Latest Discussions - COMSOL Forums

Comsol batch recovery option

Tue, 09 Jun 2026 05:56:08 +0000

Hi,

since the documentation of the comsol does not really say much about it, i wanted to ask if someone has already used the -recovery option for comsol batch command? If yes, how specifically?

I was thinking something like this:

comsol batch -recover -inputfile "inptufile_that_crashed.mph" -outputfile "output.mph"

Or should i set the .recovery as input? It would have been nice if the documentation would have shown an example usage of this option.

Best regards, Zoltan

Doube emulsion simulation

Tue, 09 Jun 2026 05:27:34 +0000

I have done my single emulsion simulation now I am trying the double emulsion simulation in comsol I am not getting the double emulsion droplets microfluidics

Double emulsion droplet generation device microfluidics

Tue, 09 Jun 2026 05:14:10 +0000

I made the single emulsion droplet generation simulation in comsol like that I tried the double emulsion simulation in comsol but double emulsion droplets are not forming and I did the design as it is in the research paper still the droplets are forming

High residuals in solid mechanics frequency domain simulations

Mon, 08 Jun 2026 17:31:17 +0000

I am relatively new to solid mechanics simulations. I am working on simple 2D frequency domain simulations of simple solid resonators, as shown below. (the blue part is silicone, the white part is steel with custom stiffness and density values). I apply a harmonic pressure load on the top surface(edge in 2D), and measure the surface response. The bottom surface(edge) is free, the side ones are fixed. The model works fine with relative tolerance of 1e-6 when the thickness of the model is large (ex: 5mm ). The width of the model is 5mm. Now when i reduce the thickness to say 0.5 to 1mm, the model is not able to converge to even a relative tolerance of 1e-3. 1)Could anyone guide me, how to debug this issue? I tried playing with the mesh, refining it further, but that doesn't work. 2)On a side note, is the aspect ratio of cells being close to 1 a stricter requirement in solid mechanics FEM than that in standard CFD problems?

3D simulation of heat and moisture generation in a stacked fertilizer reactor – non‑convergence

Mon, 08 Jun 2026 13:14:02 +0000

Hello everyone,

I am using COMSOL Multiphysics 6.0 Build 318 to simulate a 3D model of heat and moisture generation during fermentation in a stacked fertilizer reactor. The simulation always stops with a non‑convergence error around 29 s. I would greatly appreciate any advice from experienced users to help me get the simulation running to completion. Thank you in advance.

Below is a summary of the model (Attachments: Figures 1–12 (as referenced in the text)).

Simulated object

(1) Four fertilizer buckets are stacked on top of each other (like steamers). Each bucket contains a cylindrical pile of fertilizer. Microbes inside the fertilizer decompose the material and produce heat and moisture when heated. A hollow, constant‑temperature heating tube runs vertically through the center of the whole system, providing a heat source.

(2) The part of the fertilizer pile closest to the heating tube heats up first and therefore starts producing heat and moisture earlier. As heat conducts radially outward, the outer parts of the pile also start to heat. In addition, the outer wall of each bucket is also a constant‑temperature heat source – i.e., both the inner (near the tube) and outer (near the bucket wall) sides of the fertilizer pile are heated.

(3) The four layers are separated by septum plates that allow heat transfer but no mass transfer. The top bucket has a cover, and the bottom bucket sits on a base plate. Both the cover and the base plate have convective cooling boundaries on their outer surfaces.

(4) The goal is to simulate 1 day (86 400 s) of operation and obtain: temperature evolution, moisture (water vapor concentration) evolution, and the total heat generation rate [W] of the fertilizer piles over time.

Model setup

(1) Figure 1 shows a rough view of the geometry (buckets, fertilizer piles, air domains, central tube, septum plates, cover, base).

(2) Global parameters are listed in Figure 2. Parameters inside the red box are geometry‑related (not critical). The parameters outside the red box are more important, especially k_Q and k_moist – the coefficients for the temperature‑dependent heat and moisture production rates. Because I am not sure whether they cause the divergence, I have set them to 0 for now.

(3) Under Component > Definitions I created a nonlocal coupling > integration operator (intop1) to compute the total heat generated [J] after the simulation (Figure 3).

(4) Under Component > Selections I defined three explicit selections (Figure 4) – fertilizer domain (four piles), air domain (the space above the fertilizer in each bucket), and stainless steel domain (all containers and structural parts). This makes material assignment and meshing easier.

(5) The geometry (Component > Geometry) is shown in Figure 5. I do not think the problem lies here.

(6) Material definitions (Component > Materials) are shown in Figure 6. Briefly:

Stainless steel: 304 [solid, polished] – water content, permeability, porosity set to 0; other properties default.

Compost (fertilizer): custom material – density 546 kg/m³, thermal conductivity 0.3 W/(m·K), heat capacity 3200 J/(kg·K), porosity 0.34, permeability 1e‑7 m², water content 0.2 kg/m³.

Air: taken from the built‑in library – density 1.2 kg/m³, thermal conductivity 0.026 W/(m·K), heat capacity 1005 J/(kg·K).

(7) Physics interfaces (Component > Physics) – two interfaces: Heat Transfer in Solids (ht) and Transport of Diluted Species (tds).

ht handles conduction in the stainless steel, fertilizer, and air domains.

tds handles water vapor diffusion in the fertilizer and air domains.

Detailed settings:

ht – initial values: three nodes for the stainless steel, fertilizer, and air domains (initial temperatures).

ht – temperature boundaries:

central tube inner wall – constant temperature heat source (65 °C)

bucket outer walls – constant temperature heat source (30 °C)

cover and base outer surfaces – convective heat flux (cooling, external temperature 20 °C, heat transfer coefficient 5 W/(m²·K)).

tds – transport properties: two nodes –

air domain: diffusion coefficient = 2.5e‑5 m²/s

fertilizer domain: diffusion coefficient = 1e‑5 m²/s

tds – initial values: three nodes – fertilizer, air, stainless steel (initial water vapor concentrations in mol/m³). Stainless steel initial concentration = 0.

Important – ht – heat source and tds – reaction for the fertilizer domain:

Heat source expression (Q0): Q_base + k_Q*(T - T0_compost)

Reaction rate (Rc): S_moist_base + k_moist*(T - T0_compost) (Both coefficients k_Q and k_moist are currently set to 0, as mentioned.)

The moisture cap sub‑node (using the truncated expression (S_moist_base + k_moist*(T - T0_compost)) * (c < c_saturation)) has been disabled because I wanted to test the basic case first.

(8) Mesh (Figure 8) – I am wondering if the mesh might be too coarse, causing divergence. The settings are:

Mesh sequence type: User‑controlled mesh.

Size node (global): max element size 100 mm, min 20 mm, growth rate 1.5, curvature factor 0.6, resolution of narrow regions 1.

Global network (size): max 80 mm / min 10 mm, growth rate 1.5, curvature factor 0.5, narrow region resolution 0.5.

Free tetrahedral 1 – covers remaining domains.

Three “general refinement” nodes – for fertilizer, central tube, and air domains: max element size 60 mm (all three), min element size 15 mm (all three), other parameters default.

(9) Study settings (Figures 9–12) – I suspect the problem may be here. I do not have much experience tuning the solver. Current settings:

Step 1: Time Dependent

Time unit: s

Output times: range(0,10,100) (test up to 100 s)

Tolerance: user‑controlled

Relative tolerance: 0.01 (arbitrary)

Solver configuration – Time‑Dependent Solver 1

General – By study step: user defined; time unit: s; output times: range(0,10,100); times to store: interpolated output; relative tolerance: 0.01 (arbitrary).

Absolute tolerance – Global method: scaled; tolerance method: factor; tolerance factor: 0.01 (arbitrary); derivative tolerance method: automatic (Update scaled absolute tolerance checked).

Time stepping – Method: BDF; step size taken by solver: free; initial step: 0.1 s (default); maximum step constraint: constant; max step: 100 s (arbitrary); max BDF order: 2; min BDF order: 1; event tolerance: 10 (arbitrary); singular mass matrix: maybe; initial step for backward Euler: 0.001 (arbitrary).

Fully Coupled (Figure 12) – Linear solver: AMG, concentration (tds); nonlinear method: automatic (Newton); initial damping: 1; min damping: 1e‑4; step length limit: 10; step length growth limit: 1; use recovery damping: automatic; recovery damping factor: 0.75; termination technique: tolerance; max iterations: 4; tolerance factor: 1; termination criterion: solution.

Question

Could the divergence be caused by:

The k_Q / k_moist coefficients (even though they are set to 0 now)?

The mesh being too coarse (or even too fine)?

The solver / time stepping settings (I am not confident in these)?

Possibly something else that I have overlooked?

I would be very grateful for any suggestions to make the simulation run through the full 1 day period. Thank you very much for your time.

2 physics

Wed, 03 Jun 2026 14:02:14 +0000

I'm working on a cavity inside a magnetic field that I should calculate the form factor (combination between 2 physics) EMW+MF. However I created a 1 study for 2 different steps (stationary + eigenfrequency) it shows an error that there is complex number that Maxwell equations cannot solve, but whenever I separate it into 2 study it works but i cannot calculate C-factor.

Help with Definition Error: Surface Integration of Acoustic Pressure on Solid Boundaries

Mon, 01 Jun 2026 21:38:30 +0000

Hi everyone, I am running a room acoustics simulation using the Pressure Acoustics, Frequency Domain (acpr) interface. My model consists of a standard air domain with a solid concrete inertia block located inside the room. I need to calculate the total acoustic pressure acting on the block. I am doing this via Surface Integration of the total acoustic pressure (acpr.p_t) over the shared boundaries between the air domain and the solid block. However, COMSOL keeps an error, stating that the acoustic pressure variable is not defined on these boundaries. I know this happens because acoustics don’t exist within solid domains , but I still need to find a way to compute this surface integral at those eganfrequienceis. As a test, I tried converting the solid block's material properties to air. While that fixes the error, it completely alters the acoustic cavity volume and gives me wildly different acoustic modes/frequencies, so that workaround won't work for my setup. Does anyone know a way to evaluate a fluid variable like acoustic pressure right at the interface of an inactive solid domain? Would creating a very thin "shelled" air domain wrapper around the solid plinth be a viable workaround, or is there a better built-in way to only look at the fluid side of the boundary during integration? Please look at the results secion titled eganfrequecny study because that is where my egenfrequecies are from Best, Azaria

Electrochemical model building

Fri, 29 May 2026 07:45:34 +0000

How to build an electrochemical model for NOx sensor in which we can vary input parameters such as electrode materials, electrolyte concentrations, diffusion barriers, gas conc., etc to optimize the performance of sensor that includes sensitivity, response time, etc? Is there any tutorial that can be useful for such electrochemical models?

How to access heat generation rate variables (Qrev, Qpol, QSEI) in Lithium-Ion Battery interface for plotting?

Thu, 28 May 2026 03:11:50 +0000

Hello,

I am working on a 1D SEI formation model coupled with a 3D thermal model in COMSOL Multiphysics 6.3 using the Battery Design Module. I am trying to reproduce the heat generation rate plot similar to Figure 4 in Andriunas et al. (2022), Journal of Power Sources 525, 231126, which shows the breakdown of reversible heat (Qrev), polarisation heat (Qpol), and SEI heat (QSEI) over time during discharge. My model is based on the sei_formation_seed file from the Application Library, with an added Porous Electrode Reaction 2 for the SEI forming reaction and a Film Resistance defined by thickness and conductivity (dfilm_0 and kappa_film). My questions are:

What are the correct variable names to access the individual heat generation components (reversible, irreversible/polarisation, and SEI film heat) from the Lithium-Ion Battery interface in COMSOL 6.3?
When I browse the Replace Expression tree under Component 1 > Lithium-Ion Battery, I cannot find any Heat Sources subfolder or heat-related variables. Is there a setting I need to enable first, such as the Energy Equation in the Physical Model section?
Is the SEI film heat accessible as a built-in variable from the Film Resistance section of Porous Electrode 1, or do I need to define it manually using the expression QSEI = εSEI × aj × Jj × ηSEI as described in the paper?
If I need to define the heat variables manually, what are the correct COMSOL variable names for εSEI (SEI volume fraction), aj (specific surface area), iloc (local current density for SEI reaction), and ηSEI (voltage drop over SEI) within the Lithium-Ion Battery interface?

Any help would be greatly appreciated. Thank you.

Fixed Temperature Boundary Condition on Thin Structure Fracture

Wed, 27 May 2026 14:50:26 +0000

Dear COMSOL Community,

I am working with COMSOL Multiphysics 6.3 and I have a question regarding boundary conditions for heat transfer in a fracture modeled as a Thin Structure.

In my model, the fracture is not represented as a separate meshed volume, but rather as a thin structure. I would like to impose a constant temperature boundary condition directly on the fracture surface/edge, similar to prescribing a fixed temperature on the fracture boundary.

Has anyone encountered this situation before? Is there a way to reproduce such a boundary condition when the fracture is defined as a Thin Structure? Or is it necessary to model the fracture as a separate physical domain in order to apply this type of temperature condition?

Any suggestions or examples would be greatly appreciated.

Electroosmotic Flow in Cylindrical Microchannel Using Electrostatics + TDS + Creeping Flow Gives NaN/Inf Solver Error

Sun, 24 May 2026 17:11:27 +0000

Hello everyone,

I am a beginner in COMSOL and trying to simulate electroosmotic flow (EOF) inside a cylindrical microchannel.

Geometry:

Cylinder radius = 5 µm
Cylinder height = 10 µm

Physics used:

Electrostatics (es)
Transport of Diluted Species (tds)
Creeping Flow (spf)

Goal: I want to model electroosmotic flow caused by:

wall zeta potential,
electric double layer (EDL),
applied electric field.

Current setup:

Electrostatics:

Side walls: Electric potential = -10 mV (intended as zeta potential)
Inlet: 1 V
Outlet: 0 V

Transport of Diluted Species:

Two species:
- c1 = cation (+1)
- c2 = anion (-1)
Migration in electric field enabled
Diffusion enabled
Convection sometimes enabled/disabled during testing
Diffusion coefficients: D1 = D2 = 1e-9 m^2/s

Charge density: rho_v = F_const*(c1-c2)

Creeping Flow:

Volume force: F = es.rhoq*es.E

Mesh:

Boundary layer mesh added near walls.

Problem:

Electrostatics alone converges successfully.
But when TDS is coupled with Electrostatics (fully coupled or segregated), solver fails with: "NaN or Inf found" or "Initial guess leads to undefined value"

I already tried:

Fully coupled solver
Segregated solver
Linear discretization
Smaller applied voltage
Different iterative/direct solvers
Lower damping factors
Turning convection OFF
Refining mesh

Observation:

Smaller nanoscale geometry (100 nm radius/height) converged.
Larger micrometer geometry diverges.
TDS concentration sometimes becomes extremely large near walls.

My confusion:

Is applying wall zeta potential as electric potential physically correct in this formulation?
Should EOF be modeled instead using Helmholtz–Smoluchowski slip velocity?
Is full Poisson–Nernst–Planck coupling required here?
Is my charge density formulation correct?
What is the best solver strategy for this multiphysics problem?

I would greatly appreciate guidance on:

proper EOF implementation,
correct boundary conditions,
stable solver settings,
whether to use slip velocity or explicit EDL resolution.

Thank you.

Simulated Segmented Taylor (Recirculating) Flow

Fri, 22 May 2026 19:02:41 +0000

I am trying to simulate segmented slug flow of alternating nitrogen and DMSO slugs in COMSOL, specifically to reproduce Taylor flow behavior with internal recirculating flow inside the liquid DMSO slug. To model this, I am using a time-dependent two-phase laminar flow, phase-field study with alternating initialized fluid domains representing the gas and liquid segments. However, instead of sustained Taylor recirculation throughout the liquid slug, recirculation only develops near the ends of the nitrogen bubble due to buoyancy effects, after which the flow transitions into a more typical laminar profile within each segment.

So far, I have included surface tension and surface energy effects, gravitational effects, and alternating initialized gas-liquid domains, but I am curious whether additional parameters, physics settings, or numerical considerations may be important for establishing stable Taylor recirculation in this type of simulation. I am considering whether factors such as wetted wall film resolution or interface thickness parameters may play a bigger role.

Needle Piercing Simulation: Abnormal Results (Solid Mechanics, Stationary Study)

Wed, 20 May 2026 15:55:32 +0000

Dear all, For the past week, I've been trying to set up a simulation comprising an arrow-like needle piercing a skin-like substrate. As I haven't found a similar (3D) example in COMSOL's Application Gallery or Learning Center so far, I have followed analytical step-by-step instructions by three different AI chatbots (Gemini, Perplexity, Claude). After days of trial-and-error and diagnostic tests, I have reached the point where all three of them agree with each other that my method was set up correctly. Nevertheless, the results (see 1.png in attachments) indicate abnormally low pressure values on the needle, with the appearing scale being on the order of E-4 N/m2 (whereas, e.g., atmospheric pressure is at E5 N/m2), while zero pressure appears on the skin-like substrate. The AI chatbots claim that the needle is faulty (it's an imported .stp file from Autodesk Inventor), but I cannot really trust them. All my Model Builder data can be seen in the 2.png attachment. I'm also attaching the simulation file (NP.mph).

Any kind of help will be sincerely appreciated! Thank you everyone in advance.

PML in 3D Plasmonic nanostructure

Tue, 19 May 2026 07:58:18 +0000

Hello everyone,

I am writing this post because I need help on a simulation I did on a 3D plasmonic nanostructure. Indeed I wanted to know how PML works with periodic port .

Actually my nanostructure is in the bottom of a cube of air in which my periodic port is on top of the cube, below there is an another cube of air in which I put my other periodic port (port off) on the bottom. The Floquet periodicity are along x and y axis and the PML and excitation along z axis.

I put only a PML at below the cube of air so there is another last cube with half air/half PML. Please can you tell me if this configuration is correct to have the right "physical" R and T coefficient . My working wavelength is around 1 to 1.4 um.

Thank you.

How to Set Up Physics & Tune Parameters to Reproduce Gas-Liquid Two-Phase Flow with Chemical Reaction in Porous Media

Mon, 18 May 2026 14:47:59 +0000

Hello everyone,

I’m a graduate student majoring in Process Equipment and Control Engineering, focusing on trickle bed reactor simulations. I’m currently trying to reproduce a published study on gas-liquid two-phase co-current flow through catalyst-packed porous media, coupled with heterogeneous chemical reactions and heat/mass transfer. I’ve run into bottlenecks on the correct physics coupling logic and parameter tuning methods, so I’d really appreciate your guidance.

My Target Scenario

Gas and liquid phases flow through a catalyst bed, and heterogeneous reactions occur on the catalyst surface, with reaction heat coupled to the temperature field.

Model Selection

AI suggested using the Mixture model + user-defined volume force equations to couple reactions, but I'm not sure if this combination is suitable for the trickle flow regime (gas-liquid stratified flow in fixed beds). Is there any more recommended physics combination for this specific scenario?

Core Divergence Error (Basic Issues Already Excluded)

When running pure flow simulation (no reactions enabled) with the Mixture model, it converges perfectly with slip velocity disabled (homogeneous mode); However, as soon as I enable Slip Velocity (Schiller-Naumann model), the simulation diverges immediately after 3-5 iterations, with residuals spiking directly (no gradual oscillation process); I have already tried: lowering all under-relaxation factors, refining the porous media mesh (mesh quality ≥ 0.8), step-by-step flow field initialization, and double-checking all boundary conditions, but none of these worked.

My Requests

Has anyone encountered this Mixture + slip velocity divergence issue specifically in porous media zones? How to debug the root cause and fix this problem?

Are there any COMSOL reference cases, setup tutorials, or published paper appendices for gas-liquid two-phase flow + porous media + chemical reaction simulations, especially for trickle bed reactors? Paid resources are also acceptable. Any advice or resource sharing would be greatly appreciated! I will share the full setup workflow and troubleshooting guide once I resolve this issue.

Thank you in advance!

Convergence issues in Silicon-Glass adhesion simulation with curved surfaces

Fri, 08 May 2026 15:07:40 +0000

Hi everyone, I am working on a model to evaluate the load required to make a 2 mm thick silicon tile adhere to a glass substrate.

Setup Description: - Geometry: The glass substrate has a cylindrical curvature with a radius of 40 m (nearly flat) on its top surface (in contact with the tile), and a base of 32 mm x 12 mm and 3 mm thickness. The silicon tile (30x10x2) mm^3 is positioned so that it is initially tangent to the substrate at a single point.

- Boundary Conditions The flat bottom of the glass substrate has Fixed constraint. I applied a Prescribed Displacement (w) to the bottom surface of the Silicon tile to "push" it against the glass. I defined w = w_{gap} = -y^2/(2R) to follow the parabolic profile of the gap. I defined a Contact Pair between the bottom of the silicon and the top of the glass. I enabled the Adhesion feature with a gap activation criteria delta = 10^{-7}m. I even tried with " Always Active"

- Mesh: I am using 0.6 mm tetrahedral elements and 0.2-0.3 mm elements on the glass/Si interface.

- Solver: I am using a Direct Solver with default parameters.

I am struggling with two main issues: the simulation fails to converge as the the tile approaches full contact. Even when it partially solves, the adhesion condition doesn't seem to activate across the entire contact surface.I've checked the gap variables: the difference between my analytical w_{gap} and the built-in initgap_src_p1 is negligible (within 0–2 nm), so the geometry seems correct.Could this be a mesh refinement issue in the contact zone?Is the Penalty Method for adhesion sensitive to the relationship between the prescribed displacement and the contact offset?Should I be using a Boundary Method or a specific step-by-step loading (e.g., an Auxiliary Sweep) instead of a direct prescribed displacement to improve stability?Any advice on how to achieve stable convergence for this type of thin-plate bending and adhesion would be greatly appreciated.

Difference between the methods for methanol reforming

Sat, 02 May 2026 16:40:32 +0000

This is a catalytic bed section for methanol reforming, and I used two methods to simulate it.

The one side is a porous catalyst, and the other side is a porous medium.

The difference between the two is: for the former, I filled the parameter Scat into the porous catalyst, while for the latter, I directly filled it into the forward reaction frequency factor in the chemical reaction.

It feels like the product is the same for both, so why are the simulation results so different?

Calculation of core losses in a transformer

Fri, 01 May 2026 06:14:09 +0000

After referring the ecore_transformer documentation in the application library, i am trying to calculate the core losses using the Steinmetz equation using the time dependent study followed by time to freq losses.

Using the solid conductors with current excitation at input and External I vs U as coil excitations, the error as "The Current excitation is not supported for Single conductor coils in time dependent studies."

Whereas with External U vs I excitation, the Pardiso and Assembling matrices are keep progressing with the compute and solution progress as constant hence no convergence data is coming. Can you help where I am doing wrong.

Can Parametric Surfaces Be Turned Into Volumes?

Fri, 01 May 2026 02:18:05 +0000

Hello

This question is pretty simple in nature. I am attempting to model a tube with a rectangular transition in the middle. There is a screenshot of the geometry in question below for context. I am working on COMSOL version 5.2a, so to create this geometry, I opted to do so using a parametric surface. I did so using a super ellipse parametric equation, and as it was a surface I wanted to turn it into a volume by unioning endcaps and using the convert to solid function on the union.

This workflow has failed repeatedly. As such I attempted to see if I could get convert to solid to work for more simple cases, and defined a simpler hollow cylinder as a parametric surface and attempted to utilize union and convert to solid to turn it into a volume. This also failed, and subsequent attempts with trying to turn other simple surfaces into volumes like cubes have also not worked.

I have only seen videos of convert to solid being utilized in 2d, so I wanted to ask if utilizing convert to solid on 3d surfaces is fruitless as I have had no successful attempts and cannot find any examples online.

If so, does anyone have any suggestions for creating smooth transitions in 5.2a? I do not have access to a loft function in this version it seems, and I also can only import STL's, rather than another form of CAD file. I have had very poor experiences with the quality of STL imports, so the idea of creating the geometry in Solidworks and importing it also is failing me.

Thank You

Adding a Rotating Wall in the Euler-Euler Interface

Tue, 28 Apr 2026 19:24:23 +0000

Hello all,

I am trying to model a rotary tube reactor, where I am trying to model two phase gas-solid flow inside a rotating domain. I am using Euler-Euler for the two-phase flow and I have added a rotating wall to prescribe the rotary motion of the external wall.

A part of the geometry is as shown in one of the images. I want to rotate the outer wall and solve for the flow in the domain highlighted in blue. However, the Euler-Euler interface's wall setting does not include a place to enter or couple directly from frame the rotational component. Only the translational and sliding components are present.

Does anyone know why this is happening? Am I missing something or is there any other way to solve this?

Thanks!