Send gmsh mailing list submissions to
gmsh@onelab.info
To subscribe or unsubscribe via the World Wide Web, visit
http://onelab.info/mailman/listinfo/gmsh
or, via email, send a message with subject or body 'help' to
gmsh-requ...@onelab.info
You can reach the person managing the list at
gmsh-ow...@onelab.info
When replying, please edit your Subject line so it is more specific
than "Re: Contents of gmsh digest..."
Today's Topics:
1. Re: separate boundary layer settings (Felix Salazar)
2. Re: Unit cell with periodic mesh (Christophe Geuzaine)
3. Re: Refining a 4th order mesh to a 2nd order mesh
(Christophe Geuzaine)
4. Re: Unit cell with periodic mesh (Sergej Tarasov)
----------------------------------------------------------------------
Message: 1
Date: Tue, 21 May 2019 14:50:03 -0400
From: Felix Salazar <felix.sala...@polymtl.ca>
To: gmsh@onelab.info
Subject: Re: [Gmsh] separate boundary layer settings
Message-ID:
<CACsb-XkyfyfQDAPikdNfipV=s-eox00ikmvkhk1nzck8936...@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Just rereading your latest message,
According to gmsh documentation, there are several ways to specify mesh size
<http://gmsh.info/doc/texinfo/gmsh.html#Specifying-mesh-element-sizes>.
Global mesh, mesh size at a point/vertex/ curvature meshing (gives you more
nodes in curves than in straight lines), and background mesh.
The boundary layer field you are using can only be properly implemented as
a mesh size indicator if used as background mesh. My guess is that in your
case is partially working, because gmsh creates and evaluates the field and
use it as background due to the command "BoundaryLayer Field = 1". But if
other field is evaluated, the previous is overwritten. That's why you can't
get the combination of your boundary layers. You can only see the cylinder
BL when you comment lines 53-61, which is where you create the 2nd BL.
I'm not sure if using a BoundaryLayer Field = {1,2} would work. In any
case, you can create the 2 fields (BL1, BL2), and then create a 3rd field
that takes the minimum of those other 2. And set the background mesh as
these 3rd field. It will only take 3 extra lines on your .geo file. Check
the example 10 to see some examples of fields and combinations
<http://gmsh.info/doc/texinfo/gmsh.html#t10_002egeo>.
Comment lines 48, 61, and add this after line 62.
Field[3] = Min;
Field[3].FieldsList = {1, 2};
Background Field = 3;
Post the results if satisfied
------------------------------------
*F?lix Salazar**felix.sala...@polymtl.ca <felix.sala...@polymtl.ca>*
------------------------------------
On Mon, May 20, 2019 at 8:58 PM Ziad Boutanios <z...@binkz.ca> wrote:
Don't worry about it. I'll take a look at background fields, but off the
bat it seems like a substantial detour for additional boundary layers,
assuming it works.
Maybe one of the developers can comment whether separate instances of
BoundaryLayer can be implemented in a mesh at all?
On 2019-05-20 3:29 p.m., Felix Salazar wrote:
Sorry for not replying sooner. I'm using my phone to read your messages
and couldn't open gmsh. But you're right, the extrusion lines were
commented. My bad.
You need to specify a background field. That field needs to somehow
integrate the effect of your 3 boundary layers. Take a look at this example
http://gmsh.info/doc/texinfo/gmsh.html#t10_002egeo
Several fields are computed, and the background field used for meshing is
the minimum of all of them
Not sure if it will work on your version of gmsh
Hope that helps
On Mon, May 20, 2019, 1:16 PM Ziad Boutanios <z...@binkz.ca> wrote:
On 2019-05-20 11:59 a.m., Ziad Boutanios wrote:
On 2019-05-20 11:26 a.m., Felix Salazar wrote:
Based on the .geo file, what you're seeing on the top and bottom surfaces
is due to the extrusions. The combination of these plus the BL fields
creates more layers than the 5 specified in the extrusions.
Try disabling (commenting) either the extrude or the BL field and remesh.
The Extrude sections in the bottom of the provided geo file are already
commented.
Around the cylinder, there's a BL. You have smaller quads there. If you
want finer control, try using an O-grid mesh
<https://www.researchgate.net/figure/O-grid-around-the-circular-cylinder_fig11_280028215>,
similar to the one on the previous link.
Thanks for the suggestion but it doesn't help me. I'm looking for a way
to implement different instances of BoundaryLayer, each with its own
settings. Any idea how this can be done?
BTW, I have to use gmsh 3.0.6 for compatibility reasons.
Forgot to mention that what you see in the original snapshot around the
cylinder is not a BoundaryLayer. That's just the unstructured quad result
of Recombine. If you comment lines 53-61 you can see what the cylinder BL
looks like. I attached a snapshot of it.
The problem is that when I use successive instances of BoundaryLayer,
only the last one in the geo file is produced. I need to have them all work
so I can customize each with its own settings.
------------------------------------
* F?lix Salazar **felix.sala...@polymtl.ca <felix.sala...@polymtl.ca>*
------------------------------------
On Sun, May 19, 2019 at 11:11 AM Ziad Boutanios <z...@binkz.ca> wrote:
Hi,
The geometry I'm meshing is that of a cylinder in a straight rectangular
channel in 2D. The geo file is attached, as well as a screenshot of the
result. I would like to use separate settings for the boundary layer
around the cylinder and at the top and bottom surfaces. To do this I
implemented two BoundaryLayer Field sections (lines 40-48 and 53-61).
Unfortunately only the boundary layer corresponding to the last section
is produced. Can that be fixed using BoundaryLayer?
Thanks,
Ziad
_______________________________________________
gmsh mailing list
gmsh@onelab.info
http://onelab.info/mailman/listinfo/gmsh
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://onelab.info/pipermail/gmsh/attachments/20190521/b630a06c/attachment-0001.html>
------------------------------
Message: 2
Date: Tue, 21 May 2019 21:00:13 +0200
From: Christophe Geuzaine <cgeuza...@uliege.be>
To: Sergej Tarasov <tara...@pmi.lv>
Cc: gmsh@onelab.info
Subject: Re: [Gmsh] Unit cell with periodic mesh
Message-ID: <1b44b8d4-dab2-4167-985d-12ab52820...@uliege.be>
Content-Type: text/plain; charset=utf-8
On 21 May 2019, at 20:19, Sergej Tarasov <tara...@pmi.lv> wrote:
Hello Christophe,
Saturday, May 18, 2019, 2:43:04 PM, you wrote:
On 17 May 2019, at 23:56, Sergej Tarasov <tara...@pmi.lv> wrote:
Hi,
I need to create cubic unit cell containing many particles crossing the walls
with periodic mesh (the geometry itself is periodic). I do it like this:
SetFactory("OpenCASCADE");
Mesh.MinimumCurvePoints = 2;
Box(1) = { 0, 0, 0, 1, 1, 1 };
Cylinder(2) = { 0.669103076524392, 0.926238630543764, 0.380445067549797,
-0.00436647701608499, -0.00549592479612254, 0.0071224075356137, 0.25, 2*Pi };
Cylinder(3) = { 0.669103076524392, -0.0737613694562362, 0.380445067549797,
-0.00436647701608499, -0.00549592479612254, 0.0071224075356137, 0.25, 2*Pi };
BooleanIntersection{ Volume{ 2:3 }; Delete; } { Volume{ 1 }; }
BooleanDifference{ Volume{ 1 }; Delete; } { Volume{ 2:3 }; }
Periodic Surface{ 21 } = { 16 } Translate{ 1,0,0 };
Periodic Surface{ 9, 19 } = { 14, 17 } Translate{ 0,1,0 };
Periodic Surface{ 18 } = { 20 } Translate{ 0,0,1 };
which works perfectly. But in case of many particles (50-100) I need
to know the ID of each opposite surface. What is the best way to do it in Gmsh?
Or there is better way to achieve this?
CG> In this case I would use
CG> e = 1e-6;
CG> left_surface() = Surface In BoundingBox{-e,-e,-e, e,1+e,1+e};
CG> right_surface() = Surface In BoundingBox{1-e,-e,-e, 1+e,1+e,1+e};
CG> Periodic Surface{ right_surface() } = { left_surface() } Translate{ 1,0,0 };
CG> etc.
CG> If there are multiple sub-surfaces, you can compare their
CG> bounding boxes (b() = BoundingBox Surface{...};) to find the
CG> corresponding pairs.
CG> Christophe
Thank you very much. The first method you suggested doesn't work with complex
shapes,
so I tried to compare bounding boxes, which works perfectly, but unfortunately
not always. It fails in one specific situation, which occurs quite often in my
case.
Below is sample code that fails to generate periodic mesh.
-------------------------------------------------------------------
SetFactory("OpenCASCADE");
Box(1) = { 0, 0, 0, 1, 1, 1 };
Cylinder(2) = { 0.58, 0.917, 0.689, -0.004387, 0.006, 0.00668, 0.1, 2*Pi };
Cylinder(3) = { 0.58, -0.083, 0.689, -0.004387, 0.006, 0.00668, 0.1, 2*Pi };
BooleanIntersection{ Volume{ 2:3 }; Delete; } { Volume{ 1 }; }
BooleanDifference{ Volume{ 1 }; Delete; } { Volume{ 2:3 }; }
Periodic Surface{ 17 } = { 15 } Translate{ 0,1,0 };
Mesh 3;
-------------------------------------------------------------------
Is there any way to avoid this? Thanks!
Indeed, it's a known limitation of the current periodic implementation:
matching curves are identified using their bounding points - and in this case
the two periodic curves have the same bounding points.
The routine to generalize is GFace::setMeshMaster(). Can you open an issue in
the gitlab so we don't forget about it?
Thanks,
Christophe
--
Best regards,
Sergej Tarasov mailto:tara...@pmi.lv
?
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science
http://www.montefiore.ulg.ac.be/~geuzaine
------------------------------
Message: 3
Date: Tue, 21 May 2019 21:00:54 +0200
From: Christophe Geuzaine <cgeuza...@uliege.be>
To: Aman Saxena <amansaxena.i...@gmail.com>
Cc: gmsh@onelab.info
Subject: Re: [Gmsh] Refining a 4th order mesh to a 2nd order mesh
Message-ID: <520f29c5-556b-40d1-9925-0b9db0fd7...@uliege.be>
Content-Type: text/plain; charset=utf-8
On 21 May 2019, at 09:23, Aman Saxena <amansaxena.i...@gmail.com> wrote:
I have a 4th order mesh(with 5 points defining the lines) and I want to refine
it twice. As I refine it once, I get a linear mesh and all other points except
for the nodes forming the elements are lost.
So when I refine it for the second time it splits and gives me points on the
linear mesh which are not on the actual boundary. If I can get a 2nd order mesh
after refining the 4th order mesh, further refinement could have given me
points on the actual boundary.
Thus how to get a 2nd order mesh from the fourth order mesh?
There's no built-in function for this. But you could code it yourself quite
easily with the Gmsh api.
Christophe
I have attached a file to demonstrate this, figure 1 shows the boundary element
of fourth order mesh while second figure shows the boundary element(linear) I
get after one refinement and the last figure shows the kind of
element(quadratic -- degree 2) I want to have after one refinement.
I have also attached the mesh file.
Thanks.
btc0-NLR-L3.v2.m4.msh.gz
<boundaryElement.pdf>_______________________________________________
gmsh mailing list
gmsh@onelab.info
http://onelab.info/mailman/listinfo/gmsh
?
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science
http://www.montefiore.ulg.ac.be/~geuzaine
------------------------------
Message: 4
Date: Tue, 21 May 2019 21:19:23 +0300
From: Sergej Tarasov <tara...@pmi.lv>
To: Christophe Geuzaine <cgeuza...@uliege.be>, gmsh@onelab.info
Subject: Re: [Gmsh] Unit cell with periodic mesh
Message-ID: <21885659.20190521211...@pmi.lv>
Content-Type: text/plain; charset=us-ascii
Hello Christophe,
Saturday, May 18, 2019, 2:43:04 PM, you wrote:
On 17 May 2019, at 23:56, Sergej Tarasov <tara...@pmi.lv> wrote:
Hi,
I need to create cubic unit cell containing many particles crossing the walls
with periodic mesh (the geometry itself is periodic). I do it like this:
SetFactory("OpenCASCADE");
Mesh.MinimumCurvePoints = 2;
Box(1) = { 0, 0, 0, 1, 1, 1 };
Cylinder(2) = { 0.669103076524392, 0.926238630543764, 0.380445067549797,
-0.00436647701608499, -0.00549592479612254, 0.0071224075356137, 0.25, 2*Pi };
Cylinder(3) = { 0.669103076524392, -0.0737613694562362, 0.380445067549797,
-0.00436647701608499, -0.00549592479612254, 0.0071224075356137, 0.25, 2*Pi };
BooleanIntersection{ Volume{ 2:3 }; Delete; } { Volume{ 1 }; }
BooleanDifference{ Volume{ 1 }; Delete; } { Volume{ 2:3 }; }
Periodic Surface{ 21 } = { 16 } Translate{ 1,0,0 };
Periodic Surface{ 9, 19 } = { 14, 17 } Translate{ 0,1,0 };
Periodic Surface{ 18 } = { 20 } Translate{ 0,0,1 };
which works perfectly. But in case of many particles (50-100) I need
to know the ID of each opposite surface. What is the best way to do it in Gmsh?
Or there is better way to achieve this?
CG> In this case I would use
CG> e = 1e-6;
CG> left_surface() = Surface In BoundingBox{-e,-e,-e, e,1+e,1+e};
CG> right_surface() = Surface In BoundingBox{1-e,-e,-e, 1+e,1+e,1+e};
CG> Periodic Surface{ right_surface() } = { left_surface() } Translate{ 1,0,0 };
CG> etc.
CG> If there are multiple sub-surfaces, you can compare their
CG> bounding boxes (b() = BoundingBox Surface{...};) to find the
CG> corresponding pairs.
CG> Christophe
Thank you very much. The first method you suggested doesn't work with complex
shapes,
so I tried to compare bounding boxes, which works perfectly, but unfortunately
not always. It fails in one specific situation, which occurs quite often in my
case.
Below is sample code that fails to generate periodic mesh.
-------------------------------------------------------------------
SetFactory("OpenCASCADE");
Box(1) = { 0, 0, 0, 1, 1, 1 };
Cylinder(2) = { 0.58, 0.917, 0.689, -0.004387, 0.006, 0.00668, 0.1, 2*Pi };
Cylinder(3) = { 0.58, -0.083, 0.689, -0.004387, 0.006, 0.00668, 0.1, 2*Pi };
BooleanIntersection{ Volume{ 2:3 }; Delete; } { Volume{ 1 }; }
BooleanDifference{ Volume{ 1 }; Delete; } { Volume{ 2:3 }; }
Periodic Surface{ 17 } = { 15 } Translate{ 0,1,0 };
Mesh 3;
-------------------------------------------------------------------
Is there any way to avoid this? Thanks!
