Two phase flow using Level Set Interface

August 19, 2016, 6:32 am

≫ Next: meshing 3d conjugate heat

≪ Previous: Is there any possibility to model rotating porous media in comsol?

$

0

0

Dear all,

I am using COMSOL 5.0 for simulating the scour at bridge piers in a computational domain. I took the Physics of turbulent two phase flow level set interface method but the solution is not converged. After two days of simulation the progress is showing 0% without any error.

Can anybody help me ?

Thanks
Abhijit

---

meshing 3d conjugate heat

August 20, 2016, 12:01 am

≫ Next: Convergence of a huge model (10km) - bore hole heat echanger

≪ Previous: Two phase flow using Level Set Interface

$

0

0

Hi every one

Did the mesh effect the solution significantly in case of conjugate heat transfer to upper surface of duct contain fluid moving in turbulent mode ?? and which turbulent mode should I use since the duct consist of tow part one heated from upper and the second part has strong curvature>>>

should I draw the solid part or just I can simulated the it as fluid undergoes heat flux(for steady state study)

second question :If the mesh of my case required ram of 12 GB ,so what the ram that should I have to run and compute the solution >

regard ,
sadd

Convergence of a huge model (10km) - bore hole heat echanger

August 20, 2016, 12:28 am

≫ Next: The gravity term in the two-phase Darcy's law (v.5.2a)

≪ Previous: meshing 3d conjugate heat

$

0

0

Hey everybody!
I am trying to simulate 2D time dependent study of vertical borehole heat exchanger using non-isothermal pipe flow and heat transfer in solids. My model consists of rectangular which resembles the rock and 1D pipe. Water flows through the pipe from the surface to the bottom of the heat exhanger, it heats up on the way, because the temperature increases with the depth in th rock. The rock cools down as the heat transfers from the rock to the water. I have tried simulations with 5 km deep exchanger, everything worked great. However, now I amd trying to simulate 10 km one, with the same parameters, (except for the depth) and I cant manage to converge this one. I have tried to:

1. Increase density of mesh, up to a computing capabilities of my PC
2. Run steady state pre-run of the flow in pipe and then use this output as input in time dependent study
3. Increase the limit of iterations in solver to 100

Nothing worked, it is not converging and it always returns an error: failed to find consistent initial values.
I think that the only problem is the size of the exchanger and limit of the computing power. Could you help me please? How you would overcome this problem?

Thank you very much
Marian

The gravity term in the two-phase Darcy's law (v.5.2a)

August 21, 2016, 9:11 am

≫ Next: Need help in understanding how to develop Oscillatory flows inside a pipe

≪ Previous: Convergence of a huge model (10km) - bore hole heat echanger

$

0

0

Hi,

The two-phase Darcy's law of the version 5.2a doesn't have the gravity term.
Is there any way to solve this problem?

Thanks.

Youngmin

Need help in understanding how to develop Oscillatory flows inside a pipe

August 22, 2016, 3:27 pm

≫ Next: 2d input

≪ Previous: The gravity term in the two-phase Darcy's law (v.5.2a)

$

0

0

Hi,

I am solving a FSI problem where the oscillatory fluid flow inside a pipe is deforming the solid. I am using COMSOL for the numerical simulation of the system. I have a rectangular pipe (2D) and at the inlet of the pipe defined a pressure condition (forcing condition) to create oscillation in the flow.
my inlet condition is:
P(inlet)= P_mean + P_amplitude * cos (wt) -- (time dependent analysis)

Outlet condition is also kept as pressure condition having same equation as inlet but with sin term to create phase difference in the inlet and outlet pressure at each time step.
P(outlet)= P_mean + P_amplitude * sin (wt) -- (time dependent analysis)
the outlet condition was kept this way to continuously maintain pressure difference which will govern the fluid flow in the forward direction and then reverse its flow to create oscillations.

It is very much critical for me to define accurate pressure conditions as the solid deformation is dependent upon the fluid flow. Can anyone please tell me if is the right way to define pressure equation to create oscillatory flows? If not, please share your insights on the same.
Thanks.

2d input

August 23, 2016, 11:22 am

≫ Next: Defining free space as domain

≪ Previous: Need help in understanding how to develop Oscillatory flows inside a pipe

$

0

0

hello!

my module(2d) is 1 m*0.15 m and the thickness is 0.6 .

inlet velocity=3 m/s,( heat flux (q)= 1500w/m2 ,h convection =12 w/m2.k for upper side ,you can say the area that undergo heat flux and convection is 1m*o.6m )
should I multiply the three input by 0.6 ???

the second qustion how the result give that (m cp delta T + 0.5*m (va2-vb2))>q*A

Regard,
sadd

Defining free space as domain

August 24, 2016, 2:42 am

≫ Next: error when adding a velocity field to v flow battery model

≪ Previous: 2d input

$

0

0

Hi all,

i have an issue with Comsol. I have a CAD Model of an electrical box with some power components. Defining the materials and heat losses is not the problem as i can select the components without problem. To know how much the interiour heats up and how much the influence is to other components i want to simulate the heat transfer in fluid (air) inside the box, but i cant figure out how to define the free space. I already read the following topics here:

www.comsol.com/community/forums/general/thread/46155/
www.comsol.com/community/forums/general/thread/32396/
www.comsol.com/community/forums/general/thread/35578/
www.comsol.com/community/forums/general/thread/36376/
www.comsol.com/community/forums/general/thread/32396/


None of them appear to help me as i dont have a tubular object to use the "cap faces" nor a surface to use the conversation to solid function. Is there a way i can do it in comsol o is it better i reacte a negative volume in the CAD application and define it as my air domain?

To simulate the walls of the box is not of my interest as it is quite isolating and closed, so i only use it for visualization purposes to create plots that are nearer to the physical model on my desk.


Thanks,
Adrian

error when adding a velocity field to v flow battery model

August 24, 2016, 9:18 pm

≫ Next: Thermal resistance in non-isothermal pipe flow (nipfl)

≪ Previous: Defining free space as domain

$

0

0

Hi ,
I am new to Comsol. Recently, I am trying to build a vanadium flow battery model. I found the v flow model in model libraries is very useful. The model is too simple and I wanna add more to the model. I tried to add a velocity field using laminar flow or brinkman equation. However, the error comes out when I compute the model, It says undefined value found......Can anyone help me?

Thanks,

---

Thermal resistance in non-isothermal pipe flow (nipfl)

May 3, 2016, 5:30 am

≫ Next: Simulation of carbon atoms diffuse in amorphous media and collected at the interface

≪ Previous: error when adding a velocity field to v flow battery model

$

0

0

Gooday,

I'm using non-isothermal pipe flow model and heat transfer in solids physics in order to simulate a bore hole heat exchanger (BHE). the problem consist of a tube modeling by nipfl and a ground around the tube modeling as a solid heat transfer. The program seems to simulate the problem in a reasonable way but when the thermal resistance (R) is calculated, the obtained R by Comsol is higher almost 30 % compared with R obtained by formula or 2D model. I guess since this module simulate the pipe flow as a 1D problem and there is a gap between the pipe and solid which in graphics shows as a solid part, it adds the value of thermal resistance. Can anyone help or has a suggestion?

In attachment you can find the file.

kind regards,
Amin

Simulation of carbon atoms diffuse in amorphous media and collected at the interface

August 28, 2016, 10:46 am

≫ Next: Define density as function of pressure

≪ Previous: Thermal resistance in non-isothermal pipe flow (nipfl)

$

0

0

Dear, I am beginner to use COMSOL software and I am not familiar with this software. However, you guys can help me out.

I want to simulate a geometry of Nickel (300nm thick)/Silicon. Carbon atoms strike at nickel surface and diffuse at high temperature inside nickel and reach at interface of Nickel/silicon. However, silicon is crystalline which does not allow carbon atoms to diffuse any more.

So I want to find out in how many minutes nickel surface will be filled with carbon atoms and how much time it will take for a carbon atom to reach at the interface. Moreover, what will be the affect of temperature on diffusion of carbon atoms.
What will be the affect on the diffusion of carbon atoms if we change the thickness of nickel.?

Define density as function of pressure

August 30, 2016, 7:40 am

≫ Next: how to show 3D profile in 2D?

≪ Previous: Simulation of carbon atoms diffuse in amorphous media and collected at the interface

$

0

0

Hello

I model laminar flow in micro structure. My material is compressible and I know that the density is change by the pressure. How can I define the density as function of pressure?

Best regards
Mohammad

how to show 3D profile in 2D?

September 1, 2016, 1:47 am

≫ Next: average velocity

≪ Previous: Define density as function of pressure

$

0

0

Hai friends,
Can you please give an idea how to show the 3D image of the simulated velocity profile of fluid flow in the channel in 2D ?

average velocity

September 2, 2016, 12:33 am

≫ Next: Multiple-size particle tracing

≪ Previous: how to show 3D profile in 2D?

$

0

0

Hi all,

How to find the average velocity and average shear rate of a fluid in the channel for the simulated velocity and shear profile in 3D?

Multiple-size particle tracing

October 30, 2012, 4:24 am

≫ Next: Is there any possible way to use the results from a time dependent study in a stationary study ?

≪ Previous: average velocity

$

0

0

Hello everyone

I have a short question on the particle tracing module. Does anyone know that whether it is possible to simulat particles of different sizes in the same model. How to implement this since I could only add one unique particle property?

Many thanks

Yuexia

Is there any possible way to use the results from a time dependent study in a stationary study ?

September 6, 2016, 12:42 pm

≫ Next: Porous media modelling

≪ Previous: Multiple-size particle tracing

$

0

0

Hello,

I was wondering if I can use the concentration profile (from time dependent study with TDS module) as a boundary condition in a stationary study(AC/DC module).

Thanks,

---

Porous media modelling

September 7, 2016, 4:37 am

≫ Next: The difference between ALE module and FSI multiphysics module

≪ Previous: Is there any possible way to use the results from a time dependent study in a stationary study ?

$

0

0

Hi
Whats the difference between porous media Transport properties and Transport properties in Transport of dilute species. Because I can implement in both cases the porous media, but the results are different.
The porosity value is 0.78.

The difference between ALE module and FSI multiphysics module

September 8, 2016, 8:00 am

≫ Next: Poisson-continuity coupled equation

≪ Previous: Porous media modelling

$

0

0

Dear all,

I have a model attached below. The upper part and lower part are rock mechanics domain (blue color), and the intermediate part are laminar flow domain. At initial stage, a small gap exists between two half spheres. With high stress loaded on the upper boundary, two spheres contact and deformed largely. The final state is also attached below. In this process, the deformation of spheres is comparatively large. I encountered a puzzle to set up the modules, I think of two options:

1. in the physics toolbar, rock mechanics module, moving mesh(ALE) and laminar flow module are selected to set up the model. In this way, the large deformation are considered.

2. in physics toolbar, rock mechanics module and laminar flow module are selected along with activate FSI module (fully coupled selection) in Multiphysics toolbar.

In the doc prior to version 4.3, most large deformation are modeled by moving mesh (ALE), but after version 4.3, most are set up with FSI module. I want to know the difference between this two options. By the way, notice that my COMSOL is 5.0.

Poisson-continuity coupled equation

February 20, 2011, 12:50 am

≫ Next: Problems in assigning inlet-outlet boundary condition for a two phase flow

≪ Previous: The difference between ALE module and FSI multiphysics module

$

0

0

Hello,
I try to model the corona discharge in wire-cylinder geometry governed by the equations:

nabla (V) =- rho/epsilon0 (Poisson equation) and

grad (rho) * grad (V) = rho ^ 2/epsilon0 (equation of current continuity)

with boundary conditions:
V_w = V0 (on the wire)
V_cyl = 0 (on the cylinder)

As rho is unknown, I should compile until the electric field on the wire is equal to a given value Es.

knowing that I can give an initial estimate for rho on the wire --> rho_w.

My problem is I dont know how to solve these two coupled equations.

Problems in assigning inlet-outlet boundary condition for a two phase flow

November 21, 2014, 2:25 am

≫ Next: Solid to liquid phase due to external heat source

≪ Previous: Poisson-continuity coupled equation

$

0

0

Dear all,

I am trying to simulate a laminar two phase flow passing over a cylinder with a phase-field model where the cylinder is submerged into the water (heavier fluid), as a part of my project. I have tried a couple of candidates for the boundary condition at the inlet and the outlet. Initially, I gave a uniform velocity at the inlet for both phases plus a zero pressure at the outlet, then I assigned a mass flow rate for both phases and mass flow rate at the exit and a few other failed attempts. Also I have tried to give a step function at the inlet for the volume fraction at the level of my free surface and give a uniform velocity which I could not even implement.

Receiving a piece of advice, I have recently tried to give a zero pressure gradient at both sides and a body force or a small acceleration in the x-direction (as my driving force), but it seems that after sometime (4% of run-time in my case) the "reciprocal of the step size" increases and causes a failure.

I would be glad if there is a solution for the problem. I also attached the basic model here.

Amin,

Solid to liquid phase due to external heat source

September 15, 2016, 1:17 am

≫ Next: An empty cavity was created

≪ Previous: Problems in assigning inlet-outlet boundary condition for a two phase flow

$

0

0

Hi friends,

I am trying to model selective laser sintering. I was able to model the moving heat source, and I have
also implemented phase change module. But, I couldn't see the melting front due to moving heating.

I would like to know if there is a way around for modeling such solid to liquid transitions based on external heat source.

It would be really helpful, if you could help me with this regard.

Thank you in advance.

Best regards,
Adhi