Latest Discussions - COMSOL Forums

Nonlocal effects for nanobeam

Thu, 18 Apr 2024 20:09:06 +0000

good evening everyone, has anyone ever modeled a beam hand in COMSOL? If so, how did you account for non-local effects?

wave propagation

Thu, 18 Apr 2024 17:26:26 +0000

How can i simulate wave propagation in a porous media based on the Biot's theory

Temperature & SOC dependent electrical circuit components

Thu, 18 Apr 2024 16:39:23 +0000

Hi everyone,

My idea is to couple a 0D Battery Electrical Circuit (BEC) with a 2D or 3D thermal model. For a given current profile, the BEC will generate heat, which will be used as an input for the thermal model.

The tricky part: the components (resistor, RC-couple, etc.) of the Battery Electrical Circuit should depend on the SOC and the temperature. So I want to use the output (temperature) of the thermal model as an input to the BEC to adjust the values of the circuit components. The BEC will in turn calculate the heat loss, which will be used as an input to the thermal model, and so on.

Question: Is there a way to define the electrical circuit components as SOC & temperature dependent?

Thanks for any answer!

Refractive index of PML

Thu, 18 Apr 2024 12:34:50 +0000

Hi, In the mode analysis of optical fibers, the Perfectly Matched Layer (PML) is used as a boundary condition. Generally, the cladding layer of the fiber serves as the PML, and its effective refractive index is set to be the same as or slightly larger than that of the core material. This results in the effective refractive index of the designated PML being a constant real value rather than a complex-valued function of radial coordinates. For a cylindrical-shaped PML, its refractive index should ideally be a complex-valued function of radial coordinates. How should this simulation setup be understood, and how can the effectiveness of this method be explained? Your assistance is greatly appreciated.

Electromagnetic Force Distribution

Thu, 18 Apr 2024 11:15:17 +0000

I am working on an E-I electromagnet. I want to connect the forces generated by the electromagnet to a force distribution on a Solid Mechanics module to calculate the stresses on the I beam. How do I get this force distribution? I seem to only be able to get the total force generated by the electromagnet.

weak form Reynolds

Thu, 18 Apr 2024 07:31:26 +0000

Hi，how do I input the weak form Reynolds equation into the expression of the weak form partial differential equation?

large deformation of cantilever - module selection

Thu, 18 Apr 2024 07:09:05 +0000

I would like to model a cantilever consisting of three layers. Of silicon, aluminum and nickel. Of which nickel will not be along the entire length of the cantilever. The force will act on the same section as the nickel. I am interested in very high deformation. With what module it is best to model this. Is Solid mechanics ok, for this task? I tried to make a larger deformation in the plate model deflected by magnet. And there were problems with it. Is solid mechanics suitable for large deformations?

(Transient Heat Transfer) Multiple step studies with previous results

Wed, 17 Apr 2024 17:47:21 +0000

Hello,

I am trying to represent a 2D permanent mold casting situation involving molten metal and a steel mold. This means I would like to run multiple cycle casting situations. The goal is to see the temperature behaviors of mold and casting for each cycle.

Below is my B.Cs and I.Cs.

Molten metal initial temperature: 750C

Mold temperature: 400C

interface resistance between casting and mold: 1500W/m2K

Upper of the molten metal and Both sides are insulated,

and air h is applied on the bottom of the steel mold

Then, I ran the step 1 study and saw how temperatures were developing on castings and molds until the molten metal solidified.

Here is a problem. I would like to utilize the mold temperature, which was the result in Step 1, for the next step, and I could do it. However, how can I apply a molten metal temperature of 750C again for the next step?

Attached is my model.

Could you please help me?

Convergence problems in electrohydrodynamics

Wed, 17 Apr 2024 14:29:52 +0000

Using electrostatics, level sets, laminar flow, and dilute matter transfer, the calculation appears that variable constraints are found in different separation groups, try merging these groups. The last time step does not converge.!

Integrating Second-Order Susceptibility χ (2) for Nonlinear Optics simulation in THz region

Wed, 17 Apr 2024 11:49:43 +0000

I want to explore the nonlinear optic phenomena for some matamaterial such as LiNO3(LNO) and ZnTe, with second-order susceptibility. Our challenge is to incorporate a 3 by 6 matrix susceptibility to the optical properties of a material. More explicitly, the polarization elements; Px^{(2)}, Py^{(2)} and Pz^{(2)} are equal to the multiplication of susceptibility χ (2) as 3 by 6 matrix to convolution 6 by 1 matrix that its elements are: {Ex ⋅Ex}, {Ey * Ey}, {Ez * Ez}, 2{Ey * Ez}, 2{Ex * Ez} and 2{Ex * Ey}. my problem is that I dont know how to do this simulation. As one solution, I am thinking to obtain polarization using an approperiate PDE or ODE module in COMSOL. Alternatively. using Add User-Defined Variables I might be able to define separate expressions or function for each element of the χ(2) matrix if the elements have different frequency dependencies. I have attached relevant screenshots for reference. Any insights or experiences similar to mine would be greatly appreciated.

Reaction Rates

Wed, 17 Apr 2024 10:30:50 +0000

Hello, I am using COMSOL 6.1 Reaction Engineering Module. I am modeling a reaction in the form of:

A-->B+C+D+... The reaction depicts the heat breakdown of a solid substance. I obtained the reaction kinetics [A, E] values from TGA measurements. When I use these numbers to define the reaction constant [k], the reaction does not occur. I receive a graph with the concentration of [A] represented by a fixed straight line and all products at zero. Has anyone ever modeled reactions using kinetic values from TGA data? Please help.

I also want my reaction rate to be temperature dependent. my temperature is rising from 0K to 800K and then stays at that point. I tried to identify temperature as a function but I couldn't include the function in the rate constant. COMSOL wont recognize it. If anyone has such success modeling these type of scenarios. I would be grateful to hear you thoughts.

Best Regards

Modeling the deformation mechanism of a ferromagnetic cantilever using magnetic field

Wed, 17 Apr 2024 09:54:44 +0000

I would like to model a mechanism for deforming a terahertz metasurface with a magnetic field generated by a set of coils that allow magnetic field generation in three directions. In this case, I am interested in finding the value of the magnetic field density that will allow me to deform the metasurface. At first, I can assume that it will be a simple cantilever consisting of a ferromagnet with a thickness of 1 um and dimensions of 5x100 um (by default it will be 3 layers: silicon, alluminum and ferromagnetic. In the future, I plan other shapes than ordinary cantilevers (such as spirals). I am interested in very large deformations, including complete bending of the cantilever. How to model this mechanically. Is it possible not to put the coils in the model just to adjust the intensity of the magnetic field in the domain. How can this be best implemented? ? I attach my model, could you take a look at what is wrong with it?

Is it possible to intervene in the parameteric sweep of time-dependent solver?

Wed, 17 Apr 2024 07:16:01 +0000

Hi there,

I want to model the evolution of structure configuration toward its steady state under various specific parameter combinations and given initial conditions.

The problem is that the period from the initial condition to the steady state is unknown before modeling. For now, I use the parameteric sweep of time-dependent solver and specify a shared pretty long time period Ttotal to each of the time-dependent study. After a trial run, I find that for some parameter combinations, the steady state has already been reached far before Ttotal, while for some other parameter combinations, the steady state has not been reached yet.

So, I was wondering if it is possible (maybe through scripts) to intervene in the parameteric sweep of time-dependent solver, so that when a customized criterion (such as energy convergence) is satisfied, comsol can stop the calculation of current parameter combination and start to the next combination, automatically.

Best, HC L

Undefined Variable Error lambda in a whispering gallery resonator model

Tue, 16 Apr 2024 19:06:36 +0000

I am trying to find the bound modes of a particular WGM, similar to the Application found here: https://www.comsol.com/model/whispering-gallery-mode-resonator-90741. Because my structure is quite large (~mm scale), I have decided to use the Beam Envelopes interface instead of the frequency domain interface. When I run the simulation, I get an error for an undefined variable lambda that I cannot find anywhere in the model. Is there another undefined parameter that I need to work with? Note that the geometry is imported from another .mph file that I was originally working with.

Edit 17.4.2024: The issue seems to be purely in plotting the results. I am able to execute the solver completely without error, although I am not sure if it is storing valid solutions. When I try to plot a particular function, I get the error I have screenshotted above (pasted here so that others can read it)

I'm assuming this is some sort of stack trace, but as I am new to comsol I am not entirely sure how to read it.

Undefined variable. - Variable: lambda - Geometry: geom1 - Domain: 7 Failed to evaluate variable. - Variable: freq - Defined as: ((-0.5)lambda)/(pii) Failed to evaluate variable. - Variable: comp1.ewbe.freq - Defined as: freq Failed to evaluate variable. - Variable: comp1.ewbe.omega - Defined as: (2pi)comp1.ewbe.freq Failed to evaluate variable. - Variable: comp1.ewbe.iomega - Defined as: comp1.ewbe.omegai Failed to evaluate variable. - Variable: comp1.ewbe.k0 - Defined as: (comp1.ewbe.iomegasqrt(mu0_constepsilon0_const))/i Failed to evaluate variable. - Variable: comp1.ewbe.phi1 - Defined as: comp1.ewbe.k0comp1.ewbe.posz Failed to evaluate variable. - Variable: comp1.ewbe.expPhi1 - Defined as: exp((-i)comp1.ewbe.phi1) Failed to evaluate variable. - Variable: comp1.ewbe.E1z - Defined as: comp1.ewbe.expPhi1((comp1.E1rcomp1.pml1.T31)+(comp1.E1zcomp1.pml1.T33)) Failed to evaluate expression. - Expression: comp1.ewbe.E1z - Plot: Electric Field

surf plot in 3d models

Tue, 16 Apr 2024 16:31:59 +0000

Is there a feature of a surface plot as surf(x,y,z) in Matlab? In 2d models, 2d surface plots can be converted to a surf plot by a height plot but in 3d models this seems to be missing?

combination solution at different positions in the same point or line graph

Tue, 16 Apr 2024 15:51:17 +0000

Suppose I have a solution written generally as u(x,t). I need to compute u(0,t)-u(1,t) and plot this quantity as a function of t. I've found even for this trivial application in point graph or line graph under a 1D plot, I have difficulty. I need to pick the x=0 point or edge to obtain u(0,t) and pick the x=1 point or edge to obtain u(1,t) but there's no way of putting u(0,t) and u(1,t) together in the same point graph or line graph!

multiple studies in one "solve"

Tue, 16 Apr 2024 15:46:53 +0000

Is it possible to click "solve" to run multiple studies in .mph in one go?

Meshing large domain

Tue, 16 Apr 2024 12:48:28 +0000

Hi,

Does anybody know how to speed up meshing process of large domains? It takes hours to finish meshing using my i7 CPU 64GB RAM? Are there tips in COMSOL or in the hardware that I can replace? I should mention that the the domain contains two different meshes: very fine (mm size) and coarser mesh (m size).

Kind regards, Taofik

EWBE - Fiber to Waveguide Coupling

Mon, 15 Apr 2024 18:32:25 +0000

Hello,

I am new to COMSOL and doing some initial tests to get familiar with the Wave Optics model. One of my intial tests is trying to determine the coupling efficiency from an SMF-28 fiber to a slab dielectric waveguide. I am using the EWBE physics and was able to succesfully get the beam envelope for the fiber alone, then fiber dispersing into air, and now I have added the waveguide but I am getting an error when trying to run the study. I used the tapered waveguide and the fiber to fiber coupling from the COMSOL learning center as guides.

My general approach was to define three domains in the simulation. In each one I am using EWBE to find the phase value psi, and I gave separate definitions in each of the three domains. I have PML around the 3 sides that don't include the input port. I am using an input numeric port and and output domain backed slit port. In between the fiber-air and air-waveguide domains I have transition boundary conditions. My study uses two boundary mode analysis steps for each of the ports and then a frequency domain step to solve for the field overall.

I am including the MPH file and the output error log. The error log is very difficult to parse, so I don't really know where the source of the error is. COMSOL version 6.1 and running on an M1 mac with Sonoma 14.4.1 if that is of any relevance.

As a side note, my next step is to try and optimize the thickness of the waveguide to maximize coupling, so if anyone has good tips on using the Optimization Module with the EWBE physics that would be appreciated.

Thanks, Jack

Amplitude Modulated sinusoidal

Mon, 15 Apr 2024 14:53:30 +0000

Hello,

I want to amplitude modulate a sine wave. I have a simple 2D two-turn coil using the magnetic field node with a current excitation. Using the frequency domain I was able to extract good results but now I want to amplitude modulate wave.

Can anyone help me introduce amplitude modulation to my 2D coil, frequency domain, magnet field node (current excitation), simulation?

thank you