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

error when coupling reactive flow and heat transfer

April 1, 2014, 10:19 pm
$
0
0
Hello everyone,

I am a Comsol beginner and I am trying to model a turbulent reactive flow (homogeneous gasphase reaction).
I created a coupled model for the reactive flow and the heat transfer, very similar to the model "syngas combustion in a round-jet burner" in the library.
The model runs for the turbulent reactive flow (at least for some concentrations given in initial conditions and inflow) but when I try to couple it with the temperature, it first seems to converge pretty well but then gives this error:

Segregated group 4
Attempt to evaluate non-integral power of negative number.
- Function: ^
Failed to evaluate operator.
- Operator: nif
- Geometry: 1
- Boundary: 1 2 4 5 7
Failed to evaluate expression.
- Expression: nif((comp1.rspf.d_w_plus<(10/(comp1.rspf.Sc_wCO^0.33333333333333331))),nojac((comp1.rspf.d_w_plus*comp1.rspf.Sc_wCO)),(comp1.rspf.d_w_plus<sqrt(((comp1.rspf.kappav*1000)/comp1.rspf.ScT_wCO))),nojac(((15*(comp1.rspf.Sc_wCO^0.66666666666666663))-(500/(comp1.rspf.d_w_plus^2)))),nojac((((15*(comp1.rspf.Sc_wCO^0.66666666666666663))+((log(comp1.rspf.d_w_plus)*comp1.rspf.ScT_wCO)/comp1.rspf.kappav))-(((comp1.rspf.ScT_wCO*0.5)*(log(((comp1.rspf.kappav*1000)/comp1.rspf.ScT_wCO))+1))/comp1.rspf.kappav))))
Failed to evaluate variable.
- Variable: comp1.rspf.wplus_wCO
- Defined as: nif((comp1.rspf.d_w_plus<(10/(comp1.rspf.Sc_wCO^0.33333333333333331))),nojac((comp1.rspf.d_w_plus*comp1.rspf.Sc_wCO)),(comp1.rspf.d_w_plus<sqrt(((comp1.rspf.kappav*1000)/comp1.rspf.ScT_wCO))),nojac(((15*(comp1.rspf.Sc_wCO^0.66666666666666663))-(500/(comp1.rspf.d_w_plus^2)))),nojac((((15*(comp1.rspf.Sc_wCO^0.66666666666666663))+((log(comp1.rspf.d_w_plus)*comp1.rspf.ScT_wCO)/comp1.rspf.kappav))-(((comp1.rspf.ScT_wCO*0.5)*(log(((comp1.rspf.kappav*1000)/comp1.rspf.ScT_wCO))+1))/comp1.rspf.kappav))))
Failed to evaluate expression.
- Expression: d((down(comp1.rspf.rho)*comp1.rspf.beta0s^0.25*sqrt(max(comp1.k2,eps))*(comp1.rspf.wCOWall-comp1.wCO)*test(comp1.wCO)/comp1.rspf.wplus_wCO)*(dvol),{test@4})
- Feature: Stationary Solver 1 (sol7/s1)

Does anyone know what my mistake is?
BTW, it converges if I set all concentrations to zero.
Thanks for any hints!
Best regards,
Miriam

PS: The file is too big to upload, if it is any help I can try and cut it down.
↧
Search

averaging of data

April 1, 2014, 10:50 pm
$
0
0
I am a new user of comsol 3.5a . I am getting data of heat transfer coefficient from a surface plot which is non-repeatable ,can you please tell me how to averaging of this data.
↧
↧

How to solve 4th PDE in postprocessing

April 1, 2014, 11:11 pm
$
0
0
When I use Electric currents module, I want to solve 4th order PDEs in this postprocessing. For example, d(d(d(d(V,x),x),x),x). But it didn’t work.

Please give me some advise. Thanks.
↧

Trouble with Matlab functions

April 2, 2014, 12:30 am
$
0
0
Hello

I'm working with the LiveLink to Matlab and Comsol 4.4. Now it seems that either there are some restrictions in this LiveLink or I'm missing something.
First I defined global variables in Matlab, that I call later on in Matlab functions. These Matlab functions are used in the Comsol model. This does not work in Comsol. I read in a discussion that Comsol does have problems with global variables. Is it really not possible to work with them?
Second it seems that Comsol can not read Matlab functions that contain if-else statements. Is that true?
At the moment it seems the only Matlab functions I can call in Comsol are simple polynoms.
Does anybody know the correct approach to call Matlab functions in Comsol or if there are really some restrictions? Is the Matlab environment used by the LiveLink not the usual Matlab environment?

Thank you for taking time and for your help.
Anina
↧

Squeeze filme damping in FSI

April 2, 2014, 12:48 am
$
0
0
Hello everyone,

I was wondering if the Fluid-Structure Interacxtion module would be suitable to simulate the squeeze film damping effect?

To be more precise: Im simulating a vibrating cantilever in air environment and considering I am in the micrometer scale I need to take into account the damping due to squeeze film, my idea was to use the FSI module to calculate the damping ratio and use it as a raw parameter in the damping of a normal Solid Mechanics simulation. Does anyone know if FSI take into account squeeze film effect?

Another way I tried is the thin-film damping but I can´t make it work, any suggestion?

Thank and all the bests,

Edoardo.
↧
↧

Boundary condition in Weak Form PDE

April 2, 2014, 2:47 am
$
0
0
Hi!

I'm trying to implement micropolar continuum in Weak Form PDE in COMSOL.
I want to implement a "sliding" bc between to interfaces.

It is easily done in Solid Mechanics module through "thin elastic layer". There I can set the stiffness to very high in say x-direction and set stiffness to zero in y-direction.

When I want to implement the same bc in Weak Form PDE, I look at equation view under "thin elastic layer" from Solid Mechanics module, I see two lines under "weak expression":

-1000*solid.uspring1_tel1*solid.d*test(up(u)-down(u))
and
0

I guess that the first line set the stiffness in x-direction to 1000, and the other line somehow sets the stiffness to zero in y-direction.
There is no information about the constraint force, and I've been trying to implement this weak expression as "weak contribution" and also "weak constraint" with no success.

Can anyone help?

Best regards,
Alexey

↧

Electrowetting / droplet motion

April 2, 2014, 3:21 am
$
0
0
Hi,
I want to determine the physics of motion of droplets (transport, splitting and merging) using COMSOL. I found the paper: www.comsol.co.in/paper/download/37206/Cahill.pdf
but I can't implement the simulation.
Who might be able to help me, and send some examples or model (.mph) or information about e.g. the transport speed determination of a droplet between two electrodes?

Many thanks in advance
↧

Problems with Magnetic force

April 2, 2014, 4:48 am
$
0
0
Hi! I am new to COMSOL and I really need some help please.

I want to study an electromagnetic actuation system to guide magnetic microparticles in blood vessels. For that I used the AC/DC module to build some coils to generate a gradient magnetic field and then I used the Fluid Flow module to model the blood. The final goal is to use the Particle Tracing for Fluid Flow module to determine the trajectory and velocity of the particles when subjected to the magnetic force, however when I include the Magnetophoretic Force (already included in the module) nothing happens! There is no change in particle velocity.

As it can be read in COMSOL: “The Magnetophoretic Force causes motion of permeable particles toward regions where the magnetic field is stronger. The magnetophoretic force is applicable for particles which are charge neutral and have a different relative permeability than the background fluid.”

But even when I insert a high value of particle relative permeability (400000) nothing happens. I was expecting a change in particle velocity. I attach some pictures of my problem.

Maybe it's something pretty obvious, but I really need some help here. Thank you!

Best regards,
Patrícia
↧
Search

Draw à 3D triangle

April 2, 2014, 8:02 am
$
0
0
Hello, I try to draw a 3D structure representing an isosceles triangle with a regular height. I used the polygon drawing, but fails to define the pattern as a solid.

Thank you in advance,

Josephine
↧
↧

Magnetic flux density (streamlines) and current density by cylindrical condutors

April 2, 2014, 9:56 am
$
0
0
Hello,

I am trying to plot magnetic flux density (streamlines) and current density produced by cylindrical conductors shown in the attached file, for dc and ac (e.g. 50 Hz) cases. But there is a problem occurs when I click on 'compute' and comsol becomes irresponsive,
I would very happy if someone will help me in this matter.

Regards,
Akim
↧

How to deal with Contact problems, Adhesion!!

May 17, 2011, 3:40 am
$
0
0
Hello Everyone,

I am dealing with the topics of contact mechanics and adhesion. In my case, two surfaces are adhered strongly and I would like to model the detachment of contact or treat it as a crack propagation in the contact area of the two surfaces when a tensile force is applied to the non adhering surface. I searched in the model library but haven't found any information in this regard.

I got an impression from the Literature review that we can deal such cases by using Cohesive zone model (CZM) or Cohesive zone elements which is well defined in Abaqus or Ansys. I am just wondering whether we have the possibility to implement the CZM in Comsol?

It would be very very helpful for me if someone could suggest me the way to implement CZM or any other possibility to deal with this situation!!

Any suggestions or help would be highly appreciated.

Best Regards,
Ramgopal
↧

Time dependant Structural Mechanics & Magnetic Field

December 31, 2013, 10:54 am
$
0
0
Hy

Can anyone please let me know that is it possible to simulate time dependant magnetic field when force or some external conditions are there in which geometrical position of the shape is changing.
Comsol builtin model of freely falling magnet is one case of Time dependant magnetic field but is the above case mentioned by me applicable? And what if I combine moving mesh together and simulate all three physics together? Which studies I should carry out.

Its important for me, plz someone help.
Thanks in advance.
Regards
MF
↧

Problem with weak form-SBC

March 17, 2014, 6:42 am
$
0
0
Hello to everyone,
I am new to Comsol (i use the3.5a version) and i try to implement an SBC using the weak formulation in a simple structure in order to learn the weak form.Perhaps I have not understood it enough.
My first try was to use the rf-formulation of the SBC, then multiply with the test function and put it in the weak tab in the boundary setttings.
Is this a wrong way?

Thanks in advance
↧
↧

Set Stefan condition manually

March 31, 2014, 6:12 pm
$
0
0
Hello,

I'm currently struggling with stefan problem and want to implement the stefan condition L(V.n)=Kl grad(T).n-Ks grad(T).n manually on the Solid/Liquid interface. L is latent heat, V is interface velocity.

I figured out there could be 2 possible ways:
(1) HT in liquid with solid subnode defined in this physics;
(2) 2 HTs, one is HT in solid and the other is HT in liquid and set;

for (1), since the default continuity condition assigns Kl grad(T).n-Ks grad(T).n=0, how to override this to add the latent heat term?

for (2) Do I need to define temperature and flux for both HTs in their boundary settings?

Thanks in advance for any suggestion!
↧

User Controlled Mesh across two different geometries

April 2, 2014, 5:31 pm
$
0
0
Hello All,
I am new to COMSOL and would like to get some help with user controlled mesh. Attached is a figure I would like to model in COMSOL. The length is 2cm and the height is 1cm. When I try to use the user defined mesh. It meshes both bodies separately, so I'm unable to get the mesh which is in the figure.
Any help is appreciated.
Thank you
↧
Search

time dependent study for sinusoidal wave

April 2, 2014, 10:34 pm
$
0
0
Dear all,

I'm trying to simulate my module in 2D axis of COMSOL version 4.3 in joule heating physics with sinusoidal input voltage of { V0*sin(2*pi*50*t)} and for the frequency of 50 Hz it will takes 0.02 second for full wave . so i must use small time step of around 0.0001sec. . i have problem when i increase the time to more than 5 second with step of 0.0001 sec .

I need to simulate my module for time between 0 sec until 300 second with time step of 0.0001. as times: range(0,0.0001,300)s
So it needs around 3 million step to get the full result and i think it's difficult to that .

is ther are any way to get it in the transient study so i can check the time that need to arrive the steady state temperature .

best regards ,




↧

Frequency domain using two pi/4 phaseshifted Piezos

April 2, 2014, 10:52 pm
$
0
0
Hello,

I want to analyse the mechanical behaviour of a spring, that is being piezoelectrically activated. With one piezo this wasnt any problem, as i would just use 5V as electrical potential, and run a simple FrequencyDomain analysis.
But in my case the Piezos need to be phaseshifted like:
U1=U0*sin(2pi*f*t)
U2=U0*sin(2pi*f*t+pi/4)
I tried to use waveforms as global definitions and used them for my electrical potential like "wv1(t[1/s])" (found that hint on another thread), which seems to work in the first place, as the entry wont turn red/yellow.
Trying to simulate, i get this error:
--------------------------
Failed to evaluate variable.
- Variable: t
- Geometry: 1
- Boundary: 52 67
Failed to evaluate variable.
- Variable: comp1.pzd.V0
- Defined as: wv2((t/unit_s_cf))
Failed to evaluate expression.
- Expression: comp1.pzd.V0-comp1.V
-------------------------

I tried looking for solutions on the forum, but couldnt find any useable (or im just searching for the wrong topics, as im very new on comsol). I mean this is a very basic problem:
Two phaseshifted oscillations on something... it probably hardly matters in that case if its voltage, force or whatever..

Thanks Andreas
↧
↧

Heat Transfer. Surface-to-surface radiation node

April 3, 2014, 3:14 am
$
0
0
Hello everybody,

I'm a physics student and I'm trying to simulate with COMSOL the heat behavior of a chip connected to a CdTe detector. In my geometry I have domains that are not touching each other and I would like to take into account how they are heating up each other due to radiation. Since they are so close to each other, I think I should use the surface-to -surface radiation node (instead of the surface-to-ambient radiation node), but I do not understand how to use it. The surface-to-surface node does not ask me to give the distance between surfaces. How can I be sure that COMSOL does not only calculate the radiation coming from the surfaces, but also how the surfaces are heated up from receiving the radiation from other surfaces?

It would be helpful if somebody could explain me how to do this.

Thank you in advance,

David
↧

RAM memory & Dofs

April 3, 2014, 4:08 am
$
0
0
Dear all，

I am going to update my computer for a larger 3D fluid model. My computer can handle 1,400,000 Dofs with
6 RAM GB and 2 i5 cores using segregated iterative solvers.

What RAM and processors do I need if I need to solve a same problem with 10 times more Dofs? Do I just need to go from 6 GB to 60 GB ?

Thank you very much.
Best wishes
↧

Automatic time step increase in generalized-alpha solver

April 3, 2014, 5:32 am
$
0
0
Particle tracing is conducted by a generalized-alpha solver.
For many cases, when the particle motion is finished, the time step is increased drastically.
However, in current case, time step is oscillating.
In some case, when I turn off the Brownian force (which is a monte carlo modelling), this problem is solved.
However, this is not a solution for the current problem.
What should I do for avoiding this problem?
This problem causes the longer solution time.
↧
Viewing all 26512 articles
Browse latest View live
Search