Physics - Latest Discussions - COMSOL Forums

2D and 3D boundary pair?

Thu, 09 Feb 2012 18:55:11 +0000

My basic question is how does one match a solution on boundaries between a 2D model and a 3D model? I feel like this was possible on a much older version of COMSOL, but I can't seem to figure it out now.

In detail: I'm working on a simple electrostatic model to solve for the capacitance of a system. The majority of the geometry is rotationally symmetric and is being modeled in 2D (it also has drastic changes in aspect ratio, which 2D is able to handle easily). One surface, however, has radial spoke features that I need to model in 3D. I am currently imposing the "zero charge" condition at the boundaries that I would like to pair and running the 2D and 3D as independent models, but this is not necessarily physically accurate.

timeint (Time Integral) gives wrong answers for mass balance

Thu, 09 Feb 2012 12:54:58 +0000

Hello everyone!

I hope the COMSOL experts and enthusiasts here can help a fellow user out :)

The situation is that I have a tube (L=1m) with a porous medium in it and air flows in and out again and at the same time water is being evaporated from the porous solid phase to the gas phase. This 1D model uses "Species Transport In Porous Media" and "Darcy's Law". The first one calculates the concentrations, the second one gives the species mass and calculates the density and velocity. The 'drying' is modeled using a mass source of H2O that is dependent on temperature and the amount of water left (first order Arrhenius type equation). These I modeled myself and seem to be fine.

The problem is that I want to make a mass balance over the reactor. So, the time integral over all mass that flows in + the time integral over all mass that is released from a source term into the gas phase, is equal to the time integral over all mass that flows out at the exit. When the temperature is constant, this seems to be true with an 0.5% error, which is quite ok, but if the temperature changes over time (simulated using a step function) the difference in the time integral over what goes in and out is more than 50% off!! And so is the integral over the source flux, it is ridiculous. I may be doing something wrong here, but I have no idea what. Can anyone please help me?

I have included the file. It is not as complicated as it may seem. Just O2 and N2 (air) flow in without doing anything and there is 1 kg of water available for release. No water flows in, so the end point should 'see' 1kg of water come by over time. However, according to the timeint function the answer is 0.49 kg.. same for the integrated source flux.

Things I already tried:
1) Increasing the accuracy of the timeint function ( timeint(START,END,dl.rho*dl.u*A_flowthrough,1e-10) )
(1e-8 is standard), usually I use 1e-5 since it doesn't matter much and takes a lot longer to calculate).
2) fiddling around with the time stepping. Tried all kinds of settings, it converges well now, but still no good answer.
3) Changed Tolerances: Abstol is now 1e-3, Reltol is now 1e-8, quite okay I think.
4) Conservative vs Non-conservative formulation (compressible vs incompressible gas), does not matter much.
5) Changed amount of elements, currently 2000, just to be sure. But increasing # elements does not help.
6) Changed discretization order from Quadratic to Quartic and back again, no effect.

It must have something to do with temperature, because it did work when T was constant and now it doesn't since T is a function. Included in this post is the actual model (4.2a) and some graphs.
The first graph shows my own calculation of the amount of water left in the solid, this goes to 0, like it should.
The second one show the time integral over the exit and this is 0.49 kg while it should be 1 kg.
The third has the same problem.
The last one shows the H2O concentration in the entire reactor vs time. Most important is that is goes to zero, so apparently all mass flows out of the reactor.

I would be very happy with any useful input! Thanks in advance!

Lubrication Shell Physic 2D

Wed, 08 Feb 2012 16:56:28 +0000

Hi all,

I am working with the Lubrication Shell Physic (this is aleady defined):

h = h0 (initial geometry) + hw(deformation on the solid wall) + hb (deformation on the channel base)

The only problem is h0.

I am trying to simulate a tilting-pad thrust bearing. The idea is:

"A hinged bar on a pivot point(intermediate position) and it should be placed in the position that the sum of moments in the bar is null."
This "bar" is the "h" from the Shell Lubricant.

I have already found the mathematical solution so I want to include:

A FUNCTION:

ht = k*x + n

AND A GLOBAL EQUATION:

State Variable: k
Expression: intop1(pf*(La-x))=0 [=sum of moments equals 0 on the tilting pad]

In order to achieve:

h = h0 (intial geometry) + ht (tilting correction) + hw (deformation on the solid wall) + hb (deformation on the channel base)

(ht can be added in the h0 gap)

How will you do that? I tried everything and I am not able to do it... thanks in advance.

Hagen–Poiseuille flow

Wed, 08 Feb 2012 09:52:50 +0000

I would like to use Comsol for a microfluidics project I'm working on. As this is my first time working with Comsol I am trying to simulate a simple problem to compare the simulation with analytical results.
The Hagen-Poiseuille flow problem is a solution to a simple fluidic problem, the flow of a liquid through a circular tube. More infor here: http://en.wikipedia.org/wiki/Hagen%E2%80%93Poiseuille_equation

I've set up the simulation as follows:
The geometry is 2D axisymmetric (radius is 15, length is 800)
The material is water
The physics is laminar flow (spf)
I use incompressible flow as my physical model
And define the inlet and outlet
The inlet pressure is 100 Pa, the outlet pressure is 0 Pa
The mesh is 'physics controlled' and normal

I run the simulation and to obtain the flow rate I apply a surface integration spf.U with Dataset 'solution 1'.
The result for this particular simulation is: flow rate = 5.593e-11 M3/s

The problem is that the analytical result is different. I've put it in Matlab to make sure it isn't me. The result is a factor of 22 too high.

r = 15e-6; % radius
dP = 100; % Pressure
Nu = 0.001; % viscosity
L = 800e-6; % Length
Fr = (pi*r^4*dP)/(8*Nu*L) % Hagen-Poseiulle equation for flow rate

Fr = 2.4850e-012 M3/s

I've triple checked these results but I cannot find why they don't match up. A factor 22 indicates that something's wrong, either on the theory side, or on the simulation side. Because I've just started with Comsol I would like to ask anyone to look over this model and see if something is evidently wrong. If anyone can give me suggestions why the analytical model is not correct I would like to hear it as well.

Thanks in advance

Export coefficient matrix of FEM

Tue, 07 Feb 2012 22:11:45 +0000

Hello Every Body,
I would like to obtain potential distribution in my geometry that I draw in COMSOL.
Moreover, the following of my work, I need the coefficient matrix which has been obtained in FEM. I would be thankful, if you let me know how can I get this matrix?

hi from comsol forum

Mon, 06 Feb 2012 12:49:57 +0000

Hello Alexis!

I have read your question and it took my interest a little bit :) I have a new job now and we should make the thin layers on some optics. Different kind of materials. It should be done by the PVD coating. You are writing that you simulate this kind of process. I just would like to ask you what is the basic way how to do it. We would like to buy some modul of COMSOL, but we don't know exactly which kind. CFD, micfrofluidics, electrodeposition, ....
Well, if you have 2 or 3 minutes to give us a hint, we would be very greatful!

Thank you very much,
greetings, Katerina N.

3d - Rf Simulation: PML mesh

Mon, 06 Feb 2012 07:52:50 +0000

Hi people,
I must simulate a block of lossy material (10x10x15cm) in the space and a plane wave that hits it. This is the extention of the problem I had posted some time ago ("http://www.comsol.com/community/forums/general/thread/24889/").
With your help I have set the space around the box 1.5 lambda times thick and the PML zone wide 1 lambda.
Now the new question is:
how make the PML mesh?
How many plane waves can I set toghether? Their experession become very big: is possible to semplify thei expression?

*****
My settings are:
for the electric field (in Electromagnetics waves tab)

KEx*exp(i*emw.k0*(VerNx*x+VerNy*y+VerNz*z))
KEy*exp(i*emw.k0*(VerNx*x+VerNy*y+VerNz*z))
KEz*exp(i*emw.k0*(VerNx*x+VerNy*y+VerNz*z))

and with this parameters:

Fi 0 Longitude
Teta 0 Latitude
Alfa 0 polarization
Emax 1000 Electric field amplitude
GrToRad pi/180 Degrees to radiant conversion
KEn Emax*cos(GrToRad*Alfa) Normal field
KEt Emax*sin(GrToRad*Alfa) Tangential field
KEx KEt*sin(GrToRad*Fi)-KEn*cos(GrToRad*Teta)*cos(GrToRad*Fi)
KEy -KEt*cos(GrToRad*Fi)-KEn*cos(GrToRad*Teta)*sin(GrToRad*Fi)
KEz KEn*sin(GrToRad*Teta)
VerNx cos(GrToRad*Fi)*sin(GrToRad*Teta)
VerNy sin(GrToRad*Fi)*sin(GrToRad*Teta)
VerNz cos(GrToRad*Teta)
*****

Thank you for attention.

Bye, Ale.

how to couple heat transfer and chemical species transport

Fri, 03 Feb 2012 21:25:23 +0000

I have water vapor + air that are diffusing in a porous media from one side of porous media, and from other side of this porous media I have liquid water entering in. There is also a phase change for water which effects water concentrations (in gas and liquid phases) in porous media.

so my question is if I specify my geometry ( porous media) as a domain for water vapor for example. so how comsol can build a connection between water vapor and water liquid. or how can I determine the temperature of air, when I have only added the water vapor ( steam) as material for geometry? shouldn't be an option for contributing these materials instead of overriding (because when you add another material to same geometry it overrides the previous one) ?

Because comsol only allows us to add one material for porous media, but I need to apply heat transfer on all of these materials in porous media. so what should I do ?

I am new to comsol, and I am testing comsol's abilities. so please tell me whether you can have a geometry with few materials applied to all of that geometry and not parts of it ( like multi-component flow inside a pipe) ?

Thank you very much

Handover of temperature between Heat Transfer Physics Modules

Fri, 03 Feb 2012 13:30:11 +0000

Hello everybody,

I have a problem with a Comsol model. I have two heat transfer Modules. The first module is used, to compute the heating phase of a to be simulated SLM-process. There shall be a displacement due to the heat during the heating phase, but the there shall be no stress at the start of the cooling phase in the area of the powder and the molten material. Therefor the second heat transfer module ist used with a different starting temperatur. The starting temperature shall be the end temperature of the first heat transfer module. Therefor I set the starting temperature of module 2 to T2 = T. The problem is solved in two study steps, one for the heating phase and one for the cooling phase with segregated solvers. The study steps contain different time intervalls for the heating an cooling phase. The heating phase seems to be calculated correctly. When the cooling phase starts to compute, the temperature of the heating phase does not get handed over to the second module. Instead a start value of 0 K seems to be used.

I started using Comsol just a few weeks ago, so I'm relatively new to the software. I'd be thankful for every hint or advice.

Manipulating heat-transfer coefficient

Fri, 03 Feb 2012 09:56:06 +0000

Hi All,

I have 3 domains, from bottom to top iron, air and glas. Iron and Glas are defined as heat transfer in solid, Air as heat transfer in fluids. Iron being heated, the heat flow to air and glas.

i want to increase the heat flux from iron to air by manipulating the heat-transfer coefficient.

the convective cooling doesnt work here, it works only at outer boundary.

thanks. happy weekend

akmal

Modifying Darcy's Law

Fri, 03 Feb 2012 06:27:25 +0000

Hi
I'm trying to model the effect of osmotic pressure using Darcy's Law in Comsol 4.2.
As it stands the equation is
d(rho*epsilon_p)/dt + grad(-rho/mu* kappa * div(p) ) = 0

I'd like to augment the equation so that the last term on the left hand side was div( p - OS)
where OS is my calculated osmotic pressure.
I've already tried placing the terms d(d(OS,x),x)+d(d(OS,y),y) on the right hand side using the 'mass source' option, however this seemed to create numerical problems (not surprising for a double derivative of a variable I suppose..)

Any thoughts? I'm realtively new to comsol, so sorry if this is a stupid question.
Cheers
John

How to create an ambient magnetic field

Thu, 02 Feb 2012 19:08:43 +0000

I am trying to simulate my structure when immersed in a uniform magnetic field in a 3D block. I tried using inward flux density as the boundary condition of one face, and zero potential as the opposite face, but the simulation did not converge correctly, I got a message saying "Ill-conditioned preconditioner". The structure I am using is simply a piece of soft magnet with nonlinear permeability.

Geometric nonlinearity

Wed, 01 Feb 2012 08:39:44 +0000

Hi,

I have used structural mechanics to simulate an indenting sphere on a half space polymer film. I am changing the elastic modulus of the polymer film to go from hard material to soft material.
I am trying to reach a certain force using weak constraints. When I use the elastic moduli from some GPa to around 80 MPa for indentation of 100- 400 nm, very well I can use linear elastic physics without selecting the box "consider geometry nonlinearity".

However, as long as I choose the elastic modulus 60 MPa or less with indention between 500 nm to 600 nm (I want to reach to a certain force and as the material is soft indentation is high), the simulation fails. It does not stop but a big triangular max pick appears at around 500 nm. However if I turn on "consider nonlinear geometry " and change the poission ratios, it works. but then I reach to a force at much lower indentation that I cannot compare the results with the previous calculations.

I would like to know why do that happens? and how can I fix it?

Yours,
Marjan

Time Derivation (or model in general) suddenly becomes unstable.

Tue, 31 Jan 2012 21:35:40 +0000

Hello everyone,

I am modelling electrical machines with the rotating machinery interface wthin the ACDC module.

This is my thread about the same model:
http://www.comsol.com/community/forums/general/thread/25702/

My problem is that the model won't solve when the permeability is non linear. Interesting is that after 2ms the simulation becomes very unstable and won't get to the 10ms mark.

In the attached pictures I want to show how unstable the time derivation of the flux linkage suddenly becomes.
You have an overview there and zoomed pictures.

Is it somehow possible to damp that derivation a litte? I'm afraid that this is causing my simulation to become unstable. Could it be that Comsol by default changes something at ca. 2ms that affects the derivation?

Here is a small overview about how I usually define my variables:

Psiges1 = 4*(intop2(mod1.Az))*...
e1 = -d(Psi1ges,TIME)
i2 = (u13+e1-e3)/4/Rsg/(-3)-...

Everything, including the derivations, is defined in the Definitions/Variables table.

Does anybody have an idea, why my model is so unstable? : (

Thanks in advance

modelling very thin layers

Mon, 30 Jan 2012 16:13:58 +0000

Hi,

I have a model where i have a very thin metal copper layer on the top of composite, now i do not want to increase the size of the problem by modelling this 0.002 mm layer explicitly therefore I want to give it a boundary condition such that the problem is solved including it now which boundary condition is usable here:

a. electric shielding
b. contact impedance-when i do this comsol doesn't select the top boundary and copper is highly conductive
c. floating potential?

thanks,
rabia.

Can not change domains

Sun, 29 Jan 2012 14:51:08 +0000

Hello.

I want to ask why each model are pre-defined modules (as magnetic insulation, electrical insulation ...) which are pre-defined domains that can not be changed because the buttons to work with the domain are gray and locked?
Can this be changed somehow?

Thank you.

convection/diffusion problem

Fri, 20 Jan 2012 00:36:36 +0000

Hi everybody,

I'm modelling convection/diffusion mass transport in a porous media. The diffusion coefficient of the model is very small (4e-12 m^2/s) and the velocity field is quite small as well (maximum 1e-5 m/s).
The boundary condition is a constant concentration at a side of the geometry.
After I solve the model for 15 min, the concentration at some parts of the model is greater than the source concentration which doesn't make sense. These areas have higher velocity magnitudes.
I would appreciate it a lot if anybody help me to solve this problem.
Can I use artificial diffusion to solve it? Does it affect the result of my model?

Many thanks in advance

General questions about Transient Acoustics and Solid Interaction

Thu, 12 Jan 2012 11:09:27 +0000

Hello fellow Comsol users,

I would very much like to ask a few relatively simple questions regarding transient acoustic models with some regarding transient solid interaction.

In essence my models are looking at how a membrane found in animal ear's behaves to incoming sound pressure, I am only looking to model the membrane not the ear itself. As you can imagine the membranes are very thin (around 1um in thickness and 2000um in diameter). I am doing the models in 3D and am looking at how the membrane deforms in the presence of a sound pressure wave.

I have adapted the Gauss Pulse model to a 3D one. I am modelling the source of the sound as a point Gauss Pulse.

My questions?

1)
Is there a better way to model an incoming sound field? The problem is that I need to place the Gauss Pulse relatively far from the membrane such that the incoming wave has a relatively flat wave front as it comes into contact with the membrane. This however creates relatively large models.

Is there perhaps a plate that I could use as the source of the pulse?

2)
Does anyone know of any other models which I could look at? There are not that many models in Comsol that look at Acoustic-Solid interaction.

3)
Any thoughts or suggestions on how to make the model more efficient.

I have included an image of the model so make things a little clearer.

Thanks very much, I hope there is some discussion to build on the knowledge base on acoustics in Comsol.

Rob

Particle tracing in COMSOL 4.2

Thu, 01 Dec 2011 19:30:05 +0000

Hi all!

Does anybody have negative experiences with particle tracing in COMSOL 4.2 (and NOT 4.2a)? Because I do.

COMSOL simply fails to trace small particles in even very simple situations. I placed a particle with a radius of 10 microns with zero initial velocity in a reservoir filled with still water. Gravity and buoyancy forces were present, of course. The density of the particle was higher than that of water so the particle was supposed to settle down on the bottom of the reservoir. Instead, it went straight up as if it was in an anti-gravitational field. And just by increasing its radius above 40 microns (and not changing anything else) I could make the particle sink.

Isn't that crazy? The particle size should only affect the sedimentation speed but not whether the particle sedimentation actually occurs or not.

Is this "bug" still present in version 4.2a? If yes, how could I use COMSOL for reliable particle tracing in microfluidic systems and laminar flows?

Thanks,
Daniel

Modelling of ion transit in oscillating electric field

Tue, 05 Jan 2010 08:38:08 +0000

Hi,

I've recently spent quite a bit of time using SIMION to model the trajectories of ions as they pass through fine wire meshes, upon which electric potentials are applied (such that a small DC electric field is generated). In addition, I am very interested in plotting the trajectories the ions take when a high frequency oscillating voltage is superimposed on the DC potentials which are applied to the grids.

The SIMION modelling has generated some very interesting results, and I would like to validate them by performing a similar analysis using COMSOL.

I recently attended the Intensive COMSOL course, held at Hatfield University, UK. The course only briefly covered the Particle Tracing option in the Post Processing screen, however from the little amount that we covered I was left with the impression that COMSOL will only plot the trajectory that an ion would take under steady state field conditions (after a solution to the model has been generated), and would not be able to calculate the trajectory of an ion under an oscillating electric field.

I am curious to find out whether my understanding of the particle tracing option is correct, and I would be very interested to hear from anyone who has succesfully used COMSOL to plot ion trajectories in non steady state field conditions.

Regards,

Jon Atkinson