Quantcast
Channel: Latest Discussions - COMSOL Forums
Viewing all 26512 articles
Browse latest View live
↧

Crack propagation length Extraction

January 4, 2017, 3:03 am
$
0
0
Dear All,

I'm working on a decohesion problem owed to initial stresses using the Adhesion node ( Contact Zone model implementation )

I have generally set up a working model although I would like to start extracting some useful information out of it now.

I'm looking for a way to extract the crack propagation length of my interface. I have tried picking out the maximum and minimum of solid.gap using maxop1(solid.gap,r) ( axis symmetric model) and minop1 accordingly. The issue is that as my geometry scales the evaluation of the minimum gap (however I try to extract either by the preset contact options or by a coupling operator ) gives position either too close to the edge or at 0, which at leas is not what I see in the surface plots.

Essentially I need a way to find the location where contact is lost for the first time for each simulation.

Any ideas ?

Thanks in advance!
↧
Search

undefined variable error

January 4, 2017, 3:07 am
$
0
0
Dear,

I am using General Form PDE to solve my problem. Although I defined all the initial and boundary conditions, I am having such a problem with my dependent variable, Col

Undefined variable.
- Variable: Col
- Geometry: geom1
- Domain: 1
- Feature: Time-Dependent Solver 1 (sol1/t1)

Can you please help regarding with the problem?
Many thanks,
Kevser
↧
↧

Errors in Electromechanical copuling with Nernst-Planck equation

January 4, 2017, 6:53 am
$
0
0
I tried to simulate the electromechanics of polymer-metal actuator (IPMC) with involve ions migration i.e. Nernst-Planck equation.
I made 2 components i.e. Studies (time dependent) with identity mapping. The first component consists of electric Current, Transport of diluted species and General Form PDE while the second components consists of solid mechanics only.
The problem is the Study 2 failed to execute due to this error:

Extrusion coupling operator interpolation failed for at least one point: no source point found.
- Geometry: geom2
- Boundary: 11
- Destination coordinates.: [11.57961538461539; 2.630084735069649e-017], [10.3560505690097; 4.667339610677469e-017], [12.80318020022108; 5.928298594618296e-018], [14.73884615384617; 0], [13.51528133824049; 0], ...
Failed to evaluate Jacobian of operator.
- Operator: comp1.idmap1
- Geometry: geom2
- Boundary: 11
Failed to evaluate Jacobian of expression.
- Expression: dvol*(comp2.solid.d*comp2.solid.bndl2.FAy)
Failed to evaluate Jacobian of expression.
- Expression: (comp2.solid.d*comp2.solid.bndl2.FAy*test(comp2.solid.bndl2.uy))*(dvol)

I attach my file for your reference. Please anyone could suggest solution to this problem. I really appreciate it.

Thank you.
↧

Heattransport in Compressed Air Foams

January 4, 2017, 9:14 am
$
0
0
Hello together!
First of all I want to apologize for my probably bad English and some typos! I have a (some) COMSOL problem(s) and I need to discuss this with someone. First I’ll introduce me and my project after that I’ll describe you the problem(s) I’m trying to solve, so for those who are just interested in that, just skip the introduction.

Introduction
I’m student of Safety and Hazard Management at Otto-von-Guerike-University in Germany. To get my degree I need to write my master-thesis. My education about simulation-tools is little and my experience is even less. I worked with Fire Dynamic Simulator (FDS) for half an year and heard a course about ANSYS Fluent but did not use it.

Project
The topic of my Master-theses is to simulate the heat transport in Compressed Air Foams, especialy in firefighting foams. By now I’m sure about the geometry of the foam and some physical aspect’s I must consider. So now let’s get to the COMSOL questions.

COMSOL-SetUp and Settings

Geometrie
In general foams can have very complex geometries and vary, even under the same circumstances while production. The big advantage on my side is, that Compressed Air Foams (for firefighting applications) (CAFS) are so called structured foams with a quite narrow Distribution of bubble diameter. So most foams consist of 50µm bubbles in an bcc packing. I'm encounting first dufficulties in producing the Geometrie!

COMSOL
up.picr.de/27919330df.png<-Geometrie-Tree
This is an overview over the geometric node in COMSOL. I'll go through this and ask questions where I am unsure.
First the Result: up.picr.de/27919603sw.png
I started with one bubble of the size of 50µm...
up.picr.de/27919574ak.png<- Settings of the Initial Bubble
and used the array function to achieve a bottom layer
up.picr.de/27919619et.png<- Settings of the Array
After that I made a Union with all the bubbles to acchieve on strukture.
up.picr.de/27919671lg.png<- Union Settinggs
The bubbles where abstracted from a Block.
The block simulats the water in the foam.
This results in the first Geometrie!

Question: I want no Mesh and no callculation inside the bubbles. Did I acchieve that?
Explainaction: For the beginning I negelgt the heat transfer from the water to the bubbles because of the low heat capacity of the air. This results, in my opinon, in no need for a mash inside the bubbles. The shall be Walls for the flow and thermal insualtion for the heat transfer.

Question: How can I adulterate the overlaping of three or more bubbles to one strukture? I had no success with any of the Options in the Geometrie tab.
Any comments for the Geometrie so far?

Material
Again in general, the foam consists of water and air. So I added "Water" from the material tab. See teh Result: up.picr.de/27919828jt.png
Question: Why do I have 28 Domains? Did the Union-Thing not work? And what Material do I have to define guess Domain 2-28 are the bubbles. The should just be bounderys so actually they dont need any material?

Physics
In general I'm the opinion that I'm considering an non-isothermal Problem. As I'm investigationg the heat transfer in the foam I need a heat source that must be cooled and some foam. I'n my opinion there will be just natural convection and Marangoni convection in the foam.

COMSOl
To simulat the model in added "Laminar Flow" and " Heat transfer in fluid" to the physics and conected them with the Non-Isothermal flow multiphysics and added marangoni convection.
up.picr.de/27919916gn.png<- Non-Isothermal Flow
up.picr.de/27919917no.png<- Marangoni-Effect

The Laminar Flow Node is Set up like this:
up.picr.de/27919959lt.png
up.picr.de/27919960iu.png
up.picr.de/27919961gw.png
up.picr.de/27919962lt.png

Question: Which BC's should I probebly apply? On the sids I would use Wall/No Slip. Top Side is the big question! PLZ Help! At the bubbles I woud use Interior Wall?


Heat transfer node

The Setup of the Heat transfer in fluids Node
up.picr.de/27920045ry.png
up.picr.de/27920046cl.png
up.picr.de/27920047tt.png
up.picr.de/27920049ox.png

Question: Again I would apply "thermal insulation" to the bubble strcture and the Walls. The question is again the top side! There should be Environment. How do I model this?


Okay this so far. Please comment to the Setup and help me with my questions. After I completed the Setup I'll deploy the file here and would like to discusse about the solver. But for now ist late in Germany and I'm tired. I'll correct the typos tomorow.

Thanks and PLEASE help
so far Fabian!





↧

Angle of incidence in Wave optics module

January 4, 2017, 9:57 am
$
0
0
Hello,

I usually work on Wave optic module in COMSOL, the issue I have been facing is the to incident an electromagnetic wave at a certain angle.
Is it possible to set the electromagnetic wave at a particular angle of incidence. For an example I want the wave to be incident at 65 deg, how can this be achieved.

I will be highly obliged for your kind suggestions and help.

Thank you!!
↧
↧

Why the same program sometime need 4000 secs and sometimes 300 secs

January 4, 2017, 11:01 am
$
0
0
May I ask if anyone know why the same comsol program sometime need 4000 secs and sometimes 300 secs to solve? Exactly the same one, I didn't change any parameters, mesh, solver, just run it
↧

Filled Tables Re-evaluation

January 4, 2017, 11:08 am
$
0
0
After re-running a study that contains a parametric sweep; reevaluating tables clears the table settings. Converting the table settings to a filled table allows me to plot the table data the way I want in a table graph. Is there a way to keep the table settings while re-evaluating the data from the updated study? More broadly, is there a way to update all results with new study solutions without having to reassign the data set for each plot group?

Thank you,
-Matt
↧

how to plot fillers in matrix

November 15, 2016, 7:59 am
$
0
0
My question is about constructing a geometry where there are a box full of platelets ( disks in 2D)
my question is : should i build an algorithm to get the right distribution of platelets in my box or i can build my geometry without problems?
Thank you
↧
Search

Temprature effect on piezoresistive pressure sensor

January 5, 2017, 1:44 am
$
0
0
now i am working on effect of temperature on piezoresistive pressure sensor.please anyone tell me where to apply heat flux and heat source in piezoresistive model using comsol multiphysics
↧
↧

Importing 3D geometry into COMSOL

January 2, 2017, 2:26 am
$
0
0
Hello,

Happy new year everyone!

I need help using this software. I want to know how to select my flow area/domain , which needs fluid to flow through my imported complex geometry(say from AUTOCAD).

The geometry is something like a complex pipe with a specific thickness. So when i import the geometry from AUTOCAD to COMSOL, the fluid is flowing through the thickness of my solid geomtery instead of flowing through it. Can anyone help me with this? how do i select the inner area fir my fluid flow/properties.

I need the flow to flow through the inner part, instead of the thickness of the solid geometry.

Can someone please help me with it?

Thanks in Advance
↧

busbar effets

January 5, 2017, 3:06 am
$
0
0
hi,

I would like to know how we model the mutual effects between several conductors (example of the busbar).

Thank you for your help.
↧

Modelling of a Hydrocyclone

January 5, 2017, 3:41 am
$
0
0
Hello,

I'm modelling a hydrocyclone for solid-liquid seperation.

I have two inlet pipes placed opposite to each other at the bottom of a cylinder, water is pumped with a constant velocity, it swirls in the cylinder to reach an outlet placed horizontally at the top of the cylinder. The pressure at the outlet is assigned to zero as well.
I modeled the liquid flow by using Laminar flow and the study is stationary and I have added the particle trajectory interface to model the solid particles with a time dependent study.
According to my supervisor the flow of liquid shouldn't occupy all the horizontal outlet and should fill only half of it, I have assigned everything and every possible boundary condition, the problem is that I can't assign the gravity force (F=-g_const*rho), when I add this term the solution is not converging.
I have tried almost everything but still no solution.

Any ideas how can I solve these problems?

Thank you in advance
↧

Cathode aging in Li-ion battery

January 5, 2017, 4:11 am
$
0
0
Please advise if there are any tutorials/models on how to simulate cathode material aging (capacity fade) in Li-ion battery. Thank you!
↧
↧

Slope Stability Analysis Using COMSOL

January 5, 2017, 4:12 am
$
0
0
Hello, I am new to COMSOL. Has anyone here used COMSOL for slope stability analysis problem?. Pls kindly share your model or tips on the issue.
↧

How to? Inelastic wall bounce in particle tracing

December 27, 2016, 9:04 am
$
0
0
I would like walls in my simulation to produce a bounce effect on particles, but with reduced velocity coming off the wall. Ideally this effect would be velocity dependent, though I would be satisfied with particles bouncing with a set fraction of the velocity.

Currently the only option seems to be to use the "General reflection" wall condition, but it is unclear how to make use of the tangential and normal velocity components to produce a damped bounce. The only evidently available wall variables are fpt.wall1.vprimx, vprimy, and vprimz, but using them (with a fraction) results in a circular reference error. I cannot use the particle velocity components because there is no mechanism for extracting the normal vector of the wall and thus producing a reflection vector.

Is there a way to achieve this effect?
↧
Search

Residual stress in Butt welding

February 5, 2016, 6:20 am
$
0
0
Dear friends


Can any one what are the require structural boundary condition for butt welding of thin plates? And also how can I give the partuclar time step from heat transfer study to structural mechanics module?


Thanks
Dhinakaran V

↧

Advection and rotation of tensor property in fluid flow

December 5, 2016, 8:33 am
$
0
0
Hi,
I would like to use the mathematics module to define a tensor property which is advected and rotated by the flow of a fluid (Fluid flow, single-phase creeping flow). Physically, I wish to model the average alignment and orientation of small crystals at the sub-continuum scale. Since Comsol already uses tensor quantities, I wonder whether such behavior is already implemented.

I can use the PDE interface in coefficient form to define a field with 6 entries (symmetric tensor in 3D). And advection is pre-defined in the interface. But how about rotations? Would I have to define them myself?

(I thought I would check the Heat Transfer module for help; but apparently the thermal diffusivity is neither advected nor rotated along with the flow).
↧
↧

Field Emission of Carbon Nanotubes - Screening Effect and Electric field enhancement

January 5, 2017, 8:59 am
$
0
0
Hello all,

I am an absolute newbie in COMSOL. I have trying to model the field enhancement of Carbon Nanotubes (CNT) in a field emission process. The outcomes that I was anticipated is similar to these:

www.comsol.com/paper/optimizat...ube-field-emission-arrays-6288

nanoscalereslett.springeropen....ticles/10.1186/1556-276X-8-319

I started with a simple model consist of substrate-CNT cathode-anode as shown in the attached picture. I used Electrostatic study to compute the electric field distribution of the model. I have input potential of -100V at the substrate and grounded the anode. However, the outcome that I have obtained deviates from the anticipation. No field enhancement at the CNT tip and the electrical field is the same along the the CNT. At the same time, I have attempted to build a few more CNT around the initial CNT (as shown in the url link). The electrical field at the CNT tip(s) is identical despite single CNT/multiple CNT stays closely together.

Did I miss any steps in performing the simulation? You help is really appreciated. Thank you!
↧

Failed to evaluate variables and Jacobian variables

January 5, 2017, 9:02 am
$
0
0
Hi COMSOL Users,

I an new to using COMSOL 5.2a and perhaps because of that I am missing something simple. I have tried to model evaporation of water, which is part of modeling a new toilet technology we are proposing for developing countries, but I get several errors when trying to model the second study in the toilet system.

The problem that noticed is due to the domain number . Once it is removed from the geometry, the second study is done successfully, while adding this domain results in the following errors:

Undefined variable.
- Variable: comp1.c
- Geometry: geom1
- Domain: 2
Failed to evaluate Jacobian of expression.
- Expression: comp1.c
Failed to evaluate Jacobian of operator.
- Operator: mean
- Geometry: geom1
- Boundary: 9
Failed to evaluate Jacobian of variable.
- Variable: comp1.c
- Geometry: geom1
- Boundary: 9
Failed to evaluate Jacobian of expression.
- Expression: dvol*(r*(pi*((2*comp1.ht.bhs1.Qb)*comp1.ht.d)))
Failed to evaluate Jacobian of expression.
- Expression: (2*comp1.ht.bhs1.Qb*test(comp1.ht.bhs1.Tvar)*comp1.ht.d*pi*r)*(dvol)


I have attached my model to this post just in case.

Has anyone had problems like the above ones? How can fox these issues?

Thanks in advance,
Babak
↧

MATLAB LiveLink

January 5, 2017, 9:14 am
$
0
0
Hey there,

I am using LiveLink MATLAB but when running my model, the notice "Undefined function or variable 'mphopen'" is emerged on the command window. However, I have set up the file "domain_activation_llmatlab" in the MATLAB path. could it be some problem related to LiveLinke? How could I find it that my COMSOL 5.2a can LiveLink to the MATLAM R2016a.
this is one of the comsol's sample problems (www.comsol.com/model/domain-ac...ivation-and-deactivation-12060). I, however, cannot do this. what do you suggest. the MATLAB code is as follows:
------------------
model = mphopen('domain_activation_llmatlab');
domInd = [2,3,5,4];
ht = model.physics('ht');
for i = 1:8
k = mod(i,4);
if k == 0
k = 4;
end
ht.selection.set([1 domInd(k)]);
ht.feature('init2').selection.set(domInd(k));
model.study('std1').run;
if i==1
cpt1 = model.result.dataset.create('cpt1', 'CutPoint3D');
cpt1.set('pointx', '0 L/2 L');
cpt1.set('pointy', '0 L/2 L');
cpt1.set('pointz', 'L/10');
pg1 = model.result.create('pg1', 'PlotGroup1D');
pg1.set('data', 'cpt1');
ptgr1 = pg1.feature.create('ptgr1', 'PointGraph');
ptgr1.set('legend', 'on');
pg2 = model.result.create('pg2', 'PlotGroup3D');
surf1 = pg2.feature.create('surf1', 'Surface');
surf1.set('rangecoloractive', 'on');
surf1.set('rangecolormax', '336');
surf1.set('rangecolormin', '293.15');
ht.feature('init1').set('T', 1, 'T');
v1 = model.sol('sol1').feature('v1');
v1.set('initsol', 'sol1');
end
figure(1)
mphplot(model,'pg1','rangenum',1)
hold on
figure(2)
subplot(4,2,i)
pg2.setIndex('looplevel','25',0);
mphplot(model,'pg2');
time = mphglobal(model,'t','solnum','end');
model.param.set('t0',time);
disp(sprintf('End of iteration No.%d',i));
end
----------------------
↧
Viewing all 26512 articles
Browse latest View live
Search