Quantcast
Channel: Solving, solving\ - Latest Discussions - COMSOL Forums
Viewing all 3988 articles
Browse latest View live
↧

Combining different mesh types

June 26, 2014, 7:01 am
$
0
0
Hi everyone,

I'm very new to Comsol. I'm trying to model a resistor with power running through a nichrome section, generating heat. Part of my geometry was extremely thin so I created a free triangular face and then swept through the entire geometry multiple times which worked fine. However, I want to see the effects of conduction with air (however small) as the next step. I'm assuming that air goes to infinity.

I saw that it was recommended to simply put a block around it that was 'big enough'. However when I tried meshing it using free tetrahedral I got the error: "Boundary mesh must consist of triangular elements only." My geometry is too complex to create more swept features and it would be really nice if I could simply use free tetrahedral for the block of air every time I change a dimension of the resistor. Is there any way of getting around this?

Thanks!
Mate M.
↧
Search

Error: Failed to find a solution, Singular Matrix, Returned solution is not converged.-Feature: Stationary Solver 1(sol1/s1)."

June 27, 2014, 6:19 am
$
0
0
Hi,

I have a laminar flow module and it is based on an imported mesh. I changed the java heap space according to the tutorial in this COMSOL forum and also changed the iterative solver to direct solver and it allowed me to run and solve my model. Everything was good one week ago and it could be solved and obtained reasonable result. But today when I opened this file and tried to run this one, it showed
" Failed to find a solution.
Singular matrix.
Returned solution is not converged.
-Feature: Stationary Solver 1(sol1/s1)."

My biggest confusion is how is it possible to appear an error when I reopened the same file? (I didn't make any modification!!!!!) Can anyone help me??????

Best,
Nan
↧
↧

Comsol and the Intel Xeon Phi

July 2, 2013, 2:21 pm
$
0
0
Hello everyone,

We are considering building a workstation for use with Comsol. Does anyone know whether Comsol supports the new Intel Xeon Phi many-core co-processor, and whether it uses the Xeon Phi well?

Thank you
↧

COMSOL giving me wrong Ks value!

June 29, 2014, 10:22 pm
$
0
0
Hi,

I am using the Richards Equation interface and there is an equation relating kappa (permeability) to k (hydraulic conductivity) which is written in the documentation but I have calculated by hand what I should be getting and it seems COMSOL is making a mistake. (I have also googled my equation and found that my hand calculations are confirmed elsewhere.) The equation states that:

kappa = k*mu/(rho*g)

where mu is the fluid dynamic viscosity (in my case 1E-3 Pa*s) and rho is the fluid density (in my case 1000 kg/m^3) and g is the gravitational acceleration (~9.8 m/s^2). My ks (k at saturation) is 6.94E-7 m/s, so my kappa at saturation should be 7.07E-14 m^2. I set my kappa as such, and sure enough it is showing me that "kappas" (saturated permeability) is 7.07E-14, and then when I check the saturated hydraulic conductivity it says that "ks" = 2.94E-4, which is off by several orders of magnitude.

Does anyone know why COMSOL might be using a different equation than it should be? I am sure that my math is right and all of the units are as they should be - I have no idea what the problem is!

Thanks,
Shoshana
↧

Failed to find a solution

June 30, 2014, 10:31 am
$
0
0
Hi all,

I am using Heat transfer in Solids and Solid Mechanics modules to find the thermal stress of a PDMS micropump that contains a fluid reservoir, where the fluid will flow out due to heat.

I am using a stationary solver. When I tried to solve I got this error message:

"Failed to find a solution.
The relative error (1.3e+002) is greater than the relative tolerance.
Returned solution is not converged."

How can I fix that?

Best regards,
Marwan
↧
↧

Problems Solving Model - 2D Parallel Plate Waveguide w/Coupler

July 1, 2014, 9:57 am
$
0
0
All,

I'm trying to model a 2D parallel-plate waveguide that has coupling structures attached to the parallel-plate (very similar to conical horn antennae). I want to simulate the waveguide using THz frequencies. In my experiment, the incoming THz pulse is TEM with the electric field polarization vertical (between the plates). The waveguide also operates in TEM, or the TM0 mode. I need to compute the coupling of the waveguide, assuming the THz beam (bandwidth 0 - 5 THz) enters the waveguide on the left and exits on the right.

I've tried building my model 2 ways:

1) Building the waveguide structure, then leaving only the negative space between the plates and trying to place the Port 1 directly on the left of the structure and Port 2 on the right

2) Enclosing the waveguide structure with a circle of air, combining the encapsulating air space with the negative space between the waveguide plates, and leaving the plates present while setting their material definition to be copper

I am using the Electromagnetic Waves physics for this simulation. In both cases, I am unable to compute anything for the final model, getting an error everytime.

I am rather new at this. I've tried looking through models on both the Model Exchange and those created by COMSOL and I haven't determined the source of the error, although I'm sure it's because I've set a definition wrong somewhere, likely in the physics portion of the program. I'd really appreciate any assistance anyone has to offer!

NOTE: I could not attach the second file with the encapsulating circle. The file size became too large, even when I deleted the mesh. But for reference, I placed a circle with a radius of 0.1 m at the (x,y) coordinates (0.015,0)m. I then used the Union function to combine the circle with the negative space left over from the Difference function and applied the Air material to it. I also adjusted the Difference function so that the plates of the PPWG were visible again, and then I applied the Copper material to them.
↧

Sessile droplet

March 17, 2013, 9:25 pm
$
0
0

I want to put a sessile droplet inside a channel. How to create a droplet of a specific material ? and how do I choose the model ?

As a reference I was wondering about the model in one the Paper - A Dynamic Electrowetting Simulation using the Level-Set Method-- www.comsol.co.in/papers/5283/

How do I replicate this paper in simulation?

Is the model available on Comsol website?

I was not sure that I can at all do anything like this in Comsol before I came across the paper.
↧

Moving mesh for contact pair

July 1, 2014, 9:20 pm
$
0
0
Hey,

in my bachelor thesis i have to simulate a wire, which has to be pushed through a tube (they have contact) to get electricity/ heat. For this i used the physic "moving mesh" and "pdgl- interface(weak form)" and give my tube a velocity. If i simulate this i can see that my wire is getting slurred and losing his geometry form over the time. For weeks i have tried everything, but i still dont know what the problem is. I would be very happy about your support. Have a nice day.

Kind regards
↧
Search

Heat Transfer in Solids+Solid Mechanics boundary conditions

July 1, 2014, 10:57 pm
$
0
0
Hi,

I am trying to simulate the behaviour of optics components when being cooled down from 293K to 4K.I am using Solid Mechanics and Heat Transfer to Solids, where i am applying they temperature to all boundaries. I managed to get results, but am not quite satisfied yet. I am using a fixed constraint boundary condition on the surface where the part is being mounted - which isn't really correct, as this surface won't undergo any thermal contraction in the simulation. I have tried to work with a points(also edge)-fixed constraint condition, but they solution won't converge.

Thanks in advance,
Andi
↧
↧

Unpolarized light

July 2, 2014, 10:11 am
$
0
0
Hello,
I am trying to shine not polarized light on my structure, but I don't know how. I tried to do like if I polarized light but with no magnetic field. Meaning that took what is on the Fresnel equations model and change the field to zero but is gives me the following error:
Undefined value found
-Detail: Undefined value found in the constraint residual vector: NaN or Inf in vector

Please, can someone help me?

Have a great day,

Phil
↧

Unpolarized light

July 2, 2014, 10:11 am
$
0
0
Hello,
I am trying to shine not polarized light on my structure, but I don't know how. I tried to do like if I polarized light but with no magnetic field. Meaning that took what is on the Fresnel equations model and change the field to zero but is gives me the following error:
Undefined value found
-Detail: Undefined value found in the constraint residual vector: NaN or Inf in vector

Please, can someone help me?

Have a great day,

Phil
↧

How to link Two physics in Comsol

July 2, 2014, 9:06 pm
$
0
0
I am solving a problem using Two physics Shell, Shell 2 (I applied shell to particular region, Shell 2 to remaining) , Comsol Solver is Solving Two physics separately and it is showing results separately for two physics that means it is splitting the model, Can anyone tell me how to link those two physics such that entire body will be solved as single
Thanking You
↧

Define a variable that is dependent on number of timesteps

July 3, 2014, 10:45 am
$
0
0
Hi COMSOL community

I have a time-dependent optimization model which simulation time (not computational time) is dependent on a Stop Condition, i.e. my simulation time differs depending on which parameters the optimization solver changes during the computation .

My question is, if it is possible to use the simulation time in mathematical operations e.g. to define a Variable??

For example, lets say I want to define a Variable as:

A = ht.ndflux_acc_d / "simulation time [s]"


Hope someone can be of help.
↧
↧

Fluid flow in a heated flux media porous

June 23, 2014, 2:43 am
$
0
0
Hello,

I pretend to built a model where I simulate the heat transfer to air flow through a porous media (in a channel with an inlet and outlet).
My fluid is the air and the porous structure is the SiC.

I stablished the following physics and boundary conditions:
- Brinkman equation: all structure
fluid and flow matrix properties: all structure and properties from the material
wall: no slip, u=0
inlet: velocity u=0,03 m/s
outlet: pressure without viscous stress, p=0,01 Pa

- Heat transfer in solids: porous structure (outer channel)
heat transfer in solids: properties from material (SiC)
thermal insulation: outside surfaces
initial values: 800°C (outer channel)
surface to surface radiation: emissivity 0,4; Tamb=20°C (describes the absorbed radiation on the front boundary of the structure)

- Heat transfer in Fluids: inner channel (air area)
heat transfer in fluids: properties from the material
thermal insulation: no tehrmal insulation
initial values: temperature=20°C (inner channel)
temperature : inlet T=100°C
outflow: outlet, convective flow
temperature: on the boundary cup-air T=800°C

After compute, an error appears:

"Failed to find a solution.
In segregated group 1:
Divergence of the linear iterations.
There was an error message from the linear solver.
Very ill-conditioned preconditioner.
The relative residual is more than 1000 times larger than the relative tolerance.
Returned solution is not converged.
- Feature: Stationary Solver 1 (sol1/s1)"

Please could you give any tipp or advice about what I should do?

I hope you can help me.
Thank you very much,

Ana

↧

Is it possible to solve "Static Mixer" in coupled mode?

July 5, 2014, 9:54 am
$
0
0
Hello everyone,

I'd like to solve the "static mixer" example with coupled physics. In the other words, I want to solve both "particle tracking" and "fluid" physics simultaneously. In the current version, it first solves for fluid and then solves for particles.

The reason behind this is, I want to see the effects of moving particles on fluid. I mean, I want to see what happens in fluid parameters (like velocity and pressure) when a number of particles are passing through a channel.

Regards
↧
Search

Does COMSOL Utilize FEM or FDM or Both?

July 7, 2014, 7:51 am
$
0
0
This is kind of a generic question, but I cannot find any straightforward answer searching online or the documentation. Finite Element Modeling as I understand, is more especially suited for structural modeling, while Finite Difference Modeling has application in electromagnetics, for instance. Thus, I assume that COMSOL uses both, but I want to be sure. Of course, there is also CFD and other methods I am sure.

Also, if it does use both, how can I be sure I am using one approach or the other? Is there a way to select which method is used?
↧

Model convergence dependent on mesh quality, confused!

July 7, 2014, 11:57 am
$
0
0
Hi all,

My COMSOL model only solves when the mesh size is "above" the predefined Normal size. When I increase the mesh resolution to "Fine" or higher, the step BiCGStab in the progress tab stalls at around 96%, and continues iterations forever until infinity trying to converge.

I'll describe the model briefly in this paragraph, and include an attachment of it. I'm simulating a very small, simple, conductive, flattened-conical tip (sliced in half vertically for symmetry) that radiates microwave (~6GHz) energy into the surrounding region via a coaxial lumped port. I'm using the Electromagnetic Waves, Frequency Domain physics module. The model behaves exactly as desired when the lower-limit of mesh quality is met, but endlessly iterates otherwise. I've tested varying the geometry and size of the tip to no avail. It seems that the minimum element size I can reach before causing non-convergence is about 3e-7 meters. (My model size is on the order of 1e-6 meters).

I'm so confused; there's obviously nothing wrong with the model setup/geometry/physics, because it solves for lower-quality meshes. Is this a computer memory problem? (I've got 8GB). Is there a lower-limit to element size / calculation in the [EMW, Frequency Domain] module that I've reached?

Thanks for the help.

Jon Michelson
↧
↧

Modeling tuning a cavity

July 7, 2014, 2:21 pm
$
0
0
Hi, I am very new to COMSOL, but I think I am getting the hang of the system. I was wondering if anyone might have suggestions as to how I may go about modeling tuning a cylindrical cavity.

I have modeled a cylinder and run an Eigenfrequency and Frequency Domain analysis. I hoped to use the Frequency Domain to model the electric field if the cavity was very close to being tuned but not exactly tuned. I used Eigenfrequency to run a very tight frequency sweep but all that I got was that the electric field = 0 . I used aluminum as the cavity walls and set an impedance boundary condition. Are there any suggestions? Am I missing something that makes this solution realistic?

Thank you,
Enrique
↧

error using parametric sweep of COMSOL4.4

July 8, 2014, 6:32 pm
$
0
0
Hi,
i am trying to simulate the weld pool procedure of laser welding, but some error happened when the model is computing. boundary temperature of domain is set as Tref+DeltaT, and “parametric” (auxiliary sweep) in the solver is used, DeltaT :1e2 1e3 2700 [K].

However, error happened when the DeltaT is 1.4e3 [K], and error message is "Failed to find a solution for all parameters,even when using the minimum parameter step.Segregated iteration leads to undefined function value.Returned solution is not converged."

Could u help me to solve this problem?
Thank u for your help.

Youmin RONG
↧

Thermal Micropump

July 9, 2014, 6:51 am
$
0
0
Hi all,

I am trying to build a model of a PDMS micropump that pumps a liquid stored inside its reservoir due to heat. When the air inside it expands due to heat, the liquid will be pumped out.
The heat source is a copper layer attached to the PDMS external walls, with a total power of 0.2 W.

I used heat transfer in solids and solid mechanics modules to simulate this model, but I kept getting an error saying:
"Failed to find a solution. In segregated group 2: The relative error (4) is greater than the relative tolerance. Returned solution is not converged".

- What is the reason behind that?
- Do you suggest using other modules, such as laminar flow or conjugate heat transfer?

Thank you so much
Best regards,
Marwan
↧
Viewing all 3988 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>