Quantcast
Channel: Latest Discussions - COMSOL Forums
Viewing all 26527 articles
Browse latest View live
↧

Same model but Convergence Problem

February 21, 2013, 4:17 am
$
0
0
Hi

I have a model (using solid mechanics and heat) which is setup in 4.2a and I get excellent convergence. I have alos setup the exact model in 4.3a no changes been made in any boundary conditons or anything else. Why my solution is not converging in 4.3a? I have to manually cancel the calculation everytime because it doesn't converges to my prescribed relative tolerance. In both models I have used automatic scaling.

Can some one please explain what could be the reason.

↧
Search

How to see the plots after running in batch mode?

February 21, 2013, 4:27 am
$
0
0
Hello!
I managed to get comsol to run in batch mode (Linux). However, after I open my newly created output model file out.mph, I do not see any graphs available. I am solely interested in CFD module, where comsol usually builds Velocity and Pressure plots if run with GUI. I want to have access to those after I open out.mph with Comsol GUI again.

I tried to run my computation command with -graphics switch, but nothing changed.
Is there a way to get those plots, with animation and all that stuff?
If not, where are the output files and how to interpret them? Any computation is useless unless you can access some sort of output.

Any help is greatly appreciated!
↧
↧

Use of Ports to calculate the Transmittance and Reflectance of an array of resonators

February 21, 2013, 5:34 am
$
0
0
Hi everybody,

+Problem: I want to simulate an square array of gold nanorings, in order to get the graphs for the transmittance and the reflectance and find the resonant frequency.

+My first approach:
-One box (Air) with one ring in the center (Au).
-A regular Port is ON on the top of the box (E=1,0,0) so it launches plane waves along Z. The waves will pass through the hole of the ring.
-Another regular Port is OFF on the bottom part of the box.
-The lateral walls of the box of Air parallel to X are PMC the perpendicular ones are PEC.
-I made a 1DPlot of
Reflectance= abs(emw.S11)^2
and
Transmittance=abs(emw.S21)^2
versus frequency

+Questions:
-How tall should I make my box to get an accurate result?

-I've read some post about people simulating similar structures by adding two PMLs (one at the top, one at the bottom of the box of Air). These people use a surface current to generate the plane waves. Is this method better/more accurate than what I'm doing?

Thanks in advance for your help!

Ramses
↧

how to draw a rectangle on a 3D cylinder's surface ?

February 21, 2013, 5:46 am
$
0
0
hey !
we are working on the simulation of MEMS Based pressure sensor for biomedical applications. We are not able to create the heater element which will heat the liquid inside the cylinder. Kindly help us in proceeding with the project.
↧

mode analysis for 2D slab waveguide

February 21, 2013, 6:17 am
$
0
0
Hi, Everyone,

I am new for using comsol, and firstly I built a easy model to learn.
It is a 2D slab waveguide (see attached ).

By mode analysis, I found the effective mode index obtained by comsol is different with the value obtained by analytical solution.
In comsol mode, the infinite direction is set as PBC with continuity.

I appreciate you can give me some help.

Regards
Marvin
↧
↧

2D hinge in compliant mechanism

February 21, 2013, 8:21 am
$
0
0
Hi all,

I am trying to simulate a compliant mechanism as seen in the attached picture but I am having difficulties simulating the rotating pin in 2D. As see in the picture the structure is composed of a flexing member and two rigid members connected by a pin. I made the geometry into an assembly instead of a union, the rotating pins as rigid connectors and the rigid bars as rigid domains. My question is how to model this rotating pin, I saw an example in 3D but I am not sure how to do this in 2D. Let me know what you guys think, I appreciate your help.


Thanks

Oscar

↧

Not able to properly use remesh feature in multiphysics stationary simulation

February 21, 2013, 10:50 am
$
0
0
Hi,

I have simulated a steel specimen that becomes deformed due to bending. For that, I use the solid mechanics interface in a stationary study. Parametrizing the magnitude of bending also works. Overall, it works as expected.

I have also simulated an electromagnet by using the magnetic fields interface and specifying a DC external current density in a stationary study. That also works well.

I have attempted to include both of them in the same model, where I want the specimen to be deformed, remeshed, and that new mesh be used in the magnetic fields stationary study. Oh, and I want to parametrize the magnitude of bending, such that the sequence of operations is:

1. Deform 1 mm relative to rest position (calculate from initial mesh)
2. Remesh
3. Calculate DC magnetic fields (using remeshed geometry )

4. Deform 2 mm relative to rest position (calculate from initial mesh)
5. Remesh
6. Calculate DC magnetic fields using remeshed geometry

and so on...

So far, I have managed to remesh by using ALE and the model works without errors. However, I have a hunch that the magnetic fields interface only takes into account the initial mesh. As a result, the magnetic field distribution remains the same, and the only reason why it follows the shape of the deformed specimen in the plot, is because I have added "Deformation" as an addition to the plot. Could this be the case, what do you think?

I have been struggling with this for quite a while. Any ideas/suggestions would be highly appreciated. Thanks in advance for your time,

Kind Regards,
Orfeas
↧

stress corrosion cracking

February 21, 2013, 10:52 am
$
0
0
Is it possible to be stress corrosion cracking? The steel structure has tensile stresses and would be susceptible to stress corrosion cracking.

Thanks

--
Arthur Rupel
↧
Search

"comsol batch" crashed with Java errors

August 25, 2010, 8:04 pm
$
0
0
Hi,

I am trying to run a CFD model using the batch mode but crashed with some Java errors. The same model runs fine using the GUI (comsol client) and a Comsol Server.

OS: SLES 11
COMSOL version: 4.0a

The command I used:

comsol batch -inputfile in.mph -outputfile out.mph

The result:

...
Running: Transient run (turn on weak contributions)
Time-dependent solver (BDF)
Number of degrees of freedom solved for: 24144.
Current Progress: 0 %
Memory: 561/561 2072/2077
Nonsymmetric matrix found.
Step Time Stepsize Res Jac Sol Order Tfail NLfail
0 0 out 5 3 5 0
Exception:
com.comsol.util.exceptions.FlException: The following feature has encountered a problem
Messages:
The following feature has encountered a problem
- Feature: Time-Dependent Solver 1 (sol6/t1)
- Error: Could not find the service 'com.comsol.nativeutil.h'

Stack trace:
at com.comsol.solver.SolverOperation.addError(Unknown Source)
at com.comsol.solver.SolverOperation.execute(Unknown Source)
at com.comsol.model.impl.SolverSequenceImpl.b(Unknown Source)
at com.comsol.model.impl.SolverSequenceImpl.f(Unknown Source)
at com.comsol.model.impl.SolverSequenceImpl$c.a(Unknown Source)
at com.comsol.model.impl.SolverSequenceImpl$c.execute(Unknown Source)
at com.comsol.model.clientserver.ClientManagerImpl$d.call(Unknown Source)
at java.util.concurrent.FutureTask$Sync.innerRun(Unknown Source)
at java.util.concurrent.FutureTask.run(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(Unknown Source)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
at java.lang.Thread.run(Unknown Source)
Total time: 27.035 s.

Anyone knows what the "Could not find the service 'com.comsol.nativeutil.h'" error refers to and how to fix it?

Tim
↧
↧

SImulation of an electrostatically actuated beam

March 3, 2011, 1:31 am
$
0
0
Hi,

I am new to COMSOL and presently trying to simulate the pull-in instability of electrostatically actuated micro beams in COMSOL MEMS module. I need further insights into the following topics pertaining to this problem:

1) Has anyone benchmarked some typical beam problem so that one can have some idea about what is the optimum number of elements; say along beam-length, width, thickness? I have followed the tutorial problem in COMSOL and the results do match with the literature, but just checking if someone has already tried this out.

2) How do we decide the optimum dimensions of the surrounding box (air chamber) in which the electrostatic problem is solved. I have observed that making it too small changes the results substantially and making it big increases the number of elements in the electrostatic domain, and hence the computational effort.

3) How do we ensure that the charge distribution over the surface of the beam is accurately captured? I feel that turning on the weak constraints would help, however, just curious if somebody is having a better understanding and/or previous insights from their own numerical experiments.

Thanks in anticipation for your time and valuable inputs.


Dekar
↧

Heat conduction in solids - temperature dependent heat transfer coefficient

February 4, 2013, 4:31 pm
$
0
0
Hi,

I would like to solve a simple heat conduction in solids problem with a temperature-dependent heat transfer coefficient at one of the boundaries

h(T) = 1000*(300/T)^1.3

However, the result computed by Comsol is different than I would expect from analytical techniques and an equivalent Comsol model with a solid layer of very low thermal conductivity.

Are there any issues I should be aware of when using temperature-dependent heat transfer coefficients?

Thanks,
Kevin
↧

Continuing Modeling Issues with 2D & 3D piezoelectric SAW model

February 7, 2013, 6:22 am
$
0
0
Hello all,
I’m in desperate need of help with my model. The initial model setup is a 2D cross-sectional SAW device where I have silicon, a metallic electrode as my ground, a buffer layer, a piezoelectric layer (PZT-5), and gold electrodes. The current physics I have defined is with piezoelectric devices (with COMSOL version 4.2.0) with the silicon and buffer/protective layer (SrTiO3) as my linear elastic material model, the metallic electrodes and metallic layer as my electrical material layer, and piezoelectric material model with the PZT. I have defined parameters most of the materials in terms of elastic matrix D, Young’s Modulus, Bulk modulus, permittivity, and so forth. My main issue is when I run a frequency domain analysis; I consistently get the error of:
Failed to find a solution for the initial parameter. Singular matrix. There are 3906 void equations (empty rows in matrix) for…etc.
I have looked through the discussion forum consistently to see why I continue to get the errors I received and I thought it was simply a matter of fixing my boundary conditions but I thought I had accounted for all constraints for this simplified model. I do not know at this point where I have been going wrong for the past couple of months now. I’m attaching the current model. Any suggestions or comments would be greatly appreciated. Thank you.


--
Sarah
↧

Plotting the gap distance variable

February 14, 2013, 4:53 pm
$
0
0
I am trying to do something very simple, but it doesn't seem to be intuitive. I have a contact pair in the structural mechanics module and I want to plot the gap distance variable as a function of time. I know this is should be simple, but I am stuck!

Ryan
↧
↧

magnetic shielding with mumetal(nonlinear - saturation)

February 18, 2013, 5:21 am
$
0
0
Hi
I need some help with my model....
I have a 3D sim, A current line that generates my field and a mumetal infinite plane(1m*1m*0.5mm)
The 0.5mm is the thickness of my mumetal. I have 0.5T on the surface of the shield and i want to see the B behind the shield.
a.I use relative permeability constant at 7000.
b. I use a non-linear equation with saturation for relative permeability
μr= 1+[(μrmax-1)^(-1) + (μ0*/Η/ / B_sat)^(1/n)] ^(-1)
where mu_rmax=7000, B_sat=0.7T and n=1
c. I use the B-H and H-B curves filling them with my data curves.
Am I correct in any way of these models or I need some refinements to be more precise;
I can see the saturation and the B-H curve; How we can change the scales from a plot in comsol;
Also I attach some photos of my model.
Thanks in advance.
↧

Calling variable or convert T(x,y,z) = >T(x,y,t) between models

February 18, 2013, 11:23 am
$
0
0
Hi: Please help! I'm very stuck with a quite straightforward problem.

I've got two models:

1) Model 1: 3D Heat Transfer Stationary
2) Model 2: 2D Teat Transfer Time Dependent.

How can I call T(x,y,z) from the model 1 to T(x,y,t) in the model 2 If I want, for example, a source temporal variable Q(x,y,t) with the same shape. i.e. Q(x,y,t)=10*T(x,y,z=t)? The problem is how to change the z axis for the "t" temporal coordinate. How to input a source like this?

Thanks in advance.

↧
Search

How to set displacement from study 1 as prescribed displacement in study 2?

February 20, 2013, 4:31 am
$
0
0
Hey,

I have a trouble with defining a prescribed displacement.

Let's say that i received displacement of the edge caused by heating of sample. But displacement of this edge is different in every point of its length.

I want to keep this displacement as prescibed displacement condition in next study. So i can't just add condition prescibed diplacement in y=a, becouse it's more like ax*bx^2.

I tried to use mod1.u as prescribed in y, but it's not working or I doing something wrong.

How can I solve this problem?

Kind Regards

Robert
↧

Importing real data heat flux boundary condition to COMSOL

February 20, 2013, 7:28 pm
$
0
0
Hello all,
I'm working on a 2D transient heat transfer project and I can't figure out how to import real data for heat flux vs.time in COMSOL. It only allow me to input a fix value or function, not discrete data points.
Any idea?
Thank you very much,

Vinh
↧
↧

eddy current

February 20, 2013, 9:43 pm
$
0
0
how to get the imaginary component (phase) in the inductor in the high-frequency alternating magnetic field?
Comsol 4.2a
Help me please
↧

Local maxima of solution

February 21, 2013, 3:45 pm
$
0
0
Dear all,
I use comsol to solving a certain EDP problem (Laplace equation for exemple in crown domain: 2D or 3D domain). The solution is noted 'u'.
I want to calculate the all local maxima of the solution 'u' but I dont know the convenable command comsol can I use.
Please does any one can help me.
Sincerly
Moncef Mahjoub
↧

"No frame is defined for 0D applications"

February 22, 2013, 2:58 am
$
0
0
Hello again.

I am trying to simulate an E-core transformer. When I click the study 1 and select the show default solver there is an error occurs "No frame is defined for 0D applications" . What is it mean?

Thanks

Asli
↧
Viewing all 26527 articles
Browse latest View live
Search