4.0a - Latest Discussions - COMSOL Forums

Stretching+Inflation

Thu, 09 Feb 2012 19:30:51 +0000

Hi,

I have a 2D problem in which an arc of a circle undergoes two consecutive deformations (Please see the attached mph and PDF files). I use a two-step study which also happens to have two Physics in order to change some of the boundary conditions throughout the deformations. Also because the geometry is symmetric, I only simulate its upper half. The Steps try to deform the geometry as follows:

Step 1:
The right end is a symmetric boundary and the left end is stretched in circumferential direction to create a closed cylinder. Through this deformation, the central angle of the arc will change from 71.4 to 180 degrees.

Step 2:
After the solution is saved by “Store Solution”, the left end will change to Symmetry and the surface on the inner radius will go under a pressure to inflate the cylinder. I do this by a parametric study by setting an Extension in my “Step 2: Inflation” node.

I have run into a few problems:

1- Although I have used a “Store Solution” in my Solver after the Closing solution, inflation does not use the stored data and therefore does not start from the first deformation (i.e. closed cylinder); in other words it starts from the original unclosed arc. However, it correctly enforces symmetry condition for the left boundary before it solves for the inflation.

2- In the second step of the solution which is inflation, although the boundaries are symmetry, the solution never converges as if the problem is indetermined. This obviously does not occur if I wanted to simulate a full cylinder and not half of it. Adding at least one constraint such as fixing one node on the geometry makes the problem mathematically determined but the physics of the problem will be compromised. Is there any approach that can resolve this issue?

3- I have set the program so that it plots the results while solving it. I have experienced that sometimes I do not get one of those plots during the solution. I mean sometimes, the result of Step 1 is shown during solving, but not the results of Step 2, and sometimes the other way around. I was wondering if this is a bug or I am missing something.

Please see the attached mph and PDF files and let me know how I can fix these problems.

Thank you for reading my posting,
Jessica

plot versus time

Thu, 09 Feb 2012 04:48:56 +0000

Dear Sir,
I try to solve my problem in time domain

I apply sine wave on a conductor of underground cable to study the elecrtic filed in insulation

i get the distribution of electric field at 10 sec by setting time interval (0,0.1,10)

and in another run (i got it at 20 sec) by setting time interval (0,0.1,20)
and in another run ( i got it at 30 sec) by setting time interval (0,0.1,30)

finally i want to draw electric field at specific location versus time how can i get it?

thank you

semi infinte

Tue, 07 Feb 2012 04:38:48 +0000

Dear Ivar

I have infinte cable (3 m) and i want to study a sector (0.165 m) to get the temperture distribution

how can i define the boundary at the infinte end?

i.e how can i tell comsol that this end of cable is too long and infinte

thank you very much

Manipulating heat-transfer coefficient

Fri, 03 Feb 2012 09:56:06 +0000

Hi All,

I have 3 domains, from bottom to top iron, air and glas. Iron and Glas are defined as heat transfer in solid, Air as heat transfer in fluids. Iron being heated, the heat flow to air and glas.

i want to increase the heat flux from iron to air by manipulating the heat-transfer coefficient.

the convective cooling doesnt work here, it works only at outer boundary.

thanks. happy weekend

akmal

Error Sol1/S1

Thu, 02 Feb 2012 23:51:00 +0000

Hello,

First of all I would like to say I am a first time user of Comsol so please bear with me.

The error expression that I get in comsol is:

Attempt to evaluate negative power of zero.
- Function: ^
Failed to evaluate expression.
- Expression: d(d((-5.3079*(0.6+3523.3558082974137*(-294.15+mod1.T)^(1/6))^2*(293.15*unit_K_cf-mod1.T)*test(mod1.T))*(dvol),{test@0}),mod1.T)

Time dependent and non linear model takes (too?) long to solve...

Sun, 29 Jan 2012 21:52:09 +0000

Hello everyone,

I am simulating electrical machines within the rotating machinery interface in the ACDC module.

The model I calculate right now solves. But the way it solves is not satisfying in my oppinion. As soon as I choose the iron to be nonlinear, the solving takes days to calculate just 50ms with a time step of 100us.

I'm not sure if I'm allowed to upload the model in this forum, since the Geometry is not meant to be public.

The Geometry is a fourth of the motor and periodic boundary conditions are used for this. The current, stamped into the slots, is calculated by time dependent equations which involve the derivation of the flux field linkage. So comsol has to solve the equation for the current and vector potential simultaneously.

I already tried to improve the model this way:
http://www.comsol.com/support/knowledgebase/1127/
But it didn't really increase the speed. Actually with it's help the model could first be solved nonlinear...

Does anybody have general advices to get nonlinear time dependent models solved faster?

Thanks in advance.

EIS - impedance measurements with very low conductivity

Fri, 27 Jan 2012 10:34:23 +0000

I've bee working for a while on a EIS model, so models of electrical impedance spectroscopy on various shape of electrodes (2D model properly setting the d [m] )

(Terminal 1 and 2 in voltage configuration, frequency analysis)

It happens that at a very low frequency the real part of the impedance change into negative....

I can't find a physical reason to obtain this so I suppose it is something related to the precision of the calculation

I did a lot of tests and I found out that randomly, a very low conductivity setting to an interface or to a materials results in a negative real term of impedance (1/ec.Y11).

...? a bug? a not proper numerical precision?

help please
best

Small Copper Wire Coil Producing Magnetic Field

Thu, 26 Jan 2012 23:34:48 +0000

Hey all,

I'm pretty new to COMSOL, but I tried to build a very simple model which has a small, copper wire loop. The hope is that when a current is run through the loop there should be a larger magnetic field in the CENTER of the loop. Unfortunately this sort of "lensing" effect is not observed. I have consulted the model builder files both the current carrying wires model and the helmholtz coil model without being able to figure out whats wrong.

When I define the model I simply draw the parallel wires that bring current, and then a small loop and since I'm using the magnetic fields as my physics I then define the domains that make up the coil as having a current density in x,y,z under external current density. When I solve, I am able to produce a thermal map of the flux but it doesn't seem to make sense, as it implies the highest flux is along the wires and when I change it to magnetic field, it shows some nonsensical surface map that I'm having trouble believing. I would like to be able to produce data of the magnetic field gradient, the magnetic field, the heat produced by the sample when it's running, and the force produced by the wire, but I still haven't been able to make any reasonable results with the model I have. I fear that I may have somehow defined the current density incorrectly, but I'm not sure how. I also tried to build the model in 3D with similar results. I have attached the file, any help is always greatly appreciated.

Regards

How to remove the history record in the m-file exported from COMSOL

Tue, 24 Jan 2012 06:46:09 +0000

Hello all,

I hope this is not a silly question. With "Save as Model M-file", COMSOL can generate a m-file that can be used in MATLAB with LiveLink.

But I realized that the generated m-file contains every step I did in COMSOL GUI. For example, if you set the Geometry Import several times in COMSOL, specified "1.mphbin" first, then imported. And you realized you chose the wrong file. You then changed to "2.mphbin", imported. You realized you did it wrong again.....

After changing the filename several times, you export the model to a m-file. It will look like:

model.geom('geom1').feature('wp1').geom.feature('imp1').set('filename', '1.mphbin');
model.geom('geom1').feature('wp1').geom.feature('imp1').importData;
model.geom('geom1').feature('wp1').geom.feature('imp1').set('filename', '2.mphbin');
model.geom('geom1').feature('wp1').geom.feature('imp1').importData;
model.geom('geom1').feature('wp1').geom.feature('imp1').set('filename', '3.mphbin');
model.geom('geom1').feature('wp1').geom.feature('imp1').importData;
model.geom('geom1').feature('wp1').geom.feature.remove('imp1');
model.geom('geom1').feature('wp1').geom.feature.create('imp1', 'Import');
model.geom('geom1').feature('wp1').geom.feature('imp1').set('filename', '1.mphbin');
model.geom('geom1').feature('wp1').geom.feature('imp1').importData;
model.geom('geom1').feature('wp1').geom.runAll;

But it is pretty clear that the first 7 lines don't make any sense. The problem not only exists for geometry, it also exists for mesh, solver, and so on.

Is there a way to remove all the history actions so I can get a clean m-file?

Thanks.

Best,
Kan

Electrical impedance of a piezoelectric device

Sat, 21 Jan 2012 15:19:11 +0000

The electrical impedance of a piezo device, piezoelectric transformer e.g. around a resonance mode, can be electrical model with the as a LCR brance in parallel with a capacitor: "LCR||C". Equivalent to the Mason model of a piezoelectric transformer with either secondary or primary side short-circuited. The amplitude plot of the impedance around resonance will first have a resonance-peak followed by a anti-resonance-peak.

To narrow the problem down the, we look at a 2D model consisting of an simple rectangle shape.

-Using COMSOL 3.5 or 3.5a the electrical impedance of the 2D model matches the electrical "LCR||C" equivalent and a real-life measurement confirms the result.

-Using COMSOL 4.xx to simulate the same model dose not fully match the "LCR||C" equivalent. The resonance-peak is fine. The anti-resonance-peak seems to have the right frequency, however the impedance is fare from high enough. In the electrical domain a low anti-resonance-peak is equivalent to a material with to high conductivity. The material parameter nor the physics uses conductivity or similar as I'm aware of!?

What have changed in the physics from COMSOL 3.5a to COMSOL 4.xx?? Are there something that I miss in my model??

PS I have treble checked the material parameters and they are identical from 3.5a to 4.xx

Attached is plots that illustrate the problem.

comb drive actuator

Sat, 21 Jan 2012 13:22:35 +0000

Hi,

I am trying to simulate a comb drive actuator in 2D with electrostatics , ale, and structural mechanics .The model is similar to that of symmetrical comb drive of comsol 3.5 model file(which is slightly different from 4.0a).And i have cross checked the procedure, yet I am getting errors while solving.

I have checked the associated discussion forums, but i couldn't figure the problem. And I am very new user, so kindly help me out in indicating what went wrong.

this was the error message i got, and i have attached the mph file along.

Failed to evaluate variable.
- Variable: mod1.emes.nTy_Fes
- Geometry: 1
- Boundary: 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 386 388 390 392 394 396 398 400 402 404 406 408 410 412 414 416
Failed to evaluate expression.
- Expression: d((mod1.smps.d*(mod1.emes.nTx_Fes*test(mod1.u)+mod1.emes.nTy_Fes*test(mod1.v)))*(dvol),{test@1})

------------------------------
Kind regards,
Sharmila Radjasegarane

Rotation of a solid body under fluidic vortex

Fri, 20 Jan 2012 06:34:56 +0000

Dear COMSOL Community,

I am trying to simulate the rotation of a biological cell in a microfluidic vortex using Fluid Surface Interface (FSI). I have modeled my simulation based off the FSI example for comsol, however the model will not converge. The initial values can be computed, however the model reaches a convergence error before the first time step is realized.

I have attempted to constrain the input to have a smooth transition and to be parabolic (see functions and parameters) with little success.

Ideally, the circle (cell) would be contained to only move rotationally, without translation (even though I cant even get to this step).

Any suggestions would be appreciated and the model is attached.

uninstalling version 4.0a

Wed, 18 Jan 2012 13:58:10 +0000

Hello,

Long time ago I got a trial version of COMSOL 4.0a and it expired, unfortunately I don't have the money to buy it; now I would like to uninstall it but is not possible because it doesn't have a "unistall" icon or neither it appears under the list of programs installed in the control panel... (I know) I cannot open it because it says licence error or download an update since I don't have a valid license but I wonder if there is a solution for this or any advice you can give me...

thanks! :)
Elsa

Input impedance problem in a nanodipole

Mon, 16 Jan 2012 16:11:00 +0000

Hello everybody,
I'm trying to simulate a dipole with a voltage source in the middle of them. For that, I'm using a lumped port that is found in the middle of the dipoles, and it have the voltage that I want. But i'm having a very small input impedance. An I doing something wrong?
I found to a model tutorial simulate that use the lumped port, that is the "rf coil", and it's found in the rf module paste.
In the attachment is the geometry and where the lumped port is found, and my simulation too.
I'm using comsol 4.0a as my software.
Thanks

Problems generating lamb waves in an Aluminum plate

Tue, 22 Nov 2011 11:19:52 +0000

Hi,
I am trying to generate lamb waves in an Aluminum Plate using a Piezoelectric actuator placed in the center of the plate.
I am pretty sure that the physics involved are solid mechanics and Electrostatics (to apply electric potential to the PZT).
So according to theory, there should be waves generated in the plate because of the electric potential applied to the PZT, which is attached to the Aluminum plate. But I am not getting it, I am unable to couple the two physics. Can someone help !
Also, what BC should be applied to make sure that the deformations in the PZT take effect on the plate.
If it helps, I have attached the file, please take a look

RF Mode Analysis Normalization Convention

Fri, 18 Nov 2011 20:34:08 +0000

Hi all,

This question is in regards to: RF Module > Electromagnetic Waves > Mode Analysis

Does anybody know the normalization convention that is used in the solutions? For example, part of the solution are surface plots of the electric field. The shape of the field distribution is dictated by the specific eigenvalue problem; however, the actual values given by COMSOL are quite arbitrary (i.e. you can always include a multiplicative constant with impunity). How then, does COMSOL determine what numbers to give in the solution for field components, power, etc.?

My first thought was that the integrated power flow would be normalized to 1 W, but, as far as I can tell, this isn't the case. Other obvious integrals over E^2 or H^2 don't seem to indicate any type of obvious normalization either.

problem creating a model using a Matlab function (Comsol 4)

Mon, 11 Apr 2011 07:55:30 +0000

Hi,
i want to create a Comsol model using a Matlab function like:

function myFunction()

model = ModelUtil.create('Model');
...

return

When i call the function Matlab returns this error:
??? Undefined variable "ModelUtil" or class "ModelUtil.create".

I guess somehow i have to tell Matlab to use Comsol 4 because it seems not able to find ModelUtil from within a function.
What am i doing wrong?

Thank you for any advice.

Modelling of ion transit in oscillating electric field

Tue, 05 Jan 2010 08:38:08 +0000

Hi,

I've recently spent quite a bit of time using SIMION to model the trajectories of ions as they pass through fine wire meshes, upon which electric potentials are applied (such that a small DC electric field is generated). In addition, I am very interested in plotting the trajectories the ions take when a high frequency oscillating voltage is superimposed on the DC potentials which are applied to the grids.

The SIMION modelling has generated some very interesting results, and I would like to validate them by performing a similar analysis using COMSOL.

I recently attended the Intensive COMSOL course, held at Hatfield University, UK. The course only briefly covered the Particle Tracing option in the Post Processing screen, however from the little amount that we covered I was left with the impression that COMSOL will only plot the trajectory that an ion would take under steady state field conditions (after a solution to the model has been generated), and would not be able to calculate the trajectory of an ion under an oscillating electric field.

I am curious to find out whether my understanding of the particle tracing option is correct, and I would be very interested to hear from anyone who has succesfully used COMSOL to plot ion trajectories in non steady state field conditions.

Regards,

Jon Atkinson

Exporting Comsol data to Matlab?

Wed, 18 Nov 2009 11:00:06 +0000

Hello everybody,

I want to postprocess Comsol data in Matlab. Since Ascii files blow up in size and I get "out of memory error", I'm looking for some other way. I've no experience with Comsol-Matlab interface and it seems to me that this interface is not working in my system.

Could someone please tell me, how to export Postprocessing data in some Matlab-compatible format (not Ascii)?

I've converted a Comsol model (after executing) into an m-file and wanted to run it in Matlab (running separately from Comsol) but Matlab is probably not able to read the flbinaryfile (*.mphm). Any suggestion, what could be wrong here?

Thanks.
Daniel

How to set Acoustic Source

Wed, 04 Nov 2009 06:13:54 +0000

Hi!

I have a question on the Acoustics module in Comsol, regarding how to set a monopole acoustic source.

I have a point acoustic source characterized by a modulated sin wave:

In Matlab coding,

y=-sin(2.*pi.*Freq.*Ts).*hanning(length(Ts))'

Here, the acoustic source is described in terms of particle displacement, y.

However, I understand that in Comsol, to set a point source, we are supposed to go to Physics > Point settings > Pressure acoustics, and select the type of source as flow, intensity and power.

My question is, how do we convert the description of the point source from displacement to a description in terms of either flow, intensity or power?

Thanks!