von karman instability

Numerical methods and mathematical models of Elmer
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: von karman instability

Post by raback »

Hi

Yes, this instability is on the verge of happening. You need minimal space and time resolution to initialize the vortexes. When I did the case I tuned down the resolution to make it as light as possible to run. Thus if you increase the timestep just quite slightly then you may loose the instability.

-Peter
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: von karman instability

Post by kevinarden »

I found a later Elmer Tutorial manual than the one I had. The time step size is typed in as

Time Step Size = $ 8/200

with the dollar sign it does the math

with

Time Step Size = 8/200 the time step is 8.
Gary R
Posts: 162
Joined: 12 Apr 2012, 07:23
Antispam: Yes
Location: Long Beach CA, USA

Re: von karman instability (Solved)

Post by Gary R »

Thanks for the help. The shortening of the time step pretty well solved everything. I'm still grappling with units in Elmer. Additional information in that area would be helpful. For most of the units it is just a matter of keeping the system consistent (cgs, mks etc.) Time on the other hand still throws me. Does 1 mean 1 second?

Somehow I got a second solver inserted in the system that was causing the blank frames. So my original installation is now also clean.

Thanks again for you help.

Gary R
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: von karman instability

Post by kevinarden »

Time is seconds, 1 is 1 second. Elmer does not specifically use units, the user has to keep them consistent. However the built in material library values and default constant values are in SI units, with meter as the length. So if you load a material from the library and do not convert the values, the material is in SI (meters). Elmer has a scale option where you can scale the mesh when it is read in.

Kevin
Post Reply