|
[Sponsors] |
May 23, 2007, 15:34 |
Calculate a converged flow solution
|
#1 |
Guest
Posts: n/a
|
Hey,
I am currently using fluent version 6 to model the flow past a 2D circular cylinder. I have created a mesh using Gambit. after exporting the files to fluent and reading the file what do i do now ? what solver do i use ? how do i check or make the solution onverge ? What is the purpose of setting the force monitors (drag, lift and moment) ? How i do calculate a converged flow solution ? Please help me ! |
|
May 23, 2007, 19:18 |
Re: Calculate a converged flow solution
|
#2 |
Guest
Posts: n/a
|
Really, you need to read the manual. That would answer most of your questions. But, here's a quick brief on how to get started.
Fluent is set up so that you can follow the menus from left to right across the screen. First, check your grid and make sure there aren't any holes in it by hitting grid>check. The viscous model you should use is dependent on your Reynold's Number. If it below about 40,000 (for external flows, 2300 for internal flows) then the laminar flow model will work just fine. Otherwise, for a problem like this, the k-e model is probably the best. Then under define>materials you need to import your fluid properties from the fluent database. I think air is the default. Under boundary conditions, you need to set your inlet velocity and apply the proper materials to each zone. Check to make sure that FLUENT is using the same system of units that your grid was made in by clicking define>units. Then under solve> residuals, you can determine at what residual level the solution is converged. For Laminar flow, .001 is typically sufficient. For turbulent flow, you may want as small as .00001, but .0001 is the largest that is really recommended. If it doesn't converge to this level, you can set the residuals higher, or you may just have to redo the grid. Now, go to solve>initialize, set your flow velocity from the inlet (which you should have defined in your boundary conditions), determine how many iterations you want, and if you chose an unsteady temporal model, the amount of time between each time step and how many time steps you want to calculate, and then hit the iterate button. You can typically view the results visually under display>contours or display>vectors. Really, though, the manual is your friend. Good Luck! |
|
May 24, 2007, 02:33 |
Re: Calculate a converged flow solution
|
#3 |
Guest
Posts: n/a
|
Hey Donovan,
Thank you so much for your immediate response. I am studying the aerodynamically generated sound due to flow past a circular cylinder using fluent version 6.as stated in the manual, the first thing i need to do is to obtain a steady state converged flow solution. Now here are the steps that i have done. Drew and meshed the surface using gambit read it to fluent. went to grid-> check went to grid-> reorder-> domain i am currently investigating the flow past a cylinder with a reynolds number of 100. So, next i went to define->model->viscous-> laminar. next, define->materials->viscosity was set to 0.01, density was set to 1, velocity was set to 1 I then went on to define the boundary conditions. Went to solve-> control-> solution -> ok solve->monitors->residual-> activated the plot and left the default values -> ok Next i went to solve->monitors->force->activated the plot and write for the drag, lift and moment. Solve->initialise->compute from velocity inlet 1-> init iterate-> no of time steps taken was 200 with a time step step size of 0.1 s -> iterate The thing is why do i need to plot and write the drag, lift and moment ? how do i see from these 2 graphs and the residuals that the solution is converged? what is the definition of a solution being converged ? and and if it doesnt converge what do i do ? what parameters do i play with ? and finally how do i obtain this steady state converged solution ? also what solution do i use ? segregated or coupled ? implicit or explicit ? steady or unsteady ? as a beginner, i am using this case to study fluent and once i can get a hang of it ill be just fine Help me! Thanks |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Solution not Converged: Crossflow Blower | devanson | FLUENT | 1 | October 10, 2014 01:26 |
how to calculate the averaged mass flow along a line over time ? | iampolaris | OpenFOAM Post-Processing | 0 | March 10, 2011 23:03 |
calculate volume flow from a 2D simulation | SimonH. | OpenFOAM | 0 | October 27, 2009 05:39 |
Exact solution in compressible flow by RH cond. | jinwon | Main CFD Forum | 2 | May 29, 2007 03:49 |
Density based compressible flow solution | Ahmet | Main CFD Forum | 11 | May 22, 2007 04:48 |