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

Nice Challenge

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

Reply
 
LinkBack Thread Tools Display Modes
Old   August 25, 2006, 16:05
Default Nice Challenge
  #1
Bilir
Guest
 
Posts: n/a
Hi everyone,

This challenge is about the following: I am performing a simulation of a chamber (e.g. living room, including furniture), with one small inlet and one outlet. The incoming flow is CONSTANT and after a while a STEADY-STATE conditions is found. After this I give a (very) short pulse of inert material(e.g. CO2)at the inlet, and monitor at several points, inside the chamber, the CO2-concentration vs time. The area of this curve found (the integral), is also known as the DOSE (=INTEGRAL(C(t)*dt)) from 0 to infinite, unit: kg.s/m3). The interesting thing is that for EVERY spot inside this chamber, the same DOSE is observed. I checked this with my simulation and this is indeed the case. It is not easy to imagine this, as you would think that e.g. at the corner of this chamber low concentrations levels are measured, however the intergral with time results in the same dose as measured in the outlet for example. Can anyone of you prove this observation, based on the underlying transport equation? This means show that the DOSE is independent of POSITION.

Thank you, and looking forward to the outcome. Regards Bilir
  Reply With Quote

Old   August 25, 2006, 16:44
Default Re: Nice Challenge
  #2
Steve_NTUA
Guest
 
Posts: n/a
if you get the dose at steady state its logical to have that DOSE is independent of POSITION!

Else what time step did you use; and is it right with cfl critirion;;

  Reply With Quote

Old   August 26, 2006, 06:57
Default Re: Nice Challenge
  #3
Bilir
Guest
 
Posts: n/a
Hi

My time step = 0.005sec Convergece criterion: 5e-4 cfl? where does it stand for?

Regards
  Reply With Quote

Old   August 26, 2006, 08:29
Default Re: Nice Challenge
  #4
Jim_Park
Guest
 
Posts: n/a
It's not clear from your post. Do you have data as well as results of a simulation?

A thorough mixing of the CO2 might result from a really diffusive model for the convection terms in the NS equations. That could occur because of a diffusive (low order) difference approximation or because the space is not well resolved by your grid.

Just some possiblitlites to consider.
  Reply With Quote

Old   August 26, 2006, 09:30
Default Re: Nice Challenge
  #5
Steve_NTUA
Guest
 
Posts: n/a
if you are using un explicit finite difference there is a criterion namely 'clf' that gives you a minimum time step for a given Dx and Dy ... but i suspect that 0.005 sec is ok. could you send same results maybe i can help if i see them.
  Reply With Quote

Old   August 26, 2006, 14:00
Default Response
  #6
Bilir
Guest
 
Posts: n/a
Hi there,

Thanks for the replies. So when I monitor the concentration profiles at several points, I observe that near the inlet, I have high peak and more far from the inlet a low peak but wider. And all of them have the same area i.e. same DOSE. This is independent on the degree of mixing and/or diffusion, on provision that I have Steady-state velocity field and homogeneous injection across the inlet-pipe. Even in my experiments I observe this constant DOSE at all spots. The obervation of same DOSE everywhere should also be refelected by the transport equation. If someone could derive this, that would be nice.

Thank you.
  Reply With Quote

Old   September 8, 2006, 12:30
Default Re: Nice Challenge
  #7
Jack
Guest
 
Posts: n/a
Hi,

I doubt that you may have misunderstood the means of 'dose'. The dose should be an integral from inlet to the monitoring point(e.g. on streamlines or particle trajectories). So, dose close to the outlet should be highr that dose at the inlet.

The way you performed is a tracer simulation process. So if your chamber is close to a plug flow, you definitely can get the same integral value of cdt at each monitoring point, which actaully shows the recovery of the tracer is close to 100%.

To calculate the 'dose', you may need to solve a transport equation for dose with an appropriate source term.

Regards,

Jack

  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
Paraview nice animation PyGloo ParaView 4 June 7, 2012 12:34
nice plot of boundary layer with experimental values jack1980 FLUENT 0 May 5, 2010 08:59
It would be nice to have application options abbreviated! lakeat OpenFOAM Running, Solving & CFD 0 September 16, 2009 22:22
Wow.. nice to be the first one here.. :) Sean Lin FLOW-3D 2 September 30, 2008 17:46
Open Challenge for Concept of Turbulence K.K.J.Ranga Dinesh Main CFD Forum 2 June 3, 2003 13:41


All times are GMT -4. The time now is 15:14.