3.5a - Latest Discussions - COMSOL Forums

Does COMSOL treat the nodes at the boundary also as unknowns?

Tue, 22 May 2012 09:14:12 +0000

Hi Ivar,

But in COMSOL 3.5a, there is no option to clear solution, reset history etc.
The only option available is reset model, in which case everything including mesh will be deleted and the model will be saved as a .mfile.

So I have again created a new model with all the inputs, coefficients, B.C., solver settings (steady-state) and mesh.
But I have not run it as that will increase the file size. Please note that I have already entered the mesh element size and number for each subdomain, but I have not meshed it.
For subdomain 3, use mapped mesh (values already entered).
For subdomains 1 and 2, use free mesh (advancing front, values already entered in the dialogue box).

You just have to click mesh selected button for subdomain 3 in mapped mesh parameters. Click mesh selected separately for subdomains 1 and 2 in free mesh parameters. Then just click the solve button and see the results at z=0(boundary 10) using cross-section plot parameters dialogue box for subdomain 3.

Looking forward for your reply,

Regards,
Seetha

Periodic boundary conditions

Tue, 22 May 2012 02:55:47 +0000

Hi, all

Does any one knows what are the suitable boundary conditions for heat transfer scenario applied at the inlet and exit of a domain when the boundary conditions for turbulent kinetic energy (k) its dissipation (e) and axial velocity (w) are set as periodic conditions. Note that the temperature of the fluid always increases in the stream wise direction if a constant heat flux acts at the wall of the pipe. Since periodic conditions are used to reduce the computational burden only a small segment of the domain is modeled in this case which means we cannot set a Dirichlet boundary condition for the temperature at the inlet and exit.

Thanks for any sort of help

Does COMSOL treat the nodes at the boundary also as unknowns?

Tue, 22 May 2012 01:50:14 +0000

Hi Ivar,

I am using 3.5a.
The file I sent you earlier is .mfile and from 3.5a.
Isn't it not possible to open it in comsol GUI using 'comsol with matlab'?
I am attaching it again here.
I am not sure whether it has information about mesh. So I am mentioning here the mesh I used.

Domain 3: Mapped mesh- Boundary 9 & 12: no. of edge elements = 2000, element ratio =1
Boundary 10&11: no. of edge elements = 36, element ratio =1

Domain 2: Free mesh (advancing front) - Boundary 5: number of edge elements = 500, element growth rate=1
Boundary 6 & 7: element growth rate=1.1
Boundary 8: number of edge elements = 2000, element growth rate=1

Domain 1: Free mesh (advancing front)- Boundary 1: number of edge elements= 50, element growth rate =1
Boundary 2&3: element growth rate =1.2
Boundary 4: number of edge elements= 500, element growth rate =1

If you are not able to open this .mfile, how can I send my model .mph to you?

Regards,
Seetha

How can I build the structure with sinusoidal-shape interface in comsol?

Mon, 21 May 2012 03:10:46 +0000

Hi! I am a novice and have a question that how to construct a geometry object with the sinusoidal shape interface,e.g.sinusoidal grating in comsol 3.5a? Thank you very much.Here is my email:hujigang@hfut.edu.cn.

How to get the derivative of a force with respect to a self defined variable?

Mon, 21 May 2012 00:47:46 +0000

Hi All,

I am working on a simulation of a particle moving in water.

The particle moving velocity is V and its value is 0.01m/s, which we can define as a moving boundary condition.

Then COMSOL can calculate the hydrodynamic force (F) on the particle by integrating the hydrodynamic tensor (for example,T_z_chns and T_r_chns in a 2D axisymmetric geometry).

But, question is how I get the derivative of F with respect to V?

I guess coupled variable in comsol should be used as well as matlab.

Any experts can help me out?

How can I avoid the interference between the deflected structures

Sat, 19 May 2012 20:23:15 +0000

I'm using comsol to do static analysis for a cantilever beam. I used a relatively large force to press the beam toward the substrate, I expected the deformation of the beam tip would stop at the substrate, but according to the value I got, it already penetrated the substrate.
Has anybody had this problem before?
How can I avoid it?

The attached picture shows another structure. A load is added at the upper surface in the middle, the deflection will cause the upper inner surface of the channel touch the bottom when the force is large enough, but in my simulation, it goes beyond the bottom, how can I solve this problem?

Please help: 3D model of an elastic (collapsible) thin walled tube with fluid flow (3.5a)

Sat, 19 May 2012 20:08:19 +0000

Hi Everyone,

I am working with with two other engineering students trying to create this model for some of our school faculty.

Our final goal is to output a output a force - displacement curve. The model should show the deformation of the vessel walls as the fluid flows through it.

Our variables are internal flow rate ( affecting internal pressure in the tube ) and external pressure.

We created the internal flow rate just by doing boundary conditions of high pressure in ( 2989.0 [N/m^2] ) and zero pressure out. Our fluid is just water (density = 995.7 [kg/m^3] , viscosity = 0.789 Pa*s)

The dimensions of our tube are 6.35E-3m inner (lumen) diameter, 7.32374E-3m outer diameter, (wall thickness = 3.96875E-4 m)

Modulus of the tube wall is 0.8MPa, Poisson's ratio = 0.5.

Our model is for some reason saved as a .m file and we cannot figure out how to open it in COMSOL. I attached the .m code if that will help anyone understand our model, if not please ignore it.

We could only get a velocity profile to show up, but couldn't get any stress values to appear on the walls.

any help would be very appreciated. We need help with every aspect of it, including which physics to use, how to model the fluid flow, and the vessel deformation, and how to couple the two.

Thank you everyone

Thin Film BAW Composite Resonator simulation in COMOSL 3.5a

Sat, 19 May 2012 12:22:40 +0000

I'm currently engaged in the simulation of FBAR device in COMSOL 3.5a, and want to use this model.

However, "in the model library", I find that the admittance of electrode is defined as "0.5[mm]*nJs_smppn/1[V]", where 0.5[mm] is the width of the electrode, and "nJs_smppn" is the surface current density, which I don't know why, because we use the integration coupling variables, and the width of the electrode is the variable.

Then why we use the "0.5[mm]*nJs_smppn/1[V]", but not the "1.7[mm]*nJs_smppn/1[V]", where 1.7[mm] is the thickness of the 2D scheme of FBAR device.

3D Model of a thin walled collapsible tube

Fri, 18 May 2012 01:14:49 +0000

Hi Ivar,

My name is Justin I am an undergraduate biomedical engineering student at Cal Poly, San Luis Obispo. I am working on my senior project with two other bmed students. Part of our project is to create a 3D model (in comsol) of a thin walled collapsible tube. Our variables will be external pressure and fluid flow rate.

We are having a lot of trouble with our model and we would really appreciate some advice. If it would be possible, could you please take a look at our model? We need to output a stress-displacement curve.

We are all very new with COMSOL, any assistance, advice, or guidance would be greatly appreciated by all of us.

Thank you

Airfoil in Air using FSI

Wed, 16 May 2012 22:31:14 +0000

Hi,

I am trying to compute an FSI simulation of an airfoil with a constant wind flowing from the right to the left with slip wall condition. The model will converge if the Re number is low (less than 10^5). As I increase my Re, the model will not converge. In additional, my inlet ramp flow velocity will need to be increase very slowly before it reaches constant. I am wondering if there is something wrong in my setup? Or, can FSI be used for high Re.

I am wondering if anybody can point me to the right direction.

Thanks a million in advance!

Ting

What is P_chms?

Wed, 16 May 2012 09:25:04 +0000

I noticed this variable in some equations auto-generated when importing parameters from the Chemical Reaction Engineering Library. If I have an application mode that has pressure as a dependent variable (say General Laminar Flow) are "p" and "P_chms" the same thing?

Simple Acoustic

Tue, 15 May 2012 07:53:45 +0000

Hello,

I am trying to model a 2D acoustic time harmonic square (10x10[m]; left bottom corner at (0,0)).
At the top the pressure is 101325[Pa] and now I just wanted to get started with the gravity force as a kind of a load, but how can I implementate it to get a continous rise of the pressure?

The 2nd step should be to create a point-source which is located at (0,0). But this is as well unknown for me yet.

Thanks for any hints on this.
Nikolaus

error 2100

Mon, 14 May 2012 07:50:11 +0000

Hi,
I'm a comsol 3.5a user.I want analyse a cantilever composed of several layers.I draw 4 layers on my system and,when I add an other layer, error 2100 :'internal error in geometry decomposition' appears. I don't know why!
If someone can help me.

Thanks

Constitutive relations (coupling E and H)

Sat, 12 May 2012 18:54:44 +0000

I am trying to simulate a medium where the constitutive relations are given in a more general form, as follows:

D=epsilon*E + alpha*B
H=(1/mu)*B + beta*E

I do not know how to implement that in comsol 3.5a.
Can somebody help me with this?
Thanks
Ben

boundary conditions of applying an alternative current

Mon, 30 Apr 2012 10:07:03 +0000

Hello everyone,

i need to simulate a method which uses AC signal in a relative low frequency of 128 Hz. I have run the simulation of the same device (test Wenner) with DC giving not bad results, but now i need to simulate with AC, as it really is, hoping that it will give similar results to that of DC (since it works in low frequencies). So, i changed my module and i went to AC/DC module, on order to use the module Electric currents of quasi statics, Electrics as you have proposed to me in an earlier question. The thing is that i create points that they have to inject on to a block an alternative current, i.e. sinusoidal. However, when i go to the point boundaries i do not get the option for currents but only potentials.

If instead of the points i create surfaces i go to the surface boundaries and i write an expression for the inward current flow : 0.01sin(8t) for exemple, will it comsol understand automatically the form of the signal or should i also change the solver parameter to time dependent and adjust the time stepping etc? (on the same window i am using the time harmonic currents and not the transient electric currents, is that right?)

Thank you in advance!!!

Marilena

FSI+Rotating mechinery

Sat, 28 Apr 2012 06:28:01 +0000

hello, I am ready to simulate the strength of strired tank of baffle. I used the Rotating Mechinery model and the plan strain model. when I initial the model, the comsol 3.5 gives error:Messages:
Failed to evaluate expression
- Expression: cos(-2*pi*rpm*t)*X-sin(-2*pi*rpm*t)*Y

Failed to evaluate variable
- Variable: t
- Geometry: 1
- Boundaries: 14,15,16,17,18,19,20,21,22,23, ...

Can anyone help me fix this error?

Thank you!

pulse and heat flow

Wed, 14 Mar 2012 06:04:17 +0000

Dear Sir

I need urgent help !!!

i have to apply pulse train voltage with( 13kv and 1khz) on cable termination to get its effect on temp. after 3 hours.

i can not solve by time dependent solver (take very long time)

i can solve for 10 pluses and get the heat produced Q(W/m3) after 10 pluses then i can get the total energy produced by multiply (Q* (time of 10 pulses =0.01 sec). afterthat , how can i use the value of Q or the value of the energy to get the temperature after 4 hours

reaction rate for physical absorption

Wed, 01 Feb 2012 08:32:23 +0000

I am using Comsol Multiphysics 3.5a to model a physical absorption. I chose the che chemical engineering module/ mass transport/ convection and diffusion. I want to know how to set the reaction rate in Sumdomian settings for physical absorption? Hope any professional friends can give me some help. Thans a lot!

Extruding to 3d

Wed, 29 Jun 2011 18:00:15 +0000

Hi,
2 questions:

1) I am creating a ring in 2d, and it seems perfectly circular, but when I extrude it to 3d, it suddenly gets slightly jagged sides and becomes a sort of polygon. I tried to fillet it in 2d and then extrude, but it makes no difference, the 3d still has edges. Anyone know how to fix this?

2) Is there any way to put a 2d face on a 3d object? For example, on a cylinder, I would like to put a small circle on its top face to model a laser's heat deposition. I think this would let me impose heat transfer conditions on the boundary. Other suggestions are also welcome.
Thanks

Determine Arrenhius damage integral for a solution

Wed, 12 Aug 2009 18:05:07 +0000

I am trying to apply an Arrenhius damage integral DURING the solution of an electrosurgical problem using the Bioheat Transfer Module along with the Conductive Media DC module (T and V are the dependent variables) . The integral is of the form:

Adi = int(A*exp(-E/(R*T),t)

where Adi is the Arrenhius damage integral, int stands for integral and A, E, and R are constants. T is the nodal tissue temperature and t is time.

Is there a way to apply this into the Global Equations by taking the derivative of Adi? Thus making the equation:

Adit = A*exp(-E/(R*T)

I tried this, but received an error that T could not be solved for. Do I need to do something in order for Global Equations to access my dependent variables?

Thanks for any help...I am also interested in any other way that anybody can think of!