Transient or steady state CFD problem

A subforum specific to the development of the OpenFoam-based workbenches ( Cfd https://github.com/qingfengxia/Cfd and CfdOF https://github.com/jaheyns/CfdOF )

Moderator: oliveroxtoby

julieng
Posts: 40
Joined: Sun Nov 25, 2018 8:57 pm

Transient or steady state CFD problem

Postby julieng » Wed Dec 19, 2018 10:37 pm

Dear all,

I continue my CFD learning.

I don't understand when I can compute turbulent cfd problems in steady or in transient simulations ? It is impossible to know before the simulation if the flow will be steady or not.

What should I do; try first a steady state case (play with the mesh) and if it does not converge switch to transient case ?

My next step is to compare cfdOF simulations (velocity, pressure, Drag force) with another software; and with literature experimental cases.
To acheive this, I am looking for the simplest 2D case where I can find experiments data. What can I choose, can you tell me a good example for doing this? Maybe you can share your experience on this point also.
I see the Openfoam guide and tutorials which are very helpfull also.

Best regards
thschrader
Posts: 1393
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Transient or steady state CFD problem

Postby thschrader » Fri Dec 21, 2018 9:06 am

I would choose a cylinder.
-easy to design
-2D-case possible
-different phenoma to investigate (von Karman Vortex, rapid falling drag-coeff at Re=100000 ...)
-try different meshers
-simulate laminar/turbulent, transient/steady conditions
Should be enough homework over christmas :)
What other cfd-software do you use for comparison of the results?
julieng
Posts: 40
Joined: Sun Nov 25, 2018 8:57 pm

Re: Transient or steady state CFD problem

Postby julieng » Fri Dec 21, 2018 9:04 pm

Yes i have a lot of homework, my boss want answers.
I compare with elmer in 2d cases

Best regards
thschrader
Posts: 1393
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Transient or steady state CFD problem

Postby thschrader » Fri Dec 21, 2018 10:12 pm

yess i have a lot of homework, my boss want answers.
I compare with elmer in 2d cases

Best regards
Where are your answers, your results, your screenshoths ? Mmmh?
julieng
Posts: 40
Joined: Sun Nov 25, 2018 8:57 pm

Re: Transient or steady state CFD problem

Postby julieng » Sat Dec 22, 2018 7:52 am

Hello,

I work nights and days :) , give me a little bit more time please...

More seriously I try a simple 2D case of flow past a cylinder. I don't know if my the residual is normal (residual of p higher than 0.1 for all time steps)
The Reynolds number in very low 0.5.
Mesh is max 10 mm, velocity max is 5 mm/s, timestep is 0.5s Courant number should be 0.25.
See my controlDict file attached.

I don't understand if I can constraint automatically the timestep of the solver in using the line "maxCo" in the controlDict file.
Maybe if I don't specify any timestep the solver choose himself the timestep to respect maxCo?

Image

Best regards
Attachments
controlDict.txt
(1.39 KiB) Downloaded 10 times
julieng
Posts: 40
Joined: Sun Nov 25, 2018 8:57 pm

Re: Transient or steady state CFD problem

Postby julieng » Sat Dec 22, 2018 8:09 am

I found this from oliveroxtoby:

"There are a few different ways of using PIMPLE. With a fixed time step size, it is wise to solve to a particular residual tolerance. One can use a fixed number of iterations if one keeps a limit on the max Courant number. The latter is the mode we use at the moment in CfdOF, as it is not really dependent on appropriate user input of time step size. The first option can give you faster run times, however, and should perhaps be considered."

It means that cfdOF fixe a number of iterations a timestep, no matter if the solution is converged or not and the solver continue the calculation for the next timestep? So the only important value is the maxCo?

If I leave timstep = 0, I have:

"19.9: Simulation exited with error"


Best regards
thschrader
Posts: 1393
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Transient or steady state CFD problem

Postby thschrader » Sat Dec 22, 2018 12:23 pm

julieng wrote:
Sat Dec 22, 2018 7:52 am
I work nights and days :) , give me a little bit more time please...
:lol:
OK. sorry for my impatience... :oops:
it is easier to understand your problem when you can post your FC-file.
julieng
Posts: 40
Joined: Sun Nov 25, 2018 8:57 pm

Re: Transient or steady state CFD problem

Postby julieng » Sat Dec 22, 2018 1:46 pm

Yes you are right

File attached

Best regards
Attachments
flow_past_cylinder_Re05.FCStd
file
(21.06 KiB) Downloaded 20 times
thschrader
Posts: 1393
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Transient or steady state CFD problem

Postby thschrader » Sat Dec 22, 2018 9:56 pm

pimplefoam crashes.
I rebuilt this tutorial
http://www.tfd.chalmers.se/~hani/kurser ... Report.pdf
but both 2D/3D cases dont work. After 30 iterations, pimplefoam stops with error.
tutorial2D.FCStd
(23.94 KiB) Downloaded 13 times
tutorial3D.FCStd
(21.55 KiB) Downloaded 13 times
julieng
Posts: 40
Joined: Sun Nov 25, 2018 8:57 pm

Re: Transient or steady state CFD problem

Postby julieng » Sat Dec 22, 2018 11:22 pm

Thschrader, do you tried my 2D laminar case?

Best regards