Quantcast
Channel: Latest Discussions - COMSOL Forums
Viewing all 26192 articles
Browse latest View live
↧

Finding deflection and blocking force of Ionic polymer metal composite beam in COMSOL

May 9, 2017, 1:43 am
$
0
0
Dear Sir,

I am trying to find deflection and blocking force of IPMC beam in 3D and 2D but couldn't able to do so.Can you please send me the analysis file since i am very new in COMSOL.

Regards,
Sami
↧
Search

Syntax for gradient

May 9, 2017, 3:03 am
$
0
0
Hello everyone

I am building a model that has gradient of dependent variables in the source term. The syntax for any dependent variable 'u' is 'ux'. But, my source term has terms like gradient(u*v). ( u & v are dependent variables). I tried using pd(u*v), but the results are incorrect. Partial derivative gives only the scalar value. I need gradient for the direction. Is there any syntax for this - gradient(u*v) ?
↧
↧

electrostatic force time dependent

May 9, 2017, 5:43 am
$
0
0
Hello,

sorry for the simple question - I'm quite new to COMSOL - and hopefully you can quickly help me out.

I'm trying to calculate electrostatic forces using the stress Tensor in a 2D Geometry. Following some tutorials and forum discussions I managed to calculate the stationary results. (forces, electric field... )

Now I'm trying to extend my study to the quasi stationary case and calculate the dynamic electrostatic force.
First of all I added the ec module to solve for currents and applied the appropriate boundary conditions. But when I solve the time dependent study I get the same results for every time step (force and electric field)

Am I doing something completely wrong? Is the ec and es module not suitable for time-dependent simulations? How could I calculate time-dependent electrostatic forces?

thank you very much in advance!
↧

Pull-in voltage of resonator in 2D - keep getting inverted mesh elements

May 9, 2017, 5:54 am
$
0
0
Hi,

I'm trying to find the pull-in voltage of a beam in 2D, essentially exactly as is done in the "Pull-in Voltage for a Biased Resonator-2D" model in the application gallery. The only differences are the dimensions (my bar is 100 nm thick, 15um wide) and the material (aluminium). I've tried different (types/sizes of) meshes but that doesn't help. The only thing that seems to help is to make the bar unrealistically thick (like 400 nm).

Usually I get an error something along the lines of "Maximum number of Newton iterations reached" , with a ton of inverted mesh elements in the warnings.

I'm pretty much out of ideas. All I want is to get the pull-in voltage but I don't know what I'm doing wrong. I've attached the simulation file.
↧

How to give a parabolic velocity profile at inlet?

May 9, 2017, 7:38 am
$
0
0
I am solving a steady state problem, I need to give a parabolic velocity profile at inlet. How to do that?
↧
↧

CANNOT MODELING HEAT TRANSFERT

May 9, 2017, 8:01 am
$
0
0
Hello community! I am francophone and I studying in Cameroon, MASTER RECHERCHE in mechanic. I want to use comsol Multiphysics to model heat transfert in brake drum of a car but it is very difficult for me. It is a rubbing-drum-pad contact used on braking systems. The geometric model is already done but when I launch the modeling after entering the mechanical and thermal parameters, I get nothing (the drum rotates and must slow down when the pads rub on it).Please help me to solve this problem.
↧

Gas flow simulation

February 18, 2013, 5:59 am
$
0
0
Hi,
I am a rookie in comsol multiphysics and i need a little help with my model....
I have a chamber with dimensions 1m*1m*0.005m (x,y,z). I'm using a work plane to minimize the ram geometry.is that correct...;
The domain has Air and i want to loaded with Argon.
I also want to see the concentration of Argon inside my chamber and an animated flow of argon.
Actually i want to see the flow of the argon inside my chamber, it's movement .
I'm using the laminar flow (stationary and time dependent) and then the particle tracing for fluid flow.
Laminar flow
Inlet : flow rate = 5lt/h (or gas velocity, u=0.19m/s).
Oulet: pressure, no viscous
Stationary (or time dependent)
Particle tracing for fluid flow
Inlet : flow rate = 5lt/h (or gas velocity, u=0.19m/s).
Oulet: pressure, no viscous (dissappear)
Wall : bounce
I sent also a photo from my simulation-geometry. Please tell if there is a model better and more simple than mine.
am i wrong; I need to refine this modelor I need use different physics;
Thanks , a priori.
↧

Catalysis

May 5, 2017, 1:09 pm
$
0
0
Hello!

I would like to know how to turn the catalyst (any material present) used into a powder in order to get the right calculations. and How to turn the bulk materials in the libraries into small pieces.

Thanks in advance.
↧
Search

Outlet boundary condition: Outflow

May 9, 2017, 9:09 am
$
0
0
Comsol has three type of BC for outlet of a fluid flow. Pressure, Laminar outflow and velocity. I want to give a outflow condition. Since I don't know the valocity (it has to be resolved), I can give a pressure boundary condition. Then what value should I give for pressure? Suppose I am giving an initial pressure of 1atm.
↧
↧

Mass conservation

May 9, 2017, 12:35 pm
$
0
0
have a 2D fluid flow model with a flow through porous media. There is one inlet with 20 mm/s inlet velocity and I assume conservation of mass.
I want to check that the numerical mass balance of the models close with a reasonable error. To estimate if the result is reliable and not create or destroy water numerically in the calculation process.
Considering what the blog says in www.comsol.com/blogs/how-to-ca...servation-and-energy-balance/. I went to Results > Derived Values > Integration > Line Integration
The expression you use is written as: spf.rho*(vi).

Where spf.rho is the density of the fluid and vi is the velocity.

Can someone confirm that this is the correct way to calculate mass conservation in 2D models?
↧

Cannot find mli directory for livelink for matlab setup

May 9, 2017, 3:06 pm
$
0
0
Hi,

I'm trying to run a Matlab file that someone else made in COMSOL 4.3 but when I try to run mphstart it gives me the error, undefined function or variable mphstart. I had a look at the setup instructions and it says to add the path COMSOL4.3b/mli

For me COMSOL is installed in C:\Program Files\COMSOL\COMSOL52\Multiphysics but I cannot find the mli folder anywhere.

When I go into Licensed and used products inside COMSOL, LiveLink for Matlab is listed under other products.

Is there a way to check that LiveLink is installed correctly? Or is there something I'm missing?

Thanks
↧

Electrospinning_doc file attached

May 9, 2017, 5:35 pm
$
0
0
Dear Administrator/anyone in discussion form,

I have attached the doc file for electrospinning simulation.

I am unable to get correct result.

Would you help me with this?

I am using a single needle and a rotating drum.

I am not sure what mistake I have made here.

Please verify.

Thanks,
Blesson
↧

Failed to find consistent initial values.

January 10, 2012, 2:11 am
$
0
0
Hi.

I coose frekvenci transient study and get thise error.

Failed to find consistent initial values.
Last time step is not converged.

THANK
↧
↧

Magnetocaloric effect

April 22, 2015, 11:06 pm
$
0
0
Dear all,

I am currently working on my bachelor project, which involves the numerical analysis of the magnetocaloric effect (MCE). For those of you who are not familiar with this effect: When a suitable material is placed in a magnetic field, a small increase in temperature will occur in this material. Consequently, when the material is removed from the same magnetic field, the temperature of the material will drop. This effect is not built in in the COMSOL (5.0) package, and I probably need to model it using PDEs or ODEs. The governing equations for the MCE are:



or



or



In these equations, T = temperature, t = time, M = magnetisation, H = magnetic field strength, Cp = specific heat, Q = heat and is the permeability of free space. All these equations describe the same effect, but are rewritten in some kind.

My 2d model currently consists of 2 simple permanent magnets, creating the magnetic field. The magnetocaloric material is positioned in the magnetic field created by those two magnets.

I'd like to ask if someone is able to give me a push in the right direction on which equation is best to use and how to model this in COMSOL, in such way that an increase in temperature is seen in the magnetocaloric domain.

Thanks in advance,

Dirk oude Egbrink

↧

Rotating a static boundary load in the frequency domain

May 10, 2017, 4:21 am
$
0
0
I am having problems with trying to create a rotating boundary load in the frequency domain. I have a time domain model which works, but takes quite a long time to run.

In the time domain model:

I apply a static compressive boundary load of 108 Pa, to a small circular region on the top surface of a 3D cuboid (using a boundary load with an if statement). I update the location (in x and y) of this load by giving it a time dependence, which has the effect of moving the circular boundary load along a trajectory over the surface of the solid. I used a similar method as suggested here:
www.comsol.com/blogs/how-to-ma...onditional-in-your-simulation/

I have attached a diagram of the 3D model as a .JPG

In the model the load rotates around a 20 mm circle at a frequency of 70 Hz. I.e It has a fixed angular frequency, so i assumed i could model this in the frequency domain.

To try and implement this in the Frequency domain:

I changed the loads position from a time dependence to a phase dependence.
The boundary load statement in Frequency model:
Boundary load>Load per unit area > Z Direction : if(f_distance<R_f,-108,0)

f_distance - Distances of every node from point on the trajectory. This term is phase dependent.
R_f - Radius of the circle over which the boundary load is applied.

I then tried to run this at 70 Hz using a frequency domain study.

What happens?

-The if statement ensures that the load is applied to a small circle of radius 5 mm.
-The load remains at the initial position on the circular path (i.e zero phase), and does not rotate around the path as the phase changes.
- The boundary load oscillates between + and - Z,: due to the load being 108 boundary load being sinusoidal.

What i want:

To apply a static compressive load which is periodically rotated around a circular path at a constant frequency.

Is this possible using Comsol's frequency domain study?

If you made it through all that, Thanks! I look forward to hearing your suggestions.


↧
Search

include parametric sweep of stationary in time dependent study

May 10, 2017, 6:02 am
$
0
0
Hi,

I am a beginner in Comsol. In a stationary study I simulate electric fields for different voltages in a given geometry using the parametric sweep option. After that I include the stationary study to a time dependent study to look at the particle tracks. I can't figure how to do the time dependent study for each of the parametric sweep and evaluate for example the number of particles that arrived at a boundary for each of my stationary studies.

Can anyone help?
Cheers,
Francesca
↧

Non-symmetric electrical field/potential given symmetric model?

May 10, 2017, 8:10 am
$
0
0
Hi All,

I'm a beginner so please bear with me!

I'm trying to simulate an electric field within a symmetric model, but my field keeps coming up as not symmetric. It's a base with an initial voltage connected to 5 pillars that have the other end at ground. But, the field and potential are only displayed on a single pillar. Any help is appreciated. Thanks!
↧
↧

Exporting the results of a parametric sweep in separate files

May 10, 2017, 8:19 am
$
0
0
Good day,

My question is rather easy to pose, but I can't seem to figure out how it is done. What I have is a parametric sweep over N parameter values. I want to then export these via Results -> Export. There I can select each result and export it to a file with a specific filename. Say I have 10 results, then I can manually save 10 different files. What I'd like to find is an option to batch export each of these results to separate files, perhaps with some name like aa_bb.csv where aa is fixed and bb is iterated over for the N different results.

Is this possible with Comsol 5.2?

Kind regards
↧

Floating Potential

May 2, 2012, 12:35 pm
$
0
0

I have a question about floating potential condition. I have 2 metals (like electrodes) which are not directly connected to voltage source and i want to use them to measure the signal (or voltage variation in my medium). According to some of the posts here i should use floating potential option as my boundary conditions for these metals. Is that right?
And if it is correct, then what should i assign in its terminal charge value section (in floating potential option)?

Best
Bahareh
↧

Magnetic Insulation not applying correctly

May 7, 2017, 10:07 pm
$
0
0
Hi everyone,

I'm working on a figure-8 coil model that is widely used in medical treatment for neural diseases.
I have successfully modeled the kind of coil I wanted and created a boundary volume around it.

I'm interested in simulating its magnetic field intensity in both 2D planes and in 3D volume.

I have a problem that the magnetic insulation boundary conditions apply themselves to the coil itself - not to the boundary volume as planned, causing the results to appear only inside the conductors. For the boundary volume it tells that magnetic insulation is "not applicable".

I have attached a screenshot of the compiled result I got.

Would anyone have any ideas what could cause this.

Thank you for the help!
↧
Viewing all 26192 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>