Peter,
To make sure you have the correct trimmed surface I think the easiest is simply
construct your "subvolume" by starting from the solid. Here's a modified
version of example that uses that approach.
Christophe
import gmsh
import sys
import numpy as np
# set up parameters (eventually will read from file).
th_l = 10.0
th_u = 40.0
phioff = 160.0
phi = 20.0
infrac = 0.0
outfrac = 0.7
thickness = 1.0
radius = 1.0
# create rectangular coords of point from r, theta, phi.
def polars(rad,th,phi):
x = rad*np.cos(np.radians(th))*np.cos(np.radians(phi))
y = rad*np.cos(np.radians(th))*np.sin(np.radians(phi))
z = rad*np.sin(np.radians(th))
return x,y,z
# start of main code
gmsh.initialize(sys.argv)
gmsh.option.setNumber("General.Terminal", 1)
gmsh.model.add("sphere_cut")
sph1 = gmsh.model.occ.addSphere(0,0,0, radius, -1, 0, np.pi/2, 2*np.pi)
sph2 = gmsh.model.occ.addSphere(0,0,0, radius+thickness, -1, 0, np.pi/2, 2*np.pi)
sph3 = gmsh.model.occ.addSphere(0,0,0, radius+outfrac*thickness, -1, 0, np.pi/2, 2*np.pi)
outf, _ = gmsh.model.occ.cut([(3,sph3)], [(3,sph1)], -1, True, False)
origin = gmsh.model.occ.addPoint(0.,0.,0.,0,-1)
x,y,z = polars(radius+thickness,th_l,phioff-phi)
pll = gmsh.model.occ.addPoint(x,y,z,0,-1)
x,y,z = polars(radius+thickness,th_u,phioff-phi)
pul = gmsh.model.occ.addPoint(x,y,z,0,-1)
x,y,z = polars(radius+thickness,th_u,phioff+phi)
pur = gmsh.model.occ.addPoint(x,y,z,0,-1)
x,y,z = polars(radius+thickness,th_l,phioff+phi)
plr = gmsh.model.occ.addPoint(x,y,z,0,-1)
c12 = gmsh.model.occ.addLine(pll,pul,-1)
c22 = gmsh.model.occ.addLine(pul,pur,-1)
c32 = gmsh.model.occ.addLine(pur,plr,-1)
c42 = gmsh.model.occ.addLine(plr,pll,-1)
c1 = gmsh.model.occ.addLine(origin,pul,-1)
c2 = gmsh.model.occ.addLine(origin,pur,-1)
c3 = gmsh.model.occ.addLine(origin,plr,-1)
c4 = gmsh.model.occ.addLine(origin,pll,-1)
cl = gmsh.model.occ.addCurveLoop([c12,c22,c32,c42],-1)
cl1 = gmsh.model.occ.addCurveLoop([c12,c1,c4],-1)
cl2 = gmsh.model.occ.addCurveLoop([c22,c2,c1],-1)
cl3 = gmsh.model.occ.addCurveLoop([c32,c3,c2],-1)
cl4 = gmsh.model.occ.addCurveLoop([c42,c4,c3],-1)
s1 = gmsh.model.occ.addPlaneSurface([cl],-1)
s2 = gmsh.model.occ.addPlaneSurface([cl1],-1)
s3 = gmsh.model.occ.addPlaneSurface([cl2],-1)
s4 = gmsh.model.occ.addPlaneSurface([cl3],-1)
s5 = gmsh.model.occ.addPlaneSurface([cl4],-1)
sl = gmsh.model.occ.addSurfaceLoop([s1,s2,s3,s4,s5],-1)
v = gmsh.model.occ.addVolume([sl],-1)
stuff, _ = gmsh.model.occ.intersect([outf], [(3,v)])
gmsh.model.occ.fragment([(3,sph1),(3,sph2)], [stuff])
gmsh.model.occ.synchronize()
gmsh.fltk.run()
gmsh.finalize()
> On 30 Apr 2019, at 07:59, Peter Johnston <[email protected]> wrote:
>
> Dear Max,
>
> Thanks for your thoughts. However, I am not sure how this helps as the
> spheres are no longer concentric.
>
> I have made another example script, this time in python, that shows the issue
> that I keep coming up against. If you run the script, you will see that
> surfaces 10 and 13 have the same set of bounding curves: 6, 7, 8 and 9, but
> one is a bspline surface and the other is a spherical surface. Perhaps I
> simply need to delete one of these?
>
> Regards,
>
> Peter.
>
> ---------------------------------------------------------------------
>
> Associate Professor Peter Johnston (FAustMS, FIMA)
> School of Environment and Science
> Griffith University | Nathan | QLD 4111 | Technology (N44) Room 3.19
> T +61 7 373 57748| F +61 7 373 57656 Email [email protected]
> On 29 Apr 2019, 6:41 AM +1000, Max Orok <[email protected]>, wrote:
>> Hello Peter,
>>
>> I took a bit of a different approach and hopefully it is similar to what
>> you'd like to do.
>> This short script makes a "sphere patch" by subtracting one sphere from
>> another.
>>
>> SetFactory("OpenCASCADE");
>>
>> // arbitrary angle selections here
>> Sphere(1) = {0, 0, 0, 0.5, -Pi/9, Pi/9, Pi/4};
>> Sphere(2) = {-.25, 0, 0, 0.5, -Pi/9, Pi/9, Pi/4}; // shift sphere 2
>>
>> // use sphere 2 to cut sphere 1
>> BooleanDifference{ Volume{1}; Delete; }{ Volume{2}; Delete; }
>>
>> <image.png>
>> <image.png>
>>
>> I hope this is helpful! If I missed the point please let me know.
>>
>> Sincerely,
>> Max
>>
>> On Wed, Apr 24, 2019 at 7:22 PM Peter Johnston <[email protected]>
>> wrote:
>> Thanks Max and Christophe,
>>
>> I have attached a simple sample .geo file. For some reason it creates even
>> more volumes than the more complicated script I was working on. However, it
>> still displays similar problems.
>>
>> If you have any further questions, please let me know.
>>
>> Thanks again,
>>
>> Peter.
>>
>> ---------------------------------------------------------------------
>>
>> Associate Professor Peter Johnston (FAustMS, FIMA)
>> School of Environment and Science
>> Griffith University | Nathan | QLD 4111 | Technology (N44) Room 3.19
>> T +61 7 373 57748| F +61 7 373 57656 Email [email protected]
>> On 25 Apr 2019, 1:36 AM +1000, Max Orok <[email protected]>, wrote:
>>> Hi Peter,
>>>
>>> Scripts are always helpful! It is nice to be able to see the steps as you
>>> go along.
>>>
>>> Max
>>>
>>> On Wed, Apr 24, 2019 at 8:36 AM Peter Johnston <[email protected]>
>>> wrote:
>>> Hello,
>>>
>>> I appear to have a problem for which I cannot figure out a solution.
>>>
>>> I have an annular hemispherical volume created between to hemispheres
>>> centred on the origin. I would like to divide the volume into two parts in
>>> a special way using Boolean operations. To define a sub-volume of the
>>> initial volume by defining four points on the surface of the inner sphere,
>>> create a “quadrilateral” on the inner surface by joining the four points in
>>> order using circles through the origin. (I think that these should be parts
>>> of great circles on the inner surface?). Then I create a line loop and a
>>> surface, and finally extrude the surface away from the origin to create a
>>> volume. Perhaps a .geo script would help here?
>>>
>>> When I perform BooleanDifferences and BooleanIntersections I end up with
>>> three volumes instead of two. The reason I get the third volume is that the
>>> surface patch that I create does not fit exactly onto the original
>>> spherical surface. Is there any way that I can force the patch to be part
>>> of the spherical surface?
>>>
>>> A consequence of the extra volume is that I cannot create a sensible mesh.
>>>
>>> Any ideas would be greatly appreciated.
>>>
>>> Thanks very much,
>>>
>>> Peter.
>>>
>>> PS, I could send a simple script file if that would help.
>>>
>>> ---------------------------------------------------------------------
>>>
>>> Associate Professor Peter Johnston (FAustMS, FIMA)
>>> School of Environment and Science
>>> Griffith University | Nathan | QLD 4111 | Technology (N44) Room 3.19
>>> T +61 7 373 57748| F +61 7 373 57656 Email [email protected]
>>> _______________________________________________
>>> gmsh mailing list
>>> [email protected]
>>> http://onelab.info/mailman/listinfo/gmsh
>>>
>>>
>>> --
>>> Max Orok
>>> Contractor
>>> www.mevex.com
>>>
>>>
>>
>>
>> --
>> Max Orok
>> Contractor
>> www.mevex.com
>>
>>
> <sample.py>_______________________________________________
> gmsh mailing list
> [email protected]
> http://onelab.info/mailman/listinfo/gmsh
—
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science
http://www.montefiore.ulg.ac.be/~geuzaine
_______________________________________________
gmsh mailing list
[email protected]
http://onelab.info/mailman/listinfo/gmsh
