Quantcast
Channel: Latest Discussions - COMSOL Forums
Viewing all 26323 articles
Browse latest View live

Iterative solver, convergence plot outrunning iteration number shown in Progress tab

$
0
0

Hello everyone,

I have been solving some 3D RF Frequency domain problems recently, using the the default FGMRES solver.

I just encountered a weird problem: the solver iteration number shown by the Progress tab lags behind the iteration number shown on the convergence plot. When the convergence plot shows that the solver has reached the desired tolerance, I can see that the resource usage by COMSOL drops close to zero, consistent with the fact that the solver has stopped. However, the Progress tab still shows a residual error above tolerance, and I have to wait for it to catch up to the convergence plot. As I have to do many such 3D calculations, this extra time is quite annoying.

Has anybody encountered similar problems before, or have an idea of what might be causing the lag? I have been doing a bunch of similar calculations for the past few weeks and the problem only appeared recently.


3D model with PCM

$
0
0

my model is not showing the phase change what are the reasons for this?

2d axisymmetry to get the Total Harmonic Distortion of a speaker

$
0
0

Hello to all,

I have solved for 2d axisymmetry to get the Total Harmonic Distortion of a speaker {Can refer pdf of Transient Analysis of Loudspeaker Driver}. But if I want to find the same for a 3D design then how am I going to deal with it. I need some material or guidance for reference. Kindly, guide me for the same.

how to define "step velocity"?

$
0
0

Hello to all,

Could you please help me to how can I define "step velocity" as in for first 0.5 sec velocity is zero then for next 2 second it is 1 m/s and again zero for next 0.5 sec.?

Request for trial pass code

$
0
0

Hello, I'm new here! I'm trying to determine the liquid flow behavior of textile dyeing liquor which are used for textile dyeing.

As a beginner, I have need a trial pass code of version COMSOL 5.0/ 5.2.

How can I get this trial pass code?

Thank You!

How to sort parameters

$
0
0

Hi,

I have been working with COMSOL for a long time now and I realy enjoy many of its features.

However, one thing that I really miss in COMSOL is the ability to sort your parameters in groups so not all parameters are in the same list.

I work with a solid-fluid coubling where there are fluid parameters and solid parameters, so it is frustrating that I have to mix these. Then I thourght the 'materials' where ideal to define different parameters. But then I run into the next problem: In the solid, I define Young's modulus 'E' and Poisson's ratio 'nu'. Now I wanted to also define the shear and pressure wave speeds 'cs' and 'cp' which are functions of 'E' and 'nu', so I can use these in equations and to see their value. BUT: now they are stored as VARIABLES, so I have to make a PLOT to see their value!

This is very annoying and I would think that this is a problem that many users have.

Do you know any smart way to come arround this or do I simply have to put all my parameters in the same huge global list?

Connection to COMSOL Server on a remote server

$
0
0

Hi there,

I was attempting to use Matlab/LiveLink on my local computer to a COMSOL Server on a remote server. I managed to start the comsol mphserver there, but when I tried to use mphstart(ipaddress, portnumber, username, password) on Matlab to connect to the COMSOL server, I had the following error message: Error using mphstart There was a license error when connecting to the server

I am using a floating network license for COMSOL, and it is fine when I was starting comsol on the server, or connecting MATLAB to COMSOL Server on my local computer. Only this time there was an error. Did anyone have similar problem and manage to fix it in the end? What license was it referring to?

Many thanks

Save Auxiliary Sweep solutions?

$
0
0

To help the convergence of a stationary model with a varying stress-dependent elastic modulus, I am using an auxiliary sweep instead of parametric sweep. I would then like to get the eigenfrequencies of the resulting sweep solutions

However, I can't find a way to save the individual solutions from the auxiliary sweep. Parametrics sweeps have an option to "keep solutions in memory" but this option does not seem to exist for auxiliary sweeps. Does anyone know of another way to save these solutions so they can be used in a later study?

Thanks!


welding simulation

$
0
0

can someone suggest me how to model moving heat source for welding problem using guassian distribution.

boundary diffusion flux , first fick's law?

$
0
0

I did a dilute mass transfer simulation. I built a simple two-dimensional closed diffusion model with a concentration of 0 at the boundary. It is easy for me to calculate. When I get the boundary flux, I first integrate the online variable, and the integral variable is the diffusion. Flux, obtained by Fick's first law. The amount of diffusion is found to be less than the initial value minus the residual value. At this point, I found another result that expresses the flux, called the boundary flux, and the result obtained after the integral boundary flux is more ideal. The expression is dflux.spatial(c), which is a spatial function of the diffusion flux associated with concentration, but does not know how the specific expression is sought. Then my question is: (1) Why is the result of using diffusion flux not correct? (2) What is the specific expression of boundary flux? I want to know how the software counts. Thank you for taking the time to look at my question. I am very grateful for any reply.

size of matrix exported from Comsol : understanding?

$
0
0

Deal all,

can you explain how to understand the size of stiffness matrix which is exported from comsol,

for example i have a 2d cantilever beam, discritized with nel_x x nel_y=20 x 10 => nnode_x x nnode_y = (20+1) x (10+1)=231 nodes, and ndofs=231 x 2=462 degree of freedoms,

so i expect the size of stiffness matrix will be K(full)=462x462, but the size exported from Comsol is: 1322x1322

you can following the matlab file for runing in Comsol server;

COMSOL convergence problem?

$
0
0

I'm running a big CFD problem for gas storage in caverns. Now when I set the final simulation time to 20 days or less, it converges. If I increase final time to 21 days or more I get the message: Failed to find consistent initial values. Last time step is not converged. It's weird since the only thing I changed was final time, anyone familiar with this? Thanks

"Initial values of variables solved for" for t != 0

$
0
0

Dear COMSOL Users,

I am trying to use results of a stationary study as initial values for a time-dependent one. The topic was handled in the forum quite often, but nevertheless, there are three open questions left for me:

1) If I am using the "Inital values of variables solved for" option with the results from the previous study (let's say of solid1) for a thermomechanical simulation (for solid2), do I additionally need to specify inital stress values in solid2 as solid1.sx, etc. or not? What's the difference?

2) If I am using the "Inital values of variables solved for" option, are they defined at t=0 or at the first time specified in the time range of the study (e.g. "range(2,1,10)")?

3) If I am also using general extrusion due to the fact that I am adding an additional domain on top. Is it enough to just define the destination map and the source map both as x-expression: "x", y-expression: "y" and z-expression: "z"?

Thanks for your help! Best regards, Fabian

ECM (electro chemical machining)

$
0
0

if any one is working ecm process in 3d geomerty. please help me because i am also working on ecm

Point heat source in axial symmetry at origin

$
0
0

Hi all,

I want to model a simple point heat source at the interface of 2 materials. I can make finite cylindrical slabs of material in 3D with the point source in the middle but it would be much more efficient to move to an axial symmetric model with a point heat source at the origin. Unfortunately, whenever I try to place a point heat source at the origin of the axial symmetry but no heat is dissipated in the model and the entire model stays at room temperature. Is there an easy way around this? Or is this fixed in a newer version of COMSOL (I have version 4b)?

Thanks,

Erik


Streamfunctions vorticity formulation in cylindrical coordinates

$
0
0

The below post is related to an archived discussion


[natural convection in axisymmetric cavity fluid is air . Stream function-vorticity formulation. I can't find the mathematic partial equation for : 1- Stream function 2- vorticity. Please can you help me to do. Thanks.  athenaserra@gmail.com

3d design for simulating loud speaker

$
0
0

Hello,

I have solved for 2d axisymmetry to get the Total Harmonic Distortion of a speaker {Can refer pdf of Transient Analysis of Loudspeaker Driver}. But if I want to find the same for a 3D design then how am I going to deal with it. I need some material or guidance for reference. Kindly, guide me for the same.

Tumor- boundary condition

$
0
0

The below post is related to an archived discussion


Hi, I wanna simulate and analyze a tumor in COMSOL. I am a little confused about the boundary condition and some other aspects. Is there anyone can help me?

Shrinkage modelisation

$
0
0

Dear all,

I have a question regarding the setting the cast of polymer molding in Comsol. I want to simulate the shrinkage process of a plastic part in a mold, during which volumetric shrinkage happens due to the drop of temperature.

Is there a way in Comsol to set this modelisation taking into account the shrinkage effect and the disband that occurs between the casted part and the mold?

BTW, I am using both structural mechanics and heat transfer module and hoping to simulate the final shape (displacement and temperature field). I used "contact pairs" to model the contact and the Multiphysics "Thermal Expansion" and "Temperature Coupling" (Thermo-elasticity) and I’m getting convergence problems. I tried already to refine the mesh on the contact boundaries with no success.

It seems that when the disband between the mold and the casted part occurs, the nonlinear solver does not converge.

Thanks in advance!

Best regards,

Moaine

Gaussian-weighted surface temperature

$
0
0

Hello; i need to calculate the Gaussian-weighted surface temperature for surface heated by gaussian and periodic laser beam. Any idea how to find this weighted average value? Thank you, HM

Viewing all 26323 articles
Browse latest View live