Continuous streamlines across non-conformal interfaces - CFX

francois louw · February 13, 2015, 07:49

Hi all

I'm modelling a fan and my domain consists of various sub-domains 'stitched' together. Where the sub-domains meet, the interfaces are non-conformal. The entire domain is solved in the rotating reference frame. Therefore every sub-domain is static relative to the other.

I'm presently having trouble trying to plot continuous 2D streamlines on a surface through my domain in CFX. I've read all the posts on "Velocity in stn frame" and the relative "Velocity" options in CFX, but I think my problem is a little different since I do not have an MRF between my "rotor" sub-domain and the rest of the computational domain. I'm plotting the "Velocity in stn frame", but a discontinuity still occurs at the interfaces (see figure attached). Is it possible that CFX can plot continuous streamlines in this case.

Any help would be appreciated.

Kind regards
Francois

ghorrocks · February 13, 2015, 18:33

I think there is an option to plot streamlines over periodic boundaries. I would select that.

Also, make the streamlines go forwards and backwards.

What are you using as the seed points for the streamlines?

francois louw · February 16, 2015, 03:36

Hello Glenn

First of all, thank you for your faithful replies as always.

Following on the topic in question:

The streamlines are now plotted with the "forward and backward" option enabled. The option for "cross periodic boundaries" is only available for 3D streamlines and not surface streamlines. If I however switch to 3D streamlines, the streamlines are continuous, but this is not what I want. (I would like to see the surface streamlines in the meridional plane and preferably continuous across the interfaces.) Do you think a possible reason might be the fact that the simulation was solved in Fluent and only post-processed in CFX post?

The periodic boundaries of my domain are also non-conformal, since it is an unstructured poly-mesh and therefore had to be solved by means of a periodic interface (where interpolation occurs between the selected periodic boundaries).

I've tried all the seeding options available in CFX (e.g. "vertices", "rectangular grid", etc.), but still do not find the streamlines to continuously cross interfaces. I've also tried the help manual, but have not found answers yet.

Regards
Francois

ghorrocks · February 16, 2015, 04:22

No the solver used should not matter in this issue.

I understand why the streamlines are discontinuous across domains - and if you are extracting a meridonal plane then this is going to be tricky to fix.

As a wild guess have you tried generating your meridonal plane using a user surface rather than a plane? I suspect it won't work but it might work and it should be easy to do so it is worth a try.

The best thing I can recommend is to use the line generated on the boundary of the domain as a seed for the streamlines, then do forward and backward streamlines. This will make some streamlines continuous across the interface but not all.

Another idea is to draw shorter streamlines (maybe limit them to 1s of flow time or what ever a sensible time is). Then the discontinuity at the interface will not be so visible.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ICEM] Simple pipe meshing - problems with y+ in CFX	Keizers	ANSYS Meshing & Geometry	23	January 15, 2015 08:00
Pros and Cons for CFX, CFdesign, COMSOL	Val	Main CFD Forum	3	June 10, 2011 02:20
CFX 4.4 installation problem	Pandu Sattvika	CFX	1	December 1, 2001 04:07
CFX 4 for two continuous fluid flow solution	sinjae	CFX	1	May 19, 2001 03:07
CFD Short Course & CFX User Day	Chris Reeves	CFX	0	September 11, 2000 08:53

February 13, 2015, 18:33		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,696 Rep Power: 143	I think there is an option to plot streamlines over periodic boundaries. I would select that. Also, make the streamlines go forwards and backwards. What are you using as the seed points for the streamlines?

February 16, 2015, 03:36		#3
francois louw New Member francois Join Date: Jun 2010 Location: Stellenbosch Posts: 9 Rep Power: 15	Hello Glenn First of all, thank you for your faithful replies as always. Following on the topic in question: The streamlines are now plotted with the "forward and backward" option enabled. The option for "cross periodic boundaries" is only available for 3D streamlines and not surface streamlines. If I however switch to 3D streamlines, the streamlines are continuous, but this is not what I want. (I would like to see the surface streamlines in the meridional plane and preferably continuous across the interfaces.) Do you think a possible reason might be the fact that the simulation was solved in Fluent and only post-processed in CFX post? The periodic boundaries of my domain are also non-conformal, since it is an unstructured poly-mesh and therefore had to be solved by means of a periodic interface (where interpolation occurs between the selected periodic boundaries). I've tried all the seeding options available in CFX (e.g. "vertices", "rectangular grid", etc.), but still do not find the streamlines to continuously cross interfaces. I've also tried the help manual, but have not found answers yet. Regards Francois

February 16, 2015, 04:22		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,696 Rep Power: 143	No the solver used should not matter in this issue. I understand why the streamlines are discontinuous across domains - and if you are extracting a meridonal plane then this is going to be tricky to fix. As a wild guess have you tried generating your meridonal plane using a user surface rather than a plane? I suspect it won't work but it might work and it should be easy to do so it is worth a try. The best thing I can recommend is to use the line generated on the boundary of the domain as a seed for the streamlines, then do forward and backward streamlines. This will make some streamlines continuous across the interface but not all. Another idea is to draw shorter streamlines (maybe limit them to 1s of flow time or what ever a sensible time is). Then the discontinuity at the interface will not be so visible.