CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > CFX

moving grid problem

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Display Modes
Old   February 23, 2006, 23:24
Default moving grid problem
  #1
Dr C.Bhasker
Guest
 
Posts: n/a
Hollow CFX users:

I am working cfx for the problem with moving boundaries. Though the procedures applied works fine but for my problem the moving part is not translating expected distance. More precisely I wanted move 5 mm radii 180 deg arc in u bend about 100 mm along lines connected to arc edge points in the flow direction. Please suggest any hint to this aspect through reply email.

Dr C.Bhasker
  Reply With Quote

Old   February 28, 2006, 22:34
Default Re: moving grid problem
  #2
Rajit
Guest
 
Posts: n/a
whats the problem you are facing with this. Is the solver kicking you out by forming negative elements or u hav problems defining the expressions.

u should understand one important concept in CFX u cannot move a rotating element in a simoultaneous operation.so do it step by step.first move it in very small increments and small time steps and then rotate it.

Thanks Regards Rajit
  Reply With Quote

Old   February 28, 2006, 23:25
Default Re: moving grid problem
  #3
Dr C.Bhasker
Guest
 
Posts: n/a
The run is failed mostly due to negative elements. The problem I am trying move semi circle in an u duct to in flow direction towards downstream end. I have used both coarse and fine grid the problem stops due to negative elements. Please suggest a hint to over come this problem. Very similar problem in ANSYS Flotran does well but it has got some limitations for solver aspects, especially when it is subjected to geometry and physics complexity.

Dr C.Bhasker
  Reply With Quote

Old   March 1, 2006, 20:19
Default Re: moving grid problem
  #4
Rajit
Guest
 
Posts: n/a
How are you moving it.do you know the distance it has to move or you are moving it based on the pressure from the fluid.

in either case the reason for the formation of negative elements is it's moving too fast and getting out of the fluid domain.calculate the distance based on the force applied for you to have an idea how much it has to move.

once you calculate the distance move it in small increments with very small time steps.you have to create a function to do it.

still the solver will kick you out because CFX does'nt morph the mesh (it does not form new elements).

you cross the first hurdle by moving it in the appropriate directions then i shall mail you in detail to go further.

personal request:iam in australia at the moment an want to be back in india is there a vacancy in CFD that you can guide me to.i have almost 2 yrs experience.

Thanks Regards Rajit

  Reply With Quote

Old   March 1, 2006, 22:20
Default Re: moving grid problem
  #5
Dr C.Bhasker
Guest
 
Posts: n/a
Mr Rajit:

Thanks for email reply. I know the physical distance for moving part original location to its distination. I have tried with small increment and tried to run 50 global time steps. After moving 2 mtrs it, moving part element shows distortion and then fails. Some researchers have done the same problem in ANSYS Flotran using ALE approach with remeshing, it is very impressive. I am posting this animated file on today at my home page www.geocities.com/bskr2k (under current highlights). I have tried to examine this and runned moving cylinder problem. Though solver writes multiple file both geometry and solution, but I am unable to animate across these files using ANMRES command. Please know me your email id and I will mail the cfx def file later. Concerned to your CFD Analyst job, I advise you to log on to cfd-online and post your cv to ANSYS-CFX openings.

Regards,

Dr C.Bhasker
  Reply With Quote

Old   March 2, 2006, 00:22
Default Re: moving grid problem
  #6
Rajit
Guest
 
Posts: n/a
Alright your problem is pretty simple.

try this first consider a simple piston and cylinder arrangement. move the piston downward.if you have to move the piston say 2m then intially move the piston for the first few meters say .5.you have your first.res file.

then you hav to create the geomtry again with 1.5meters more to go.in this case you subtract .5m since you have already moved this distance.

then recreate the mesh file and move by another .5m.you hav your second .res file.while running the solver for the second file in the solver definition form there is an option called interpolate the mesh on to the original file if iam right.

this means the results from your first file will be taken into consideration for the current solver process.

now open the first .res file in the post create the animation at various time steps.your last time step should read zero frames.

then without closing the post just open the second .res file with the first .trn file reading zero. create the animation file as usual.

i hope these hints are helpful.
  Reply With Quote

Old   March 5, 2006, 01:07
Default Re: moving grid problem
  #7
Dr C.Bhasker
Guest
 
Posts: n/a
Thanks for your reply message! I am in the processing of doing the same. However, if any problem, I will reach you again.

Hopy you seen other projects highlights of animated pictures.

  Reply With Quote

Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
moving mesh conceptual problem cesco FLUENT 2 September 13, 2008 07:22
modification of fractional step for moving grid Wei Main CFD Forum 3 May 13, 2006 21:30
About the moving grid system paul Main CFD Forum 0 October 25, 2004 00:47
Moving Mesh Problem in Unix Platform Arnab CD-adapco 3 August 16, 2004 08:59
moving boundary problem Andrea Panizza Main CFD Forum 1 February 26, 2002 15:54


All times are GMT -4. The time now is 22:46.