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

Exciting a meta-surface with a point source

January 5, 2015, 8:25 am
$
0
0
Hi there,

I'm completely new to COMSOL, so I still struggle with it a lot. Bear with me! I've already looked in the existing discussions but found nothing helpful.

What I want to do is model the power flow over a metal-dielectric patterned surface (it's patterned with complimentary split ring resonators or CSRRs). I've already produced a model that solves the eigenmodes of the surface using floquet boundary conditions, which produces the correct result for what the resonant frequency of the material should be.

However, I need to model what the power flow looks like on this surface. Ideally, what I want to do is excite the surface with a point source electric field. But how do I do that?

I should mention, if I don't use periodic boundary conditions, I think I'm going to struggle computationally. The CSRRs I'm working with are spaced along a 2mm square pitch and the materials I'm investigating are A3-sized surfaces covered with these. So I sure couldn't model the whole material.

Any help at all is immensely appreciated.
↧
Search

Volum integration using the derived values

January 5, 2015, 1:27 pm
$
0
0
Hi,

I am trying to evaluate an integration over the volume of a domain and the integral expression is (int (div(B*E)) where div stands for divergence and * is an outer product, B and E are two vectors. I tried to include them in integration in the results but it seems they are not supported in comsol or atleast I dont know how to do it. I m wondering if there is any other way to perform such operation on derived values. Its a great help If anyone could write me a line.

Regards,
Ehsan
↧
↧

spherical harmonic source

January 5, 2015, 3:39 pm
$
0
0
I have a 3D beam model in the frequency domain. I am trying to impose a harmonic spherical source. A point load that acts as a spherical harmonic source. I am not sure how I can do this.

I defined a spherical coordinate system; however, when I assign a point load in that coordinate. It is merely a directional load.

What I need is spherical source that applies load along radial direction in all angles.

Also is it possible to prescribe a harmonic displacement rather than harmonic load.

Thanks
↧

Hole in plate

February 23, 2012, 2:02 pm
$
0
0
Hello all! I am new to comsiol and ha ve a basic question. I am trying to put an inlet hole in a modle and am having a bit of trouble on how to put a hole in a surface. I also need to creat a plate with 40 3/8 in holes in it. If there is a quick way to do this it would be very helpful.
Thank you,
Joe bob
↧

How to simulate WOOD FIRE with Comsol?

November 12, 2013, 6:46 am
$
0
0
Hello dear COMSOL community,
I have this phenomenon I want to simulate with COMSOL to understand it and improve the way we use wood in my region: WOOD FIRE
I need ideas in the approach to take using chemistry (if needed), heat (conduction, convection & radiation), ... My biggest problem is the simulation of the wood fire itself maybe taking in account the phase changes, etc. I am very interested in the amount and types of gases emitted and the quantity of heat generated.
Thank you very much for your precious help.
- Regis
↧
↧

Problems with static mixer simulation (diffusion vs convective mixing)

December 9, 2014, 11:43 am
$
0
0
I am trying to simulate a static mixer to assess its mixing quality through multiple iterations. It is meant for liquids with minimal diffusion such as latex paint in the realm of 750 Pa*s viscosity. I am having difficulty reducing the diffusion within the liquid as i would like convection to be the main mixing mechanism. I have tested higher velocity, higher viscosity, and lower diffusion coefficient and none of them seem to affect the results of the chds simulation. I would like some guidance as to how to remedy this problem or where i am going wrong. Attached is the Comsol simulation file since it may be helpful if a mistake has been made in the setup. Any help would be greatly appreciated.

Thanks greatly for your time and help,
Jon O
↧

3D Boundary mode analysis for hybrid directional coupler

January 6, 2015, 3:26 am
$
0
0
Greetings.
The proposed coupler consists of two layers. the bottom layer comprises one-single mode silicon waveguide embedded in sio2. the upper layer comprises two parallel symmetric metal (Ag) plates isolated by 50nm wide gap and upper clad-with EO polymer to form an (MDM plasmonic waveguide) as shown in the attached file
To guarantee a strong coupling between two waveguides, the effective refractive index of the photonic and plasmonic mode should be close. After coupling between photonic and plasmonic modes, The two Eigen modes(quasi-even and quasi-odd) are formed in the coupler
the interference pattern is induced by beating between the quasi-even and quasi-odd mode , and at the coupling length , a nearly complete power transfer from silicon waveguide to plasmonic waveguide

First, I have made a 2D mode analysis (by comsol 4.3b) and received an effective refractive index for photonic mode, plasmonic mode, quasi-even , and quasi-odd mode . I obtain the real part of refractive index for photonic mode and plasmonic mode are 2.1131 and 2.1134 , respectively. this is done by using perfect match layer

when I design 3D coupler (using comsol version 4.3b) using electromagnetic wave beam envelope and boundary mode analysis study . the power is transfer from silicon waveguide to plasmonic (i.e. confined in polymer ) but not completely. what is the problem ? is the perfect match layer (pml)is necessary for 3D boundary mode analysis, becaue in 3D I didn't use pml condition
I excite the 3D design by numerical port , is this right? what is the type of numerical port TE or TM mode

i.e I have added two port , numerical port ,(on silicon layers) I have also created two boundary analyses, one for each port , I search mode around 3.48 silicon, when I compute boundary mode analysis 1, I received 2 modes which Is not same as the photonic mode (in 2D mode analysis)
when I compute the study(boundary mode analysis 1, 2 and frequency domain)and choose power flow time average y component , the problem is (only part of power is transferred from silicon to plasmonic waveguide)

Is the problem in the boundary conditions and PML?? if I must use what is the thickness of pml?
is the physics and study is suitable for this case
Is the problem in mesh???? because I use the default Fine mesh

The other problem when I apply the same steps (3D coupler) above but in comsol version 5, the design is not work (no any power is transferred , no solution) and appear problem
(Failed to find a solution for the initial parameter. The relative error (3.5e+005) is greater than the relative tolerance, returned solution is not converged)
what is the origin of the problem , I attached my work built in comsol version 5 below

Thank you
keep in touch

↧

Bend a sheet of metal to cylinder

January 6, 2015, 5:41 am
$
0
0
Hello,

I would like to bend the following plane structure (attached) and form a cylinder along the x-axis direction. I have tried many several ways, but nothing worked.

Thank you alot in advance.

↧
Search

Frequencies on a squared plate (membrane)

January 6, 2015, 5:48 am
$
0
0
Hi guys, i'm sorry if i'm doing this wrong but i'm kinda new on COMSOL and I'd like to know if you guys could help me on this exercise:

i need to see how is the behavior of the frequencies on a squared membrane, like chladni figures experiment. There's a model for circular ones that works pretty well but i can't adapt for myself, because there's data like pre-tensions that i won't need right now. My plate is only fixed on the center by a screw on a base made by me. I just would like to see how would be this behavior on some frequencies (6 is fine).

Plate:
size - 16 cm
thickness - 3 mm
poisson - 0.33
young - 69 GPa
↧
↧

Warning sign in "Step 2: Time Dependent" node in COMSOL

January 6, 2015, 10:14 am
$
0
0
Hello guys,

I am solving a ODE in COMSOL which is coupled with AC/DC module through the velocity (Lorentz term) in "Magnetic and Electric fields" interface. When I choose "study controlled" in Equation form of this interface, as shown in picture 1.1 attached, the solver converges quickly and I get good results, at least their trends are reasonable. However, in this case, a "Waning sign" appears in the node "step 1: Time Dependent" which the picture 1.2 shows that. Is there any problem with this "Warning sign" or not?

Could you please tell me the here what does "study controlled" mean? I checked the COMSOL help but did not find related comments about this.

In another case, when I choose "Stationary" in Equation form of "Magnetic and Electric fields" interface, it can be seen in picture 2.1, the "Warning sign" disappear and I can see the "Green tick sign", see picture 2.2 please. But the problem is that my model in this case does not converges !!!

Any help or idea would be appreciated.

Thanks,
Mohsen
↧

hallo fans pretty little liar are you ready baby

January 6, 2015, 4:07 pm
↧

Non-Newtonian models in COMSOL

January 3, 2015, 6:48 pm
$
0
0
HI,

I am re-posting this as I didn't receive a lot of attention last time. I have only seen Carreau and PowerLaw models for non-Newtonian flows in COMSOL. However, I want to implement a different constitutive relation of my own that involves some constants and has it's own unique functional form in the CFD module for non-Newtonian flow! The independent variable is still shear rate (gamma dot) but it is a different function than Carreau.

Does anyone how to define this function and then how to call it for calculation of the viscosity for a non-Newtonian flow simulation?
↧

meshing

January 6, 2015, 11:34 pm
$
0
0
hi

I am new to the comsol 3.5 , I am working on two pass cooling channel with ribs .now what I want is that I want fine meshing of the bottom plate of channel and normal mesh on rest of the geometry ,can anybody tell me how to do different meshing of different parts of geometry.
↧
↧

high mach number flow simulation (CFD)

January 6, 2015, 11:52 pm
$
0
0
Hi
i want to simulate an airfoil in high mach flow .
i am using high mach number flow module k-epsilon method .also adaptive mesh method employed to finer meshing near shock wave region .
for the special case of airfoil, simulation for all velocity converged(even for higher than 10 mach number flow) and result are acceptable .but when i change slightly dimension of the airfoil ,simulation cannot be converged,even in very low mach number(nearly 2 mach ). I am really confused.nearly two month i am working on this simulation but i can not identify the problem .
to solve the problem please help me.
best regard
↧

Acoustic-Solid Interaction physic relational problem

January 7, 2015, 3:42 am
$
0
0
Hi all, can you tell me how can I set the sound attenuation in the solid model?
Someone told me that it is in the damping that is located in linear elastic material.
But I still donot know how to set the parameter for the competition of sound propagation loss.
Looking forward to your replay.
↧
Search

Laser Heating of a Silicon Wafer Model ID: 13835

January 7, 2015, 5:20 am
$
0
0
I think I should have written this somewhere else but since it took me at least half a day to understand that something is wrong with this model, I thought it will worth writing here. This model has a fundamental error since the rotational frequency is wrongly accepted as angular velocity. The difference is only a factor of 2pi. What I mean is: in the model v_rotation is defined in terms of rpm and it is converted to Hz by COMSOL automatically. On the other and the rotation is defined in terms of translation vectors (u's in COMSOL) as ux = -y*v_rotation, uy= x*v_rotation, which is in fact should be ux = -2*pi*y*v_rotation, uy= 2*pi*x*v_rotation. Try yourselves and see.. Thanks a lot.
↧

Assigning material

January 7, 2015, 6:51 am
$
0
0
How can I assign the Young modulus (E) of the solid material which has different E (in a specified time) over the volume. I tried to use of the interpolation function (spreadsheet) but it dosen't work. The spreadsheet data is E for (x,y,z). The type of erroe it shows is 'failed to evaluate variable Jacobian'. thanks
↧
↧

heat and mass transport in porous media

January 7, 2015, 6:58 am
$
0
0
hi
plz . can any one answerme about
if COMSOL can simulate heat and mass transfer in porous media or not ?
witout need to any other programm
best regards
↧

Diffuse Double Layer Model - Interpretation of results

January 7, 2015, 8:16 am
$
0
0
I have been trying to recreate this model :

www.comsol.com/model/model-of-a-diffuse-double-layer-13525, which is based on the Paper web.mit.edu/bazant/www/papers/...2005_SIAM_J_Appl_Math_MTF2.pdf .

This might sound like an unusual question, but I'm regarding the dimensionless charge density -Why is the Charge density across the domain asymmetrically distributed? i.e, close to the Cathode, it reaches 0.6, while on the Anode, it reaches -1.8. I'm not really able to understand what is the physical reason behind this. Mathematically, I can see that the distribution of Anions is an exponential curve dependent on Potential, which is inherently asymmetric, but even then, shouldn't the distribution of Cations and the distribution of Anions be 'inverses' of each other? I have attached an image of the plot that I'm referring to.
↧

How to generate scattered field for multiple angle of incidence ( or sum parametric sweep solutions)

January 7, 2015, 10:56 am
$
0
0
Hi all,

i'm working on a scattering problem. I need to find the scattered field over a range of angle of incidence.
I know that i can plot the scattered field for each angle of incidence , but i want to plot the total field for all angle of incidence added together. So basically, i want to find out the total scattered field from multiple plane waves. I figured i could join the results for this parametric sweep, i could obtain the total scattered field, but i'm not sure how do to so in comsol, since i don't have a 2nd dataset. Your help is greatly appreciated!

thanks,

Yuchu
↧
Viewing all 26323 articles
Browse latest View live
Search