Latest Discussions - COMSOL Forums

Exporting data from COMSOL while computing electric field magnitude and direction in an electrospinning machine

Thu, 10 Jul 2025 14:49:05 +0000

I am building a simulation of an electrospinning machine on COMSOL, and my objective is to study the electric field magnitude and direction within the machine. I would like to export the magnitude data that shows on a 1D plot. Is there a way to obtain a table of values that I can open in Microsoft Word, Excel, or another similar platform?

COMSOL not using all memory available

Thu, 10 Jul 2025 14:27:36 +0000

Hello,

I was trying to run a COMSOL simulation and it would always go over the available amount of RAM and start using Virtual Memory, which was extremely slow. Therefore, I bought more RAM (went from 64gb to 128gb), but COMSOL doesn't seem to want to use the full amount it has available. Indeed, once it reaches 64gb, it starts to write on the disk again even though there is still twice as much RAM available. Thinking that this was because COMSOL didn't "recognize" that there is more RAM, a clean install was done of COMSOL MultiPhysics, but this didn't solve the problem. Furthermore, I also tried to disable the use of Virtual Memory to force COMSOL to use the available RAM, but the simulation ends when COMSOL reaches the 64gb mark, with an error saying that there wasn't enough memory.

Why does COMSOL not use the full amount of RAM available?

Thanks in advance for your help!

Marek

Failed to evaluate negative error for Heat Transfer in Porous Media

Thu, 10 Jul 2025 02:17:20 +0000

HI, I am trying to simulate a flow through packed bed and transfer between the fluid and porous bed using thermal non equilibrium case, but I am getting this error it is because of the model input in the Fluid section, when I enter a number then the error is not coming, but when I set the default ht/porous/fluid/T value then it is giving error, i have coupled the Brinkmans equation with the HTPM using Non Isothermal Flow Multiphysics

Choice between Transport of Diluted Species and Transport of Concentrated Species

Wed, 09 Jul 2025 02:50:07 +0000

My physical scenario is as follows: At room temperature and atmospheric pressure, a gas mixture of 80% Ar and 20% CO₂ is introduced at a flow rate of 500 sccm through two small inlets (each with a diameter of 1 cm) into a metal enclosure filled with air. The enclosure has dimensions of 3.14 m × 1.76 m × 0.24 m, and there are two outlet holes (also 1 cm in diameter) on the opposite side.

My question is: Given this setup, should I choose the "Transport of Diluted Species" or the "Transport of Concentrated Species" physics interface in COMSOL?

Thank you in advance for your guidance!

Continuity between two domains with different physics

Tue, 08 Jul 2025 17:21:48 +0000

Hello all,

I am developing a model for the combustion of a fuel pellet placed inside a furnace. There is air flow enabled in the furnace, which I would think "seeps" oxygen inside the porous pellet. Before I went into the complete heat, mass and reaction coupling, I just wanted to to run a study with just fluid flow interfaces defined. I am using the laminar flow interface for the furnace domain and brinkman equations for the porous pellet domain.

Since I defined the geometry as a union, I would assume that COMSOL automatically treats internal boundaries with continuity conditions, however, when I checked the results I see that the continuity is not maintained.

Currently, the interfaces between the pellet and the furnace are defined as walls, which I assumed would be overridden automatically. Now I think I need to add some additional B.Cs to let COMSOL know that there is an interface between the pellet and the furnace where continuity needs to be maintained. Or do I need to define identity pairs? Or is there anything else that I am missing. I am also having some trouble determining the source and destination boundaries while defining identity pairs.

I am attaching an annotated image with some information that might be useful.

Any help is much appreciated, thank you!

'Results While Solving' in Iterativ Stationary Solver

Mon, 07 Jul 2025 16:30:34 +0000

Hello!

Is it possible in an Iterativ Stationary solver calculation to see the successive converging results due to 'Results While Solving' every iteration or every few iterations??

I tried activating 'Results While Solving' in 'Fully Coupled' node or changing 'Update at', but I don't see any converging behaviour, though it's definitely there. I just see the graphical result of the previous calculation, though I know there is a noticable converging behaviour by testing with different 'Maximum number of iterations.'

Is such or anything like that possible???

A simple 'Yes' would be enough; then I figure out for myself.

Nice greetings

Eddy Simulation as Inspection

Mon, 07 Jul 2025 00:26:23 +0000

Hi, I'm a doctoral student using the License number COMSOL® 6.3, in my thesis. So, my objective is to simulate an eddy current inspection for a heat exchanger tube. You can understand this kind of test by watching the video on the YouTube link (https://youtu.be/MYd92lzhsX8?si=mtOjR_rgwhkP2LJP). How can I assemble this kind of model in COMSOL? Could you please help me? Basically, I need to insert a bobbin coil sensor probe inside a tube, where the probe should be moving inside the tube along time. In the end, I want to see the impedance plan for each step, XL versus XR, and all steps as voltage versus amplitude versus probe displacement. Please, I need your help!!!

Two sided Wall (line) boundary conditions - Form Assembly

Fri, 04 Jul 2025 15:04:19 +0000

Hello, when there is a line which is wall in form assembly, so, there is 2 sides of this wall. Can I use different boundary conditions in this situation? As example: one side is No Slip and the other Slip condition?

I make a simple test and When I did this, It was give different velocities when I compare the side of Slip condition. So, when I use the 2 sides as Slip it give me correct solution. When I use No slip in one side and Slip in another, the Slip condition is giving different value compared the prior test.

Using current dependance in Joule heating module

Fri, 04 Jul 2025 11:43:52 +0000

Hello everyone, I've just started learning how to use COMSOL.

I'm using the Joule Heating module to model a conductor connected to an electric potential on one side and grounded on the other to analyze heat transfer. Everything works perfectly when I use only the electric potential. However, when I try to analyze the heat dependence of the conductor using a constant current (by applying a Terminal condition), nothing changes after computing in terms of temperature.

How can I change the model depends only on the current instead of the voltage? How should I modify the model to reflect this?

Any ideas, suggestions, or advice would be greatly appreciated!**

Boundary Error in meshing

Thu, 03 Jul 2025 21:41:09 +0000

I am modeling a laser cathode stem that uses a heater coil for joule heating.

My issue is that the bottom portion of my coil has 0 spacing between them, the windings are directly on top of each other. When I am meshing in COMSOL the boundaries are giving the error of "intersecting face elements." I have this on "extremely fine" free tetrahedral. I tried ignoring edges and some boundaries but it just seemed to cause more issues that I don't know how to fix.

This is my first COMSOL simulation so apologies if things are a bit unclear. Couldn't seem to reduce file size down even after removing meshing. Added the stp file to show geometry, though.

Requesting Guidance for Vacuum Chamber Analysis – New User Struggling

Thu, 03 Jul 2025 14:44:49 +0000

Hi everyone,

My name is Vijay, and I am currently working on my master’s thesis, focusing on vacuum chamber analysis using COMSOL. I am new to this field and have been struggling to get meaningful results for the past 30 days despite trying my best.

If anyone is available to guide me or provide advice on the workflow, boundary conditions, or setup strategies until I can get my first meaningful results, it would be a huge help. I am open to having a call to discuss my setup in detail if needed.

Simplest CUDA context

Thu, 03 Jul 2025 12:56:44 +0000

Hello!

I have a 1D Weak Form PDE with only one dependent variable u, which has shape-function 'Nodal discontinous Lagrange'. In Compile Equations Statistics it is seeable, that this dependent variable u creates 45 Deegrees of Freedom, and in total there is also 45 Deegrees of Freedom.

But when computing with Hardware Acceleration (CUDA) a warning still says: "GPU calculations disabled. Not all dofs are DG.". But Compile Equations Statistics says otherwise.

How is this possible?

Nice greetings

How can we find the Current distribution on the top and bottom metal surfaces of the absorber

Thu, 03 Jul 2025 11:17:35 +0000

Hi, i am working on metamaterial absorber and i want to know that how can we find Current distribution on the top and bottom metal surfaces of the absorber.

meshing in axisymmetric models

Thu, 03 Jul 2025 09:33:34 +0000

I've noticed that vertical boundaries in axisymmetric models are meshed very coarsely, whereas if I create the same geometry in 3D the equivalent boundaries are meshed more finely.

Is there some explanation for this?

Divergence h(div) as shape-function; which physics interfaces?

Wed, 02 Jul 2025 16:35:09 +0000

Hi!

Which physics interfaces use divergence h(div) as shape functions?

I want solve a 'grad div A' or 'divA*test(divA)' context and want to know how to configure the solver for divergence-shape-function contextes.

Thank you. Nice greetings

Coil turns have the same current ?

Wed, 02 Jul 2025 15:26:24 +0000

Hello,

I am working on a 2D-Axisymetric model, where i am modeling a coil. If i draw each turn individually, spaced, will the different turns have the same current (as they should be connect in series)? If not, and if I am trying to induce current in my coil, how can i ensure this?

Thanks, Vitor

Moving magnets study

Wed, 02 Jul 2025 02:53:22 +0000

I'm new to COMSOL (so new that I just downloaded it yesterday) , and I want to study movements from magnetic forces.

I've been trying to do a 2 magnets study, but I kept getting the " relative error higher than tolerance" error.

Now I've come to seek help from the community.

Error in Coupled Electrostatics-Heat Transfer Simulation: Divergence in Linear Iterations?

Tue, 01 Jul 2025 22:45:23 +0000

Model Description: I am simulating power dissipation in a 3D chip with RF and DC electrodes embedded beneath insulating layers. The simulation uses:

Electrostatics (FEM) coupled with Heat Transfer in Solids.

Stationary study

Applied Voltage: 100 V on the RF electrodes.

Key Assumption: Since the electrode dimensions are in millimeters (mm), I approximated the 20 MHz RF signal as a DC field (quasi-static assumption) and opted for electrostatics instead of full electromagnetics.

Error Message:

Failed to find solution.
In temperature: Divergence of the linear iterations.
Returned solution is not converged.
Not all parameter steps returned.
Note: I kept mesh normal in size. I need the solutions and susggestions asap.

Electric interference

Tue, 01 Jul 2025 08:24:42 +0000

Dear Experts,

I got the following error message:

Parameters: "time","0"
The following feature has encountered a problem:
Feature: Time-Dependent Solver 1 (sol75/t1)
Failed to find consistent initial values.
Last time step is not converged.

The physics: Electric currents
The equation I used: (the polarity changes)
Io*sin(2*pi*f1*t)
- Io*sin(2*pi*f2*t)
Io*sin(2*pi*f1*t)
- Io*sin(2*pi*f2*t)

See at physics.png
Note: though the are two current sinks under the electric current, these are Current source as well. (I know, in Electrochemistry Module there is a designated module to current sink, but I do not have access to it. )

The solutions that did not work: - A Stationary step was added before the Time dependent study
- Using the following equations to avoid the value 0:
- Io*sin(2*pi*f*t)*(1 - exp(-t/tau)) - Io*sin(2*pi*f*t)*(t<ramp_time) + Io*sin(2*pi*f*t)*(t>=ramp_time)

Changing the time range
redo the whole study/physics

Magnetic tumbling robot simulation

Sat, 28 Jun 2025 14:16:24 +0000

Hello, everyone. I wanna achieve a simulation of the tumbling motion of magnetic layer under magnetic field, as shown in the attached figures. Can anyone help me? Thank you very much.