CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > FLUENT

DPM Model. Too much particle injected.

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

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   December 16, 2018, 06:55
Unhappy DPM Model. Too much particle injected.
  #1
New Member
 
Jacob
Join Date: Sep 2018
Posts: 7
Rep Power: 4
Legolas is on a distinguished road
Hi Everyone,

I am currently doing a ventilation project for a small office room. There are two occupants. One the pollutant, which the "tracer" gas is expelled out through his mouth. The other is the exposed which is basically there to see if the tracer gas will actually flow to him.

I am tasked to do it using DPM. However, I am having troubles with injecting less particles. As of current, the number of particles being injected out of the mouth is 60. Which is too much for my case.

Another question I have is, will having too much particle injected affect the turbulence model (k-e)?

Thank you everyone. Hoping to hear from you soon.
Legolas is offline   Reply With Quote

Old   December 17, 2018, 04:24
Default
  #2
Senior Member
 
Cees Haringa
Join Date: May 2013
Location: Delft
Posts: 607
Rep Power: 0
CeesH is on a distinguished road
How do you set the injection? If you do it via a surface, 1 particle is emitted from every gridcell every timestep. If you want to control the number of particles, it may be better to use group-injection where you can set the number of streams explicitly.

When it comes to turbulence, particles will only affect turbulence in case 2-way coupling is enabled (interaction with continuous phase). If this is disabled, turbulence may affect particles but not the other way around. Since you are just using particles to study flow profiles (massless particles?) and not to physically affect flow, I suppose you are not using 2-way coupling.
CeesH is offline   Reply With Quote

Old   December 18, 2018, 04:16
Angry
  #3
New Member
 
Jacob
Join Date: Sep 2018
Posts: 7
Rep Power: 4
Legolas is on a distinguished road
Quote:
Originally Posted by CeesH View Post
How do you set the injection? If you do it via a surface, 1 particle is emitted from every gridcell every timestep. If you want to control the number of particles, it may be better to use group-injection where you can set the number of streams explicitly.

When it comes to turbulence, particles will only affect turbulence in case 2-way coupling is enabled (interaction with continuous phase). If this is disabled, turbulence may affect particles but not the other way around. Since you are just using particles to study flow profiles (massless particles?) and not to physically affect flow, I suppose you are not using 2-way coupling.
Hi Sir,

one more question, because im really lost. Do you perhaps know the difference between the three different k-epsilon model? The standard, RNG and realizable. I am currently running on realizable, however, I am not getting convergence. I have checked my boundary conditions and am pretty sure it is right.

background: My project is slow velocity particle injection from one human in a small office room. Air is from the inlet at 0.5m/s and the particle is "breathe out" at 1.4m/s.

My target for this project is to see how the particle trajectory is in the room from the effect of this setup (Displacement Ventilation)

I am not getting any nice convergence. and am pretty worried.
Thanks sir.

Sincerely,
Legolas
Legolas is offline   Reply With Quote

Old   December 18, 2018, 05:37
Default
  #4
Senior Member
 
Cees Haringa
Join Date: May 2013
Location: Delft
Posts: 607
Rep Power: 0
CeesH is on a distinguished road
Yes, I know the difference, but I'm not sure how relevant that is for now. Not getting convergence can be due to a lot of things, and when using k-epsilon type models, it's often not the turbulence model that is the issue (they are all relatively easy to converge, compared to alternatives). It can be wrong boundary conditions, it can be a poor mesh, it can be a too aggressive solution strategy (high under-relaxation factors), it can be that you're trying to cast an unsteady problem into a steady solution, it can be you added additional physics (e.g. temperature) without specifying conditions for that, so something irrelevant blows up. You will need to specify more information for help on such an issue.

But a more fundamental question, since you apparently don't know the difference between the various k-epsilon formulations (conceptually, not necessarily mathematically), how much time did you spend on learning CFD theory? The challenge in CFD is often knowing what you are doing; anyone can click buttons in ANSYS and get something to run - but you need to know which buttons, why, and how to interpret the outcome. How much time did you spend on that? On this forum, you can find technical assistance for many specific options in FLUENT, but it is assumed you have some training in basic CFD - we can't teach you the entire theory here.
CeesH is offline   Reply With Quote

Old   December 19, 2018, 04:54
Default
  #5
New Member
 
Jacob
Join Date: Sep 2018
Posts: 7
Rep Power: 4
Legolas is on a distinguished road
Quote:
Originally Posted by CeesH View Post
Yes, I know the difference, but I'm not sure how relevant that is for now. Not getting convergence can be due to a lot of things, and when using k-epsilon type models, it's often not the turbulence model that is the issue (they are all relatively easy to converge, compared to alternatives). It can be wrong boundary conditions, it can be a poor mesh, it can be a too aggressive solution strategy (high under-relaxation factors), it can be that you're trying to cast an unsteady problem into a steady solution, it can be you added additional physics (e.g. temperature) without specifying conditions for that, so something irrelevant blows up. You will need to specify more information for help on such an issue.

But a more fundamental question, since you apparently don't know the difference between the various k-epsilon formulations (conceptually, not necessarily mathematically), how much time did you spend on learning CFD theory? The challenge in CFD is often knowing what you are doing; anyone can click buttons in ANSYS and get something to run - but you need to know which buttons, why, and how to interpret the outcome. How much time did you spend on that? On this forum, you can find technical assistance for many specific options in FLUENT, but it is assumed you have some training in basic CFD - we can't teach you the entire theory here.
Dear Sir,

I have been learning about CFD theory since two months ago. However, it has been a steep learning curve for me because I have to learn it by myself. Therefore, I do have lots of queries about how CFD works. I have done lots of literature review on similar projects and have been trying to incorporate similar boundaries conditions and meshes. Hence, with the inputs from professionals here, I am hoping to learn more and possibly towards a more specific application regarding this project. Hope you see where I am coming from.

Sincerely,
Legolas
Legolas is offline   Reply With Quote

Reply

Tags
dpm, hvac, injection, particle

Thread Tools Search this Thread
Search this Thread:

Advanced Search
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 Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
problem with solving lagrange reaction cloud Polli OpenFOAM Running, Solving & CFD 0 April 30, 2014 07:53
In DPM model, with rosin-rammler particle size distribution described problems, thank fa84a8g779 FLUENT 1 July 11, 2013 21:30
UDF for particle injection in DPM model sohail FLUENT 0 July 22, 2008 10:12
DPM model w/ Wave model - errors in documentation HS FLUENT 0 April 12, 2006 04:37
steady or unsteady? (in dpm) winnie FLUENT 1 April 28, 2003 11:30


All times are GMT -4. The time now is 12:41.