Latest Discussions - COMSOL Forums

Moving Mesh Simulation – Fatal Error During Assembly/Calculation

Fri, 13 Jun 2025 07:39:29 +0000

Dear colleagues,

I would like to seek your advice on a moving mesh simulation issue.

Current Model: I am simulating "the motion of a current-carrying conductor rod under the influence of a coil's magnetic field."

Physics Interfaces Used: Moving Mesh Magnetic Fields (AC/DC Module) Global ODEs and DAEs (for equation-based modeling)

Implementation Details: In the Magnetic Fields interface, I added a Force Calculation condition to compute the electromagnetic (Lorentz) force. The calculated force is passed to a global differential equation: F=ma=mxtt The displacement is then fed into the Moving Mesh interface to update the conductor rod's position. Since the electromagnetic force changes as the rod moves, I use Transient Analysis in the study step.

Issue: Despite multiple adjustments, the simulation consistently fails with the error: "A fatal error occurred during assembly or calculation."

Request: Has anyone encountered similar issues or have suggestions for troubleshooting? Any insights on: Correct coupling between Magnetic Fields, ODE, and Moving Mesh? Solver settings for such coupled electromechanical problems? Common pitfalls in force-driven moving mesh simulations? Thank you in advance for your expertise! Best regards,

Liu

Setting of anisotropic conductivity of carbon fiber braided composites

Fri, 13 Jun 2025 05:03:12 +0000

I studied the cases in the official case library, trying to define the anisotropic conductivity of carbon fiber with curvilinear coordinates（cc）, but I don't know how to correctly quote the curve coordinates in the current (ec) and magnetic field (mf), do you have a partner who has done similar research?!!!

Modelling of B-Field with AC/DC Module

Thu, 12 Jun 2025 21:48:35 +0000

I am currently working on building a model to simulate the B-Field of an electro-magnet. This is my first project on COMSOL and I am self taught so I am know i am making errors in my simulation approach for sure. I've attached an outline of my approach showing my Boundary conditions, Mesh and results. The results are not too far of the measured data (in a small location - I still need to check the rest of the field area) but i don't think my model is very stable.

RLC circuit not giving correct resonance

Thu, 12 Jun 2025 21:05:29 +0000

Hi, I am trying to test a circuit with one lumped port and 3 lumped elements, one for R, L and C respectively. I am using 1 mohm, 1 nF and 1 nH as my values, so I should get a resonance at 159.15 MHz, with a Q of 1000 giving a width of 0.2 MHz. But when I sweep the frequency around this, I do not see any resonance. Why might this be? Thanks.

Possible to Obtain Current/Current Density in Electrical Currents, EC, from different Physics?

Thu, 12 Jun 2025 18:15:05 +0000

Hello,

Is it possible to impose a current/current density in a Electrical Current domain, from a different physics? In my model, i have a coil, hts superconducting coil, using Magnetic Field Formulation, mfh, where i have a current density in the hts coil domain. How can i have the same current density, seen in the mfh domain, in the Eletrical Currents side? How can i impose it/connect it?

My goal is to then have a terminal so i can connect it (hts coil) to an Electrical Circuit. Is there any other/possible way to connect this superconducting coil to an electrical circuit?

Thanks

Incorrect S11 resonant frequency

Thu, 12 Jun 2025 16:24:23 +0000

So based on previous discussion, I have managed to obtain better S11 values in negative range. However, I am unable to get the correct resonant frequency.

I am not sure if it because of improper modelling, or that of wrong lumped port definition.

Based on the image below https://gusbertianalog.com/content/images/2024/11/MEAS_ROUND_CC.png

from the author of the article https://gusbertianalog.com/loop-antenna-design/

I am supposed to get the resonant frequency at 434 MHz, but instead i get it at 460 MHz.

I have uploaded the comsol file.

(P.S. I tried to upload this question, but i got a message saying i cant view this question in the forum, so i am uploading the question again. )

Activated carbon fiber packed bed adsorption

Thu, 12 Jun 2025 08:39:32 +0000

How to couple chemical adsorption and physical adsorption when both exist?

Model amorphous silicon crystallization with COMSOL

Wed, 11 Jun 2025 20:28:13 +0000

Hello, I am interested in modeling laser induced crystallization of amorphous silicon (a-Si). The chain of phase transformations I am interested in is: a-Si -> Liquid Si -> polycrystalline Si. Is there a good way to model this in COMSOL? The main difficulty I have with Phase Change Material subnode of Heat Transfer in Solids module is that upon cooling the material drops back into a-Si phase which is not what should happen.

Thank you!

CFD - Phase Field in Fluids - Solver Settings

Wed, 11 Jun 2025 14:48:39 +0000

Hey Comsol Peeps,

I struggle with setting up a CFD simulation in which a small amount of liquid (micro liter regime) should be rotated within a hollow shell of 2mm radius. I always run into convergence errors after less than 1000 time steps no matter my settings.

The simulation is 2D and so far I only try to simulate the fluid resting inside the shell (no rotation). The used physics are the lamina flow interface with gravity included (the gravity will ramp up over time to make the initiation less demanding) and the Two-Phase Flow, Phase field interface. As phase field settings I chose for the interface thickness the default: pf.ep_default as mobility tuning parameter 0.5 m*s/kg as external free energy 0 J/m^3 (these values are rather guesses)

When computing I run into the error "Time adaptation interrupted." This happens at the time 0.057244s Looking at the velocity surface plot one can see two velocity hotspots at the fluid interfaces. However, checking the mesh at these hotspot , the quality doesn't seem to be too bad here. (see pictures) Does someone have an idea how to fix my simulation?

Fluid solution mapping error in Fluid-Acoustic Simulation

Wed, 11 Jun 2025 14:40:13 +0000

Hello everyone,

I am currently doing an Acoustic-fluid simulation with stationnary turbulent flow, convected wave equation and background fluid flow coupling modules. I have only one geometry for the whole study, in the first place i do the fluid simulation, then I map the fluid fields onto my acoustic mesh and finally I do my acoustic simulation with the background flow set up as a variable not solved for. My problem is that i already did many simulation using a simple geometry for my conduct (3D cylinder) which worked totally fine.

Now that I am using a more complex one based on CT-scans, I face issues during the mapping of my fluid fields onto the acoustic mesh. The error message is the following :

" Extrusion coupling operator interpolation failed for at least one point: no source point found. - Geometry: geom1 - Domain: 1 - Destination coordinates: [-0.00146266445760077; -0.0021864553895055991; 0.00060947754522279061], [-0.0014856891031490996; -0.0021811204460443394; 0.0006247936128450256], [-0.001450351221716216; -0.002194843619005469; 0.0006247936128450256], [-0.0014270113859349211; -0.0021981395225136669; 0.00060828294733559498], [-0.0014966022294479824; -0.00217111449601754; 0.00060828294733559509], ... Failed to evaluate expression. - Expression: withsol('sol1',comp1.spf.pA) Failed to evaluate expression. - Expression: (comp1.bffc1.p_map-withsol('sol1',comp1.spf.pA))dvol_spatial Failed to evaluate Jacobian of expression. - Expression: ((comp1.bffc1.p_map-withsol('sol1',comp1.spf.pA))test(comp1.bffc1.p_map))*(dvol_spatial) "

I understand that COMSOL struggles to find source point in my fluid domain but i can't understand why since my CFD simulation works fine and i have good results. Even with a mesh refinement of the source domain i keep having this problem, do you have any idea why and how I could solve this ?

I sincerely thank you for your time and your answers !

Acoustic streaming due to bubble oscillations

Tue, 10 Jun 2025 21:23:04 +0000

Hi,

I'm trying to simulate the acoustic streaming due to the bubble oscillations in a cylindrical cavity under acoustic fields from literature. I'm unable to get the expected streaming. It would be of great help if someone could help me figure out what is wrong in my setup. I expected the flow away from the bubble as in the reference but mine is reverse.

Thanks,

Learning Center Simulation: Modeling a Coil Connected to a Circuit

Mon, 09 Jun 2025 23:36:29 +0000

Hello,

I am following along to the COMSOL Learning Center video titled Modeling of a Coil Connected to a Circuit and have come across an issue. I am able to follow along seamlessly with the tutorial until 41:25, where the instructor changes the core to instead be a hollow solid instead of just being off to the side. When it's done in the video, COMSOL immediately understands what all the domains should be assigned to and computes the solution. But in my file (COMSOL 6.2.0.339) domain 5, which is the air inside the hollow core, can not have any domains applied to it as it is "Not applicable." Magnetic insulation is applied to both the coils and the core by default and I get a trivial solution. Is there something that I'm doing wrong when trying to follow along?

I attached both a .mph file from just before the hollow core is added, and a .mph file with the trivial solution (this one has Issue appended to the title).

surface impedance and sound absorption coefficient

Mon, 09 Jun 2025 11:51:30 +0000

How to calculate surface impedance and sound absorption coefficient for a two cavity parallel MPP system.

Expression for 2nd moment of area for domain probe?

Sun, 08 Jun 2025 12:30:51 +0000

Hi to all,

I was wondering if there is any built-in expression for calculating the second moment of area (area moment of inertia) using the Expression Tool. I’ve tried searching with keywords such as "inertia", "second moment", and "moment of inertia", but haven’t been able to find a suitable expression so far.

Any guidance on how to access or define such an expression would be greatly appreciated.

Best regards, Mirhan

Coupling physics, H-A formulation

Sat, 07 Jun 2025 11:49:45 +0000

Hello, i am trying to link two physics, applying the H-A formulation. My objective it to induce current in an hts, superconductor, pickup coil, using a copper coil. I wanted to use the H formulation for the hts coil, using COMSOL Magnetic Fields Formulation, mfh, and the rest (air+copper coil) solved with A, using Magnetic Fields Formulation, in a 2D axisymetrical model.

My question is how the linking/coupling of the two physics should be done, so there is continuity in the electromagnetic properties. I'm guessing that is should be done via boundary conditions, but i am a bit lost where, what should be done in each side (mf and mfh), and using what parameters/values.

Thanks

Fiber to Waveguide

Fri, 06 Jun 2025 16:55:56 +0000

Hi I am having trouble figuring out how to couple a fiber optic to a waveguide. I keep getting an error of having an undefined value in a residual vector and I don't know how to resolve it. I have attached my model. Any advice will be welcome!

Convective Flow

Fri, 06 Jun 2025 12:58:30 +0000

I am working on a coupled heat and water flow. The physics involved are Richard's equation for water flow and heat transfer in porous media for heat flow. In my model setup, I applied free drainage (pressure head equal 0) at the bottom boundary which means water is leaving the system. This is confirmed from the total velocity result however for the convective flow the values are negative and the sign convection is upward. Since water is leaving the system, convective flow should have arrow downward add to the heat loss from the system via water removal in addition to the heat removed via conduction.

Please any idea on what could be wrong and how to fix the model so the convective flow becomes positive and sign convection turns downward.

Please see the attached for more details.

Lumped port question

Thu, 05 Jun 2025 13:32:25 +0000

Hello,

I am trying to model a simple 3D wire with MHz AC current running through it using the magnetic fields interface with lumped ports.

I am printing out the parameters that the solution gives me for the lumped port, and I am a bit confused as to how to interpret them. I get an impedance, is this the impedance of just the region of the lumped port itself, or everything that is attached to the lumped port. Same for the power and voltage, is this for the whole system or only the lumped port itself?

Thanks

How to Plot Effective Relative Permittivity (Tensor Form) vs Wavelength

Thu, 05 Jun 2025 12:48:09 +0000

Hi, I am working on a metamaterial structure and have determined the effective relative permittivity components along the x, y, and z directions. However, I am unsure how to properly plot the real and imaginary parts of these permittivity components as a function of wavelength in COMSOL.

I attempted to define the anisotropic permittivity as a tensor in the Materials node, but after simulation, I am unable to access the material function to generate a plot of effective permittivity versus wavelength.

I would appreciate guidance on how to correctly define and visualize the wavelength-dependent effective permittivity components (both real and imaginary parts) in COMSOL.

Area moment of inertia (2nd moment of area) calculation

Wed, 04 Jun 2025 14:31:59 +0000

Hi everyone,

I’m trying to calculate the area moment of inertia (Ixx and Iyy) for an arbitrary 2D shape. As a simple example, consider a cross-section composed of two circles—I'd like to determine the area moment of inertia for this small circle.

Since the geometry can be arbitrary in other cases, I’m looking for a general method to compute Ixx and Iyy. I’ve noticed that COMSOL (e.g., through the "mass properties" function) provide mass-related properties like mass moment of inertia, but what I need are the area moments of inertia.

If anyone has suggestions or knows how to extract these values, I’d really appreciate your help.

Best regards, Mirhan