Latest Discussions - COMSOL Forums

Modelling of carbonate dissolution

Sat, 20 Dec 2025 20:31:30 +0000

Hi all,

Thank you for your assistance. I am trying to model moving of boundaries as a result of carbonate dissolution in COMSOL. Is it possible anyone can help me with a file to get started with or a step by step approach to get a simple model (single pore model) done?

Sincerely, E.A

Union/Assembly

Fri, 19 Dec 2025 07:48:30 +0000

I am doing magnetic field study and made one U structure rectangular coil with an Air Domain. When I am forming a union, current is flowing is 1 or 2 legs of the coil and not in all 3,and forming magnetic field according to those conductors. When I am forming assembly, magnetic field is not being generated in Air but current is flowing in every conductor, How to resolve this issue!!!

The symmetry precision problem

Thu, 18 Dec 2025 01:52:20 +0000

Recently I'm using Comsol for BEM of static electric field. I found when the geometry is symmetric, there will always be a ~0.0001V asymmetry of the calculation results when the voltage of a specific electrode is set 1V. It remains even if I set the physical field as a symmetric one, and I also tried to reduce relative tolerance from 1e-3 to 1e-4, and it failed to fix that as well. Is this inherent to Comsol? I'll appreciate it if someone could help me with that.

Stationary study of rotating thrust magnetic bearing

Wed, 17 Dec 2025 10:04:22 +0000

Hi Everyone, I have been working on a simulation of a typical active thrust magnetic bearing trying to simulate the losses in such bearing. All physics was modeled using Magnetic and Electric field module. The stationary study worked well until I began to introduce rotation to the thrust disc, using velocity Lorentz term, above the RPM of 5 the solution doesn't converge. I tried several approaches to fix this issue and none of the succeeded:

Increasing mesh progressively
Changing non linear magnetic material to linear
Replacing physical air gap with thin low permeability gap
Changing iterative solver from GMRES to FGMRES
Changing Iterative solver to direct solver (MUMPS)

Unfortunately, none of these approaches worked. Do you maybe have an idea for another approach to solve this convergence problem? I have atatched my current model as reference.

RF Module (emw): Frequency-domain spectroscopy of a single biological cell

Wed, 17 Dec 2025 09:07:24 +0000

Hello, I am working with COMSOL Multiphysics (v. 6.4) – RF Module (Electromagnetic Waves, Frequency Domain) because my frequency of interest is 2.45 GHz. My model consists of a single biological cell, with a realistic geometry imported from CAD. The objective is to study the electromagnetic response of the cell (e.g. transmembrane electric field, SAR, polarization) when exposed to an RF field at 2.45 GHz. I am trying to understand how to properly “deliver” or excite the electromagnetic field on the cell using the RF Module, in a way conceptually similar to how Terminal conditions are used in the AC/DC Module. In other words, I would like the field to be present, but without explicitly modeling a physical RF source (antenna, waveguide, coaxial cable, etc.) and without imposing boundary conditions directly on the cell membrane. I have considered the following approach: * Electromagnetic Waves, Frequency Domain (emw); * Scattered field formulation; * Background electric field (plane wave excitation); However, I am still unsure how to correctly apply and interpret this excitation in practice, especially with respect to the physical meaning of the background field and how it relates to a real RF exposure at 2.45 GHz. An alternative idea I initially had was the following: * Surround the cell with a box representing the extracellular medium * Apply a Port condition on one face of the box * Apply Perfect Electric Conductor boundary conditions on the other faces The intention was to generate a propagating electromagnetic field inside the box and expose the cell to it. However, I am not sure whether this setup is physically correct or recommended, and whether Ports can be used in this way when no actual transmission line or waveguide is modeled. Therefore, my questions are: 1. What is the recommended way to excite an RF field at 2.45 GHz on a single cell using the RF Module, without explicitly modeling an RF source? 2. Is scattered field formulation with background electric field the correct approach for this type of problem, and if so, how should it be properly implemented? 3. Is the idea of using a box with a Port on one face and Perfect Electric Conductor on the others physically meaningful in this context, or should Ports only be used when modeling real waveguides or transmission lines?

Any guidance, best practices, or references to similar models would be greatly appreciated.

Defects extending along the z-direction in BEM of static electric filed simulation

Tue, 16 Dec 2025 04:16:55 +0000

I'm trying to use BEM to calculate the Spatial electric potential field distribution of specific electrodes design, and applied that to trap some charged particles. However, I found that there seems to be some defects of the field that results in extension of the particle arrays in z-direction, which is not physical. And these defects will appear at specific sites, so I suppose it's caused by the defects of field calculation by Comsol. I used user-specified grid, and the grid looks regular, and I calculated the spatial potential distribution, found that there is a 0.001V error for symmetric positions. Are there any solutions to this problem? Thank you!

phase field method

Sun, 14 Dec 2025 07:18:37 +0000

I am a master's student currently studying the fracture phase field method. I plan to use COMSOL to simulate phase field fracture. I have read some papers on phase field fracture and some case studies, and noticed that the Solid Mechanics module, Poisson Equation module, Helmholtz Equation module, etc. were used. However, I find many of the settings difficult to understand. Therefore, I would like to know if your company has any operation documentation specifically explaining the use of COMSOL for phase field fracture analysis? If it's possible to create such an operation document, I would be extremely grateful! Thank you！

Certification

Sat, 13 Dec 2025 19:54:48 +0000

I know in the past, COMSOL had certifications for consultants and individuals.

What is the status of that now and are there any references. I tried to search before posting but I couldn't find any so that's why I'm posting in the discussion forum now.

Thanks in advance for your comments.

Skin effect not appearing with union of two cylinders in COMSOL

Thu, 11 Dec 2025 11:09:08 +0000

I am modeling a conductive disk in COMSOL under AC conditions (14 GHz and 80 GHz) with the following setup.

Geometry and Physical Setup

Geometry: The conductor consists of two thin cylinders stacked on top of each other and merged using the Form Union geometry operation. Radius (per cylinder): 0.015 m Height (per cylinder): 0.0005 m Total Geometry: A disk with R=0.015 m and total height 1×10^−3 m.

Current direction: The current flows radially across the disk (from one side of the diameter to the opposite side, i.e., perpendicular to the stacking direction).

Physics: Coupled interfaces: Magnetic Fields (mf) and Electrical Circuit (cir). Coupling: The mf interface uses Lumped Ports for input and output, coupled to the cir interface via the External I vs. U node. The components in the cir circuit are connected in series.

Problem / Observation

When analyzing the current density ? in a cross section: - With the Union of the two cylinders: The expected skin effect (concentration of current at the outer edges of the disk, parallel to the current direction) is not visible. The current density is uniform across the cross section.

As a control, using only one cylinder (R=0.015 m and total height 0.001 m): The skin effect is observed correctly.

Question I don’t understand why, even though I apply a union to the two cylinders, I don’t get the expected skin effect, just like with a single cylinder?

Inclination Flow

Mon, 08 Dec 2025 02:18:38 +0000

Hi, I have a some problem to do the inclination in my geometry. I already put all the parameters and I run the program. But the result error.

Flow splitting at Comsol 6.2 Version

Sun, 07 Dec 2025 21:15:00 +0000

Hello Everyone,

I have a COMSOL Multiphysics geometry as shown in the figure below. The model consists of a main pipe with an inlet at the upstream end and an outlet at the downstream end.

The inlet flow rate is set to 40 L/h. At the first junction, where a U-shaped branch begins, I would like to split the flow such that half of the inlet flow rate is diverted into the U-shaped pipe (20 L/h).

Could you please advise how this flow split can be defined or enforced in COMSOL?

An image of the complete setup is also attached.

Desiccant Cooling System Simulation

Sun, 07 Dec 2025 03:56:27 +0000

Hi, my name is Edward Akpee, a masters student working on a project that requires simulation. i what to ootimize a solar Assisted Desiccant Cooling System. its my first time using comsol, I don't know how to go about it. i need your assistance.

collision between two gases

Sat, 06 Dec 2025 12:05:38 +0000

Hi i want to model the collisions between the particles of a steady gas (distribution is maxwellian) and a moving one with a defined velocity (the distribution is maxwellian but i sum the known velocity in direction z). the model should be axysimmetric. i know the densities of both gas, particles mass and their volume. i want to know how the particles of the first gas change density and i want to mesure the pressure waves that form. how do i do? i'm a beginner but i dont think that would be too hard

Optimization of an electron beam radius

Sat, 06 Dec 2025 11:52:08 +0000

Good morning, I am performing a simulation for an electron accelerator which consists of a cathode, an initial accelerating potential and then the electron beam is guided by six coils on a target which is almost 1m far from the cathode. The simulation itself is ok and I can see the beam.

What I'd like to do next is performing an optimization of the radius of the beam with respect to the currents in the coils. To do so, I defined a boundary probe on the surface of my target with the expression called R_spot, where qy and qz are the position of particles in y and z.

Then I set up an optimization with objective function (spot_target - comp1.R_spot) where spot_target is defined as 1[mm] in my parameters. I ask for a minimization with control variables the six coils currents. But I keep getting this error and I don't understand how to resolve:

Error in sweep. - Parameters: "I_coil_1","20","I_coil_2","20","I_coil_3","20","I_coil_4","20","I_coil_5","20","I_coil_6","20" Cannot evaluate expression. Undefined variable. - Variable: comp1.qy - Geometry: geom1 - Boundary: 87 Failed to evaluate expression. - Expression: sqrt((comp1.qy^2)+(comp1.qz^2)) Failed to evaluate expression. - Expression: comp1.bnd1_operator1(sqrt((comp1.qy^2)+(comp1.qz^2))) Failed to evaluate variable. - Variable: comp1.R_spot - Defined as: comp1.bnd1_operator1(sqrt((comp1.qy^2)+(comp1.qz^2))) Failed to evaluate variable. - Variable: opt.obj1

I also watched the Optimization tutorials in the Learning Center. If someone can help me figuring out what I'm doing wrong I will be grateful since I'm few days into Comsol and I'm learning it by myself. Thank you in advance, Emanuele

How to prevent Model Remeshing while solving after Updating Boundary Selections (No Geometry Change)

Thu, 04 Dec 2025 20:14:42 +0000

Hi,

In my workflow, I use a union of ball selections(created in geometry, after Form Union) to identify and select new boundary surfaces. Then, I update the boundary conditions with these newly created selection sets. Importantly, I do not modify the actual geometry, just create an union of different ball selections to use as boundary.

Here’s a simplified snippet of my code where the BCs are assigned and then the study is run:

model.component('comp1').physics('ec').feature('term1').selection.named('geom1_termUnion'); model.component('comp1').physics('ec').feature('gnd1').selection.named('geom1_groundUnion');

model.study('std2').run;

termUnion and groundUnion are unions of several ball selections.

However, when I run the study, COMSOL remeshes the entire model before solving. I would expect that only the solution should update, there should be no reason to regenerate the mesh if the geometry hasn’t changed. That is the entire idea of using ball selections.

Why does COMSOL trigger a remesh in this situation, even though only boundary selections have changed? Is there a way to prevent unnecessary remeshing when updating boundary conditions via selection sets?

Any tips would be greatly appreciated! Thank you in advance for your help.

Multiple lumped ports with excitation

Thu, 04 Dec 2025 13:54:31 +0000

I want to simulate a chip with 4 independent CPW resonators, each having an input and output lumped port at the end of the respective resonators to calculate the two-port transmission S-parameters. I add "EMW - Frequency Domain" Study and add total 8 lumped ports (4 pairs with excitation ON and OFF) and compute. I expect getting S-parameters in the Results but I don't get it and instead all I can get is "Reflection coefficient with multiple excitation" only for 4 ports which had excitation ON. How can I get S-parameters for 4 CPW like, (S00, S10, S01, S11), (S22, S32, S23, S33), etc.?

Efficient Selection of Multiple surfaces as boundaries

Tue, 02 Dec 2025 22:17:19 +0000

Hello everyone,

I’m modeling a system in COMSOL where two rectangular boundary regions must be dynamically moved across a surface, as guided by an optimization algorithm. My current method updates rectangle coordinates on a workplane for each optimization step, but this requires running form union and remeshing every time boundaries change, which is very computationally expensive.

As an alternative, I’m considering pre-generating hundreds of small rectangles on the surface as potential boundary areas. For each optimization iteration, I would simply select subsets of these rectangles to define "terminal" and "ground" regions for electric current physics, without touching the geometry or mesh modules.

To clarify:

Geometry and mesh remain unchanged; only boundary conditions are reassigned among rectangles. I need an efficient way to programmatically select which rectangles to use at each step, ideally automated via MATLAB LiveLink. Currently, I reference rectangles by their surface numbers, which works for a simple test case, but managing hundreds will quickly become unwieldy. I’d like to avoid manually tracking indices.

My main questions:

Is this approach feasible for avoiding repeated form union and remeshing while constantly changing boundary conditions? Is there a robust way in COMSOL to tag or systematically reference large numbers of geometric entities for flexible boundary condition management, especially with MATLAB automation?

I've attached a sample model showcasing this approach, with hundreds of small rectangles and the previous large-rectangle method for reference. For this model, my optimization algorithm will guide me to shit my terminal and based on that I want to select another set of rectangles.

Thank you in advance for any insights!

Optical generation rate

Tue, 02 Dec 2025 16:23:51 +0000

i have computed total generation rate from ewfd module in a varialbe named G_tot that is wavelength dependent and is a spacial data. now i want to integrate the data over wavelength. so the data remains only spatial non dependent on lemda. but when i am usign the integrate operator it just reduces a dimension. as a result the unit of generstion rate becomes (1/m2.s) instead of (1/m3.s) l. what is the solution .

Silver faces and material assignment crisis

Tue, 02 Dec 2025 15:46:42 +0000

Hello, I am struggling with a simple modulator 3D structure. I tried to create the geometry in many different ways but the cleanup wizard always appears to bother me. I tried to extrude a 2D surface, or directly create the 3D blocks, but I always encounter problems such as silver faces, and I experience a constant struggle with material assignment (many boundaries are not correctly assigned). I had a 2D model with the identical structure, and never encountered these kind of problems. WIth the 3D, it is a mess. Please help me understand how to cope with this. I want to compute the S parameters and characteristic impedance. I've also tried to insert some boundary conditions such as impedance boundary condition but I am not even able to select the correct faces! This model is simple, there is nothing sketchy about it. Please help me. Thank you

Inconsistency in Theory manual 6.0 and 6.2 - Multiplicative plasticity

Tue, 02 Dec 2025 14:25:51 +0000

Dear Community and Moderators,

I am seeking clarification on the flow rule for finite plastic strain with multiplicative decomposition. In COMSOL 6.0 documentation, the flow rule is:

and in COMSOL 6.2 documentation, the flow rule is:

I am confused about which one COMSOL actually uses. Or are they equivalent? Any help will be useful, thank you. Please see the attached screenshots as well.