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

Problem with negative velocity values

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

Reply
 
LinkBack Thread Tools Display Modes
Old   September 13, 2012, 08:58
Default Problem with negative velocity values
  #1
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Hi all,
I am trying to simulate a transient analysis on blood flow in an artery. I have imported into CFX a velocity profile that includes a point where there is backflow of fluid. However, when I run it in solver and set up a monitor point to evaluate the inlet and outlet velocity, it does not plot the negative velocity values.
Can someone tell me how to deal with this problem?
Also I have used an "opening" for the outlet rather than an "outlet", how will this impact my results?
Many Thanks
Andris is offline   Reply With Quote

Old   September 13, 2012, 15:28
Default
  #2
Member
 
Benny
Join Date: Apr 2012
Posts: 40
Rep Power: 5
Benfa is on a distinguished road
Hi,

if you monitor the velocity it will show you the absolute value of it. If you want to see a change in the sign in front of the velocity you need to monitor the u,v and w velocity components

Ben
Benfa is offline   Reply With Quote

Old   September 14, 2012, 04:47
Default
  #3
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Thanks Ben,
I was wondering also, if it is possible to use the outlet option for my outlets if I have backflow? I am using an opening at the moment instead and I wondering if this will adversely affect my results.
Andris
Andris is offline   Reply With Quote

Old   September 14, 2012, 04:56
Default
  #4
Senior Member
 
Lance
Join Date: Mar 2009
Posts: 522
Rep Power: 11
Lance is on a distinguished road
Quote:
Originally Posted by Andris View Post
Thanks Ben,
I am using an opening at the moment instead and I wondering if this will adversely affect my results.
If the outlet is far away from the region of interest it might not affect your results, but it is impossible for anyone else but yourself to know that. Why dont you try it and look at the results?
Lance is offline   Reply With Quote

Old   September 14, 2012, 12:34
Default
  #5
Member
 
Benny
Join Date: Apr 2012
Posts: 40
Rep Power: 5
Benfa is on a distinguished road
Hi,

if you get backlflow at your outlet you should use the opening bc. That is the flow that will develop. If you use outlet bc you are disturbing your solution in that region. Why is it a problem to use opening?
Benfa is offline   Reply With Quote

Old   September 17, 2012, 04:47
Default
  #6
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Hi,
Thanks for your help, its not a problem, I was just wondering what repercussions there would be if I used an opening rather than an outlet seeing as there are important distinctions between the two.
Regards,
Andris
Andris is offline   Reply With Quote

Old   September 17, 2012, 07:14
Default
  #7
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Also, while setting up monitor points, I get the problem that after a certain number of timesteps, the values are no longer plotted and I cannot access them. how can I remedy this problem? Is there a limit on the number of timesteps CFX solver can hold?
Andris
Andris is offline   Reply With Quote

Old   September 17, 2012, 18:51
Default
  #8
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,805
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
No limit to my knowledge, I have had tens of thousands of points and it works fine. Are you sure you have the x axis set to all time steps, not just the latest few?
ghorrocks is offline   Reply With Quote

Old   September 18, 2012, 04:04
Default
  #9
Senior Member
 
Lance
Join Date: Mar 2009
Posts: 522
Rep Power: 11
Lance is on a distinguished road
Glenn, just out of curiosity, what is the computational overhead when using tens of thousands of monitor points?
Lance is offline   Reply With Quote

Old   September 18, 2012, 06:12
Default
  #10
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Yep, I go to monitor properties, set 'range settings' to 'display all', put 'plot data' by 'simulation time', the run is for 5 seconds at a time step of 0.01. Yet for every simulation, it stops plotting the monitor points even though the run keeps on going. I exported the results to excel to see if perhaps I could access them there but still no luck.
Andris is offline   Reply With Quote

Old   September 18, 2012, 07:18
Default
  #11
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
One of the errors i sometimes read is "Size mismatch, expected 76 values but found 77", it is at this point that the monitor points stop plotting. Can anyone help please?
Andris is offline   Reply With Quote

Old   September 18, 2012, 08:36
Default
  #12
Senior Member
 
Lance
Join Date: Mar 2009
Posts: 522
Rep Power: 11
Lance is on a distinguished road
I would guess that the solver manager expects to find 76 monitors but finds 77 and decides to not plot any of them. Have you tried to "clear generated data" (or whatever it is called in workbench) to reset the monitor displays in the solver manager?
Lance is offline   Reply With Quote

Old   September 18, 2012, 08:52
Default
  #13
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Hi,
No not yet, I have changed the path file name of my file to something simpler to see if it makes a difference based on a previous post. If that doesn't work, I will try what you recommended. I don't really use Workbench atm, just CFX, is there an equivalent option in CFX and if so where can it be found?
Thank you,
Andris
Andris is offline   Reply With Quote

Old   September 18, 2012, 12:15
Default
  #14
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Hi,
I still get the same problem, the error message comes up randomly at different times, sometimes it goes to 1.2 seconds, sometimes to 0.4 seconds with the same mesh and the same settings. I'm really confused as to why this is happening? Could it be maybe to do with the license settings that my institution has?
Andris
Andris is offline   Reply With Quote

Old   September 18, 2012, 19:36
Default
  #15
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,805
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
Quote:
Glenn, just out of curiosity, what is the computational overhead when using tens of thousands of monitor points?
Sorry, my comment was unclear - From a small number of monitor points I have tens of thousands of readings and that works fine.

I do recall a few years back trying to generate a few hundred monitor points and the CEL parser failed. So I think the CEL parser is the limiting factor, not the solver. When I cut it back to something the parser could handle I cannot recall any significant slowing of the solver. But this was a long time ago and my memory might not be right.
ghorrocks is offline   Reply With Quote

Old   September 19, 2012, 05:41
Default
  #16
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Ok, so the CEL parser can hold only a certain number of monitor points? So far it plots a monitor point for every timestep that I go through. How can I change this?
Regards,
Andris
Andris is offline   Reply With Quote

Old   September 19, 2012, 06:05
Default
  #17
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 10,805
Rep Power: 85
ghorrocks has a spectacular aura aboutghorrocks has a spectacular aura aboutghorrocks has a spectacular aura about
I am not sure precisely, but I suspect it is the overall count of "bits" in the CCL which is the limit, rather than monitor points specifically.

BTW: I meant to say CCL parser, not CEL parser.
ghorrocks is offline   Reply With Quote

Old   September 19, 2012, 06:10
Default
  #18
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Ok, how do I remedy this problem?
Andris is offline   Reply With Quote

Old   September 19, 2012, 06:12
Default
  #19
Senior Member
 
Lance
Join Date: Mar 2009
Posts: 522
Rep Power: 11
Lance is on a distinguished road
Quote:
Originally Posted by ghorrocks View Post
Sorry, my comment was unclear - From a small number of monitor points I have tens of thousands of readings and that works fine.
Ok, I thought so
Didnt really see the purpose of having that many monitor points. I've done a couple of hundred thousands of iterations with monitors and that works fine, but the solver manager can be a bit slow to open.

So, yes, the limit is in the CCL rather than the number of iterations (or monitor points).
Lance is offline   Reply With Quote

Old   September 19, 2012, 06:29
Default
  #20
Member
 
Join Date: Aug 2012
Posts: 35
Rep Power: 4
Andris is on a distinguished road
Hi,
This is very confusing. What can I do ensure that the monitor points keep on plotting throughout my simulation? Your help would be greatly appreciated.
Andris is offline   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
[ICEM] negative mesh volume problem (icem-cfd/cfx) adam2008 ANSYS Meshing & Geometry 5 April 16, 2010 12:21
Plot a graph of velocity & boundary condition problem wanie Fluent UDF and Scheme Programming 0 December 11, 2009 11:40
Problem Interface Solid Fluid with wall velocity Solver v12 hills1 CFX 2 October 12, 2009 05:36
Variables Definition in CFX Solver 5.6 R P CFX 2 October 26, 2004 02:13
Terrible Mistake In Fluid Dynamics History Abhi Main CFD Forum 12 July 8, 2002 09:11


All times are GMT -4. The time now is 23:44.