users
Post to : yade-users@lists.launchpad.net
Unsubscribe : https://launchpad.net/~yade-users
More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680334]: cohesion between polyhedral particles
Question #680334 on Yade changed: https://answers.launchpad.net/yade/+question/680334 Status: Open => Answered Jan Stránský proposed the following answer: Hi, sorry for late answer. A MWE: ### ### HTH Jan -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680636]: how to choose the unbalanced force ratio
Question #680636 on Yade changed: https://answers.launchpad.net/yade/+question/680636 procrastinator posted a new comment: Hi, Yeah, I mean "how to choose the threshold value". do we have some criteria for all the simulations? or it depends on different simulations? thanks -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680641]: Segmentation fault(core dumped)
Question #680641 on Yade changed: https://answers.launchpad.net/yade/+question/680641 Jan Stránský posted a new comment: @Janek: could you please post a duplicate link? @Cloud: from atchsegv output, the problem is in O.bodies.erase. Without a complete script, we cannot tell much more.. Jan -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #679544]: Best way to fix typos in documentation: source edit + git push?
Question #679544 on Yade changed: https://answers.launchpad.net/yade/+question/679544 Janek Kozicki posted a new comment: Solution 2. is for people who prefer to have "group runners" we can provide that. -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #679544]: Best way to fix typos in documentation: source edit + git push?
Question #679544 on Yade changed: https://answers.launchpad.net/yade/+question/679544 Janek Kozicki proposed the following answer: Oh good. Now you are part of yade-dev team. Welcome :) You will push to a branch on gitlab and make merge requests. Along the way you will see what gilab-runners are doing, when you will click on a 'pipeline' on your first MR. -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
[Yade-users] [Question #680661]: How to add cohesive force between contact particles
New question #680661 on Yade: https://answers.launchpad.net/yade/+question/680661 hello i am going to use yade to simulate effect of particle shape on particle accumulation morphology, this requires different particle shapes to eventually have the same stacking patterne , but the shape of the particles affects the Angle of repose of the final accumulation.Is there a way for particles of different shapes to end up in the same heap. The method I came up with was to add cohesion between the contact particles and then subtract the final accumulation body to make the accumulation form the same shape. If you have a better method, i would be appreciate for your advice, if not , would please tell me how to add cohesive force between contact particles. thank you -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680642]: Segmentation fault(core dumped)
Question #680642 on Yade changed: https://answers.launchpad.net/yade/+question/680642 Janek Kozicki requested more information: also make sure that it has no mixed indentation (use either tabs or spaces) and use brackets ( ) for your print function. -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680642]: Segmentation fault(core dumped)
Question #680642 on Yade changed: https://answers.launchpad.net/yade/+question/680642 Status: Open => Needs information Janek Kozicki requested more information: please make sure that yuor MWE reproduces the issue. ``` File "./test.py", line 180 from yade import pack ``` SyntaxError: invalid syntax -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680653]: Boost Pointer Assertion when running yade in parallel
Question #680653 on Yade changed: https://answers.launchpad.net/yade/+question/680653 Janek Kozicki requested more information: We need a MWE to reproduce this. Not part of it. https://yade- dev.gitlab.io/trunk/user.html#questions-and-answers Also would be useful if you compile yade with -DDEBUG=1 then you could produce a complete stacktrace. Maybe you will need to fix permissions for that: https://yade- dev.gitlab.io/trunk/introduction.html?highlight=ptrace#id4 -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680432]: Can't install yadedaily in Ubuntu16.04
Question #680432 on Yade changed: https://answers.launchpad.net/yade/+question/680432 Status: Answered => Solved Xu Mengqian confirmed that the question is solved: Thanks Manfred Hampl, that solved my question. -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680653]: Boost Pointer Assertion when running yade in parallel
Question #680653 on Yade changed: https://answers.launchpad.net/yade/+question/680653 Minh Huynh requested more information: Hi Duriez, I am the one who has this problem and asked my friend to help me to post this topic. I can run YADE code with spherical packing. However, when try to make clump packing, this problem occurs when I use parallel simulation (-j4) (I have no trouble with single core). The code to generate Clump is below. Could you please help me to solve this problem?! Thank you very much for your time and consideration! Sincerely, Nhat Minh Huynh sp=pack.SpherePack() sp.makeCloud(mn,mx,psdSizes=psdSizes1,psdCumm=psdCumm1,num=num_spheres,periodic=False,seed=1,distributeMass=True) O.bodies.append([sphere(center,rad,material='spheres') for center,rad in sp]) templates=[] templates.append(clumpTemplate(relRadii=[.65,1,.65],relPositions=[[0.75,0,0],[0,0,0],[-0.75,0,0]])) O.bodies.replaceByClumps(templates,[1.0],discretization=50) -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680653]: Boost Pointer Assertion when running yade in parallel
Question #680653 on Yade changed: https://answers.launchpad.net/yade/+question/680653 Status: Open => Needs information Jérôme Duriez requested more information: Hi, Hard (impossible ?) to tell without Clump.py I guess... Note that you(r colleague) do(es) need to shorten the script at most to help us help you. Jérôme -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680636]: how to choose the unbalanced force ratio
Question #680636 on Yade changed: https://answers.launchpad.net/yade/+question/680636 Status: Open => Answered Jérôme Duriez proposed the following answer: Hi, You mean "how to choose the threshold value (for quasistatic conditions to happen)" ? As for me, I usually aim for values around 0.5 -- 1 %, and I'm quite sure this always ensures quasi staticity for my simulations (which is kind of what the doc says). I mentioned this choice of threshold e.g. in https://onlinelibrary.wiley.com/doi/full/10.1002/nag.2774, but there should be many other examples. There also are some examples and counter-examples in my PhD thesis, e.g. Fig. 2.25 and 2.26. It's for interface tests, being able to read French might help.. -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680624]: why python 3.6 when I define python2 as executable
Question #680624 on Yade changed: https://answers.launchpad.net/yade/+question/680624 Robert Caulk posted a new comment: I recommend posting this as an issue on the gitlab website [1] where they are actively working on the python 3 merge. [1]https://gitlab.com/yade-dev/trunk/issues -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #680608]: mpi4py MPI terminates Yade
Question #680608 on Yade changed: https://answers.launchpad.net/yade/+question/680608 Status: Open => Invalid Jérôme Duriez rejected the question: Duplicate of https://answers.launchpad.net/yade/+question/680609 -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
