Hello,
I am simulating a basic model in electromagnetic frequrency domain physics. The material which i am using has three values electrical conductivity, relative permmitivity, relative permeability, and density. when i put the value 40 in the electric condictivity option, the simulations converges. However, when i put 1 in the electric conducitvity option, the simulation doesn't converge.
Can anyone tell why is it happening?
Thankyou so much.
Hello, I would like to understand how to simulate a volume force (in my case I am trying to simulate electrotheral body force) in a microfluidic channel using laminar flow multi physics. I would appreciate if anyone kindly share any model similar to it or direct me towards the right procedure.
Thanks in Advance
I'm using version 5.3a, I've built a uniform rectangluar geomerty, with a uniform mesh but the results are not shown for all the domain (I exported surface results into excel, but didn't get equal x points for every y coordinate. Is there something that I can change in the settings to generate uniformly distributed results?
Hi,
I need to create a geometry in which there are several small cylinders that intersecate other bigger (~50 times bigger in size) objects, usually blocks. For example, some operations involve the difference between a big block and a bunch of those small cylinders. If the size of the big block involved in the difference is too big, I get an error "Internal Error in Geometry decomposition", even when I press Build selected. If the block is not as big, I do not get the error immediately, but still I get it when I try to run Form Union. If the size of the block is smaller, I can run Form Union without problems. Moreover, the computation time used (both in the difference and in the Form Union) increases significantly as the size of the block increases. It seems that the problem is not a small angle or a small domain somewhere, but it seems to be the fact that I am subtracting two objects that are significantly different in size. I also noticed the the size of the virtual memory keeps increasing as try sequentially different sizes. This slows down also operations that were faster before.
Is there a way around? Thanks!
There is great example in Comsol blog: "How to Calculate a Capacitance Matrix in COMSOL Multiphysics" (https://www.comsol.com/blogs/how-to-calculate-a-capacitance-matrix-in-comsol-multiphysics/).
All seems ok beside one point, i will be very thankful if someone can make it clear.
The question is: – According to equation for total charge of 1st conductor (Q1 = Cm11V1 + Cm12(V1 – V2) + ...), if we remove all conductors from the model beside 1st one, we obtain simple equation for capacity of single conductor Q1 = Cm11*V1.
– Consequently self-capacity of the first sphere Cm11 should be equal capacity of isolated sphere, that according to C = 4piEps0*r equals 11.3pF. But in example it equals 5.4pF (form mutual matrix calculation).
– Can someone tell please what the reason of this difference? Why Cm11 don’t equals capacity of single (isolated) sphere? Is there а something I am misunderstood?
Thanks for response.
In COMSOL Multiphysics 5.4, I had run two simulations using the Time-dependent Study. In them, the Iterative solver was used. According to the Complete Reports that were written from these simulations, there were three additional Segregated Study steps done in Simulation #2 that were not done in Simulation #1. These additional steps, as they are described in the Complete Report that was written from Simulation #2, is attached to this post as a JPG image ("AdditionalSegregatedStudySteps.JPG").
May you explain how the execution of these three additional Segregated Study steps in the Time-dependent Study affects the results in terms of their accuracy?
For answering this question, the following information may be useful:
The most basic scenario is that a collimated beam of light rays is incident upon one lens. As the rays travel through the lens, their energy gets absorbed progressively by the lens's material. This absorbed energy is a source of heat within the lens. Thereby, the temperature of the lens increases.
In Simulation #1, there was one lens, as described exactly in the paragraph immediately above.
In Simulation #2, there were two lenses, where the second one was placed immediately after the first one, i.e., with no gap in between them. That means that they were in physical contact with each other, although I do not know if this physical contact was perfect. The second lens was made by using the "Array" subnode in the Geometry node. The value I inputted into the "Displacement" setting in the "Array" subnode was the length of the first lens, which had 14 decimal places in 15 significant figures (1.26582278481013 mm). The use of the "Array" subnode means that these lenses are identical to each other.
To give you an idea of what the physical contact between the lenses looks like, please see the JPG image attached to this post named "TrianglesInContact.JPG". Then extrude these triangles into the screen/page such that there are now two triangular prisms in contact with each other at one of their edges. Along the direction of the extrusion, there are 13 linearly-spaced rays that travel into the prisms in the direction shown in that JPG image.
In the Geometry of the two lenses, there are 114 such edges of physical contact.
In both simulations, I was trying to calculate the spatial distribution of the surface temperature of the lens(es) in the scenario where the light rays were a heat source. I used the Time-Dependent Study and calculated this distribution at a time t at which I knew that the maximum and minimum surface temperatures were at steady-state. These calculations were done successfully.
The maximum and minimum surface temperatures calculated in two lenses (Simulation #2) were lower than those calculated in one lens (Simulation #1). This is a point of concern. May this difference be explained? Please note that the heat transfer physics that were considered were:
The progressive absorption of light rays by the lens material as the rays travelled through the material. This is the origin of the heat in the simulation.
Thermal conduction within the lens material.
Emission of thermal radiation from the lenses, i.e., surface emissivity.
Please note that there was no medium surrounding the lenses - i.e., vacuum surrounded the lenses.
In both simulations, the same settings were used, except that, in Simulation #2, there were around 2 times the number of Mesh elements that were in Simulation #1. The Mesh Statistics from both simulations are attached to this post ("MeshStatistics_Simulation1.JPG" and "MeshStatistics_Simulation2.JPG"). This is because of the extra lens in Simulation #2. Also, by "same settings", I mean that the same values were used for all parameters in all Physics modules, with the same subnodes in the Physics modules, and with the same Multiphysics node activated. The Physics modules that were used in both simulations were the "Geometrical Optics" and the "Heat Transfer in Solids" modules. These were coupled via the Ray Heat Source Multiphysics node.
In the "Geometrical Optics" module, the subnodes that were activated were: Medium Properties; Material Discontinuity; Ray Properties; Release from Grid; Deposited Ray Power (domain); Ray Termination, which defined a rectangular prism that contained all the lenses. The beam that was simulated was monochromatic.
In the "Heat Transfer in Solids" module, the subnodes that were activated were: Solid; Initial Values; Thermal Insulation; Surface-to-Ambient Radiation. For all boundaries except two, the "Surface-to-Ambient Radiation" subnode overrode the "Thermal Insulation" subnode. These two boundaries were flat and had surface normals perpendicular to the travel direction of the rays.
I am not sure if, with only the physics settings mentioned above, thermal conduction works in the software such that it is possible for heat to flow from one domain to another domain through edges that are in contact with each other. May you confirm whether or not this is possible?
Please note that, in both simulations, a Swept Mesh was used. The values of the all the parameters of the Swept Mesh were kept the same over both simulations.
Hello,
I built a 2D multi-layer structure. When high-frequency magnetic field goes through this structure, I want to consider the attenuation of this field so that I can get different magnetic field dietribution for different layers. To this end, I chose H-formulation Physics and set up the prescribed background field in Dirichlet Boundary Condition. However, I am not sure if this prescribed field will force the field distribution to be the same for each layer? Should I set different background fields for distinct layers? But how?
Thanks for your help in advance!!!
Napo
I wrote a simple application with one input field. If I enter in the input box 1.0 (through dot) everything works. If I enter 1,0 (through comma), it gives an error (Syntax error in expression). Is it possible to somehow force the application to correctly read a number written with a comma?
Hi,
I wanna find the transport properties of electrons as a function of reduced electric fields;As Plasma User's Guid suggestion , I have used Plasma Madule>>Boltzman Equation, Two-Terms Approximation>>Reduced Electric Fields .
Now there are 2 questions ? 1- What range is good for "Reduced electric fields" at Study section ?
2 - What does this error : "Failed to find a solution for the initial parameter. Singular matrix. There are 1 void equations (empty rows in matrix) for the variable comp1.be.lam. at coordinates: (0), ... Returned solution is not converged. Not all parameter steps returned." mean ?
What parameter did I miss to specify at "Intial Values" section ?
Hi, I am trying to do time dependent study with Electrowetting droplet on COMSOL. My problem is after doing all the settings, the simulations goes on and on and never ends. What can be the possible problem?
Thanks & Regards,
I am trying to model a hydrodynamic journal bearing using COMSOL and I can get the static performance of the bearing, but I don't know how to obtain the dynamic performance of this bearing in terms of stiffness coefficient and damping coefficient. Is there any suggestion to deal with this problem?
Hi, I am currently try to simulate a chemical reaction reactor. The figure is attached. The gas flow into the reactor through a tube. This gas tube permit the gas flow inside the reactant region, the gas and the chemical powder react. The tube acts as a porous fliter and permit gas flows in and out , preventing the powder flow out. I want to know how to simulate the mesh tube? The only function of this tube is allowing the gas in and out and preventing the powder out. What physics module or what kind of material character should I add to this region?
Hello,
How to resolve "mumps is unable to switch to out of core mode: insufficient disk space for out of core mode" error? I am wokring on a workstation, windows 7, 48gb RAM.
Any help would be appriciated.
Hi,
I'm simulating air flow through a tube and looking at difference in presure along the system. I want to graph the maximum pressure along different areas of the system. To do this, I would like to find the maximum value of cut planes along the system. I have only found a way to find a single maximum but not each maximum for each slice. Anyone have advice?
Thanks!
Hi
I'm currently working on a mass induced natural convection 2D problem in porous media (single phase miscible displacement).
System description:
At the initial condition, component 1 fills a rectangular domain. On top, a Dirichlet BC is set with mass fraction(component 2)=1. A non-monotonic density-mass fraction function is implemented, such that when component 1 is pure, it is lighter than component 2. However, when they mix due to diffusion, the mixture becomes heavier forming instabilities which lead to the classical fingering patterns.
In order to do so i'm solving the two components mass conservation. The conservation of momentum in done via the Darcy law for porous media. Hydrodynamic dispersion is represented by the scheidegger model which states linear dependency between the mechanical dispersion and velocity (multiplied by the dispersivity coefficient). Finally, a linear viscosity-mass fraction function is implemented.
So far so good. I currently use Comsol (finite element and implicit scheme) to do so, and manage to solve the system to get reasonable results (compared to experimental results).
The conundrum:
Problem starts when i increase the dispersivity coefficient. For some reason, which is counter intuitive to me, when i increase dispersivity, for the same element size, convergence becomes worse up to a point where it doesn't converge at all. I assumed that increasing dispersivity would play a stabilizing factor (decreasing the elemental Peclet number), but for some reason it is not.
Any insights?
Thanks!
The below post is related to an archived discussion
[start here]I have a resistive coil from a MEMS. The only data I have to simulate the resistive heating is the input voltage which I can use between 0.5V to 20V and the material of the coil(GOLD). I am getting error like [Failed to find a solution. System matrix is zero. Returned solution is not converged.]. How would someone perform simulation with this amount of data and the given cad model.
I have a resistive coil from a MEMS. The only data I have to simulate the resistive heating is the input voltage which I can use between 0.5V to 20V and the material of the coil(GOLD). I am getting error like [Failed to find a solution. System matrix is zero. Returned solution is not converged.]. How would someone perform simulation with this amount of data and the given cad model.
I am a little confused about the difference between contact pair and identity pair when I do some simulatation of solid machanics . And are there any restrictions of the study type? When we do stationary study, what is the difference between assembly with identity pairs and union? Thanks very much!
Dear all,
I am simulating a pm_motor that I defined the magnets with Ampere Law. I used the same value of Remenant flux density. When I reached the result, the result in magnetic field is normal but in rmm field is zero. The magnetic insulation is in same condition. I uploaded the pictures of the results. Have you met this problem? How can I solve the problem.
Regards
Joey
Can we simulate light propagation in plane tiO2/SiO2 waveguide on silicon substrate? Please help regarding the problem. Thank you.