Fluid Structure Interaction with Two Phase Flow

September 13, 2017, 6:39 am

≫ Next: Electrostatic (es) to solid mechanics (plain strain) domain coupling issue

≪ Previous: Import a matrix from mpheval operator.

$

0

0

Hello to everybody,

I am working on a simulation trying to reproduce the flow of two fluids( gas and liquid) flowing on a silicone membrane which gets deformed due to the fluid flow.

I saw the tutorial about FSI here linked (https://www.comsol.it/model/fluid-structure-interaction-361) where is simulated the flow of only one fluid. I have problems to simulate the same physic with two different fluids. I tried to couple FSI and Two phase flow, phase field but it didn't work.

How can I solve this problem?

Thanks in advance

---

Electrostatic (es) to solid mechanics (plain strain) domain coupling issue

September 14, 2017, 1:05 pm

≫ Next: multiple domain filed avaverge in matlab with comsol

≪ Previous: Fluid Structure Interaction with Two Phase Flow

$

0

0

Hi,

I am trying to simulate a simple cantilever beam using electrostatics and solid mechanics coupled by moving mesh domain. The space charge densities are assigned to different domains within beam and electrostatic results are verified using theoretical calculations as well as other FEA tools. I am trying to calculate force generated by electrostatic domain and couple to solid mechanics by Maxwell's stress tensor in both X and Y direction using boundary load. The resultant output displacement is much smaller from theory and other FEA tools. Could you please help me to understand the mistake I am doing here? I have attached my simulation file here for reference.

Thanks in advance

multiple domain filed avaverge in matlab with comsol

September 16, 2017, 6:25 am

≫ Next: How to solve wear ODE equation

≪ Previous: Electrostatic (es) to solid mechanics (plain strain) domain coupling issue

$

0

0

Hi, I am recently using the average function in the matlab with comsol for the homogenization of the filed in the domain, I know that I can use the command in matlab mphmean(model,'emw.Ex','volume','selection',18)
for the volume average of Ex in domain 18, but my question is how can I get multiple domain filed average like the Ex volume average of both the domain 18 and 41?

How to solve wear ODE equation

September 19, 2017, 4:24 pm

≫ Next: Antenna tissue interface-temperature and boundary conditions (Electromagnetics + Bioheat)

≪ Previous: multiple domain filed avaverge in matlab with comsol

$

0

0

Hello Everyone,

I want to solve the following ODE equation, is comsol capable of solving this?

My wear equation is Wd=Cd*P(t),

where my Cd= kV/H K= wear coefficent, V= sliding velocity and H is hardness of material. My P(t) is contact pressure changing with wear or Time.

So, here my P(t) is changing with wear or time so its a post process result or value changing with time. How can, I give the contact pressure as function of time in ODE?

can anyone please help me on this?

Thanks in advance.

Antenna tissue interface-temperature and boundary conditions (Electromagnetics + Bioheat)

October 9, 2017, 9:44 am

≫ Next: Interpolating external Data to Certain Parts of the Geometry

≪ Previous: How to solve wear ODE equation

$

0

0

Hi,

I am trying to simulate a simple antenna radiating into a block of a tissue using EMW and Bioheat physics and have followed the tutorial from https://www.comsol.com/model/microwave-heating-of-a-cancer-tumor-30

Initially, I inserted the antenna end into the tissue for about 5 mm and subtracted it from the tissue. The antenna end is cylindrical and is made up of ceramic with metal coatings on the side faces. I am facing these major issues and would be really helpful if someone could highlight and/or point out any missing items relating to any of the issues:

1) Setting up “Perfect Electric” (PE) boundaries to the metal coatings on the side faces of the ceramic end situated in the tissue provides expected S11 (~-12dB) however the temperatures soar to 70-80 degrees in 2 seconds or so (the expected value of T is around 45 degrees). On the other hand, not setting up PE boundaries yields expected temperatures but that is perhaps due to the fact that S11 is poor (~-5dB)

2) I have set up “Temperature” boundary conditions on the tissue boundaries including the common boundary shared by the ceramic tip and the tissue, however this results in a bit unrealistic temperature distribution pattern. Excluding the ceramic-tissue interface boundary in the Temperature boundary condition results in an expected temperature distribution pattern but the temperature values are unrealistic (~100 deg in 2 sec). I have meshed the antenna-tissue interface boundaries a bit finer as well.

3) Finally, and perhaps the strangest of all, if I keep every other variable same, and move the antenna position i.e. go deeper than 5 mm or bring it back to apply it just on the edge of tissue, the temperature values change significantly. In general, antenna tip on the edge of the tissue yields much higher values than when it is located deeper in the tissue. Considering the fact that the ceramic end is subtracted from the tissue, it should not make any difference? I have checked the material properties, geometries, meshes, radiation boundaries, port plenty of times but no success yet. I would really appreciate your comments.

Many thanks,

Shailesh

Interpolating external Data to Certain Parts of the Geometry

October 24, 2017, 11:59 pm

≫ Next: Pair thermal contact

≪ Previous: Antenna tissue interface-temperature and boundary conditions (Electromagnetics + Bioheat)

$

0

0

Good Morning,

I need to import externally computed datapoints into COMSOL multiphysics.

I have a geometry which looks like a sandwiched structure, consisting of a metal - insulator - silicon - metal. I want to load my externally calculated electron distribution onto the silicon itself.

The reason why I need to do this, is because Comsol is unable to calculate the electron distribution as soon as the structure looks like a metal - insulator - silicon - insulator - metal, which I also need to analyze.

I tought I could just create an interpolation (int1), load the datapoints (x,y,n(x,y)), check "Use Spatial coordinates as argument" and change the Frame to Material. Then I would just call the function from the material properties as int1(x,y).

This doesn't throw me an error, however the simulation does not converge. What am I doing wrong?

Thank you for your time and ahve a nice day, Killian

Pair thermal contact

October 27, 2017, 9:43 am

≫ Next: Heating a sphere with a uniform flux

≪ Previous: Interpolating external Data to Certain Parts of the Geometry

$

0

0

Hello,

I want to use pair thermal contact in my simulation ( two polymeric film that press toghether under the defined temperature). actually I get good results that have good agreement with experimental reasults at first run but when I chanegd the parameters ( Q, hg , hc, r) some times there is no difference or some times there is a error and can not be solved.If any one have an idea or information on these parameters Please help me.

Thank you in advance, Zahra

Heating a sphere with a uniform flux

November 20, 2017, 4:17 am

≫ Next: Evaporating droplet by level-set (How to shrink the drop due o evaporation?)

≪ Previous: Pair thermal contact

$

0

0

Nevermind

---

Evaporating droplet by level-set (How to shrink the drop due o evaporation?)

November 29, 2017, 3:23 pm

≫ Next: How to transfer desired x,y,z coordinates to MATLAB function?

≪ Previous: Heating a sphere with a uniform flux

$

0

0

Hi, I would lilke to simulate an evaporating sessile droplet. I tried the Level-set module in COMSOL but it seems it does not work! At the water-gas interface of a evaporating droplet, the flow is not divergence free. This feature shrinks the droplet during evaporation. But in the COMSOL interface, I did not find a place to add the velocity jump term to the Continuity Equation which is already del.u = 0 (divergence free). I know that I may be able to do this project by Moving Mesh module, but I still would like to understand whether it is possible to add the velocity jump term to the continuity equation (i.e., del.u = X). Any help is appreciated. Thanks Mohsen

How to transfer desired x,y,z coordinates to MATLAB function?

December 1, 2017, 2:03 am

≫ Next: The Java application works fine but the JAR file doesnt work.

≪ Previous: Evaporating droplet by level-set (How to shrink the drop due o evaporation?)

$

0

0

Hello, I am working on a MATLAB function to calculate my pressure for Comsol. Now my function is working. How can I tell Comsol that i want the x,y,z position? So: In the Global Definitions now I do have a Matlab function called MATLAB1. In this function I have defined my MATLAB function like this:

function name: pressure

Arguments: x,y,z

I tried to call the function in the components via: pressure(x,y,z), but I keep getting the error: unexpected unit of Input. Do I need to declare my Matlab function otherwise? And: How do I tell Comsol that I want the x,y,z coordinates there? Any suggestions?

Kind regards,

The Java application works fine but the JAR file doesnt work.

December 1, 2017, 6:58 pm

≫ Next: Making a executable exe file from BeamModel.mph in windows

≪ Previous: How to transfer desired x,y,z coordinates to MATLAB function?

$

0

0

Hi,

I'm working on a Comsol GUI application with the Java API. I used NetBeans and for begining i tried the beam model. I created the DemoBeamModel.java according to the Comsol Documentation. It works fine when i run it in NetBeans but When i export the program to a runnable JAR file, it doesnt work. Anybody got the same issue?

The errors comes from this part of the source code:

ModelUtil.initStandalone(true);

This is part of the errors as follows:

Exception in thread "AWT-EventQueue-1" java.lang.UnsatisfiedLinkError: Can't load library: /home/m/Documents/lib/glnxa64/libcssystemutil.so at java.lang.ClassLoader.loadLibrary(ClassLoader.java:1827) at java.lang.Runtime.load0(Runtime.java:809) at java.lang.System.load(System.java:1086) at com.comsol.model.util.ModelUtil$15.run(Unknown Source) at java.security.AccessController.doPrivileged(Native Method) at com.comsol.model.util.ModelUtil.initStandalone(Unknown Source) at com.comsol.model.util.ModelUtil.initStandalone(Unknown Source)

I'm trying to fix it but if anybody already got this issue solved let me know. Thanks.  

Making a executable exe file from BeamModel.mph in windows

December 4, 2017, 7:55 am

≫ Next: Comsol Java API

≪ Previous: The Java application works fine but the JAR file doesnt work.

$

0

0

Hi,

I'm working on Comsol Java API and trying to make a exeutable exe file in windows from BeamModel.mph as i can double click and run it in windows or linux. any idea? is there any way to make it?

Comsol Java API

December 6, 2017, 12:35 am

≫ Next: How to use Matlab to manipulate PDEs and directly input it to PDE solvers in comsol.

≪ Previous: Making a executable exe file from BeamModel.mph in windows

$

0

0

Hi,

Anybody knows if comsol Javs API models (like BeamModel.java) are runnable by NetBeans/Ecilipse in Linux(ubunto)? I tried and i got some errors. I still couldnt figure it out how to fix it. Any help will be greatly appreciated. Thanks in advance.

Regards, Mohammad

How to use Matlab to manipulate PDEs and directly input it to PDE solvers in comsol.

March 1, 2018, 3:54 am

≫ Next: How to solve PDE with dependent variable?

≪ Previous: Comsol Java API

$

0

0

I am trying to solve a set of non linear coupled PDEs and the results are usually known be not converging without some modifications to the set of equations. These modifications usually makes the equations really long. Hence I would like to transfer the new long equations directly from some mathematical softwares like matlab after certain modifications. How is this done in comsol, so that it directly reads the set of equations from say a matlab script file?

How to solve PDE with dependent variable?

March 7, 2018, 11:01 pm

≫ Next: The value of Electrolyte conductivity

≪ Previous: How to use Matlab to manipulate PDEs and directly input it to PDE solvers in comsol.

$

0

0

Hi, all,

I am trying to solve two PDEs regarding phase field.

The common PDEs include one equation related to phi and another related to chemical composition (c). Here, I non-dimensionlized the composition by using U=U(c,phi), so U is a function related to c and phi.

Now the question is how I set U expression. Because when I set U as a variable with the expression, when I tried to solve the 2nd PDE, there were two problems. Fristly, since I have not defined c, the U expression is actually in orange color (warining). Secondly, since I already define U as a variable, while in the PDE, I have to set U as the dependent variable to solve PDE, and the error (duplicate variable) shows.

So how do I set U to let the PDE solver knows that U is a function of phi and c, and I am actually trying to solve c.

Thanks, Xing

---

The value of Electrolyte conductivity

March 13, 2018, 5:09 am

≫ Next: Stokes-Darcy Coupling

≪ Previous: How to solve PDE with dependent variable?

$

0

0

Hello, I'm begginer for comsol, and i'm working on fuel cell modeling. I started with the example PEM GDL 2D. From the parameters table , I found the electrolytic condutivity of membrane = 9 s/m, i'm wondering if this value refer to a specific material? if so what is it ? Thank you in advance,

Stokes-Darcy Coupling

March 21, 2018, 3:20 am

≫ Next: I want a domain not to override in the free at the Solid Mechanics

≪ Previous: The value of Electrolyte conductivity

$

0

0

Hello everyone,

Working on Stokes-Darcy coupling of a simple 2D geometry; two rectanglar domains where the liquid flows first in the Stokes domain then penetrates into Darcy’s through the interface.

The three interface conditions are shown in the image attached (velocity, stress and BJS).

How to typically introduce these conditions in COMSOL in order to well define flow through the interface ?

Any help would be greatly appreciated.

I want a domain not to override in the free at the Solid Mechanics

April 6, 2018, 6:20 am

≫ Next: Implementing the boundary equations in Stokes-Darcy Interface

≪ Previous: Stokes-Darcy Coupling

$

0

0

I have a immersed beam and want to find the natural frequency. However, for the solid beam the domain is overridden at the Free in Solid Mechanics. Therefore, It does not give me the results I want.

And I also have chosen the domain to be the beam in the Solid Mechanics settings.

Implementing the boundary equations in Stokes-Darcy Interface

April 26, 2018, 12:39 am

≫ Next: Particle photon(using DAC converter connected with 2 Wire Speaker) + play audio files using smart phone

≪ Previous: I want a domain not to override in the free at the Solid Mechanics

$

0

0

Hi everyone, Working on Stokes-Darcy coupling of two simple rectanglar 2D domains, the fluid flows in the Stokes domain first then penetrates into Darcy's, there are three simple equations that must be applied in the interface between the two physics : 1/ Continuity of the normal flux/velocity, 2/ Continuity of normal Stress and finally 3/ Beaver-Joseph-Saffman (BJS) ... the attached image shows the simpliest forms of these equations. My question is, what is the most efficient way to implement these three equations ? Thanks in advance.

Particle photon(using DAC converter connected with 2 Wire Speaker) + play audio files using smart phone

May 4, 2018, 12:10 am

≫ Next: Retrieving Boundary Topology in Simulation Applications

≪ Previous: Implementing the boundary equations in Stokes-Darcy Interface

$

0

0

Hey I have two Wire speaker which I wanted to interface with Particle photon using DAC, I wanted to play audio files through smart phone wireless and listen to the output through 2 wire speaker interfacing with photon, As I am new to control DAC with photon your expertise suggestion to make it work will be very helpful for me