|
[Sponsors] |
July 16, 2007, 00:25 |
Moving Mesh
|
#1 |
Guest
Posts: n/a
|
Hi All, I would like to generate a moving mesh and my problem is of turbine rotation and i know the velocity and i want to know the rpm for the corresponding velocity. and i culdnt find the mesh deformation option in my CFX10 so can you please tell how shud i apply the mesh deformation.
Thanks in advance |
|
July 16, 2007, 12:27 |
Re: Moving Mesh
|
#2 |
Guest
Posts: n/a
|
What do you want to deform? I hope you do not want to use deformation to let your turbine rotate?
If you want to rotate your turbine, you have to create two domains with a rotating mesh and a stationary mesh. Put the rotating parts in the rotating mesh and the rest in the stationary part. Then let the rotating part rotate with the known velocity and find out what the fluid does....... Good luck, Gert-Jan |
|
July 16, 2007, 15:20 |
Re: Moving Mesh
|
#3 |
Guest
Posts: n/a
|
hello
can i ask something????? if we already know the velocity of the air... and we would like to know the rotational velocity of the turbine??? how can we set this up??? i mean the opposite that gert says.. thanks a lot hayate |
|
July 16, 2007, 16:24 |
Re: Moving Mesh
|
#4 |
Guest
Posts: n/a
|
hi,
I'm doing the same simulation for rotor. Gert my problem is the solution converge really good and fast for the steady state solution but for transient solution it does not converge at all, when i change the time step and trying to keep the coarent number around 1 for water as a lubricant the solver keep fluctuating and i will not have any result at all. Do you know any solution or trick.?! i don't know how i should find an answer for transient simulation. |
|
July 17, 2007, 00:45 |
Re: Moving Mesh
|
#5 |
Guest
Posts: n/a
|
Thanks Mr.Gert, i did a problem for my micro air vehicle with flapping wings in that we used only mesh deformation option to flap the wings. so i am not much aware of propeller blades and i dont ve any idea of rotating mesh. is that we ve to apply domain rotation? option in cfx10
Thanks and regards Siva.K |
|
July 17, 2007, 16:16 |
Re: Moving Mesh
|
#6 |
Guest
Posts: n/a
|
Try a tutorial with rotating domains first. You will find out that two meshes are required, a stationary and rotating domain and that you need sliding interfaces between both domains.
Gert-Jan |
|
July 17, 2007, 17:36 |
Re: Moving Mesh
|
#7 |
Guest
Posts: n/a
|
You want to set the velocity and find out how fast your turbine will rotate. Not sure how you can do that using CFD. Out of the blue..............
First you have to know how much force on the shaft is required to let your turbine rotate. If you know that, I think you should do: - perform a simulation with your velocity on the inlet and let the turbine rotate with a well-guessed omega; - describe a relation omega = omega(Force), based on your dynamo data. - Set the rotational velocity omega on your tubine while Force is the force the fluid act on the turbine. This can be calculated using areaAve(Force)@turbine. This way, I guess that the final rotational velocity and the acting force will give an equilibrium with the force of your dynamo. Momentum and force are mixed up a little but this is the way to go, I think. Good luck, Gert-Jan www.bunova.nl |
|
July 17, 2007, 17:50 |
Re: Moving Mesh
|
#8 |
Guest
Posts: n/a
|
Well, what is converged? You need to be sure that within a timestep you obtain convergence. How many coefficient loops do you use in each timestep?
I guess that within a timestep you don't obtain convergence. What can be done? - reduce time step? - improve the grid on the GGI's? - are you sure you can obtain convergence? Is the flow around the turbine converged but in other regions not? In other words, where is the solution fluctuating? - are all BC and interfaces well defined? Gert-Jan |
|
July 18, 2007, 11:38 |
Re: Moving Mesh
|
#9 |
Guest
Posts: n/a
|
Hi,
I have to simulate a rotating cranckshaft.I tried with rotating mesh but it doesn't work. So I'm trying with two different domain: fluid domain and the crank (defined as a solid domain).My idea is to put crank in a FRF and crank case in a stationary frame of reference, but I encountered some problems: - How can I define a RFR, when I define a new frame of reference this option isn't activate - For solid domains is not possible to set any frame of reference. - It isn't possible to define a rotating solid domain Have you some hints? Thank you very much! |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
how to set periodic boundary conditions | Ganesh | FLUENT | 15 | November 18, 2020 06:09 |
[snappyHexMesh] SnappyHexMesh for internal Flow | vishwa | OpenFOAM Meshing & Mesh Conversion | 24 | June 27, 2016 08:54 |
How to let the mesh motion solver just solve a small region near a moving boundary? | zhajingjing | OpenFOAM Running, Solving & CFD | 9 | April 28, 2016 04:15 |
[snappyHexMesh] external flow with snappyHexMesh | chelvistero | OpenFOAM Meshing & Mesh Conversion | 11 | January 15, 2010 19:43 |
Waves halted by moving mesh | Piotr Czajko | COMSOL | 1 | November 24, 2007 02:00 |